Chief Justice Ma and Mr Justice Fok PJ: 1. We have had the benefit of reading the judgment in draft of Mr Justice Tang PJ as well as the concurring judgments of Mr Justice Ribeiro PJ and Mr Justice Spigelman NPJ. We, too, would dismiss this appeal. The central issue in the appeal concerns the proper construction of section 300 of the Securities and Futures Ordinance (Cap.571)(“the SFO”). The main contentions of the appellants are set out in the judgment of Mr Justice Ribeiro PJ. For the reasons given by him, by Mr Justice Tang PJ and by Mr Justice Spigelman NPJ, those contentions cannot be accepted. We make the following observations: (1) Given the analysis of the meaning of the word “transaction” contained in the judgment of Mr Justice Tang PJ, it is unnecessary to explore whether that word can also be construed in the plural: see paragraph 26 below. We express no views on this aspect. (2) We also adopt the reasoning of Mr Justice Spigelman NPJ in relation to the proper approach to section 300 within the insider dealing scheme in the SFO as set out in paragraphs 62 to 70 of his judgment. Mr Justice Ribeiro PJ: 2. I have had the benefit of reading in draft the joint judgment of Chief Justice Ma and Mr Justice Fok PJ as well as the judgments of Mr Justice Tang PJ and Mr Justice Spigelman NPJ. Subject to the observations made in the joint judgment of Ma CJ and Fok PJ, and by Spigelman NPJ, with which I respectfully agree, I am in respectful agreement with the judgment of Tang PJ. I gratefully adopt his Lordship’s recitation of the facts and wish to add a few words directed at the way in which the case was developed at the hearing by Mr Gerard McCoy SC. 3. The case turns on the true construction of section 300 of the Securities and Futures Ordinance,[1] which provides: (1) A person shall not, directly or indirectly, in a transaction involving securities, futures contracts or leveraged foreign exchange trading— (a) employ any device, scheme or artifice with intent to defraud or deceive; or (b) engage in any act, practice or course of business which is fraudulent or deceptive, or would operate as a fraud or deception. (2) A person who contravenes subsection (1) commits an offence. (3) In this section, a reference to a transaction includes an offer and an invitation (however expressed). 4. The objective of Mr McCoy SC’s argument was to establish that the conduct of the defendants fell outside the terms of section 300. To that end, the construction that he advanced involved three main propositions: (a) First, that the words “a person shall not directly or indirectly, in a transaction involving securities” must be read to require that “person” – ie, the defendant – to be a party to the “transaction” referred to. (b) Secondly, that, since the only transactions that the defendants (meaning Patsy and, through her as their agent, Eric, Betty and Stella) entered into were the contracts to purchase the Hsinchu shares and then to sell them to SCB in accepting SCB’s Tender Offer, the relevant transactions in the present case were those share dealing transactions, ie, the purchase and then the sale of the shares, taken as separate transactions. (c) Thirdly, that the fraud or deception also had to be “in the transaction”, meaning, they had to be practised by the defendant on the counterparty to the relevant transaction. 5. Applied to the facts of this case, Mr McCoy SC’s argument was that the defendants’ share dealing transactions (i) did not involve any fraud or deception practised on their counterparties, that is, the shareholders in Taiwan or the SCB, when contracting to purchase or sell the Hsinchu shares respectively; and (ii) those transactions in any event took place outside of Hong Kong and could not found jurisdiction here. He did not have leave to argue point (ii), but sought impermissibly to raise it, purportedly as an aspect of his construction argument. 6. It will be evident that the lynchpin of the appellants’ argument is the proposition that section 300 requires the defendant to be party to the “transaction involving securities” in question. In my view, it is an unwarranted construction of the section. To produce the result desired by the appellants, section 300 would have to say something along the lines of: “A person, being a party to a transaction involving securities, shall not directly or indirectly” employ a fraudulent or deceptive scheme, etc. That is obviously not what section 300 says. The words “in a transaction involving securities” are most naturally read to mean “in connection with” or “in relation to” a transaction involving securities. There is no requirement that the defendants be parties as long as their fraudulent or deceptive scheme or course of business is employed in connection with or in relation to the transaction. 7. The reality of the defendants’ scheme is comfortably accommodated within section 300 so read. The relevant “transaction involving securities” as engaged by their fraudulent scheme encompassed Betty’s misuse and disclosure to the defendants of inside information regarding SCB’s takeover plans; their misuse of that information by purchasing, through Patsy and Hong Kong brokers, Hsinchu shares with a view to selling them to SCB at the higher tender price; and their acceptance of SCB’s offer and their fraudulent or deceptive realisation of large profits derived from their misuse of the inside information. They were indeed parties to the share dealing transactions. But those dealings formed merely a part of the overall transaction. 8. Once the premise that the defendant must be a party to the transaction referred to in section 300 is removed, Mr McCoy SC’s argument is entirely undermined. But even on his argument, it is hard to avoid the conclusion that in taking advantage of SCB’s tender offer without disclosing that they had accumulated the shares being sold through misuse of inside information obtained in breach of duty to SCB by a solicitor working on the deal, they had practised a fraud or deception on SCB, a party to the sale transaction. Mr Justice Tang PJ: Introduction 9. The facts are simple. They are not or can no longer be disputed. I will state them briefly. At the material time, the 1st defendant, Betty, was a solicitor in the employ of Messrs Slaughter & May (“SANDM”). The 2nd defendant, Eric, was also a solicitor and was employed by Messrs Linklaters. He was Betty’s good friend and one-time lover. Patsy, the 3rd defendant, is Eric’s elder sister. Stella, the 4th defendant, is Patsy’s younger sister and Eric’s elder sister. 10. Hsinchu International Bank Co Ltd (“Hsinchu Bank”) shares were listed on the Taiwan Stock Exchange. In 2006, Hsinchu Bank was acquired by the Standard Chartered Bank (HK) Ltd (“SCB”) pursuant to a friendly takeover which began on 29 September 2006 when SCB made a recommended tender offer for all its shares. Earlier, on 20 April 2006, Betty was seconded by her employer, SANDM to SCB’s Group Legal Department to assist with the work which led to the offer. In the course of such work, Betty learned on 14 September 2006 that the recommended tender price would be NT$24.50. This was confidential material price sensitive information (“inside information”). The SFC’s case was that Betty shared the inside information with Eric[2] about the impending offer and the proposed tender price. In other words, Betty was the tipper and Eric, the tippee. On 20 September 2006, Patsy[3] opened an account with Tai Fook Securities Co Ltd (“Tai Fook”) for the purpose of trading in shares listed in Taiwan. Between 22 and 29 September, using the Tai Fook account, 1,576,000 shares at the average price of NT$16.99 were purchased. The purchase money HK$6,381,000 was contributed by the four defendants.[4] The tender offer was made public on 29 September and the tender price became publicly known. Patsy accepted the tender via Tai Fook and made a profit of HK$2,685,000. The profits were distributed as follows: Betty $1,000,000, Eric $1,300,000, Patsy $175,000, Stella $210,000. 11. In proceedings under s 213 of the Securities and Futures Ordinance (Cap 571)(“SFO”), brought by the plaintiff, the Securities and Futures Commission (“SFC”), Mr Justice Anthony Chan found that Betty, Eric, and Patsy being persons within s 213(2)(b) had contravened s 300 of the SFO in that they, directly or indirectly, in transactions involving securities, namely the shares of Hsinchu Bank listed on the Taiwan Stock Exchange: “(a) employed a scheme with intent to defraud or deceive or; (b) engaged in acts which were, or a practice which was, fraudulent or deceptive or would operate as a fraud or deception, in that in September 2006, they engaged in dealings in Hsinchu Bank shares for personal profit whilst in possession of and misusing confidential material price sensitive information obtained in the course of Betty’s employment with Messrs. Slaughter and May and/or her secondment to Standard Chartered Bank (Hong Kong) Ltd, without those principals’ informed consent.” 12. Section 300 provides: “(1) A person shall not, directly or indirectly, in a transaction involving securities, futures contracts or leveraged foreign exchange trading – (a) employ any device, scheme or artifice with intent to defraud or deceive; or (b) engage in any act, practice or course of business which is fraudulent or deceptive, or would operate as a fraud or deception. (2) A person who contravenes subsection (1) commits an offence.[5] (3) In this section, a reference to a transaction includes an offer and an invitation (however expressed).” 13. Also, pursuant to s 213(2)(b), the 1st, 2nd and 3rd defendants were ordered to, inter-alia, disgorge or account for the profits made in their dealings in Hsinchu Bank shares in September 2006. Although the 4th defendant Stella was not found to have contravened s 300 of the SFO, a similar order was made against her, pursuant to s 213(2)(b) because she had been involved in the contravention of s 300 by the 1st, 2nd and 3rd defendants. The learned judge was satisfied that it is desirable that these orders be made and that they would not unfairly prejudice any of them.[6] 14. On appeal by the 2nd, 3rd and 4th defendants, the Court of Appeal affirmed the learned judge’s decision. All four defendants appealed to the Court of Appeal but the 1st defendant withdrew her appeal before the hearing. 15. Leave to appeal was granted to the 2nd, 3rd and 4th defendants by the Court of Appeal on 6 March 2018 on the following questions of great general or public importance, namely: “(i)	In the context of s.300 of the SFO, how should the word ‘transaction’ be construed? In particular: a.	Was the CA correct in giving the word ‘transaction’ a wide interpretation to give effect to s.300 as a ‘general catchall provision’, giving s.300 an even wider application than Rule 10b-5 of the Securities [Exchange] Act 1934 (from which our s.300 originated)(CA Judgment §§25-34)? b. Was the CA correct in construing the word ‘transaction’ independently from how that word is used in other parts of the SFO, such as s.271(8)(a)(ii), 292(8)(a)(ii), 295(3) & (4), and Schedule 5 Part 2 (CA Judgment §30)? c. Whether the scope of the phrase ‘transaction involving securities’ should extend to conduct other than the purchase and sale of securities, and the offer or invitation to trade in securities (s.300(3) of the SFO)? In particular, is the concept of a ‘transaction involving securities’ capable of covering ‘the whole deceptive scheme or the whole course of dealings’, including acts such as the disclosure of inside information for the purpose of trading in securities, the opening of a securities account for the purpose of trading in securities, the depositing of money for the purpose of trading in securities, and the giving of instructions for the purpose of trading in securities (CA Judgment §§25-49)? (ii) 	In the context of s.300 of the SFO, how does one determine whether the alleged fraudulent or deceptive act or scheme occurred ‘in a transaction involving securities’, particularly where the transaction in issue concerned securities traded on a stock exchange? Was the CA correct to adopt a ‘nexus’ approach, requiring simply that there be a ‘real and substantial’ connection between the fraud or deception and the transaction (CA Judgment §41)?” The Questions 16. The questions turn on the construction of s 300, question (i) in relation to the word “transaction”, and question (ii) the words “in a transaction involving securities”. Question (i) 17. The defendants argued that the purchase of the shares was a transaction and their sale when the tender was accepted was a separate transaction,[7] and that it would strain the natural meaning of the word to cover preparatory steps antecedent to the dealing in securities such as the use or disclosure of the inside information, or the deposit of money into the Tai Fook account.[8] 18. It is not clear from the appellants’ printed case why that mattered. Suppose one reduces “transaction” to the narrowest unit of offending, the inclusive definition of “transaction” under s 300(3) includes “an offer and an invitation (however expressed)”. [9] Thus, a bid or an offer could be a transaction. But it does not follow that a purchase which followed a bid, or a sale following an offer, could not also be a transaction under s 300(1). Nor do the defendants so contend. They merely contend that purchases and sales are separate transactions. Indeed, given the number of Hsinchu Bank shares purchased, the purchases most probably ranged over a number of days but the defendants appeared to be willing to accept that they could be one transaction. But, if so, why should “transaction” not include a purchase and sale, or a sale and purchase, or a series of both? 19. In HKSAR v Yeung Ka Sing Carson (2016) 19 HKCFAR 279, in connection with the crime of money laundering,[10] this court said: “137. In making a judgment as to whether acts are so connected that they can fairly be regarded as forming part of the same transaction or criminal enterprise it is necessary to keep in mind the purpose for which the question is asked.” 20. That statement followed what Lord Diplock said in Director of Public Prosecutions v Merriman:[11] “Where a number of acts of a similar nature committed by one or more defendants were connected with one another, in the time and place of their commission or by their common purpose, in such a way that they could fairly be regarded as forming part of the same transaction or criminal enterprise, it was the practice, as early as the eighteenth century, to charge them in a single count of an indictment.” 21. I am sure that if the defendants had been prosecuted in connection with their purchase and sale of shares in Hsinchu Bank, the charge would not have been bad for duplicity for the reasons given by Lord Diplock and by this court in Yeung Ka Sing. 22. For the same reason I do not believe in civil proceedings under s 213, “transaction” in s 300 could not cover both the purchase and sale of the shares. 23. The Court of Appeal observed[12] and I respectfully agree, it is contrived and artificial to split the purchase and sale into two or more separate transactions. That the purpose of the defendants was not the mere acquisition of the shares, their purpose was to make a profit by purchasing and then selling them by accepting the tender offer,[13] and that the scheme or course of business planned by the defendants[14] was to make a profit by purchasing and then selling the shares by accepting the tender offer.[15] 24. As the Chief Justice has said, in interpretation, a word must be given a meaning that is required by the context of the section and such as would achieve its purpose.[16] Adopting this approach, I am sure the entire enterprise could be regarded as a transaction. 25. That being the case, I do not believe the argument (question (i)(b)) based on the use of the word “transaction” in the singular as opposed to in the plural in some of the provisions helps the defendants. Counsel for the respondent has referred us to even more provisions where “transaction” appears in the singular. I would not trawl through them. 26. In any event, I agree with the Court of Appeal that there is nothing in the context of s 300 which displaces s 7(2) of the Interpretation and General Clauses Ordinance (Cap 1), namely, that words and expressions in the singular include the plural and words and expressions in the plural include the singular, which applies “save where the contrary intention appears either from [Cap 1] or from the context of any other Ordinance or instrument”. 27. Question (i)(c) raises the question whether the scope of the phrase “transaction involving securities” should cover or be capable of covering “the whole deceptive scheme or the whole course of dealings”[17] which includes acts such as the disclosure of inside information for the purpose of trading in securities, the opening of a securities account for the purpose of trading in securities, the depositing of money for the purpose of trading in securities, and the giving of instructions for the purpose of trading in securities. As I have said, Yeung Ka Sing shows clearly that such acts or conduct could fairly be regarded as forming part of the same transaction. 28. Moreover, it makes no sense to consider “transaction involving securities” in isolation, they must be construed in the context of s 300 which makes it an offence in any transaction involving securities to “(a) employ any device, scheme or artifice with intent to defraud or deceive; or (b) engage in any act, practice or course of business which is fraudulent or deceptive, or would operate as a fraud or deception”. It defies all sense to say that in such proceedings, evidence relating to such device, scheme, act, practice or course of business etc should be disregarded. 29. Question (ii), I think, concerns the question whether any of the circumstances covered by s 300(1)(a) or (b) had been shown. I think the submission is that that the fraud or deception must be practiced on a counterparty to the transaction before it can be regarded as being “in a transaction”.[18] In other words, no victim no fraud. Let me say at the outset that I agree with the learned judge and the Court of Appeal that fraud was practiced on SCB both in respect of the misuse of the inside information and the tender of the shares to SCB. 30. Mr McCoy SC relied on the fact that the origin of s 300 can be traced back to s 10(b) and Rule 10b-5 of the Securities Exchange Act 1934 (“SEA”).[19] These US provisions have been characterized by the US Supreme Court as a catchall.[20] There is no dispute about that. However, Mr McCoy would not accept that characterization for s 300. That does not matter. Essentially, Mr McCoy relied on the dissenting judgment of Justice Thomas in US v O’Hagan 521 US 642, which was decided in 1997, and sought to persuade us that we should construe s 300 in accordance with Justice Thomas’s dissent. At the risk of simplification, Justice Thomas equated the use of inside information with the theft of say, cash from an employer to buy shares, and held that in such a case there was no fraud in the purchase of the shares. But, the majority espoused what may be called the misappropriation theory, and was of the view that a fiduciary who misused inside information for gain or avoidance of loss had dishonestly misappropriated that information which makes the conduct fraudulent. 31. Section 300 is a general provision and its effect does not depend on the metaphor used to describe it. What it catches or covers should be considered in the context of Hong Kong’s legislation and according to our circumstances. Given the big difference between the treatment of insider dealing in the US and Hong Kong, I think it is unhelpful to consider how Rule 10b-5 had been construed since 1934 by different US courts.[21] In any event, both the learned judge[22] and the Court of Appeal[23] agreed with the majority in O’Hagan. Moreover,I don’t think it was, and in any event, cannot be disputed that fraud had been practised on SCB by the misuse of the inside information in the purchase and subsequently when the shares were tendered to SCB. 32. It is pertinent to mention at this juncture that in these proceedings, the SFC also alleged that in 2007 there was insider dealing contrary to s 291(5) by the defendants in connection with shares in Asia Satellite Telecommunications Holdings Ltd (“AsiaSat”). The evidence[24] showed that Eric’s then employer, Linklaters, was involved in the privatization of AsiaSat. Eric was not a member of the team involved in such work, but because of the proximity of Eric’s office to the office of the team and that they shared the same printers, photocopiers, and fax machine, Eric was able to work out that a proposed privatisation was imminent. Anthony Chan J held, in respect of AsiaSat, Eric was the tipper and Betty, the tippee. That led to frantic purchases of AsiaSat shares by Betty and Patsy between the opening of trading on 9 February 2007 and 11:19 am when trading was suspended as the result of a request by AsiaSat because of the fluctuations in the share price, when their trading accounted for 73% of the entire turnover of AsiaSat on the Stock Exchange of Hong Kong (“SEHK”).[25] 33. In respect of AsiaSat, the learned judge made orders similar to those made in respect of the Hsinchu Bank shares. There was no appeal in respect of the AsiaSat shares dealings. 34. Because of the definitions of listed securities and listed corporation under s 285, s 291(5) does not apply to shares listed on the Taiwan Stock Exchange. But “securities” under s 300 is defined in wide terms and as defined under the Interpretation and General Provisions,[26] is not confined to shares listed in Hong Kong. It can cover shares not listed in a recognized stock exchange.[27] I think it would be in keeping with the purpose of the SFO and Hong Kong’s position as an international financial center, that provided “substantial activities constituting the crime” occurred within Hong Kong,[28] s 300 should cover the insider dealing in shares listed in Taiwan. I have no doubt that substantial activities constituting the complaint under s 300 occurred in Hong Kong. That was the view of the Court of Appeal, with respect, I agree.[29] 35. Since Hsinchu Bank shares were not listed in Hong Kong, there was no insider dealing under s 291(5), but if these shares are covered by s 300, might the transaction which involved them as found by the learned judge come within s 300 (1)(a) or (b)? 36. Here too, s 300 should be interpreted in the context of the SFO. If an insider dealing transaction under s 291(5) would be regarded as a transaction in which “any device, scheme or artifice with intent to defraud or deceive” has been employed or “any act, practice or course of business which is fraudulent or deceptive, or would operate as a fraud or deception” has been engaged, I see no reason why a different conclusion should apply to a similar transaction which is covered by s 300. 37. In HKSAR v Du Jun [2012] 6 HKC 119, which concerned insider dealing which took place in early 2007, the Court of Appeal after dismissing the appeal against conviction and when dealing with an appeal against a sentence of 7 years’ imprisonment and total fine of $23,324,117 said at para 156 “… Insider dealing is a crime. It is a crime of dishonesty. It is cheating”. The Court of Appeal endorsed the categorization of insider dealing by Lord Judge CJ in R v McQuoid[30] as a “species of fraud; it is cheating”.[31] I would also note that Lord Judge also said the offence was “not to be treated as a victimless crime”[32] emphasizing that “[t]he person who sold the shares in TTP at 13 [pence] may have been determined to sell on that date at that price, or at any price. However, he would not have sold at that price if he had known that the takeover was already agreed and would become public within 48 hours.” [33] 38. In this court, in HKSAR v Chan Pak Hoe,[34] Ribeiro PJ said:[35] “50. The courts recognize that insider dealing is a fraud on the public and, [in some cases], that it also involves a breach of trust.” 39. It is clear from the above that insider dealing under s 291(5) of the SFO is a crime, a species of fraud and cheating. Moreover, it is a fraud on the public and not a victimless crime. 40. That being the case, I am of the view that conduct which would have amounted to insider dealing, but for the fact that the shares were not listed in Hong Kong, should be regarded as a crime, a species of fraud or cheating, thus coming within s 300(1)(a) or (b). It is unnecessary to distinguish between (a) and (b). 41. Also, I would note that s 305(1) provides that a person who contravenes, for example, s 291(5) or s 300: “shall … be liable to pay compensation by way of damages to any other person for any pecuniary loss sustained by the other person as a result of the contravention, whether or not the loss arises from the other person having entered into a transaction or dealing at a price affected by the contravention.” 42. In Du Jun, the Court of Appeal reduced the fines imposed on the defendant because the fines would deprive the defendant’s trading counterparties of compensation pursuant to s 213, in respect of which the SFC had commenced proceedings. The Court of Appeal also noted that a claim under s 305 might also be made by a losing counterparty.[36] Du Jun was concerned with shares in China Resources Holding Limited which were listed in Hong Kong. On 18 August 2015, the SFC announced that the court-appointed administrators had completed distributions of restoration payments to all but 3 of the 297 counterparties to the insider dealing by Du Jun, a total of $23,086,314 had been paid and a balance of $813,686 due to the remaining three investors returned to Du Jun with the approval of the court.[37] Just as a claim might be made by victims of s 291(5), I see no reason why a claim might not be made by victims of insider dealing which fell outside because the shares were listed in Taiwan. I would add that in the Court of Appeal Mr Shieh SC rightly accepted for the defendants that the fraud or deception was practiced on the vendors of the Hsinchu Bank shares when they were purchased on the Taiwan Stock Exchange.[38] 43. Mr McCoy also submitted that if s 300 was construed so as to cover insider dealing, then a person who is prosecuted under s 300 for insider dealing might, for example, be deprived of the defences available to him under s 292. The concern is misplaced.[39] To my mind, conduct for which the defences afforded by s 292 are available would not satisfy s 300(1)(a) or (b). Section 291 prohibits any dealing by an insider and those whom I would loosely call tippees subject to defences provided by ss 292, 293 and 294. It is clear that insider dealings which the courts would regard as a crime, a species of fraud or cheating, are dealings in respect of which none of the defences under s 292 could be established. 44. For the above reasons, I would dismiss the appeal. 45. For completeness sake, I would answer the questions, as follows: Question 1 (a) The word transaction has a wide meaning and covers in the present case, the appellants’ scheme to profit by the use of inside information. (b) This exercise is unhelpful in view of the answer to (a). (c) The phrase “transaction involving securities” must be considered in the context of s 300, and as such covers, inter alia, dealings with a view to profit or avoidance of loss by the use of inside information. Question 2 In the context of s 300, the question is whether “in [any] transaction involving securities”, any of the matters outlined in sub para (a) or (b) had been employed or was engaged and the words should be construed in its context. Mr Justice Spigelman NPJ: 46. I have had the advantage of reading the judgments of Tang PJ and Ribeiro PJ in draft. Subject to one matter in the judgment of Tang PJ, where I reach the same result by a different route, I agree with both judgments. I wish to state my own reasons with respect to certain discrete issues raised by the submissions. “Catch All” 47. The Court of Appeal adopted the description of s 300 of the Securities and Futures Ordinance (“SFO”) as a “catch all” provision. That is inappropriate terminology for a criminal offence. In the Court of Appeal, the words were said to have been applied by the Supreme Court of the United States to the similarly worded offence under s.10(b) and Rule 10b-5 of the Securities Exchange Act of 1934 and which, probably indirectly, was the origin of s 300 and its predecessors in Hong Kong. 48. This attribution to the Supreme Court was derived from an express reference in one of the leading American texts on securities law: Loss, Seligman and Paredes Fundamentals of Securities Regulation (6th ed. p1288. Now see 7th ed. p1442). They rely on the Supreme Court judgement in Ernst & Ernst v Hochfelder 425 U.S. 185, 203 (1976). 49. The Supreme Court was dealing with a submission that a company’s auditor could be liable in negligence for failing to detect an underlying fraud and had, thereby, aided and abetted the contravention. This was a form of accessorial civil liability. 50. It appears that the words were first used by Thomas G. Corcoran, characterized by the Supreme Court as a “spokesman for the drafters”. Probably correctly so characterized, as Corcoran was regarded as the leader of the “New Dealers”, a group of influential young lawyers in the FDR White House, when the Securities Exchange Act became law. The Court interpreted the words of the statute to conclude that it was concerned only with knowing and intentional conduct. 51. It was in this context that the Court referred favourably to the terminology of a “catch all” provision as the only aspect of the legislative history of any assistance to the issue before the Court. The majority reasons stated: “This brief explanation of §10(b) by a spokesman for its drafters is significant. The section was described rightly as a ‘catchall’ clause to enable the Commission ‘to deal with new manipulative (or cunning) devices.’ It is difficult to believe that any lawyer, legislative draftsman, or legislator would use these words if the intent was to create liability for merely negligent acts or omissions” (emphasis added). 52. Whilst the Court accepted the terminology, the politically charged advocacy in the origins of the phrase makes it quite inappropriate to apply it to a criminal offence. Further, the use made by the Supreme Court of the comments by the “spokesman for the drafters” of the legislation in 1934, is not appropriate under Hong Kong law (see HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at [15]–[17]). 53. It is the words of s 300 that must be applied, not a characterization expressed at a high level of generality, in terms that are likely to misstate the scope of the offence. The Counterparty Issue 54. I wish to add two observations to reinforce the analysis of Ribeiro PJ with respect to the submission that a person cannot be found guilty of an offence against s 300 unless that person is a party to the transaction. This proposition is not consistent with the legislative history. 55. The predecessor provision was s 136 of the Securities Ordinance (Cap 333) which provided: “A person shall not, directly or indirectly, in connection with any transaction with any other person involving the purchase, sale, or exchange of securities— (a) employ any device, scheme, or artifice to defraud that other person; or (b) engage in any act, practice, or course of business which operates as a fraud or deception, or is likely to operate as a fraud or deception, of that other person.” 56. Virtually identical provisions appeared in the then separate regulation of futures trading and leveraged foreign exchange trading (see s 63 of the Commodities Trading Ordinance (Cap 250) and s 40 of the Leveraged Foreign Exchange Trading Ordinance (Cap 451)). The words “with any other person” appeared in all three. The words “involving the purchase or sale (or exchange in Cap 333) of securities/a futures contract” appeared only in Cap 333 and Cap 250. These disparate schemes of regulation were consolidated into s 300 of the SFO. 57. As indicated above, s 136 of Cap 333 expressly referred to a “transaction with any other person”,as did Cap 250 and Cap 451. These words do not appear in the successor section s 300, which replaces the three former regulatory schemes. That formulation is now expressed as a “transaction involving securities, futures contracts or leveraged foreign exchange trading”. There is no express reference to a counterparty. The effect of the Appellants’ submission is to write back into the section the words which the legislature removed. 58. Secondly, the legislature also removed the reference to the kinds of transactions which must be “involved”, by not repeating the reference to “purchase, sale, or exchange”. These words of limitation no longer appear. The generality of the word “involving” is no longer restricted in this, or any other way. 59. The word “involving” suggests a wide range of connection. As Bokhary PJ noted in Mariner International Hotels Ltd v Atlas Ltd (2007) 10 HKCFAR 1 at [51], the word “involving” was “one of the broadest words of association known to the English language”. (Referred to with approval in Moody’s Investors Service Hong Kong Ltd v Securities and Futures Commission [2018] HKCFA 42 at [35]). 60. In any event, “transaction” is not a word that can be confined to a single arrangement, like a contract for sale or purchase. As the Ontario Court of Appeal put it in R v Canavan and Busby [1970] 3 OR 353 at 356, per Schroeder J.A.: “[a] ‘transaction’ may and frequently does include a series of occurrences extending over a length of time”. The preposition “in” can clearly be applied to such a sequence. 61. For present purposes it is sufficient to conclude that conduct can involve “securities”, and have occurred “in a transaction”, if the events said to constitute the transaction consist of a series of inter-related, but discrete, steps. That was the case here. The Insider Dealing Scheme 62. The one respect in which I would reach the same conclusion as Tang PJ, but by a different route, arises from para 43of his reasoning. This paragraph deals with the Appellant’s submission that the interpretation of s 300 adopted by the Court of Appeal, would permit the Securities and Futures Commission (“SFC”) to prosecute for an insider dealing offence that would otherwise fall within s 291 of the SFO. They submitted that such a course would deprive an accused of the defences for which that legislative scheme provides. 63. Tang PJ states that a contravention of s 300 would not be upheld unless none of the ss 292–294 defences could be established. I prefer to analyse this issue by applying the frequently deployed interpretive technique of reading down general words, relevantly, to s 300. 64. I adopt the principle of statutory interpretation that general words will be read down so as not to apply when the same instrument contains a particular provision, which would otherwise wholly fall within the wider provision but which, unlike that provision, contains exceptions, restrictions, conditions or procedural requirements. In such a case, interpreting the instrument as a whole leads to the conclusion that the legislature intended only the particular provision to apply (see e.g Anthony Hordern & Sons Ltd v Amalgamated Clothing & Anor (1932) 47 CLR 1 at 7; R v Wallis (1949) 78 CLR 529, at 550; Minister for Immigration and Multicultural and Indigenous Affairs v Nystrom (2006) 228 CLR 566 at [2], [54], [59]; Saraswati v The Queen (1991) 172 CLR 1 at 23-25; R v J [2005] 1 AC 562 at [21], [35], [48] and [63]. See also D.C. Pearce and R.S. Geddes, Statutory Interpretation in Australia (8th ed, LexisNexis 2014) at [4.36]–[4.39]). 65. Most relevantly, Saraswati in the High Court of Australia and R v J in the House of Lords applied this approach in a criminal context. Both courts were faced with similar provisions for sexual offences. An offence of indecent assault was subject to a time bar, but an offence of committing an act of indecency (which always occurs in a sexual assault) was not. Both courts held that it was impermissible to charge a person with the act of indecency offence, when the facts constituted an assault and, in the circumstances, the time bar applied. 66. Division 2 of Part XIV of the SFO contains a comprehensive and detailed scheme for the prohibition of insider dealing directed to shares listed on the Hong Kong Stock Exchange. The provenance of this inside dealing regime is the Securities (Insider Dealing) Ordinance (Cap 395), a different Ordinance than Cap 333, where s 300 of the SFO originated. Section 300 is now found in Division 4 of Part XIV of Cap 571. Interposed in Division 3, which is entitled “Other market misconduct offences”, are provisions relating to false trading and price rigging. Like s 300 they were transferred from Cap 333. 67. Focusing on Division 2, specifically the defences set out in ss 292-294, it is apparent that the Division constitutes a self-contained, comprehensive scheme. Those sections provide protection for: ● shares acquired to qualify as a director; ● good faith performance of an underwriting agreement; ● good faith performance of functions of a liquidator, receiver or trustee in bankruptcy; ● acquisition by a corporation when the persons who made the decision did not have the inside information that other directors or employees did have; ● acquisition or disposal which was not for the purpose of making a profit or avoiding a loss by using inside information; ● a person who acted as an agent, without knowledge that the principal had inside information; ● an off market transaction between persons who both had the inside information; ● where the inside information was “market information” or a “market contract”, both as defined; ● acquisition by a trustee or personal representative acting in good faith on advice, and ● acquisition by exercise of a right to subscribe attached to securities acquired before the person became aware of the inside information. 68. The scope and detail contained in these defences indicate an integrated scheme, intended to make comprehensive provision with respect to the insider dealing offence created by s 291. That conclusion is reinforced by the fact that the defences are generally introduced by the words: “for the person to prove”. 69. The separation of the two schemes is further reinforced by s 306. That section empowers the SFC to makes rules prescribing circumstances in which conduct, that would otherwise offend Part XIV, including s 291, do not constitute an offence. Section 300 is specifically excluded from this power. Where such rules had been made, a prosecution for such conduct under s 300 must be impermissible. 70. The Appellant’s submission to the effect that a prosecution under s 300 for conduct constituting an offence under s 291 may be permissible, should be rejected. The legislature intended that conduct, which constitutes an offence under s 291, should be prosecuted under Division 2, to the exclusion of s 300. Chief Justice Ma: 71. For the above reasons, the appeal is dismissed. As to costs, we would make an order nisi that the Appellants pay the costs of the Respondent in this appeal, such costs to be taxed if not agreed. Should any party seek a different order as to costs, written submissions should be lodged with the Registrar (and served on the other parties) within 14 days of the handing down of this judgment, with liberty on the other parties to lodge and serve written submissions in reply within 14 days thereafter. If no written submissions are received seeking a different order as to costs before the expiry of the relevant period, the order nisi will become absolute. Mr Gerard McCoy SC, Mr Derek Chan SC and Ms Cherry Xu, instructed by Wellington Legal, for the 2nd to 4th Defendants (1st to 3rd Appellants) Mr Benjamin Yu SC and Mr Laurence Li, instructed by the Securities and Futures Commission, for the Plaintiff (Respondent) [1] (Cap 571). [2] CFI, para 132. [3] The judge found that Patsy knew that Betty was the source of the inside information. CFI, para 141. [4] Betty $2,250,000, Eric $3,280,000, Patsy $351,000, Stella $500,000. [5] The seriousness of the offence can be gathered from the fact that on conviction on indictment, inter alia, there could be imprisonment for 10 years and a fine of $10,000,000, s 303(1)(a) as well as payment to the government of an amount not exceeding the amount of profit made or loss avoided (s 303(2)(d)). No criminal prosecution was brought against any of the defendants. [6] Section 213(4). [7] Appellants’ case, para 3.3. [8] Appellants’ case, para 3.4. [9] Possibly because I believe it is probable that the extended meanings were already covered. [10] Section 25(1) Organized and Serious Crimes Ordinance (Cap 455), in connection with the question whether charges which involved multiple transactions in multiple bank accounts were bad for duplicity. [11] [1973] AC 584, 607. [12] Court of Appeal, paras 26 & 29. [13] Court of Appeal, para 38. [14] Court of Appeal, para 25. [15] Court of Appeal, para 38. [16] Fully Profit (Asia) Ltd v Secretary for Justice (2013) 16 HKCFAR 351. [17] Court of Appeal, para 41. [18] Court of Appeal, para 17. [19] This is raised in question (i)(a) but I think this is a better place to deal with it. [20] Fundamentals of Securities Regulation, by Loss, Seligman and Paredes, 6th ed, vol 2, p. 1288; Ernst & Ernst v Hochfelder 425 US 185 (1976), 203. [21] Insider dealing in the US had its origin in the common law, whereas in Hong Kong it has always been statutory. [22] CFI, paras 208-213, 218-219. [23] Court of Appeal, para 42. [24] CFI, paras 38-52. [25] CFI, para 52. [26] Schedule 1 of the SFO. [27] At one time, Mr McCoy submitted that it would cover shares in an unlisted company in Hong Kong. Given that the definition expressly excluded private companies as defined now in s 11 of the new Companies Ordinance Cap 622, Mr McCoy may well be right that s 300 might also cover shares in a public company as defined in s 12 of Cap 622, notwithstanding that they were not listed. But it is unnecessary to decide the point and I would not do so. [28] HKSAR v Wong Tak Keung (2015) 18 HKCFAR 62, para 33. [29] Court of Appeal, paras 53-68. [30] [2009] 4 All ER 388, giving the judgment of the English Court of Appeal. [31] Para 9. [32] Para 7. I think two victims can be readily identified. The person whose inside information was misused and the person who traded with the impugned person in ignorance of the inside information. [33] Para 7. [34] (2012) 15 HKCFAR 244, 258. [35] With the agreement of Ma CJ, Chan PJ and Lord Collins of Mapesbury NPJ. [36] Para 171. [37] https://www.sfc.hk/edistributionWeb/gateway/EN/news-and-announcements/news/doc?refNo=15PR85 [38] Court of Appeal, para 17. [39] One would not be prosecuted for insider dealing under s 300. Any prosecution or claim will be made on the basis that the relevant conduct came within s 300(1)(a) or (b). 
Chief Justice Ma : A. INTRODUCTION 1. In the normal course of civil litigation in the courts, any question regarding the costs of the main proceedings, as opposed to interlocutory matters, will generally be dealt with at the conclusion of those proceedings. In an application for judicial review, where there is a substantive hearing of that application, the costs of the proceedings will usually be dealt with at the end. Further, the usual rule as to the incidence of costs is that costs follow the event, meaning that in adversarial proceedings, the successful party will generally recover costs from the unsuccessful party. However, this general rule is subject to exceptions where it would be fair and just. The present appeal involves a consideration of an order in relation to costs whereby a party seeks at an early stage of public law proceedings to obtain an order to the effect that in the event that that party is unsuccessful, there will be no requirement to pay costs to the successful party. This is known as a protective costs order (a PCO)[1] and this provides an exception to the usual approach as to costs just stated. The principal point for determination in this appeal involves the examination of one aspect of the discretion to be exercised when considering whether or not to grant a PCO, namely, the aspect of the financial ability or resources of an applicant for a PCO, and in particular the position when that applicant is a company. There are other points which I shall presently identify. A.1 Facts 2. The appellant, Designing Hong Kong Limited (DHKL) is a company limited by guarantee.[2] There were four founding members of the company, being Ms Christine Loh, Mr Markus Shaw, Mr Peter H Y Wong and Mr Paulus Johannes Zimmerman. Mr Shaw and Mr Wong remain as directors of DHKL, Mr Zimmerman is the current Chief Executive Officer. It is assumed that Ms Loh no longer has an active role within DHKL. 3. DHKL is a non‑profit organisation dedicated to Hong Kong’s environment. One of its objects is to identify ways and means of enhancing the quality and sustainability of the environment for the benefit of residents and visitors. It was formed in 2007 when the Government announced plans for the Tamar Site and Central Waterfront Reclamation. On 14 February 2014, the Town Planning Board (the TPB), the respondent in this appeal, decided not to amend what was known as the Amended Draft Outline Zoning Plan No. S/H24/8 (the ADOZP).[3] The relevant area is a strip of waterfront land comprising a 150 metre stretch located at the north shore of Victoria Harbour near the People’s Liberation Army Garrison Headquarters. Prior to the ADOZP, the intention was that there was to be a continuous waterfront promenade along this area. The ADOZP had the effect of changing this so that the 150 metre stretch would be rezoned to allow for a military dock to be built.[4] This affected the original planning intention of a continuous waterfront promenade through the Central area. 4. The TPB refused, despite representations to the contrary made by DHKL and others, to reconsider and amend the ADOZP. This led to an application by DHKL for leave to apply for judicial review of the said decision not to amend. The grounds for the application for leave are contained in the Form 86, raising issues of abuse of power, failure to carry out statutory duties and legitimate expectation. 5. Also included in the Form 86 was an application for a PCO protecting DHKL from liability to bear any of the TPB’s (the Respondent’s) costs of the proceedings, whether substantive or interlocutory; alternatively, that the TPB’s costs be capped at $10,000.00 and that any costs that should be borne by the TPB (if it was unsuccessful in resisting the proceedings) also be limited to the reasonable costs of a solicitor and junior counsel. It is this application with which we are concerned in this appeal. A.2 The proceedings below 6. Leave to apply for judicial review was granted on 21 July 2014 after hearing submissions from both parties. The court heard DHKL’s application for a PCO over two days.[5] By a judgment given on 30 April 2015, the application was dismissed with costs. In the judgment, Au J remarked that this was the first time that the court had to deal with a PCO and examine the relevant principles applicable thereto.[6] It was for this reason that the court had invited submissions from amici curiae,[7] as well as hearing submissions on behalf of the Director of Legal Aid. 7. DHKL appealed. The appeal was heard over the course of four days[8] and was dismissed. In the CA Judgment, it was said that as the questions regarding PCOs were raised for the first time for detailed consideration, the court would take the opportunity to lay down general guidance. Owing to the importance of the matter, leave was given to the Secretary for Justice to intervene and make submissions. There were no amici curiae before the Court of Appeal. Notwithstanding that the appeal was dismissed, no costs were ordered. A.3 The questions for determination 8. The Court of Appeal refused leave to appeal to this Court,[9] and ordered DHKL to pay costs of $100,000.00 to the TPB. On 30 October 2017, the Appeal Committee[10] granted leave to DHKL to appeal to the Court of Final Appeal. The following questions were stated to be of great general and public importance :- “(1) When considering whether to grant a Protective Costs Order (“PCO”) in a case which raises an issue of general public importance, the resolution of which by the Court is in the public interest, to an applicant that has no special personal or pecuniary interest in the outcome of the proceedings (and, in the case of a corporate applicant, the directors or members of which have no such interest): (a) Should the Court consider, as relevant factors in the exercise of its discretion, whether: (i) having regard to the financial resources available to the applicant and the respondent and the amount of costs likely to be involved, it is fair and just to make the order, and (ii) that the applicant will probably discontinue the proceedings and that it would be acting reasonably in doing so? (cf. Corner House Research at para. 74(iv) and (v)) Or is the applicant required to prove that it is “genuinely not in a position to fund the litigation or to bear the respondent’s costs in the proceedings” per Sections D5, F and para. 81(c) of the CA Judgment. If neither, what is the correct approach or principle to be applied to the financial resources available to the applicant? (b) Where the applicant is a body corporate: (i) Are the private financial resources of the directors and/or members of the applicant to be treated as relevant and, if so, should they be treated as available to the applicant or as if they are the financial resources of the applicant? (ii) If the private financial resources of the directors and/or members are relevant, is an applicant for a PCO required to obtain and disclose the financial resources of its directors and/or members? If so, how is such evidence (or its absence) to be taken into account in deciding upon a PCO? (c) Further, what is the relevance of the availability of legal aid to persons who are not applicants to the principles upon which the jurisdiction should be exercised, both generally and in particular to applicants who are themselves ineligible for legal aid? (d) Further, what is the relevance of the financial resources of the respondent and how should that be considered and taken into account when deciding on an application for a PCO? (2) Is an application for a PCO to be considered at the leave stage on the principles set out in Corner House Research at paragraphs 78 to 81 inclusive, suitably adapted for the procedure under Order 53 of the Hong Kong Rules of the High Court, and applying the principles set out in Corner House Research as to the incidence of costs on such applications, or are such applications to be heard inter partes in some longer procedure as envisaged by the Court of Appeal at paragraphs 71 to 81 of its judgment and if so, on what basis as to costs?” The relevance and effect of these questions will be apparent later in this judgment. 9. We are grateful to counsel for the assistance they have given in this appeal and like the Court of Appeal, we would commend in particular those representing DHKL who have acted pro bono in this important matter.[11] 10. Before I deal with these specific questions, it is important first to set out some general principles regarding the granting of PCOs and the context in which the application for such orders is made. B. PROTECTIVE COSTS ORDERS B.1 Statutory context 11. Section 52A(1) of the High Court Ordinance[12] provides that subject to rules of court,[13] the costs of civil proceedings are in the discretion of the court. There is a similar discretion in the Court of Final Appeal.[14] RHC Order 62 deals with costs and applies to all proceedings except non‑contentious or common form probate proceedings and proceedings in matters of prize.[15] It applies to judicial review proceedings. 12. I mentioned earlier that in the normal course of civil proceedings, it is usual that the question of costs is dealt with at the conclusion of those proceedings. This makes good sense as it is only at the conclusion of proceedings that the court is able to assess the overall justice of the proceedings when determining the question of costs. Thus, for example, the court will be able to take into account the conduct of the parties over the course of the litigation, whether certain stances taken by the parties were justified, whether certain arguments ought to have been pursued and so on. RHC O.62 r.5 sets out a number of relevant factors for the court to take into account and these include the underlying objectives set out in RHC O.1A r.1 and guidance is given to the types of conduct of the parties that may be relevant to be considered.[16] 13. I have also made reference earlier to the general rule that costs should follow the event. RHC O.62 r.3(2) states that if a court sees fit to make an order as to costs, it shall order costs to follow the event. This starting point is also recognized in cases involving the public interest : see the decision of this Court in Leung Kwok Hung v President of the Legislative Council (No. 2).[17] The Court said[18] :- “An appeal against a judge’s refusal of leave, which in practice proceeds on an inter partes basis, should therefore be subject to the usual rules as to costs, namely that the starting point in civil litigation, even that involving the public interest, is that costs should follow the event.” This must, however, only be seen as a general rule and is subject to the discretion to take into consideration other relevant matters. RHC O.62 r.5(1)(f) states, for example, that notwithstanding that a party has not been wholly successful, account should be taken of any success on a part of the case.[19] B.2 Public interest litigation (PIL) 14. It is clear from the statutory context that the court has a wide discretion as to costs. In the context of public law cases,[20] the fact that matters of public importance are determined may have an important bearing on costs. Sometimes, the public interest nature of the litigation is such that notwithstanding that a party may be unsuccessful, the court will not order costs against that party. 15. In Chu Hoi Dick and Another v Secretary for Home Affairs (No. 2),[21] Lam J (now Lam VP) gave useful guidance on the question of costs where PIL was involved. Reference was made to the judgment of Li CJ in Town Planning Board v Society for the Protection of the Harbour Limited (No. 2)[22] where it was said[23] that the “fact that proceedings are commenced to vindicate the public interest, more particularly to protect a public asset which is a central element in Hong Kong’s heritage, rather than to assert or enforce some private right or interest, is plainly relevant to the exercise of the discretion.” After going through a number of authorities from the United Kingdom (R (Corner House Research) v Secretary of State for Trade and Industry[24]) and the High Court of Australia (Oshlack v Richmond River Council[25]) as well as Hong Kong authorities (including Scott v Government of the Hong Kong Special Administrative Region[26]), Lam J formulated[27] a number of criteria relevant to the court’s consideration of whether it was appropriate to disturb the usual order of costs following the event :- “29. I shall therefore formulate the three criteria as follows, (a) A litigant has properly brought proceedings to seek guidance from the court on a point of general public importance so that the litigation is for the benefit of the community as a whole to warrant the costs of the litigation be borne by the public purse as costs incidental to good public administration; (b) The judicial decision has contributed to the proper understanding of the law in question; (c) The litigant has no private gain in the outcome. 30. I should also mention that the public interest element is only one of the factors that is relevant for the exercise of the discretion as to costs. Even if all these criteria are satisfied, the court must also have regard to other relevant factors such as the conduct of the litigants in the proceedings in coming to a final decision on what is just in the circumstances. It is ultimately a matter of discretion, hence the use of the word ‘occasionally’ in the dicta of Kirby J.”[28] 16. Though reference was not made in these passages to the merits of the case, elsewhere in the judgment[29] Lam J referred to the necessity of demonstrating merits, the point being that however important the point in issue may be in a PIL case, a lack of merits will not often (if ever) save a party from an adverse order for costs. The importance of merits was emphasized by the Court of Appeal in Chan Noi Heung v Chief Executive in Council.[30] After referring to the statements of principle in Chu Hoi Dick and Oshlack, the court dealt with the facet of merits. It was said :-[31] “9. In my judgment, it is clear from the cases (not to mention as a matter of plain commonsense) that, however important the subject matter of the relevant litigation or however important any particular legal point may be, it is highly relevant for a court (when considering the incidence of costs) to evaluate the merits of the failed challenge before it. In other words, the court has to ask itself: how meritorious were the issues raised before it? If the issues that were raised by the unsuccessful applicant were, upon analysis, really quite hopeless, then it is difficult to conceive of a court making any order other than costs following the event.” 17. Both Chu Hoi Dick and Chan Noi Heung were approved by this Court in Leung Kwok Hung (No. 2).[32] 18. In all these cases, the determination of costs was dealt with at the conclusion of proceedings. A considered evaluation of merits and a resolution of the conduct of the parties are, as stated earlier, best undertaken at the conclusion of proceedings. Even a consideration of whether the litigation really concerned a matter of sufficient public importance (or not) may sometimes only become clear at the conclusion of the proceedings : see R v Lord Chancellor Ex parte Child Poverty Action Group.[33] 19. Having referred to the relevant statutory context and the general position regarding the treatment of PIL cases, I now deal with the approach to PCOs. B.3 Court’s approach to Protective Costs Orders 20. It is in the above context that the approach in granting PCOs needs to be considered. The lower courts found that much guidance was given by the decision of the English Court of Appeal in Corner House. This case was referred to in Chu Hoi Dick, Chan Noi Heung and Leung Kwok Hung (No. 2) in the consideration of the question of costs in a PIL case. All parties were agreed that the general approach to PCOs was to be found in Corner House. I am also in agreement that Corner House does provide useful guidance in Hong Kong. 21. It is first important to note that there are three underlying aspects that ought to be firmly borne in mind in considering the grant of PCOs. 22. First, exceptionality. A PCO is an exceptional order to make. It is made at an early stage of the proceedings whereby the party applying for it seeks to be protected in any event from what otherwise may be a liability to pay all or any part of the other side’s costs in the relevant proceedings. It is therefore pre‑emptive. Not only is the question of the incidence of costs dealt with at the beginning of proceedings (rather than at the end of the case), in granting a PCO, the court makes a determination having the effect of denying costs to a potentially successful party even at this early stage. This is irrespective of the future conduct of the case by the parties and irrespective of the merits of the case as the court may ultimately find them. The discipline in the conduct of PIL cases can be expected to be the same as in any other type of case and often it is only at the conclusion of the case when this aspect can properly be assessed. With the importance now of observing the underlying objectives of civil litigation in Hong Kong, the conduct of proceedings becomes even more relevant. As far as merits are concerned, even where an applicant for judicial review has obtained leave to apply for judicial review, thus satisfying the threshold test of demonstrating a reasonably arguable case which enjoys a realistic prospect of success,[34] this does not mean that ultimately a court will find that an applicant has sufficient merits in pursuing the case.[35] In a sense, it is already exceptional to make an order along the lines discussed in Chu Hoi Dick whereby a successful party is deprived of an order for costs in his favour. A PCO is all the more so given the stage at which it arises. 23. That a PCO is an exceptional order is shown in the cases. In Corner House, the Court of Appeal agreed with Dyson J in the CPAG case that “the jurisdiction to make a PCO should be exercised only in the most exceptional circumstances.”[36] The Court of Appeal in the present case also described the jurisdiction to grant a PCO as “exceptional”. In the course of its analysis, reference was made to the judgment of Hoffmann LJ in McDonald v Horn[37] in which the point was made that the rule as to costs following the event presented a “formidable obstacle” to a PCO being granted as it was difficult to have a case in which it was possible for a court to exercise its discretion properly in advance of the substantive decision. 24. Secondly, rationale. It is necessary to identify a principled basis for the making of such an order. The basis or rationale for making a PCO can really only be this, put simply : a court should consider granting a PCO in PIL cases involving an issue or issues of great public importance which should be determined and which would be stifled through lack of financial means to meet the potential costs liability to the other party or parties in such proceedings in case the applicant is unsuccessful. In this respect, I have found particularly useful the analysis undertaken by the Court of Appeal in Corner House.[38] I have also found useful the observation of Haddon‑Cave J in R (on the application of The Plantagenet Alliance Limited) v Secretary of State for Justice.[39] Ms Kaufmann QC submits this is a fundamental access to justice issue and reference was made to Article 35 of the Basic Law and Article 10 of the Bill of Rights.[40] I agree that in a broad sense access to justice is involved, although it is important to bear in mind that in the present context of PIL cases, it is access to the court for the determination not so much of private rights or to further private interests as of matters of general public importance. 25. Thirdly, fairness and justice. As with all exercises of discretion in a procedural context, it is important for the court to bear in mind the overall fairness and justice of the application for a PCO. Whenever flexibility is urged upon the courts in the exercise of discretion (and this was similarly urged on us in the application of the relevant principles), it is fairness and justice that lie at the heart of this. 26. With these essential aspects in mind, one can then proceed to examine the principles regarding the grant of PCOs. Here, as the lower courts have accepted and as the parties have also accepted, the position is summarized in Corner House in the following passage[41] :- “74. We would therefore restate the governing principles in these terms. (1) A protective costs order may be made at any stage of the proceedings, on such conditions as the court thinks fit, provided that the court is satisfied that: (i) the issues raised are of general public importance; (ii) the public interest requires that those issues should be resolved; (iii) the applicant has no private interest in the outcome of the case; (iv) having regard to the financial resources of the applicant and the respondent(s) and to the amount of costs that are likely to be involved, it is fair and just to make the order; and (v) if the order is not made the applicant will probably discontinue the proceedings and will be acting reasonably in so doing. (2) If those acting for the applicant are doing so pro bono this will be likely to enhance the merits of the application for a PCO. (3) It is for the court, in its discretion, to decide whether it is fair and just to make the order in the light of the considerations set out above.” 27. It will presently be necessary to discuss the aspect of financial ability – the main focus of this appeal – but before doing so, I would just like to make the following observations regarding these principles :- (1) The starting point is that PIL is involved. This is reflected in paras. 74(1)(i) and (ii) of Corner House. (2) It can be seen that paras. 74(1)(i) to (iii) largely resemble the factors that are relevant when considering to make an order for costs as discussed in Chu Hoi Dick.[42] It must therefore follow that for a PCO to be granted, it is not enough simply to contend that PIL is involved. Something more needs to be shown. That “something more” is reflected in paras. 74(1)(iv) to (v) of Corner House; in other words, it becomes essential to look at the financial ability of the applicant for a PCO in order to determine whether it would be fair and just to make an order. This point was made by the Court of Appeal in the following passage[43] :- “81. To overcome the potential injustice to a respondent occasioned by a PCO being made in a case which after the substantive hearing turns out not to be a PIL case, there has to be some additional justification compelling such an exceptional measure to be adopted. Thus, at the interlocutory stage when the court is asked to assess whether a PCO should be made without knowing with some certainty that the PIL criteria would be satisfied at the end of the day, we are of the view that the court should bear in mind the following additional considerations (on top of those set out at [70] and [71] above) which are germane in assessing whether a PCO should be made in the public interest : (a) Because of the lack of full understanding of the issues and their merits at the interlocutory stage, a PCO represents a more exceptional departure from the general rule of costs following the event as compared with a Chu Hoi Dick order made after judgment on account of PIL; (b) The mere raising of a point of general public importance by an altruistic litigant (viz. one having no private gain or interest in the proceedings) is not a ground for the making of a PCO. The issue raised must be sufficiently special so that it would be contrary to the public interest and the interests of justice to deprive the court of the opportunity to adjudicate on the same in this particular case; (c) As the applicant would have to fund his legal representation and as appropriate legal representation is necessary for the proper determination of an issue of such importance, and since a conditional fees agreement is not permitted in this jurisdiction, the court should be provided with information on how the applicant’s litigation is to be funded.” (3) DHKL is critical of this passage. It was contended that by the reference in para. 81(b) of the CA Judgment to the need to demonstrate “something special”, the jurisdiction to grant PCOs had somehow been narrowed down by the Court of Appeal. This was attributed to a confusion that had arisen in the cases in England after Corner House. The confusion was whether it was necessary to demonstrate some exceptionality beyond the Corner House principles referred to above. It is neither desirable nor necessary to go into detail as to how the controversy developed in England. It is sufficient merely to refer to the judgment of the English Court of Appeal in R (Compton) v Wiltshire Primary Care Trust.[44] In that case, the majority of the Court of Appeal held there was no additional element of exceptionality over and above the Corner House principles; exceptionality was only a “prediction as to the effect of applying the principles [of Corner House].”[45] On the other hand, Buxton LJ held that there was an additional requirement to show exceptionality.[46] The views of the majority have been accepted in subsequent cases : see for example R (Buglife – The Invertebrate Conservation Trust) v Thurrock Thames Gateway Development Corporation.[47] (4) Whatever the position that may have transpired in England and whether or not it was at any stage intended to restrict the principles set out in Corner House, I do not read the judgment of the Court of Appeal in the present case as in any way imposing restrictions on the Corner House principles set out above. On the contrary, the judgment of the Court of Appeal lays emphasis on the importance of the Corner House principles. (5) I should also deal at this stage with another criticism of the Court of Appeal made by DHKL, this being one of the matters raised in the questions in this appeal.[48] It was submitted that by its reference to the system of legal aid in Hong Kong, the Court of Appeal was somehow of the view that there ought to be a more restrictive approach to the granting of PCOs.[49] Where an applicant[50] for a PCO is ineligible for legal aid, the fact that a generous system of legal aid is available in Hong Kong is, it was contended, simply irrelevant. Moreover, an applicant could not be expected to embark on a search for a person who would be eligible for legal aid and who could institute or continue proceedings in his place. I do not think that the Court of Appeal was, by the reference to legal aid, in any way seeking to restrict the application of the Corner House principles. The reference to legal aid in the judgment of the Court of Appeal was merely to explain why there have hitherto been hardly any applications for a PCO in Hong Kong. As the evidence before the court showed, Hong Kong’s relatively generous system of legal aid (compared with many other jurisdictions) has ensured that most cases of public importance have over the years been determined by the courts. This has also been the Judiciary’s experience. In the vast majority of PIL cases, particularly since 1997, legal aid has played a significant part. Put another way, as far as PIL cases are concerned, it is on the whole unlikely that a lack of means will prevent a PIL case from being heard. I say “on the whole” because there are no ready figures on any cases which may have slipped through the legal aid net. Lastly in this context, the reference to legal aid in the judgment of the Court of Appeal was also for the purpose of exercising caution when looking at some of the English authorities post Corner House. It had been urged by DHKL that the Hong Kong courts should follow what was said to be a more liberal approach to PCOs in England. The context of legal aid was therefore relevant in this respect. (6) Ultimately, it is for the court to assess after looking at the various matters contained in paras. 74(1)(i) to (v) and (2) of Corner House whether it would be fair and just to exercise its discretion in favour of granting a PCO. However, given that the rationale for granting a PCO is, as I have earlier stated, the stifling through lack of financial means to pursue a case in the public interest (the public interest not being by itself sufficient), the question of the applicant’s financial ability becomes an important matter to consider. It is to this facet I now turn. C. FINANCIAL ABILITY 28. This is essentially the point raised in question 1 of the questions on which leave was given by the Appeal Committee.[51] C.1 Assessment of the financial ability of the applicant : individual and corporate applicants 29. Paras. 74(1)(iv) and (v) of the Corner House principles underline the importance of considering not only the public interest when looking at the question of costs, but also at the respective positions of the parties to the relevant proceedings themselves to see how a PCO would or might affect them. The facet of financial resources or financial ability accordingly becomes essential to be considered. I start with the position of the applicant. Before the court can come to a conclusion as to whether it would be fair and just to grant a PCO or, if an order is not made, whether the applicant will probably discontinue the proceedings and be reasonable in doing so (thus incorporating an objective element in the discretion), it must first be apprised of the financial ability of the applicant. 30. The burden is on an applicant to provide details going to financial ability to bear the likely costs of the other side should an adverse order be made against the applicant. There should be no difficulty in understanding what needs to be shown. An applicant will need to disclose his financial resources. There are no technical rules here; it is an exercise involving common sense. Much argument was devoted in the case of corporate applicants (such as DHKL), and the position of shareholders and directors of the company, but the exercise remains the same with corporations as it does with individual applicants : the fundamental question remains, what financial resources does the applicant have at his disposal? The inquiry here undertaken by the court is similar to that undertaken in other proceedings. Proceedings for ancillary relief in matrimonial proceedings, determining whether security for costs should be ordered readily come to mind. Reference was made to Wing Hing Provision, Wines and Spirits Trading Company Ltd. v Hanjin Shipping Company Ltd.[52] where, in an application for security for costs, Godfrey JA asked whether a company could raise funds from directors, shareholders and other backers or interested parties.[53] Reference was also made to the decision of Fok PJ in Tsit Wing (Hong Kong) Company Ltd. v TWG Tea Company Pte Ltd.[54] where, in an application for security for costs, it was said,[55] “The Court will need to know not only why the appellant cannot provide security from its own resources, but also why security cannot be raised by the appellants from some third party source”. 31. It is clear from these and other cases that in examining financial ability, it is legitimate, in the case of a corporation, to inquire not only as to assets belonging to the company, but also to other sources of funding to which the company would have access. For example, a company may have lines of credit or other sources of funding available to it. But what of the shareholders or directors or (in the present case) the guarantors of a company? How far does the court associate the financial ability of a company with the financial resources of its shareholders, directors or guarantors or other persons who may provide financial support to that company? This is central to DHKL’s submissions in this appeal. It is contended that the court should not look at all to the financial resources or ability of shareholders, directors or other persons who may support the company or who may in the past have supported the company. Such persons are not to be regarded as being synonymous with the company. We were reminded that a company is to be seen as having a separate legal personality from its shareholders. This contention overstates the position. Some flexibility and realism need to be applied to the position of shareholders, directors, guarantors and other supporters of a company. In some situations it may be entirely appropriate to look closely at the financial ability and resources of such persons, in other situations perhaps not. There is no fixed approach one way or the other. In the present case, for instance, the position of the directors and guarantors is simply stated : whatever their financial worth, they are unwilling to contribute any money to the litigation. In this situation, it would be pointless to expect details of financial ability or resources. The assumption that the court would make is that these persons have the means but are simply unwilling to commit any resources to the litigation. 32. Ms Kaufmann QC referred to the decision of the United Kingdom Supreme Court in Goldtrail Travel Ltd. (in liquidation) v Onur Air Taşimacilik AŞ.[56] In that case, judgment had been given against a company but permission to appeal was given[57] to appeal to the Court of Appeal. On the application of the respondent to the appeal, an order was made ordering the company to pay into court the judgment sum as a condition of appealing. The company applied to set aside this order on the basis that it did not have the financial ability to comply with it and that its appeal would be stifled if the order was made. The issue for the Supreme Court was whether the financial position of a third party such as a shareholder or director could be taken into account. In that case, the owner of the company was a person of substantial means. 33. In the judgment of Lord Wilson JSC,[58] it was stated that as a general rule where an appellant had permission to appeal, it was wrong to impose a condition which had the effect of preventing him from bringing it or continuing it.[59] Where a company is involved, it was always important to look at the financial ability or resources of that company as a starting point, and not assume that the financial ability or resources of a shareholder automatically reflected those of the company. It was said in this context[60] :- “18. It seems that, in particular and as exemplified by the present case, difficult issues have surrounded the ability of a corporate appellant, without apparent assets of its own, to raise money from its controlling shareholder (or some other person closely associated with it); and this is the context of what follows. When, in response to the claim of a corporate appellant that a condition would stifle its appeal, the respondent suggests that the appellant can raise money from its controlling shareholder, the court needs to be cautious. The shareholder’s distinct legal personality (which has always to be respected save where he has sought to abuse the distinction: Prest v Prest [2013] 2 AC 415, 487, para 34) must remain in the forefront of its analysis. The question should never be: can the shareholder raise the money? The question should always be: can the company raise the money?” 34. Ms Kaufmann QC relies also on the following passage in the judgment of Lord Wilson :- “In this context the criterion is: “Has the appellant company established on the balance of probabilities that no such funds would be made available to it, whether by its owner or by some other closely associated person, as would enable it to satisfy the requested condition?” 24. The criterion is simple. Its application is likely to be far from simple. The considerable forensic disadvantage suffered by an appellant which is required, as a condition of the appeal, to pay the judgment sum (or even just part of it) into court is likely to lead the company to dispute its imposition tooth and nail. The company may even have resolved that, were the condition to be imposed, it would, even if able to satisfy it, prefer to breach it and to suffer the dismissal of the appeal than to satisfy it and to continue the appeal. In cases, therefore, in which the respondent to the appeal suggests that the necessary funds would be made available to the company by, say, its owner, the court can expect to receive an emphatic refutation of the suggestion both by the company and, perhaps in particular, by the owner. The court should therefore not take the refutation at face value. It should judge the probable availability of the funds by reference to the underlying realities of the company’s financial position; and by reference to all aspects of its relationship with its owner, including, obviously, the extent to which he is directing (and has directed) its affairs and is supporting (and has supported) it in financial terms.” (emphasis added) 35. This judgment provides useful guidance albeit in a different context. It is not to be automatically assumed that the financial ability and resources of a shareholder, director or other persons are to be equated with and regarded as relevant when one is considering the financial ability or resources of a company. Criticism was made of the judgment of the Court of Appeal in this respect, particularly of the following passage at the end of its judgment[61] :- “125. Where the applicant is a limited company said to be of no or scanty financial ability of its own, financial resources available to it would necessarily include financial assistance that its members or shareholders, directors or backers of the litigation can reasonably be expected to provide. It follows that the applicant must provide information on the financial resources of these individuals or bodies and if they are financially capable but are unwilling to fund the litigation, the reasons why. Requiring the corporate applicant to provide such information does not involve lifting its corporate veil as such. It only ensures that the court is provided with all the necessary financial information for it to make an informed decision on whether a PCO should be granted.” 36. I agree that the Court of Appeal can be said to have been too robust if it meant to say that shareholders, directors and other supporters “must” always provide information of their financial ability or resources. The position is really this. Whether or not it is appropriate to look at the financial ability or resources of the shareholders, directors or other supporters of a company depends on the circumstances of any given case. This may, for example, involve looking at the history of support given to the company in question by the shareholders, directors or other people. Usually, the court can expect evidence of why persons who would normally be expected to support the company or who have supported the company in the past (such as shareholders) no longer wish to do so or, if they are willing, why they cannot do so. Thus, it will in some cases be necessary to provide details of the financial ability or resources of such persons; in other cases where the evidence demonstrates, this may not be appropriate or necessary at all. It is also important to look at the “underlying realities of the company’s financial position”[62] and to adopt a common sense view of the matter. Where I depart from Ms Kaufmann QC is her submission that in every case it will be impermissible to look at the financial ability or resources of the shareholders, directors or other people supporting the company, and the court is confined to looking only at the resources and assets in the company’s name or belonging to it. There is no justification for this very narrow approach and with respect, it is not commensurate with common sense. 37. Finally, I would emphasize the need for an applicant to be full and frank when dealing with questions of financial ability. It is important for an applicant properly and fully to explain his financial position to justify a claim for exceptional treatment in applying for a special costs order. C.2 Consequences of discontinuance of proceedings 38. The consideration of the financial ability of an applicant for a PCO is an important part of the court’s discretion. If an applicant can be shown to have the necessary financial capacity of meeting an adverse order as to costs in the proceedings, I cannot readily conceive of any court granting a PCO in such circumstances. This would be because if an applicant did have the financial ability or resources to meet an adverse costs order, it will hardly be reasonable for him to discontinue the proceedings in the absence of a PCO. Para. 74(1)(v) of Corner House will not be satisfied. C.3 Assessment of the financial ability of the respondent 39. Another of the criticisms levelled against the judgment of the Court of Appeal in the present case was that it was suggested that insufficient or no regard was paid to the resources of the respondent. This criticism is not justified as the Court of Appeal did take this into account. The financial resources of the respondent is clearly a relevant consideration in arriving at a fair and just outcome. C.4 Fairness and Justice 40. I have earlier alluded to the importance of looking at the overall fairness and justice of an application for a PCO. These terms are referred to in para. 74(1)(iv) and repeated in para. 74(3) of Corner House. RHC O.1A r.2(2) reminds us that in giving effect to the underlying objectives of rules of court, the court shall always recognize that the primary aim of exercising the procedural powers is to secure the just resolution of disputes in accordance with the substantive rights of the parties. 41. Some flexibility is therefore inherent in any exercise of discretion. The Corner House principles are to be applied with the necessary flexibility in order to reach a fair and just outcome. For example, the principle in para. 74(1)(iii) that the applicant should have no private interest in the outcome of the case may, in certain circumstances, be relaxed; after all, a matter may be of immense public importance notwithstanding that the applicant may have some personal benefit in the litigation. This point was touched upon by the Court of Appeal[63] and has also been discussed in a number of English cases.[64] As the point does not arise on the facts of the present case, it is not necessary to go into this further. 42. The assessment of financial ability is not a purely arithmetical exercise in which one compares the financial resources available to an applicant with the costs liability that the applicant might have to meet in case he is unsuccessful simply to see whether an applicant would in theory be able to meet such a liability. Sometimes it will be fair and just to make an order even though an applicant’s resources might appear to be adequate to meet a potential adverse costs order. For example, an applicant’s resources may be committed to other identified liabilities or expenditures and it would be unreasonable to divert them to a costs liability instead. Ms Kaufmann QC referred us to R (Howard League for Penal Reform) v Lord Chancellor whereby a PCO was ordered even though the applicant had the financial resources in theory to meet a substantial adverse award of costs. The court accepted the applicant’s contention that it would not be financially prudent to run the risk of an adverse order as to costs, given the resources needed for its other activities.[65] The critical question here is not whether, as a matter of dollars and cents, an applicant may be able to meet a potential costs order liability, but whether it would be fair and just to make a PCO. 43. In other situations, it may be fair and just not to make a PCO even though an applicant’s resources are limited and where it has been shown that his resources may be inadequate to meet an adverse costs order liability. The example given during the course of submissions was of a limited liability company being used for the purpose of instituting proceedings. Even though its resources would be limited and there was no possibility of the company meeting a costs liability, and even though the absence of a PCO may cause the company to discontinue the proceedings, it may be inappropriate to grant a PCO. In such a situation, the court may regard it as neither fair nor just to make a PCO merely to enable persons behind the company (whether shareholders or otherwise), however wealthy they may be, to become insulated from costs by the device of utilizing a limited liability company. Much depends on the circumstances. Where the shareholders, directors and other supporters of a company are wealthy and who can in the circumstances be expected to support the company, but simply choose not to do so without any plausible or acceptable explanation, the court may take the view that it would not be fair or just to make a PCO. It is one thing to say : “I caused the company to bring the proceedings and don’t have the money to meet an adverse costs order if it fails and so cannot see the litigation through without a PCO” and quite another to say : “I caused the company to bring the proceedings and I do have the money to meet an adverse costs order if it fails, but choose not to do so”. 44. Fairness and justice mean that the position of the other party to the proceedings is always taken into account. Hence, it becomes relevant to look at whether those acting for the applicant are doing so on a pro bono basis (para. 74(2) of Corner House) as this will mean that the costs liability of the other party, in the event he is unsuccessful, is also limited. This is also the justification for a PCO taking many forms, including the imposition of a condition capping the costs to which an applicant may be entitled if successful in the proceedings. As was said in Corner House,[66] “A PCO can take a number of different forms and the choice of the form of the order is an important aspect of the discretion exercised by the judge.” D. RESOLUTION OF THE APPEAL ON THE FACTS 45. With the above principles in mind, I now turn to the facts of the present case. 46. There has been no dispute over the principles contained in paras. 74(1)(i) to (iii) of Corner House : it is accepted that the present application for judicial review raises issues of general public importance, that these issues should be resolved by the court and that DHKL has no private interest in the outcome of the case. The dispute between the parties has been over the financial resources of DHKL and whether it is fair and just to make a PCO in its favour. 47. DHKL’s case in this respect is straightforward. It is contended that DHKL simply does not have the financial ability and resources to meet a potential liability as to the TPB’s costs in the event it is unsuccessful in the proceedings. And so, it is further contended, if a PCO was not made, it would have no choice but to discontinue the present judicial review proceedings. The following facts and matters are contained in the affirmation evidence placed by DHKL before the court[67] :- (1) The guarantors of DHKL are those persons identified above. The principal officers of the company are Mr Shaw and Mr Wong (who are the directors) and Mr Zimmerman (who is the Chief Executive Officer); they are also the guarantors of the Company. (2) Throughout the course of the present proceedings, details of the company’s bank balance have been provided and also audited financial reports. We are told that as at 7 February 2018, the bank balance of DHKL stood at $451,299.42. This is well short of the likely costs exposure in the event DHKL is unsuccessful in the judicial review proceedings. (3) The funding for the Company, apart from the directors, is somewhat unpredictable, relying on sponsorship and donations. (4) As far as the directors are concerned, no details whatsoever are provided as to their financial worth. Mr Zimmerman also provides no details of his own financial ability and resources. Yet these three persons would usually be expected to support the Company. They are the instigators of the present proceedings, who have chosen to use a corporate vehicle for this purpose. However, no financial details have been provided because as Mr Zimmerman repeatedly states in his affirmations[68] (and this was the effect of Ms Kaufmann’s submissions as well), whatever their worth, they are simply unwilling to countenance contributing anything towards the costs of the present litigation. As Mr Zimmerman deposes,[69] “Neither the Directors nor myself were eligible for legal aid or willing to expose ourselves to the risk of substantial legal costs for a case in which we had no personal interest (other than that shared by the public and [DHKL] itself), even though it was clearly a community interest championed by [DHKL].” Again, in his latest affirmation,[70] Mr Zimmerman states that the directors “are unwilling to contribute their personal finances towards this litigation.” This must also remain Mr Zimmerman’s position as well. (5) It is said that without a PCO, DHKL was “at serious financial risk” and this would be so even if it were to discontinue or withdraw from the present proceedings. Much of its resources have been used to pay existing costs orders and also to provide security for costs to appeal to this Court.[71] Funding for its projects has been cut and even amounts spent on salaries and overheads have been reduced. 48. In my view, as a matter of fairness and justice, a PCO should not be granted in the present case :- (1) Both Au J and the Court of Appeal were critical of DHKL’s failure to provide information as to the financial ability and resources of the directors.[72] These criticisms require some qualification. While it is true that both directors and Mr Zimmerman have consistently refused to provide such information, their position was that this information is irrelevant. Their position, as stated above, was a simple one : whatever their worth and it can even be assumed that they are persons of substantial means, they are unwilling to finance the litigation and certainly unwilling to meet any adverse costs orders against them. In addition, they assert that if a PCO was not granted, the Company would discontinue the proceedings and the proceedings, which do involve matters of public interest, will not be heard. (2) DHKL’s stated position is an unattractive one that cannot be accepted to justify the grant of a PCO. As observed earlier, shareholders, directors and (in the present case) guarantors can be expected in the normal course of things to be the main supporters, financially and otherwise, of a company. When such persons refuse to do so, the court is entitled to inquire as to the reasons why this is so. In the present case, the directors and the CEO, all guarantors and, as stated earlier the instigators of the present litigation, upon analysis wish to be insulated from an order for costs through the means of a limited liability and under resourced company. By reason of their unwillingness to vouchsafe details of their financial ability, they can be assumed to have the means to finance the Company and meet its obligations regarding costs, but are unwilling to do so and choose effectively not to do so. This is despite the fact that quite clearly, they are the main driving forces behind the present litigation. (3) The obvious potential unfairness to the TPB is not addressed, only that it is asserted that the TPB, through being supported out of the public purse, would not be too seriously prejudiced. However, even where the respondent to an application for a PCO is funded by the public purse, this is not to say there is no prejudice in the event it is unable to recover costs. (4) DHKL’s position is tantamount to saying that where paras. 74(1)(i) to (iii) of Corner House are satisfied, without more, a PCO ought to be granted. Financial ability is relegated to a relatively unimportant factor, particularly in the case of a company which has insufficient funds in its own name to meet an adverse costs order. Yet, as shown above, the whole rationale of a PCO is to ensure that proceedings in which points of great public importance arise, are not stifled through a lack of financial means. Financial ability or inability is important. This has to be approached meaningfully as a matter of substance and above all, approached consistently with fairness and justice. E. PROCEDURE IN DETERMINING APPLICATIONS FOR PCOs 49. This is question 2 of the questions on which leave was given. None of the parties however has really addressed the court on this question and in my view it is not necessary in this judgment to set out detailed guidance on matters of procedure regarding PCOs. There is an evident advantage in hearing such applications at an early stage involving all relevant parties but as to what is suitable will depend on the view the court takes regarding what is the most convenient case management course to adopt. The hearing should not take the time that it took in the lower courts in this case, but the length of the hearings in the present case was attributable mainly to having to deal with matters of principle. F. CONCLUSION 50. For the above reasons, this appeal is dismissed. On costs, given the importance of the matters involved in this appeal, I would make an order nisi that there be no order as to costs. Should any party wish for a different order for costs, written submissions should be lodged with the Registrar and served on the other parties within fourteen days of the handing down of this judgment, with liberty to the other party or parties to lodge and serve written submissions in reply within fourteen days thereafter. In the absence of any written submissions seeking to vary within the time limited for their service, this costs order nisi will become absolute. Mr Justice Ribeiro PJ : 51. I agree with the judgment of the Chief Justice. Mr Justice Tang PJ : 52. I agree with the judgment of the Chief Justice. Mr Justice Bokhary NPJ : 53. Despite the arguments so ably advanced by Ms Phillippa Kaufmann QC for the appellant Designing Hong Kong Ltd, I agree with Chief Justice Ma that this appeal should be dismissed with, as he proposes, an order nisi that there be no order as to costs. 54. Designing is a company limited by guarantee and without any shareholders. It obtained leave to apply for judicial review. But the judge refused it a protective costs order (“PCO”). That refusal was affirmed by the Court of Appeal. 55. In seeking a PCO, Designing filed evidence that it would not be able to meet any order for costs that might be made against it in the judicial review proceedings. But it provided no information on its directors’ finances. Its stance is that such information is irrelevant since (i) its directors have not pledged their support, (ii) they do not have any personal interest in the judicial review proceedings and (iii) it is not a corporation established to conceal or evade anyone’s financial liability to fund litigation. 56. The courts below felt that without information on the finances of Designing’s directors it was not possible to decide whether the judicial review proceedings would probably be discontinued unless a PCO is granted. Designing’s chief executive officer Mr Paulus Johannes Zimmerman has said on affidavit that those proceedings would be discontinued unless a PCO is granted. No application to cross-examine him having been made, it might well be appropriate to find that those proceedings would probably - not certainly but probably - be discontinued if no PCO is granted. However that may be, such a finding would not end the matter. The courts below also felt that without information on Designing’s directors’ finances it was not possible to decide whether any such discontinuance would be reasonable. 57. Information on the finances of the individuals in question was not sought merely because they are directors. They are the directors who authorized the bringing of the judicial review proceedings. I underline that hopefully to avoid, but at least to reduce, the “chilling effect” which Ms Kaufmann warned against. 58. In a contested application for a PCO the court has of course to do justice between the parties. Suppose a corporate applicant says that being refused a PCO would cause it to discontinue its judicial review proceedings. Then it is sometimes - not always but sometimes - justified to hold that deciding whether such discontinuance would be reasonable requires information on the applicant’s director’s and/or shareholders’ finances. 59. The fact that Mr Zimmerman went to the directors concerned for funds is some indication that they are in a position to provide the funds requested. In any event, Ms Kaufmann is content to have it assumed that they can afford to provide those funds. They are not willing to do so. There is no clear (if any) indication why not. 60. As to that, the view taken by the courts below comes to this. Information on the finances of those directors would shed light on why they are unwilling to provide the funds requested. So without such information it was not possible satisfactorily to decide whether discontinuance absent a PCO would be reasonable. That view, taken by the first instance judge and affirmed by the intermediate appellate court, involved a question of fact and degree in the context of an exercise of discretion. I am not persuaded that it ought to be overturned on final appeal. That disposes of the present appeal. But there remains a number of things to be said. 61. Access to the courts is an arterial right, being the channel by which judicial enforcement of legal entitlements is sought. In public interest litigation, access to the courts has an added dimension. Typically, such litigation involves elucidating public law points of wide implications. Where a party’s access to the courts for such elucidation is inhibited by the risk of having to pay the other side’s costs, the resulting injustice would be not only to the party so inhibited but also to the public. PCOs are designed to remove such inhibition by protecting against or at least limiting the potential liability to pay the other side’s costs. 62. That being so, the appropriate time to entertain an application for a PCO in judicial review proceedings is, at least in general, when leave to apply for judicial review is sought. There is a range of procedural options open to the court when deciding whether to grant such leave. So is there a range of such options open to the court when deciding whether to grant a PCO. 63. Circumstances naturally vary from one place to another. But I see no reason why Hong Kong’s PCO regime should be more restrictive than those of comparable jurisdictions. (1) In Hong Kong the availability of legal aid is wider than it appears to be in Britain. But that is no reason why PCOs should be more difficult to obtain here even in cases in which legal aid is not available. The relatively wider availability of legal aid in Hong Kong makes it correspondingly rare for a PCO to be sought here. That reduces the demand that PCOs make on public money. Such reduction is hardly a reason why PCOs should be more difficult to obtain. (2) As for the discretion exercisable at the end of a case not to order costs against an unsuccessful judicial review applicant, it does not appear to be unique to Hong Kong. In any event, it serves a purpose different from the one served by the discretion to make a PCO. (3) In Hong Kong where democracy still has a long way to evolve, persons might well resort to public interest litigation in circumstances under which persons in evolved democracies would resort to the political process instead. That difference is anything but a reason why PCOs should be more difficult to obtain here. 64. The law on the funding of public interest litigation may in time develop to become more conducive to access to the courts than it is at present. Such development may come by way of legislation, judicial decision or both. In the present case, we have not been invited to adopt an approach more favourable to PCO applicants than the approach laid down in R (Corner House Research) v Secretary of State for Trade and Industry [2005] 1 WLR 2600. Having regard to the questions on which leave to bring the present appeal was granted, the constitutional argument of which there is a foretaste in Designing’s supplemental printed case was not pursued. The resolution of any such argument would be for some future occasion. Meanwhile, close attention is to be paid to the particular circumstances of the present case. Even if this decision is not one which NGOs, for example, would particularly welcome, it is, properly understood, not a decision of which anyone resorting to public interest litigation need be fearful. 65. In thanking all the lawyers in the case, I am confident that no one would begrudge my making such thanks especially to Designing’s lawyers, they having acted pro bono. Lord Collins of Mapesbury NPJ : 66. I agree with the judgment of the Chief Justice. Chief Justice Ma : 67. For the above reasons, this appeal is unanimously dismissed. A costs order nisi is also made that there be no order as to costs. Ms Phillippa Kaufmann QC, Mr Nigel Kat SC, Mr Azan Marwah and Ms Katherine Olley, instructed by Boase Cohen & Collins, for the Appellant Mr Johnny Mok SC and Mr Jenkin Suen, instructed by the Department of Justice, for the Respondent Mr Wong Yan Lung SC and Mr Abraham Chan SC, instructed by the Department of Justice, for the Intervener [1] The order is also known as a pre‑emptive costs order. [2] This is a company where the liability of its members is limited by the company’s articles of association to such amount as the members undertake to contribute to the assets of the company in the event of its being wound up : see sections 9 and 84(2) of the Companies Ordinance, Cap. 622. Such a company has no share capital and is an appropriate form of limited liability corporation when no profits are intended to be distributed. Accordingly, companies limited by guarantee are usually charities or non‑profit organisations. [3] For a brief description of so‑called draft plans, see Town Planning Board v Town Planning Appeal Board (2017) 20 HKCFAR 196, at para. 2 fn 5. [4] The planned Central Military Dock. [5] On 16 and 17 December 2014 before Au J. [6] Para. 8 of the Judgment (the CFI Judgment). [7] Mr Stewart Wong SC and Ms Bonnie Cheng. [8] Before Cheung CJHC, Lam VP and Jeremy Poon JA. The judgment of the Court of Appeal (the CA Judgment) was handed down on 16 February 2017. [9] On 7 June 2017. [10] Ma CJ, Ribeiro and Tang PJJ. [11] Ms Phillippa Kaufmann QC, Mr Nigel Kat SC, Mr Azan Marwah and Ms Katherine Olley represented DHKL, instructed by Messrs Boase Cohen & Collins. The TPB was represented by Mr Johnny Mok SC and Mr Jenkin Suen. Mr Wong Yan Lung SC and Mr Abraham Chan SC acted for the Intervener. [12] Cap. 4. Applications for judicial review are heard in the High Court. [13] The relevant rules are contained in the Rules of the High Court (Cap. 4A) (the RHC). [14] Section 43 of the Hong Kong Court of Final Appeal Ordinance Cap. 484. [15] RHC O.62 r.2(1). The discretion in relation to costs contained in s. 52A of the principal ordinance is to be exercised in accordance with RHC O.62 : see O.62 r.2(4). [16] See also in this respect Hong Kong Civil Procedure 2018 Vol. 1 at para. 62/5/6. [17] (2014) 17 HKCFAR 841. [18] At para. 17(8). [19] See also the commentary in Hong Kong Civil Procedure 2018 Vol. 1 at para. 62/5/7. This general rule of costs following the event is not applicable to interlocutory proceedings : RHC O.62 r.3(2). [20] Which includes applications for judicial review. [21] [2007] 4 HKC 428. [22] (2004) 7 HKCFAR 114. [23] At para. 19. [24] [2005] 1 WLR 2600. It will be necessary to discuss this case in greater detail later. [25] (1998) 193 CLR 72. [26] [2004] 2 HKLRD 989. [27] At paras. 29 and 30. [28] This was a reference to the judgment of Kirby J in Oshlack (at para. 136). [29] At paras. 23 and 42. [30] [2009] 3 HKLRD 362. [31] At para. 9. [32] At paras. 17(10) and (11). [33] [1999] 1 WLR 347, at 357A-D. I shall hereinafter refer to this case as the CPAG case. [34] This is the test in applications for leave to appeal for judicial review : see Po Fun Chan v Winnie Cheung (2007) 10 HKCFAR 676, at para. 15. [35] Chan Noi Heung at paras. 11 and 12. [36] At para. 72. [37] [1995] ICR 685, at 694. [38] In the section headed “Protective costs order : the historical setting” at 2613-2617 (paras. 28 to 43). [39] [2014] EWHC 3164 at para. 7. [40] Contained in the Hong Kong Bill of Rights Ordinance Cap. 383. [41] At para. 74. [42] See paras. 15-16 above. The order whereby no adverse order for costs is made against the unsuccessful party at the conclusion of proceedings is popularly known as a “Chu Hoi Dick Order”. [43] At para. 81. [44] [2009] 1 WLR 1436. [45] At para. 24. Smith LJ agreed with the judgment of Waller LJ : see paras. 80-82. [46] At paras. 64-66. [47] [2009] 1 Costs LR 80. [48] See Question (1)(c) set out in para. 8 above. [49] Para. 19b of the Appellant’s Case. [50] Or, in the case of a corporation as in the present case, its directors, members or guarantors. [51] See para. 8 above. [52] [1998] 4 HKC 461. [53] At 464E-F referring to Keary Developments Limited v Tarmac Construction Ltd. [1995] 3 ALL ER 534, at 539‑40. [54] (2015) 18 HKCFAR 283. [55] At para. 12. [56] [2017] 1 WLR 3014. [57] Under the Civil Procedural Rules applicable in England. In Hong Kong, the term used in our Rules is still leave to appeal. [58] This was the judgment of the majority of the court (together with Lord Neuberger of Abbotsbury PSC and Lord Hodge JSC). Lord Clarke of Stone-cum-Ebony and Lord Carnwath JJSC were in the minority. The dissent was on the facts. [59] At paras. 12 and 16. [60] At para. 18. [61] A similar criticism was made of that part of Au J’s judgment in which it was said (at CFI Judgment para. 67) that for a company “it is legitimate and proper for the court to look at the financial means of not only the company itself but also of its directors and shareholders as well as the ability of the company to raise funds from other sources.” [62] Goldtrail at para. 24, set out in para. 34 above. [63] At CA Judgment para. 106. [64] Such as Compton (para. 27(3) above) at para. 23; Morgan and Baker v Hinton Organics (Wessex) Ltd. (2010) 1 Costs LR 1, at paras. 29, 35‑40; Plantagenet at paras. 21‑28. [65] The case is reported at [2017] 4 WLR 92, although the reasons for making a PCO are not contained in this report. It was separately provided to us. [66] At para. 75. [67] This evidence was unchallenged by the TPB in that no cross examination was sought of the deponent to these affirmations, being Mr Zimmerman. [68] He has provided 11 affirmations. [69] In his 7th affirmation dated 22 October 2016. [70] The 11th affirmation. [71] DHKL, despite claiming it was unable to raise the money, was ordered to provide security for costs in the sum of $400,000 : see Decision dated 20 December 2017. [72] See CFI Judgment at para. 68, CA Judgment at paras. 124‑126. 
The Court : A. INTRODUCTION 1. In the court below, the Magistrate convicted the 1st respondent of Charge (2) – taking part in an unlawful assembly; the 2nd respondent of Charge (3) – inciting others to take part in an unlawful assembly; and the 3rd respondent of Charge (4) – taking part in an unlawful assembly; and sentenced the 1st respondent to a community service order of 80 hours; the 2nd respondent to a community service order of 120 hours and the 3rd respondent to 3 weeks’ imprisonment suspended for 1 year. 2. By a judgment handed down on 17 August 2017 (“Judgment”), we allowed the Secretary for Justice’s application for review against sentence brought under section 81A of the Criminal Procedure Ordinance (“CPO”).[1] We held that the original sentences imposed by the Magistrate on the respondents were wrong in principle and were manifestly inadequate. We accordingly set them aside and substituted : (1) 6 months’ imprisonment for the 1st respondent in respect of Charge (2); (2) 8 months’ imprisonment for the 2nd respondent in respect of Charge (3); and (3) 7 months’ imprisonment for the 3rd respondent in respect of Charge (4). 3. The respondents now seek leave to appeal to the Court of Final Appeal. Under section 32(2) of the Hong Kong Court of Final Appeal Ordinance,[2] leave to appeal shall not be granted unless (1) it is certified by the court below that a point of law of great and general importance was involved in the decision or (2) it is shown that substantial and grave injustice has been done. All the 3 respondents have already directly applied to the Court of Final Appeal under the second limb.[3] Invoking also the first limb, the 2nd and 3rd respondents by their respective notice of motion dated 4 and 27 September 2017 applied to this Court for a certificate that the points of law of great and general importance as set out therein were involved in the Judgment. 4. By the parties’ consent, the applications by the 2nd and 3rd respondents are to be disposed of on paper without an oral hearing. We have duly considered the parties’ written submissions and all the authorities that they seek to rely on. We now hand down our judgment. 5. For the reasons to be articulated below, we dismiss both applications. The reasons may be summarized as follows. 6. As will be seen shortly, it is well established by both Hong Kong and English appellate authorities that : (1) On an application for review of sentence, the Secretary for Justice is not entitled to change the factual basis for sentence that it had relied on at trial and ask the Court of Appeal to inquire into facts which had not been pursued below. Even if the sentencing court had proceeded on a wrong factual basis for sentence, it is not open to the Secretary for Justice to make such an assertion before the Court of Appeal if the prosecution had the opportunity to have it corrected at trial but they had not sought to do so. The Court of Appeal will adopt the same factual basis as it was before the sentencing court. (2) Subject to (1), where the ground for review is that the sentencing court acted on an erroneous factual basis, the Court of Appeal is entitled to examine the evidence adduced below to consider if, based on the facts proved, admitted or not in dispute, the sentencing court did make the error as asserted by the Secretary for Justice. If so, the Court of Appeal is entitled to correct the factual error and to consider if, based on the factual basis as corrected, the sentence imposed is wrong in principle or manifestly inadequate. If so, the Court of Appeal is entitled to interfere. 7. As a corollary of the proposition in [6(2)] above and according to first principle, where it is shown that, based on the facts proved, admitted or not in dispute, the sentencing court has failed to take into account certain matters which are relevant to sentence, the sentencing court has proceeded on an incomplete factual basis, rendering the sentence imposed wrong as a matter of law and principle. In such circumstances, the Court of Appeal is not bound by the findings made by the sentencing court for the purpose of sentence because those findings are incomplete for such purpose. The Court of Appeal is entitled to consider all the relevant matters including those wrongly ignored by the sentencing court to determine if in the overall circumstances of the case, the sentence imposed is wrong in principle and/or manifestly inadequate. If so, the Court of Appeal is entitled to interfere. 8. Applying the above principles, we held in the Judgment that according to the facts proved, admitted or not in dispute based on the evidence adduced before the Magistrate : (1) the unlawful assembly was a serious and large-scale unlawful assembly, involving violence; (2) the Magistrate had failed to take into account those relevant factors as identified by the Court; (3) the Magistrate had made the errors as identified by the Court; and (4) the sentences imposed by the Magistrate were wrong in principle and were manifestly inadequate. We then set aside the sentences imposed by the Magistrate and substituted them with immediate custodial sentences as stated. Neither the 2nd nor the 3rd respondent has been able to demonstrate an arguable case that we had erred in allowing the application for review as we did in the manner as described above. 9. Since the answers to the questions sought to be certified by the 2nd and 3rd respondents are well settled, and they have failed to show an arguable case that we had erred, we dismiss their applications. B. THE POINTS OF LAW SOUGHT TO BE CERTIFIED 10. By his notice of motion filed on 4 September 2017, the 2nd respondent applied for a certificate that the following points of law of great and general importance are involved in the Judgment (“R2’s Question”) : “ (1) Whether on an application by the Secretary for Justice for review of sentence pursuant to section 81A of the CPO, the Court of Appeal has the power to make any factual findings not made at the trial or contrary to findings made by the trial judge, where by reason of or in consideration of the same the sentence is increased? (“R2’s Sub-Question 1”) (2) If the Court of Appeal has such power, under what circumstances and to what extent should the Court of Appeal exercise such power?” 11. By his notice of motion filed on 27 September 2017, the 3rd respondent applied for a certificate that the following point of law of great and general importance is involved in the Judgment (“R3’s Question”) : “ In a Secretary for Justice’s sentence review application under section 81A of the CPO, is it correct that the Court of Appeal does not have the power to make any factual findings adverse to the defendants different from or in addition to those made by the tribunal of fact at trial?” C. THE BASIS FOR ALLOWING THE APPLICATION FOR REVIEW 12. To put the parties’ submissions and our discussion in context, it is necessary to briefly recap the basis of our judgment in allowing the Secretary’s application for review against sentence. 13. In Part B of the Judgment, we noted that the defence did not dispute the background of the incident, or the acts of the respondents as recorded in the video clips which were produced by way of admitted facts; and that they did not seriously dispute the evidence of various prosecution witnesses.[4] We then summarized the prosecution case based on the evidence, which was either admitted or not in dispute. 14. After summarizing the respondents’ evidence (Part C) and referring to the reasons for verdict (Part D), the reasons for sentence (Part E) and the parties’ submissions (Parts F and G), we in Part H set out the legal propositions and sentencing principles applicable to unlawful assembly involving violence. We subsequently summarized those propositions and principles in Secretary for Justice v Leung Hiu Yeung and Others, CAAR 3 of 2016, unreported, 11 September 2017 at [87] – [108]. We shall not repeat them here. 15. In Part I, we explained why, based on the evidence, the facts of the present case were serious and why it was a large-scale unlawful assembly involving violence by reference to the matters discussed at [157] – [164]. We went on to observe at [165] that according to the respondents’ conduct after the incident they had shown no genuine remorse for the offences they had committed. We concluded at [166] that given the seriousness of the offences, it was necessary for the court to place more weight on punishment and deterrence and correspondingly less weight on the respondents’ personal circumstances, motives and rehabilitation. The appropriate sentence must be sufficiently deterrent to prevent them from re-offending and to warn others not to follow suit. The appropriate sentence must be one of immediate custodial sentence.[5] 16. For the errors that the Magistrate made as listed out at [167(1) – (5)], we at [168] held that it was wrong in principle for her to sentence the respondents to community service orders or suspended sentence, as the case may be, and that the sentences she passed were manifestly inadequate. 17. We at [169] stated the correct starting points for sentence to be : (1) 8 months for the 1st respondent in respect of Charge (2); (2) 10 months for the 2nd respondent in respect of Charge (3); and (3) 8 months for the 3rd respondent in respect of Charge (4). We then exercised our discretion in accordance with customary practice to reduce the sentence by one month for all the respondents on account of the fact that it was the Secretary’s application for review. In respect of the 1st and 2nd respondents, since they had already completed their community service orders, we further reduced their sentence by one month. That further reduction was made because of the principle of double jeopardy : see Attorney General’s Reference Nos 4 and 7 of 2002 [2002] 2 Cr App R (S) 77, per Lord Woolf LCJ at [34]. We finally imposed on the respondents the sentences as stated at [2] above. 18. To conclude, we reiterated at [171] of the Judgment that the respondents were not convicted of and sentenced for the offences for exercising their rights to assembly, demonstration and expression, which rights the law fully recognizes and protects. They were so convicted and sentenced simply because they had trespassed the boundaries laid down by the law for peaceful assembly by conducting themselves in the seriously unlawful manner as established by the evidence. D. THE PARTIES’ SUBMISSIONS D1. The 2nd respondent’s submissions 19. Mr Robert Pang SC (leading Mr Michael Chai and Ms Priscilla Chow), for the 2nd respondent, submitted that in reaching the conclusion as we did and holding that the Magistrate overlooked certain facts, we effectively increased the sentence of the 2nd respondent on facts which were not made at the trial or contrary to the findings made by the Magistrate. Counsel submitted that whether the Court of Appeal is entitled to do so, thus R2’s Question, is a question of law involving great and general public importance. The great and general public importance of R2’s Question is self-evident. Mr Pang went on to submit that the Court of Appeal is not entitled to do so and R2’s Sub-Question 1 must be answered in the negative. 20. Mr Pang relied on what he considered to be the practice of the English Court of Appeal by citing Archbold on Criminal Pleading, Evidence and Practice 2017, §7-445 and Taylor on Criminal Appeals (2nd Edn), §13.47. (1) The relevant passage in Archbold states that : “ The Court of Appeal will consider a reference on the basis of the facts proved or admitted; it would not constitute itself as a court of first instance inquiring into the facts which had not been pursued or proved in the Crown Court: Att. Gen.’s Reference (No. 95 of 1998) (R. v. Highfield), The Times, April 21, 1999, CA.” (2) The relevant passage in Taylor states that : “ The Court must review any sentence referred to it on the basis of the facts as they were before the sentencing judge. It is not open to the Attorney-General to invite the Court to sentence on a different factual basis, even where the sentencing judge has proceeded on a wrong basis if the prosecution could have but failed to correct it at the time. Nor is it the function of the Court to substitute their own view of the facts in light of new material.” Mr Pang submitted that the same practice is adopted in Hong Kong by relying on Attorney General v Li Ah-sang [1995] 2 HKCLR 239. 21. Mr Pang fairly recognized that in the recent judgment of Leung Hiu Yeung, supra, we at Part H2 set out the general principles concerning the factual basis of a review application before the Court of Appeal. However, he submitted that since R2’s Question has not been considered by the Court of Final Appeal, a certificate that a point of law of great and general importance should be granted. D2. The 3rd respondents’ submissions 22. The thrust of the submissions of Mr Edwin Choy (together with Mr Joe Chan), for the 3rd respondent, is similar to Mr Pang’s. He submitted that when reviewing a sentence, the Court of Appeal must not act as a fact-finding body and make any factual findings adverse to the defendants different from or in addition to those made by the sentencing court at trial. 23. Mr Choy cited R v Sin Yau Ming [1992] 1 HKCLR 127; Raymond Chen v HKSAR (2010) 13 HKCFAR 728; Chou Shih Bin v HKSAR (2005) 8 HKCFAR 70; and 香港特別行政區　訴　張育強, HCMA 540/2011, unreported, 6 October 2011 to support the general proposition that in criminal appeals, an appellate court will be extremely slow to interfere with the findings made by the trial court. 24. As to the practice of the Court of Appeal in dealing with an application for review against sentence, Mr Choy similarly relied on Li Ah-Sang, supra and R v Highfield, supra where Judge LJ observed that : “ Whether a sentence referred to the Court of Appeal by the Attorney General was unduly lenient was a question which had to be decided not in the light of what was alleged but of what was proved or found to have been established. The Court of Appeal could not constitute itself as a court of first instance inquiring into facts which had not been pursued or proved in the Crown Court.” D3. The Secretary’s submissions 25. Mr David Leung, SC, DDPP (leading Ms Annie Li, PP), for the Secretary, took two main points : (1) In allowing the Secretary’s application for review, we had not admitted any fresh evidence. We only examined the evidence which was proved or admitted at trial to determine if the Magistrate, when passing sentences on the respondents, had erroneously considered or overlooked such evidence. Thus both R2’s Question and R3’s Question were simply not involved in the Judgment. (2) What the 2nd and 3rd respondents seek to certify are not points of law, let alone points of great and general public importance. 26. In elaborating on point (1), Mr Leung submitted that if the ground for review is that the sentencing court made an erroneous factual finding or ignored certain findings of fact in sentencing, the Court of Appeal certainly has the power to determine the proper factual basis from the evidence proved or admitted below in order to consider the proper sentence that should have been imposed. Were it otherwise, an absurdity would arise in that even when it is demonstrated that the original sentence was imposed on a wholly erroneous basis, the mistake in sentencing could never be rectified by the Court of Appeal on the review application. In support, Mr Leung cited Attorney General’s Reference Nos 4 and 7 of 2002, supra; Attorney General’s Reference Nos, 90 and 91 of 2003 [2004] EWCA Crim 1839; Secretary for Justice v Au Chi Hang & Others [2006] 2 HKLRD 310; and Leung Hiu Yeung, supra. 27. Mr Leung distinguished Li Ah-sang and R v Highfield by submitting that because of the prosecution’s particular conduct before the sentencing court in those cases, the Attorney General was not allowed to seek a review on the basis which was different or which had not been pursued or proved below. Mr Leung next submitted that the proposition mentioned in the passages in Archbold and Taylor only applies to cases where the prosecution had the opportunity to have the mistake corrected at trial but they had not sought to do so. By drawing these distinctions, Mr Leung submitted that the present case is different as in the Judgment, we had not changed the factual basis identified and established by the prosecution at trial. 28. Concerning point (2), Mr Leung submitted that by reason of his submissions in respect of point (1), we in the Judgment did not assume the role of a trial court to carry out a fact-finding exercise on our own. All we did was, on the basis of the evidence adduced below which was proved or admitted or was not in dispute, to give those facts and evidence the proper weight in order to consider if the sentences imposed by the Magistrate were wrong in principle and/or manifestly inadequate and if so, the proper sentences. Thus R2’s Question and R3’s Question are not points of law, let alone points of law of great and general importance. E. DISCUSSION 29. In reviewing a sentence under section 81A of the CPO, the Court of Appeal must act on a proper factual basis to determine if it is wrong in principle and/or manifestly inadequate. For passing a sentence and reviewing the correctness of the sentence passed is always a fact-sensitive exercise. E1. The general principles 30. Recently in Leung Hiu Yeung, supra, we had had the opportunity of considering the general principles on how the Court of Appeal would approach the factual basis for the purpose of reviewing a sentence. We stated :[6] “ 124. The respondents submitted that in dealing with an application for review of sentence, the Court of Appeal must take the factual findings made by the trial magistrate as the only basis. The Court of Appeal cannot act beyond his factual findings. They cited Attorney General v Li Ah-sang [1995] 2 HKCLR 239 in support. In our view, the respondents’ submissions are questionable. And if one reads Li Ah-sang carefully, one can readily see that it does not support their contention. 125. The general principle is that in dealing with an application for review of sentence, the Court of Appeal will proceed on the basis of the facts proved or admitted; it would not constitute itself as a court of first instance inquiring into facts which had not been pursued or proved in the court below. See Archbold on Criminal Pleading, Evidence and Practice 2017, §7-445. Accordingly, the Court of Appeal will generally proceed on the basis of the facts as found by the court below. However, it does not necessarily follow that in every scenario, the Court of Appeal is, in terms of facts, bound by the lower court’s factual findings. (1) Pursuant to section 81B(3) and section 83V of the Criminal Procedure Ordinance, the Court of Appeal may, if it thinks it necessary or expedient in the interests of justice, admit fresh evidence. Although the Court of Appeal will be slow to admit fresh evidence and section 83V(5) also provides that in no case shall any sentence be increased by reason of or in consideration of any evidence which was not given at the trial, it does not detract from the power of the Court of Appeal to admit fresh evidence in an application for review. (We have not exercised this power in the present case : see [128] below.) (2) If the ground for review is that the lower court had made an erroneous factual finding, the Court of Appeal may examine if the finding in question is correct. Secretary for Justice v Au Chi Hang & Others [2006] 2 HKLRD 310 is an example. There, the first instance judge found that the weapons concerned were not offensive weapons as mentioned in the relevant authority. In applying to review the sentence, the Secretary for Justice submitted that the judge erred in making such a finding and had thereby failed to apply the relevant authority in sentencing. The Court of Appeal rejected the Secretary for Justice’s submission. The Court further pointed out the wooden pole and iron pole in question had not been seized as exhibit. Their dimensions were only described in the Admitted Facts ([33]). That being the case, the Court of Appeal said that as the Court did not have the advantage of the judge having the “feel” of the case, it was unable to overturn the judge’s finding that the weapons were not offensive weapons ([39]). As Au Chi Hang demonstrates, if it is the ground for review, the Court of Appeal may examine the factual findings made by the lower court although the Court of Appeal will be slow to interfere. (3) If based on evidence which had been proved or admitted or which had been adduced and was not in dispute, certain facts which are plainly relevant to sentence, whether they are in favour of the defendant or not, have been overlooked by the lower court, the sentence imposed by the lower court is erroneous as a matter of law and principle. In such circumstances, the Court of Appeal is not bound by the factual findings of the lower court because such findings are incomplete, having failed to take into account relevant facts. When the Court of Appeal considers those facts which ought to be considered, it does not depart from the general principle that in an application for review against sentence, the Court of Appeal will proceed on the basis of facts proved or admitted and will not inquire into facts not pursued or proved as if it were a first instance court. For the Court of Appeal only considers those evidence proved, admitted or not in dispute which are relevant to sentence but have been wrongly ignored by the first instance court. 126. The facts in Li Ah-sang are special. (1) There (the 2nd Case), the defendant admitted two charges of employing persons not lawfully employable. The magistrate fined him HK$5,000.00 on each charge. The Attorney General applied for a review on the ground that the employees employed by the defendant were illegal immigrants. The sentence was hence wrong in principle and manifestly inadequate. However, before sentencing the defendant, the same magistrate at the trial concerning the employees (the 1st Case) found that they were not illegal immigrants. Thus when the magistrate sentenced the defendant, he proceeded on the basis that the employees were not illegal immigrants. The prosecution there and then did not raise any objection and in fact did not allege that the employees were illegal immigrants. (2) The Attorney General submitted that since whether the employees were illegal immigrants was a live issue as a matter of law, the Court of Appeal might rule on it. The Court of Appeal however disagreed. It took the view that as the magistrate did not make any ruling on this point in the 2nd Case, there was no legal question arising for the Court of Appeal to consider. (3) The Court of Appeal went on to say that the magistrate in sentencing the defendant assumed that the employees were not illegal immigrants on the basis of his finding in the 1st Case. The prosecution did not raise any objection or ask to hold a Newton hearing to determine their status. In such circumstances, it is not open to the Attorney General to seek a review other than the facts on which the magistrate chose to find. The Court of Appeal would therefore adopt the same factual basis for the review. 127. Properly understood, the way which the Court of Appeal in Li Ah-sang dealt with the factual basis for review is consistent with the general principle stated in [125] above. As the Court of Appeal would proceed on the basis of the facts proved or admitted, it would not constitute itself as a court of first instance inquiring into facts which had not been pursued or proved in the court below, the Court of Appeal therefore rejected the Attorney General’s submission and held that it would adopt what the magistrate had accepted and not disputed by the prosecution as the factual basis for review. That is to say, the employees employed by the defendant were not illegal immigrants. Li Ah-sang did not establish the proposition advocated by the respondents, namely, in dealing with an application for review against sentence, the Court of Appeal must take the factual findings made by the trial magistrate as the only basis and the Court of Appeal cannot act beyond his factual findings.” 31. We take this opportunity to further elaborate on the above propositions with the benefit of the additional authorities cited to us by the parties. E1.1 The proposition in [125(2)] 32. In Leung Hiu Yeung, we cited Au Chi Hang, supra, as an example for the proposition stated in [125(2)] above that if the ground of review is that the lower court had made an erroneous factual finding, the Court of Appeal is entitled to consider if the finding is correctly made. There are other local and English authorities illustrating the same point. 33. In the local case of Lau Chiu-tak, supra, there were two applications for review before the Court of Appeal. In the first case, the respondent was charged with wounding under section 19 of the Offence Against the Person Ordinance[7] and was sentenced to a suspended prison sentence and ordered to pay compensation to the victim. In the second case, the respondent was charged with possession of tetrahydrocannabinol for trafficking. On the assumption that the drug was cannabis, the district judge sentenced him to 12 months’ imprisonment. The Attorney General sought a review of sentence (1) in the first case on the grounds that the sentence was wrong in principle and manifestly inadequate; and (2) in the second case on the ground that the sentence was manifestly inadequate. 34. Delivery the Judgment of the Court of Appeal, Sir Alan Huggins VP stated the general principles at p 26D – F : “ It is enough to say that where the application is based upon an error of principle it is incumbent on the Attorney General to identify the nature of the error alleged, and it goes without saying that he is entitled to cite authority to establish the error. Where it is alleged that a sentence is manifestly inadequate or manifestly excessive, he must clearly be allowed to say which, although it would be strange if that matter were in doubt. He may properly draw attention to any matter appearing on the record, but may not adduce new evidence (for example as to the incidence of a particular crime) with a view to an increase of sentence : s.83V(5).[8] Previously decided cases may be cited for the purpose of showing that the sentence under review is substantially and unjustifiably out of line with the usual tariff for that kind of offence.” 35. In allowing the applications for review, Huggins VP held that in the first case the judge erred in suggesting that, because the respondent was not violent in nature and acted on the spur of the moment, he did not intend the wound he in fact caused and that he might not have intended to cause such a serious wound, but he did intend to cause the wound (p 26H). In effect, Huggins VP held that the judge erred on the finding concerning the respondent’s intention. 36. In the second case, the judge sentenced the respondent on the erroneous assumption that the drug was cannabis when in fact it was something else. Huggins VP held that the sentence was manifestly inadequate and substituted a sentence of 2½ years’ imprisonment on the correct factual basis that the drug was tetrahydrocannabinol (p 27 C- D). 37. In the English case of Attorney General’s Reference Nos 90 and 91 of 2003, supra, the defendant was sentenced to 5 years’ imprisonment for possession of a controlled drug of class A (ecstasy) with intent. Initially the defendant pleaded not guilty. Later, he indicated his wish to plead guilty but only on the basis that he believed the drug to be amphetamine but not ecstasy. The prosecution did not accept this basis of plea. In the end, the defendant did not pursue his basis of plea and no Newton hearing was held on his alleged belief. The defence did not mention his alleged belief in mitigation. However, in the reasons for sentence, the judge sentenced the defendant on the basis that he believed that the drug was amphetamine. On review of sentence, the defendant submitted that the Court of Appeal should not go behind the conclusion of the judge that he should be sentenced on the basis that he believed the drug to be amphetamine. R v Highfield was cited in support. 38. In rejecting the submission, Latham LJ said : “ Further, it seems to us, having regard to the history of the matter, that we can and should take the view that the judge fell into error in sentencing the offender McFadden on the basis that he did, namely believing the tablets to be amphetamine. The history of the proceedings that we have related makes it quite plain that although that was a matter which was put forward by the offender when he first changed his plea, it was not pursued by him at the date when the judge sentenced him and was no part of his mitigation. In those circumstances it is abundantly clear that the judge simply made a mistake as to the basis upon which he fell to be sentenced. That is quite a different situation from the envisaged and dealt with in the case of R v Highfield. We do not consider that we are precluded from dealing with this offender on the basis upon which he should have been dealt with, namely that he did not have that basis of plea as any mitigation of his sentence.” We will return to R v Highfield shortly. 39. In another English case of Attorney General’s Reference Nos 4 and 7 of 2002, supra, the English Court of Appeal considered that the sentence under review was unduly lenient. Woolf LCJ said : “ 33. A question arose during argument as to whether it was inappropriate in this situation to take into account the fact that the judge who was responsible for sentencing was not aware of the extent to which the offences were committed on bail. This was a case where the sentence, even without that factor, was undoubtedly unduly lenient. Mr Pownall, on behalf of the Attorney General, said that in order to achieve fairness we should not look at matters of which the judge who passed the unduly lenient sentence was not aware. We disagree. We consider that it would be unfortunate, once we decided to intervene, if we were to deal with an offender on other than the actual facts. That is what we propose to do.” In other words, Woolf LCJ held that once the Court of Appeal concluded that the sentence under review was manifestly inadequate, it was entitled to pass sentence on the actual facts, although those facts were not before the sentencing court. E1.2 The proposition in [125(3)] 40. The proposition in [125(3)] is a corollary of the proposition in [125(2)]. Further, according to first principle, a sentencing court must take into account all relevant circumstances before passing the sentence. If the sentencing court overlooks certain relevant matters which had been proved, admitted or which had been adduced and not in dispute, the sentence passed is liable to be reviewed because of the failure to take into account all the relevant circumstances rendering it wrong as a matter of law and principle. We have also explained why in considering those facts which the sentencing court had wrongly ignored, the Court of Appeal is still acting within the permissible bounds. E2. The authorities relied on by the respondents 41. The 2nd and 3rd respondents principally rely on Li Ah-sang, R v Highfield, and the passages in Archbold and Taylor. Properly understood, they do not detract from the general propositions we have stated. 42. The point that Mr Pang and Mr Choy seek to derive from Li Ah-sang is effectively the same as advocated by the respondents in Leung Hiu Yeung, namely, that in reviewing a sentence, the Court of Appeal is always bound by the factual findings made by the sentencing court. For the same reasons that we have stated in Leung Hiu Yeung, at [126] and [127], this proposition is based on a misunderstanding of effect of Li Ah-sang. It is erroneous and does not accurately represent the approach that the Court of Appeal would adopt in dealing with the factual basis for reviewing a sentence. It must be rejected. 43. In R v Highfield, supra, the defendant was charged with an offence of causing death by dangerous driving. After trial had commenced, she pleaded guilty and was sentenced to a community service order of 240 hours and disqualified from driving for 5 years. At the trial, when the prosecution opened the case, nothing was said about either alcohol or prolonged bad driving. Rather, counsel said there was no information to show why the defendant drove onto the wrong side as she did. The judge concluded that there was an isolated or momentary error with the most tragic consequences and he identified as the main aggravating feature that the defendant did not stop and help after the accident. The Attorney General applied for a review of sentence on the ground that the sentence was manifestly inadequate and highlighted the aggravating features of drinking alcohol before driving and the defendant having driven badly for a significant distance. In rejecting the application, Judge LJ observed that it could not constitute itself as a court of first instance inquiring into facts which had not been pursued or proved in the Crown Court. 44. Judge LJ’s observation was later explained by the English Court of Appeal in Attorney General’s References Nos 114-116 of 2002 and Nos 144-5 of 2002 [2003] EWCA Crim 3374. There, the sentencing judge erred as a matter of law in making no confiscating order on the purported ground of want of jurisdiction. Had he not erred, he would have to hear evidence to decide if a confiscation order should be made. On review, the offender argued that it is the normal practice that all factual matters necessary for the purpose of deciding an Attorney General’s reference are set out in the application for leave and the Court of Appeal had normally refused to entertain any issue of fact on the hearing of a reference. R v Highfield and Judge LJ’s observation was cited. Rose VP at [28] said : “ …What Judge LJ was correctly observing was that the Attorney General could not seek to re-open the way in which the case had been put by the Crown in the court below to require this court to enquire into facts which, whatever the explanation for the decision made by the Crown at that stage, had not been pursued. In this case, the facts giving rise to the reference are not in dispute: the Crown has always wanted to pursue confiscation and has only been deprived of the opportunity of so doing by an error of law in the sentencing process.” 45. Thus understood, what the English Court of Appeal had decided in R v Highfield is effectively the same as what our Court of Appeal did in Li Ah-sang as we have explained in Leung Hiu Yeung, supra, at [126]. That is to say, when in the proceedings below, the prosecution chose to adopt a particular stance which was accepted by the sentencing court as the factual basis for sentencing, the Court of Appeal will not allow the prosecution, in an application to review the sentence, to pursue a different factual basis not pursued below. It is so even if the stance adopted by the court below was erroneous when the prosecution had the opportunity to have it corrected but had not done so. In such circumstances, the Court of Appeal will not permit the prosecution to re-open the matter and will simply proceed on the same factual basis as adopted by the sentencing court. 46. As noted at [20(1)] above, Archbold, §7-445 cited R v Highfield in support. When R v Highfield is properly understood (as explained above), it does not lend any further support to the respondents. 47. The passage in Taylor, §13.47 consists of two statements : (1) “The Court must review any sentence referred to it on the basis of the facts as they were before the sentencing judge. It is not open to the Attorney-General to invite the Court to sentence on a different factual basis, even where the sentencing judge has proceeded on a wrong basis if the prosecution could have but failed to correct it at the time.” (2) “Nor is it the function of the Court to substitute their own view of the facts in light of new material.” (1) In its proper context, the first sentence and the second sentence in the first statement should be read together. So read, the effect of the first statement is no different from R v Highfield properly understood. Even if the first sentence in the first statement is to be read independently, it must be qualified by the English cases discussed above. It cannot be cited as a free-standing proposition that in every scenario, the Court of Appeal must review any sentence referred to it on the basis of the facts as they were before the sentencing judge. (2) The second statement is irrelevant for present purpose as we have not adduced any new evidence or considered any new material. 48. Finally, the authorities cited by Mr Choy on the general position that an appellate court would adopt on findings of facts by the lower courts are decisions made in different contexts other than a review of sentence ([23] above). They do not address the question of how the Court of Appeal would approach the factual basis for reviewing a sentence. E3. Summary of the applicable propositions 49. To summarize, in our view, the authorities discussed above firmly establish the following propositions : (1) On an application for review of sentence, the Secretary for Justice is not entitled to change the factual basis for sentence that it had relied on at trial and ask the Court of Appeal to inquire into facts which had not been pursued below. Even if the sentencing court had proceeded on a wrong factual basis for sentence, it is not open to the Secretary for Justice to make such an assertion before the Court of Appeal if the prosecution had the opportunity to have it corrected at trial but they had not sought to do so. The Court of Appeal will adopt the same factual basis as it was before the sentencing court. (2) Subject to (1), where the ground for review is that the sentencing court acted on an erroneous factual basis, the Court of Appeal is entitled to examine the evidence adduced below to consider if, based on the facts proved, admitted or not in dispute, the sentencing court did make the error as asserted by the Secretary for Justice. If so, the Court of Appeal is entitled to correct the factual error and to consider if, based on the factual basis as corrected, the sentence imposed is wrong in principle or manifestly inadequate. If so, the Court of Appeal is entitled to interfere. (3) As a corollary of the proposition in (2) and according to first principle, where it is shown that, based on the facts proved, admitted or not in dispute, the sentencing court has failed to take into account certain matters which are relevant to sentence, the sentencing court has proceeded on an incomplete factual basis, rendering the sentence imposed wrong as a matter of law and principle. In such circumstances, the Court of Appeal is not bound by the findings made by the sentencing court for the purpose of sentence because those findings are incomplete for such purpose. The Court of Appeal is entitled to consider all the relevant matters including those wrongly ignored by the sentencing court to determine if in the overall circumstances of the case, the sentence imposed is wrong in principle or manifestly inadequate. If so, the Court of Appeal is entitled to interfere. 50. We reject any submissions of Mr Pang and Mr Choy which are contrary to the above propositions. We would also add that if the Court of Appeal does not have the power to correct the factual basis for reviewing a sentence under propositions (2) and (3) above, the absurdity as identified by Mr Leung in his submission would arise (see [26] above). That simply cannot be right. E4. Had we made any impermissible findings of fact? 51. Mr Pang submitted, without explanation, that we had increased the sentence of the 2nd respondent on facts which were not made at the trial or contrary to the findings made by the Magistrate. Likewise, Mr Choy submitted that we had made findings adverse to the 3rd respondent different from or in addition to those made by the Magistrate without elaboration. 52. In Part I of the Judgment, we detailed the reasons why based on the evidence, admitted or not in dispute, the present offences were serious, why it was a large-scale unlawful assembly involving violence; why the Magistrate had erred in failing to have proper regard and/or give due weight to various factors relevant to sentence as identified by us; and why she had made the errors as discussed. For the reasons we articulated and based on the applicable principles, we held that we must interfere and, proceeding on the correct and complete factual basis, set aside the sentences imposed by the Magistrate and substituted them with immediate custodial sentences as stated. Neither Mr Pang nor Mr Choy has been able to show an arguable case that we have erred in how we dealt with the matter in such manner. F. CONCLUSION 53. For the above reasons, we consider that the answers to both R2’s Question and R3’s Question are well settled by authorities. They do not give rise to any question of law of great and general importance that warrants leave to be given to the 2nd and 3rd respondents to appeal to the Court of Final Appeal : see Chim Pui Chung v HKSAR [1999] 1 HKLRD 836. In any event, the 2nd and 3rd respondents have failed to demonstrate an arguable case that we had erred in the way as they had contended. 54. We accordingly refuse to certify R2’s Question and R3’s Question as one of great and general public importance. The notices of motion filed by the 2nd and 3rd respondents are all dismissed. Mr David Leung, SC, DDPP and Miss Annie Li, PP, of the Department of Justice, for the Applicant Mr Robert Pang, SC, Mr Michael Chai and Ms Priscilla Chow, instructed by M/s Ho, Tse, Wai & Partners, for the 2nd Respondent Mr Edwin Choy and Mr Joe Chan, instructed by M/s Ho, Tse, Wai & Partners, for the 3rd Respondent [1] Cap 221. [2] Cap 484. [3] The 1strespondent did so on 11 September 2017; the 2ndand 3rd respondents, 14 September 2017. [4] Save and except PW11, whose evidence the trial magistrate rejected. [5] As noted at [166], Mr Shek, counsel for the 1st respondent, confirmed that insofar as the 1strespondent was concerned, there was no need to consider other sentencing options under section 109A of the CPO should we decide to allow the application for review against sentence. [6] The quoted paragraphs are taken from the English translation of the Judgment. [7] Cap 212. [8] That is section 83V(5) of the CPO, which provides that in no case shall any sentence be increased by reason of or in consideration of any evidence which was not given at the trial. 
Chief Justice Ma: A. INTRODUCTION 1. The present appeal requires the Court to determine the constitutionality of a legislative provision which affects the right to stand for election (also called the right to participate in public life) contained in Article 26 of the Basic Law and Article 21 of the Hong Kong Bill of Rights. The relevant legislative provision is s 39(2A) of the Legislative Council Ordinance (“the LCO”).[1] Broadly speaking, to address the issue before us, the particular point of principle for consideration can be distilled into the following: where considerable political debate and controversy have given rise to a measure (in the present case, a statutory provision) and that measure is said to be unconstitutional, what is the correct legal approach in dealing with the issue of constitutionality and what limits should the court observe to avoid being drawn into a political debate? The determination of constitutionality is of course at all times a legal question for the courts to answer, having regard to the importance of constitutional rights. A.1 The relevant statutory provision and the constitutional provisions 2. The LCO deals with numerous aspects in relation to the Legislative Council, Hong Kong’s legislature under the Basic Law, chief among them the facet of elections to that Council. Part 6 of the Ordinance deals with the conduct of elections and within that (in s 39) there are provisions setting out disqualifications from nomination as a candidate or from being elected as a member of the Legislative Council. Among the disqualification provisions is the one under challenge in relation to by‑elections, s 39(2A):- “(2A) A person is also disqualified from being nominated as a candidate at a by‑election if— (a) within the 6 months ending on the date of the by-election— (i) the person’s resignation under section 14 as a Member took effect; or (ii) the person was taken under section 13(3) to have resigned from office as a Member; and (b) no general election was held after the relevant notice of resignation or notice of non-acceptance took effect.” The effect of this provision is to prevent a member of the Legislative Council who has resigned from standing for election at the by‑election consequent on that member’s resignation. The 6‑month period is intended to cover that by‑election. As the provision makes clear, however, this bar does not apply in the case of a general election. It is confined to by‑elections. 3. The challenge is based on Article 26 of the Basic Law and Article 21 of the Bill of Rights:- (1) Article 26 of the Basic Law states:- “Permanent residents of the Hong Kong Special Administrative Region shall have the right to vote and the right to stand for election in accordance with law.” (2) Article 21 of the Bill of Rights states:- “Right to participate in public life Every permanent resident shall have the right and the opportunity, without any of the distinctions mentioned in article 1(1) and without unreasonable restrictions— (a) to take part in the conduct of public affairs, directly or through freely chosen representatives; (b) to vote and to be elected at genuine periodic elections which shall be by universal and equal suffrage and shall be held by secret ballot, guaranteeing the free expression of the will of the electors; (c) to have access, on general terms of equality, to public service in Hong Kong.” A.2 The parties’ positions in brief 4. The parties’ respective positions can conveniently be summarised at this point:- (1) The applicant (in the judicial review proceedings leading to the appeal in this Court)[2] submits that the right to stand for election is an important constitutional right not only for persons wishing to stand for election but also for the electorate. Voters ought to be given the widest possible choice of candidate. Accordingly, the importance of the right should not be lightly interfered with. In the present instance, s 39(2A) goes too far and does not satisfy the proportionality test.[3] (2) The respondent[4] contends on the other hand that all aspects of the proportionality test are satisfied. The provision in question satisfies all four facets of the test. In particular, it is submitted that s 39(2A) was the result of extensive public consultation, and also considerable discussion by the Government and by the Legislative Council itself before it was enacted. The provision had political origins and was the product of a political judgment and assessment. Accordingly, the courts should accord what has been called a wide margin of appreciation to the Legislature in the application of the legal principles in the present case. A.3 Background facts 5. It is now convenient to go into the facts in some detail in order to underline the political origins of s 39(2A). 6. In July 2009, a political party (The League of Social Democrats) (“the LSD”) raised the proposal that one member from each of the five geographical constituencies within the Legislative Council[5] should resign in order to trigger by‑elections in such constituencies in which the five resigning members would stand. The intention of the LSD was a political one: the by‑elections would be a de facto referendum in relation to the political manifesto of the LSD pressing for universal suffrage and the abolition of functional constituencies. 7. There was much political controversy caused by this proposal but the LSD and another political party, the Civic Party, adopted it. 8. On 9 December 2009, there was defeated in the Legislative Council a motion that all members returned in the five geographical constituencies should resign to promote the said referendum. However, on 25 January 2010, one member from each of the five constituencies[6] did resign with effect from 29 January. The result of these five resignations was to trigger by‑elections in the relevant constituencies.[7] 9. These took place on 16 May 2010. The five resigning members all took part and were re‑elected. Voter turnout was, however, low[8] and other major political parties did not take part in the elections. The cost of the by‑elections was about $126 million. 10. What followed was intense debate and discussion within the community as to the appropriateness of members resigning from the Legislative Council in order to trigger by‑elections in which the resigning members intended to stand. In June 2011, the Government introduced a Bill proposing that where a member of the Legislative Council resigned, his or her place would be filled by reference to the candidate with the largest number of votes who was not elected in the previous general election. However, after a number of meetings of the Bills Committee of the Legislative Council and after receiving outside submissions, this Bill was not pursued. Instead, the Government decided to consult the public to assess the extent of the perceived problem. 11. On 22 July 2011, a Consultation Paper on Arrangements for Filling Vacancies in the Legislative Council was issued by the respondent seeking the public’s views on what was termed a “mischief”.[9] It was said that there was considerable concern in the community over the incident of the five resigning members. The Government regarded it as an “abuse” for a member to resign in order to cause a by‑election in which that member intended to stand. Two extracts from the Consultation Paper make the point:- “1.04 The Administration considers that it is an abuse of process for a Member to resign in order to trigger a by-election in which the Member intends to stand and seek re‑election. This mischief needs to be addressed. In this regard, the Administration is concerned with the adverse impact such resignations and seeking to be re-elected through by-elections will have on the electoral system and on the public. 1.05 Apart from the manpower and financial resources involved, between a Member’s resignation and the by-election, the LegCo will be deprived of the service of a Member, and the constituents will be deprived of the service of the Member as their representative. Moreover, if the phenomenon of a Member resigning in order to trigger a by-election in which he or she seeks to stand becomes a common occurrence (and there have in fact been threats by some to repeat the resignation and by-election exercise), not only will the operation of the LegCo be adversely affected, the integrity of the Legislature will also be undermined and respect for the electoral process lowered.” 12. Against this was the contrary view that there was nothing inappropriate in such resignations being a proper way to convey a political message.[10] 13. Four options as to possible changes were canvassed in the Paper to deal with the perceived mischief, together with the identification of pros and cons for each option. It is not necessary to go into these options, save to refer only to the one that eventually found favour. This involved restricting the resigning member from participating in any by‑election within the same Legislative Council term. 14. The consultation period was two months. In addition to the Paper, the Government organized forums to engage the public. The results of the consultation were published in a report.[11] There were as one would expect vastly divergent views, but the consultation seemed to suggest that the majority of respondents favoured some action to be taken to “plug the loophole”.[12] 15. The report concluded with a proposal that any member who resigned from the Legislative Council would be prohibited from taking part in any by‑election within six months of resignation. This was said to be a more focused way of addressing the problem. The element of deterrence was very much behind this proposal. It was said:[13] “However, we consider that the proposal would transmit a strong signal indicating the community’s disapproval of such kind of resignation and could have considerable effect in preventing Members from abusing the system.” 16. Following the Report, on 3 February 2012, the Government introduced the Legislative Council (Amendment) Bill 2012. This Bill introduced what is now s 39(2A) of the LCO. A Bills Committee was set up to examine the Bill. A Paper was produced in February 2012 providing the Government’s view on the legality of the Bill containing the legal opinion of external legal counsel.[14] In this Paper were constant references to the justification of the amendment being to deter members from resigning in order to trigger a by‑election in which he or she intended to stand.[15] In the Legislative Council Brief dated 1 February 2012 from the respondent,[16] it was said that following the public consultation, there was “strong public support for the Government to address the mischief.”[17] The Brief also mentioned the need to “deter abusive conduct”.[18] 17. The political debate in the Legislative Council was heated. One indication of this can be seen in the legal proceedings that were sought to be instituted by one of the five members of the Legislative Council who had resigned in January 2010. On 17 May 2012, Mr Leung Kwok Hung sought leave to apply for judicial review against the President of the Legislative Council. The President had curtailed proceedings in the Council on the amendment Bill after debate at the second reading had taken over 33½ hours with over 1,300 amendments suggested by two legislators (another two of the five resigning members) and after the President had ruled 75 times when Mr Leung spoke on the matter that his speech was irrelevant or repetitious. The application for leave was dismissed.[19] The tactics of these legislators were described by the Judge as “filibustering”. In the judgment of the Court of Final Appeal, it was said:[20] “The avowed intention of the two legislators and their ally, the Appellant (also a legislator), for the introduction of these numerous amendments was to filibuster the Bill, which they opposed and which they apprehended would otherwise be passed by the majority in the Council.” 18. The Bill was eventually passed in the Legislative Council on 1 June 2012 and s 39(2A) came into force. A.4 The proceedings below 19. Consequent on the passing of s 39(2A), the applicant applied for judicial review on the basis that it was unconstitutional. The application was dismissed by Au J[21] primarily on the basis that although constitutional rights were engaged, the provisions in question nevertheless satisfied the proportionality test. In particular, the Judge accorded a wide margin of appreciation to the Legislative Council. 20. The applicant appealed to the Court of Appeal which dismissed the appeal.[22] Similar to Au J, the Court of Appeal accorded a broad margin of appreciation. Much weight was given to the fact that the Court was, in the present constitutional challenge, dealing with matters of political judgment. As Cheung CJHC said:[23] “Generally speaking, the Court is neither constitutionally positioned nor institutionally equipped to deal with a political issue, that is, an issue essentially involving political rather than legal judgment.” A.5 The certified question before this Court 21. On 29 September 2016, leave to appeal was given by the Appeal Committee[24] to appeal to the Court of Final Appeal[25] on the following question of great general or public importance:- “Is section 39(2A) of the Legislative Council Ordinance, Cap. 542 inconsistent with Article 26 of the Basic Law of the Hong Kong Special Administrative Region and/or Article 21 of the Hong Kong Bill of Rights, and therefore unconstitutional?” B. THE CONTEXT OF ELECTIONS TO THE LEGISLATIVE COUNCIL UNDER THE BASIC LAW AND THE LEGISLATIVE COUNCIL ORDINANCE 22. The wording of Article 26 of the Basic Law and Article 21 of the Bill of Rights have been set out.[26] Notwithstanding the absence of express qualifications to the right set out in Article 26 of the Basic Law, it is clear that this article must be read together with Article 21 of the Bill of Rights which does contain qualifications.[27] It is accepted that the right to stand for election is not an absolute right. It is also accepted that the words “without unreasonable restrictions” in Article 21 of the Bill of Rights require the application of the proportionality test.[28] This was the way Cheung J (Cheung CJHC as he then was) analysed the words in Chan Kin Sum[29] and this is consistent with textbook authority.[30] 23. The following points in relation to Article 21 of the Bill of Rights are also of note:- (1) Article 21(b) refers to both the right to vote and the right to be elected as guaranteeing the free expression of the will of the electors. This reinforces the point made by the applicant that the right to stand for election is a right directly linked to the interest of the electorate being given the widest choice of candidate and for this reason, the right ought not to be unduly restricted. I agree with this general statement. (2) The Article 21 right refers specifically to “genuine periodic elections” and although it is clear that the right to stand for election extends as well to by‑elections, nevertheless these words suggest the somewhat exceptional nature of by‑elections in the context of periodic elections. Article 68 of the Basic Law states that the Legislative Council shall be constituted by elections. Article 69 of the Basic Law states the term of office of members of the Legislative Council to be four years – this reflects the periodic cycle for elections in Hong Kong. This is also reflected in the LCO which states that the term of office of the Legislative Council is as prescribed in Chapter IV of the Basic Law.[31] 24. The LCO sets out detailed provisions regarding elections to the Legislative Council. In the specific context of by‑elections, the following provisions are of note:- (1) Where a member of the Legislative Council resigns, whether expressly or impliedly,[32] a vacancy arises[33] and once a declaration to this effect is made,[34] a by‑election must be held.[35] (2) Any member who has resigned is eligible for re‑election as a member but this is subject to the disqualification provisions contained in s 39 of the LCO.[36] Prior to the enactment of s 39(2A), a resigning member could, as the five resigning members did in January 2010, stand for re‑election at the ensuing by‑election. The only disqualifications were those set out in the unamended s 39.[37] (3) Section 39(2A), enacted in June 2012, amended s 39 by adding to the disqualification categories. 25. The powers and functions of members of the Legislative Council are set out in Articles 73 and 74 of the Basic Law. It is unnecessary to set these out in full; it is sufficient merely to observe that there are important public duties to be discharged by members of the Legislative Council (the Legislature of the HKSAR as it is referred to in the Basic Law). 26. Lastly in the present context, I draw attention to Article 79(2) of the Basic Law which states as follows:- “The President of the Legislative Council of the Hong Kong Special Administrative Region shall declare that a member of the Council is no longer qualified for the office under any of the following circumstances: …… (2) When he or she, with no valid reason, is absent from meetings for three consecutive months without the consent of the President of the Legislative Council;” 27. The relevance of this provision will be seen below in the context of reasons which a member of the Legislative Council may have for resigning.[38] 28. Having set out the context of elections to the Legislative Council and the position of members of the Council under the Basic Law and the LCO, I now deal with the relevant approach to constitutional challenges on provisions such as s 39(2A). C. CONSTITUTIONAL CHALLENGES – THE LEGAL APPROACH C.1 The general approach 29. The determination of constitutionality is, as is well established now, a question of law for the courts to determine. The general approach in cases involving challenges to legislation or other measures said to contravene constitutionally guaranteed rights is set out in the recent decision of this Court in Hysan Development Co. Ltd. v Town Planning Board:-[39] “43. This Court has recognised that certain constitutionally guaranteed rights, such as the prohibition against torture and cruel, inhuman or degrading treatment or punishment, are absolute and that in such cases, there is no room for any proportionality analysis. 44. Where the guaranteed right is not absolute, the law may validly create restrictions limiting such rights. It is for the Court to determine the permissible extent of those restrictions and it does so by a process referred to as a proportionality analysis.” 30. The proportionality analysis referred to in this passage of course does not arise unless three prior steps are satisfied by the person[40] asserting unconstitutionality: the identification of a constitutionally guaranteed right,[41] the identification of the relevant legislation or measure said to infringe such constitutional right and the infringement itself. 31. Once these three initial steps are satisfied, the next step in the analysis is to look at the constitutional right itself to see whether there are any in‑built qualifications. Where the right is contained in the Basic Law, there may be some qualifications that are expressly stipulated.[42] Qualifications to rights also appear in the Bill of Rights. 32. In the present case, the relevant right is the right to stand for election and this right has been infringed by the restriction contained in s 39(2A) of the LCO. The relevant constitutional right is contained in Article 26 of the Basic Law and Article 21 of the Bill of Rights. The former provision contains no express qualification, while the latter does. In any event, the proper analysis to be adopted is the proportionality test.[43] C.2 Proportionality 33. The proportionality analysis is not expressly set out in constitutional or statute form as such, but has been developed by the courts drawing on the jurisprudence of other jurisdictions.[44] 34. This Court having recently undertaken a study of the proportionality analysis in Hysan, it is not necessary to repeat the exercise in this judgment. It suffices merely to draw attention to some facets of it that have relevance in the present appeal. 35. The proportionality analysis involves four steps for the Court to determine. These were set out in Hysan as follows:-[45] “134. In Hong Kong, such a proportionality assessment has been viewed as involving a three-step process of asking: (i) whether the intrusive measure pursues a legitimate aim; (ii) if so, whether it is rationally connected with advancing that aim; and (iii) whether the measure is no more than necessary for that purpose. 135. A fourth step should be added. In line with a substantial body of authority, where an encroaching measure has passed the three-step test, the analysis should incorporate a fourth step asking whether a reasonable balance has been struck between the societal benefits of the encroachment and the inroads made into the constitutionally protected rights of the individual, asking in particular whether pursuit of the societal interest results in an unacceptably harsh burden on the individual.” C.3 A closer consideration of margin of appreciation 36. I ought to observe here that in the majority of constitutional challenges, the first two steps will provide little controversy[46] and would be relatively simple to apply. It is the third step, in its reference to “no more than necessary” that has generated the need for clarification by the courts, particularly when viewed against the concept of margin of appreciation. An extensive analysis of the third step was undertaken in Hysan and I would respectfully draw attention to that.[47] A summary of the relevant principle can be stated as follows:-[48] “136. At the third stage, assessing the permissible extent of the incursion into the protected right, two main standards have been applied. The first is the test of whether the intruding measure is “no more than necessary” to achieve the legitimate aim in question. This must be understood to be a test of reasonable necessity. If the Court is satisfied that a significantly less intrusive and equally effective measure is available, the impugned measure may be disallowed. 137. An alternative standard which may be applied at the third stage is one which asks whether the encroaching measure is “manifestly without reasonable foundation”, being a standard closely related to the concept of “margin of appreciation” in ECtHR [European Court of Human Rights] jurisprudence. 138. At the supra-national level of the ECtHR, the margin of appreciation doctrine involves the recognition that on certain issues, the Court should allow Member States latitude to decide on the legitimacy of their societal aims and the means to achieve them since they are better placed to make the assessment. Similar considerations have led the Court at a domestic level to allow the legislative and executive authorities latitude or a “margin of discretion” to do the same, applying the “manifestly without reasonable foundation” standard in such cases. 139. The “manifest” standard has been used in cases where the Court recognises that the originator of the impugned measure is better placed to assess the appropriate means to advance the legitimate aim espoused. This has occurred in cases involving implementation of the legislature’s or executive’s political, social or economic policies but the principle is not confined to such cases. 140. The location of the standard in the spectrum of reasonableness depends on many factors relating principally to the significance and degree of interference with the right; the identity of the decision-maker; and the nature and features of the encroaching measure relevant to setting the margin of discretion. 141. The difference between the two standards is one of degree, with the Court in both cases, scrutinising the circumstances of the case and the factual bases claimed for the incursion.” 37. As can be seen therefore, the difference in approach of the courts at the third stage varies depending on the particular circumstances of any given case and this is critical to bear in mind when looking at the impugned measure to see whether (a) the stricter test of the measure being “no more than necessary” to deal with its legitimate aim; or (b) the test of the measure merely being “manifestly without reasonable foundation”, ought to be applied. One should not of course be pre‑occupied with labels and instead adopt a flexible, though principled and structured, approach. Paragraphs 140 to 141 in the passage quoted in the previous paragraph are instructive here. 38. Though a matter of degree,[49] there are three aspects to consider:-[50] (1) The nature of the right in question and the degree to which it has been encroached on. (2) The identification of the relevant decision‑maker (in the case of legislation, this will be the Legislature). (3) Relevance of the margin of appreciation. 39. I have already referred to the position of absolute rights.[51] The treatment of other, non‑absolute rights varies depending on the context. Accordingly, where socio‑economic or general policies are involved, the Court’s consideration will be quite different to the position where core‑values – and a fortiori, absolute rights – are involved.[52] Although I have used the term “stricter test”,[53] this is slightly misleading if what is conveyed is that the courts will somehow take a laxer approach to rights depending on circumstances. The matter was put in the following way in Fok Chun Wa:-[54] “81. It is sometimes said that in cases where core-rights or fundamental concepts are concerned, the courts will adopt an “intense” or “more intense” level of scrutiny. These terms, like the terms “margin of appreciation” or “deference”, are used for convenience only and not to be taken literally. They are used to convey the principles identified earlier in this section of the Judgment. There is of course, no question of a court taking a laxer or less vigilant approach whenever any questions of constitutionality arise. Each case is of course approached seriously, only that the legal approach will inevitably differ depending on the circumstances of the case. Where core values or fundamental concepts are involved, these are areas where the courts have (for want of better terms) expertise and experience, and it is part of their constitutional duty to protect these values or concepts. In policy matters not involving these matters, the courts do not have this expertise or experience and, more important, it is not within its constitutional remit to determine matters of government or legislative policy, save where questions of legality arise. As Sir Anthony Mason NPJ put it in HKSAR v Lam Kwong Wai, 601E [45] “the weight to be accorded to the legislative [and I would add, the executive and other authorities’] judgment by the court will vary from case to case depending upon the nature of the problem, whether the executive and the legislature are better equipped than the courts to understand its ramification and the means of dealing with it.” 40. The term margin of appreciation refers to that area of discretion which the Court will accord to a decision‑maker, or, in the case of legislation, to the legislature. It reflects the separate constitutional and institutional responsibilities of the judiciary and other organs of government.[55] The concept of margin of appreciation was clearly articulated in Hysan:-[56] “116. On the other hand, a decision-maker’s views resulting in the promulgation of the impugned measure may be given much weight and thus afforded a wide margin of discretion reflected by use of a “manifest” standard where the decision-maker is likely to be better placed than the Court to assess what is needed in the public interest. The Court may for instance, be satisfied that he had special access to information; special expertise in its assessment; or an overview enabling him to assess competing and possibly prior claims for scarce resources. The Court might also refrain from intervening because the measure reflects a predictive or judgmental decision which it was the institutional role of the decision-maker to take and as to which no single “right answer” exists.” 41. The “predictive or judgmental decision” referred to in this passage is a facet which assumes some importance in the present case. In the passage set out in the previous paragraph, reference was made to a passage in the judgment of Lord Sumption JSC in R (Lord Carlile of Berriew and others) v Secretary of State for the Home Department.[57] Fok Chun Wa was a case involving socio‑economic policies of the Government and a wide margin of appreciation was given by the Court in that case. Similarly, matters of national security, defence and foreign policy are also matters where much leeway will be permitted by the Court.[58] 42. Lord Pannick QC argues that political decisions or legislative provisions reflecting political judgments are often precisely those areas where the courts are likely to afford a large margin of appreciation. I agree. Where electoral laws involve political or policy considerations, a wider margin of appreciation ought generally to be accorded.[59] The authorities from the United Kingdom[60] and the European Court of Human Rights[61] are consistent with this approach when politics and political judgments are involved. In particular, where there has been active political debate on an issue or piece of legislation, the Court will again be inclined to give a wider margin of appreciation.[62] The reason for this approach is evident: the courts are generally not equipped (certainly not better equipped than others) to determine political questions, although of course there are limits. The two tests earlier identified, for example, lay down obvious limits. 43. Mr Pun SC argues to the contrary and submitted that where a fundamental right such as the right to vote is concerned, no margin of appreciation should be accorded to the Legislature. He relies on the decision of the Supreme Court of Canada in Richard Sauvé v The Attorney General of Canada and Others.[63] The case involved the consideration of a statute[64] completely denying the right to vote to all persons serving sentences of imprisonment of two years or more. This was said to be unconstitutional.[65] The particular passage relied on is contained in the judgment of Chief Justice McLachlin (in a 5:4 decision of the Court):- “The core democratic rights of Canadians do not fall within a “range of acceptable alternatives” among which Parliament may pick and choose at its discretion. Deference may be appropriate on a decision involving competing social and political policies. It is not appropriate, however, on a decision to limit fundamental rights. This case is not merely a competition between competing social philosophies. It represents a conflict between the right of citizens to vote – one of the most fundamental rights guaranteed by the Charter – and Parliament’s denial of that right. Public debate on an issue does not transform it into a matter of “social philosophy”, shielding it from full judicial scrutiny. It is for the courts, unaffected by the shifting winds of public opinion and electoral interests, to safeguard the right to vote guaranteed by s.3 of the Charter.”[66] 44. This passage must of course to be seen in context. As mentioned above, that case dealt with a situation where the right to vote was entirely lost to a sizeable class whereas the present case (dealing with right to stand for election) affects a relatively small class. In Sauvé, the majority regarded as significant the fact that the Government had failed to identify particular problems to justify the denial of the right to vote. The reasons that were given were said to be “vague”.[67] It was perhaps for this reason that the majority (unlike the minority) did not accord any margin of appreciation. Chief Justice McLachlin said this:-[68] “My colleague Justice Gonthier proposes a deferential approach to infringement and justification. He argues that there is no reason to accord special importance to the right to vote, and that we should thus defer to Parliament’s choice among a range of reasonable alternatives. He further argues that in justifying limits on the right to vote under s.1, we owe deference to Parliament because we are dealing with “philosophical, political and social considerations”, because of the abstract and symbolic nature of the government’s stated goals, and because the law at issue represents a step in a dialogue between Parliament and the courts. I must, with respect, demur. The right to vote is fundamental to our democracy and the rule of law and cannot be lightly set aside. Limits on it require not deference, but careful examination. This is not a matter of substituting the Court’s philosophical preference for that of the legislature, but of ensuring that the legislature’s proffered justification is supported by logic and common sense.” This passage places what was said about margin of appreciation (or deference)[69] in proper context. In my view, this passage does not exclude the importance of considering margin of appreciation in cases involving social, economic and political matters, to which reference has already been made. It merely draws attention to the necessity of adopting a more focused and critical view of the concept in the context of an exclusion of the right to vote. 45. It reiterates the approach that any encroachment on constitutionally guaranteed rights must be carefully scrutinized by the Courts. The margin of appreciation to be accorded is but one factor in the overall consideration by the Court of proportionality, albeit depending on the circumstances it may assume a greater or lesser degree of importance. It will usually be determinative in the sliding scale as to whether the Court will veer towards applying the “no more than necessary” approach or the “manifestly without reasonable foundation” one. 46. In either situation of course, a consideration of the extent of the encroachment on the constitutional right will be important. At all times, it will be essential for the Court to keep firmly in mind the value of the right in question. As stated earlier, the question of constitutionality is a legal question for the courts to determine. As was stated by this Court in Mok Charles:-[70] “56. There are, however, obvious limits to the utility of this concept [margin of appreciation]. It is important to acknowledge that while the views of the legislature are to be considered, it is the court that has the ultimate responsibility to determine whether legislation is constitutional. This is a matter of law, only for the courts to determine.” This is the rationale behind the fourth step in the proportionality analysis. C.4 The fourth step 47. The desirability of this step was gone into in Hysan.[71] It requires, in any determination of whether constitutional rights have been infringed, to take an overall, balanced view. Without such a step, “the proportionality assessment would be confined to gauging the incursion in relation to its aim. The balancing of societal and individual interests against each other which lies at the heart of any system for the protection of human rights would not be addressed.”[72] The Court is required to make a value judgment. 48. So how are these principles to be applied in the present case? D. DETERMINATION 49. It is convenient to deal with the determination of the present appeal by reference to the said four step approach in the proportionality analysis. D.1 Legitimate aim 50. In Section A.3 above, I have gone into the background facts of the present case in some detail in order to highlight a number of important aspects:- (1) The context in which the enactment of s 39(2A) of the LCO arose was a controversial, highly political one. (2) The political debate emanating from the resignation of the five members[73] involved the public (by way of consultation and forums), the Government and the Legislative Council. Within the Legislative Council itself, the political debate was heated and led to filibustering by legislators (which in turn led to an application for leave to institute judicial review proceedings against the President of the Council).[74] (3) The enactment of s 39(2A) manifested an attempt to deal with a perceived mischief thought by many members of the community[75] (as well as by the Government and the Legislative Council) to exist, namely, resignation by members of the Legislative Council from the Council in order to trigger by‑elections in which the resigning members would stand. (4) This perceived mischief led to a concern about the adverse impact that such resignations might have on the electoral system. The fear was that apart from the cost of such by‑elections,[76] the public would be deprived of the services of a member of the Legislative Council during the period between the member’s resignation and when the member’s replacement would be able to take up office. As we have seen,[77] a member of the Legislative Council has important constitutional duties to perform. The adverse impact that was feared was that were such resignations to become a common occurrence, the integrity of the Legislature would be undermined and respect for the electoral process lowered. (5) The ultimate aim of the amendment that became s 39(2A) was deterrence in order to avoid the adverse impact identified above. I have already referred to the deterrence aspect.[78] (6) All the above matters were before the Legislative Council when the amendment Bill leading to the enactment of s 39(2A) was debated and considered. Different sides of the argument were before the Council and a balancing exercise had therefore to be undertaken to determine what measures were appropriate to deal with the perceived problem. 51. Lord Pannick QC submits that given this background to s 39(2A), its aim was clear and this aim was a legitimate one. There is much force in these submissions. The identification by the Legislative Council of the perceived mischief was based on the results of the public consultation following the resignation of the five members and also the views of the Government. There was also extensive debate on the matter. The stated aim of the legislation was therefore clear and this was accepted by the Courts below.[79] The legitimacy of the aim does not of course depend on the Court agreeing with it. Indeed, Lord Pannick was at pains to emphasise that the Court does not have to be convinced that the aims, and in particular where the aims are political ones, are politically correct or even that it agrees with these aims from a political standpoint. The responsibility of the Court is to be satisfied from a legal point of view that the aim is first, identifiable and secondly, legitimate in the sense that it lies within constitutional limits. 52. Mr Pun’s answer to the question of legitimate aim was simply to emphasise the importance of the right to stand for election. Mr Pun SC also asserted there was a lack of cogent evidence to support the said aim of the legislation. With respect, this was difficult to accept. The resignation of the five members, followed by the public consultation and the debate within the Legislative Council, all demonstrated that the aim of the legislation was to deter the mischief earlier identified. If what he meant was that there was no concrete evidence to demonstrate that if measures were not taken, this would definitely undermine the electoral system,[80] this misses the point. The point was not that an undermining of the electoral system would definitely take place; this was the mischief that was thought to exist and that was sought to be deterred. 53. In my view, there was clearly a legitimate aim sought to be achieved by the enactment of s 39(2A) of the LCO. D.2 Rational connection of s 39(2A) to the legitimate aim 54. The legitimate aim being as identified above, there is no doubt that this part of the proportionality test is satisfied. The respondent’s position was simply that s 39(2A) sought to make it less likely that a member of the Legislative Council would resign voluntarily in order to provoke a by‑election. I accept this argument and to be fair, Mr Pun did not really contend otherwise. D.3 No more than necessary: the proportionality of the restriction 55. The background facts and their significance[81] were all matters that were before the Legislative Council in determining what measures were appropriate to deal with the perceived problem. No doubt a number of options were available for consideration and the facts set out above refer to a number of options on which the public were consulted.[82] In my view, in these circumstances, a wide margin of appreciation ought to be accorded. The consequence of a wide margin of appreciation given to the Legislative Council means that in the present context, the appropriate test regarding the legislative choice made (that is, s 39(2A)) should be the manifestly without reasonable foundation test. Where, as in the present case, there are involved matters of political judgment or prediction, some leeway should be permitted to the Legislature to determine what would be an appropriate way of dealing with the perceived mischief. It is not appropriate to adopt a strict “no more than necessary” test in the present case. The Court is not in a position nor is it equipped to apply this test in the circumstances of the present case, involving as it does matters of political judgment and assessment. 56. An associated point raised by the applicant was this. It has all along been accepted by the respondent in the consultation exercises and debates within the Legislative Council, as well as in argument before the courts, that there were situations in which it would appear justifiable for a member of the Legislative Council to resign, for example, where a member had changed his or her party allegiance and could not in all good conscience continue as a member. Many may think in this type of situation that it would be unfair to disqualify that member from standing in the resulting by‑election. The point made by Mr Pun drawing on this example was that even if it were accepted that a legitimate aim existed, the line had been drawn far too widely so that every member who resigned, whatever the reason and even if such were (so to speak) a good reason, would be caught by the disqualification provisions. Seen in this way, s 39(2A) represented a disproportionate response to the problem: it went far beyond what was necessary or was at all reasonable or (in the language of Article 21 of the Bill of Rights) it was an unreasonable restriction. It was submitted that the line could be drawn much more precisely so that certain defined situations could be identified. Mr Pun provided two more examples: where a member of the Legislative Council simply resigned for the purposes of provoking a by‑election, he could be excluded; and where a member did not know how he or she should vote on an important issue, resignation should be permitted without any disqualification in a by‑election. 57. I do not accept the argument that the line drawn in the present instance (s 39(2A)) can be faulted:- (1) By reason of the margin of appreciation to be accorded, the line drawn fell within the range of reasonable options open to the Legislature to adopt in order to deal with the perceived mischief. It was within its political judgment or assessment to adopt this option. (2) The line suggested by Mr Pun to take into account those specific situations mentioned above, is itself far from clear. All involve a significant political dimension which would be in practice difficult, if not impossible, to operate. Certainly, the Court should not be put (nor ought it to be placed) in a position where it had to decide on political matters of this kind or have to be drawn into a political debate. As mentioned above, the context of the enactment of s 39(2A) was a controversial, highly political one. (3) Where, as in this case, the Legislature is entitled to draw the line it has, even though this may result in ‘hard’ cases, this does not mean that the line is impermissibly drawn. In R (Animal Defenders International) v Secretary of State for Culture, Media and Sport,[83] Lord Bingham of Cornhill said this:- “A general rule means that a line must be drawn, and it is for Parliament to decide where. The drawing of a line inevitably means that hard cases will arise falling on the wrong side of it, but that should not be held to invalidate the rule if, judged in the round, it is beneficial.” This passage was expressly adopted in Fok Chun Wa.[84] 58. It was at one stage argued by the applicant that other situations where it was unfair to apply the total disqualification in s 39(2A) included where a member of the Legislative Council resigned through poor health. It was said that where the member recovered in time for the by‑election, it would be unfair if he or she were not permitted to stand. Here, apart from the matters set out in the previous paragraph, I would also refer to Article 79(2) of the Basic Law.[85] If a member were to be in the unfortunate position of being ill, he or she could consider seeking the consent of the President of the Council or that member would have in any event a valid reason to be absent. 59. For these reasons, s 39(2A) satisfies the third step of the proportionality analysis. D.4 Has a reasonable balance been struck? 60. This is the fourth step identified in Hysan.[86] 61. There is no doubt, as Mr Pun has impressed on this Court throughout his submissions, that the right to stand for election is an important constitutional right. He reminded the Court that the right to vote has been described as “without doubt the most important political right”.[87] A court must therefore always consider carefully any encroachment on the right. The fourth step requires a court to take an overall view in the proportionality exercise to ensure that any encroachment on a constitutional right is fully justified. In the present case, however, this fourth step is satisfied by the respondent:- (1) The first three steps of the proportionality exercise are satisfied and in most cases, this will point towards the fourth step being satisfied as well.[88] (2) The encroachment on the constitutional right to stand for election is a relatively small one. It only applies to by‑elections (and not general elections) and the bar is solely against the resigning member. As far as he or she is concerned, s 39(2A) cannot be said to bear harshly on the resigning member since, having been elected on a four‑year mandate and perfectly entitled to stay in office as a legislator, he or she has chosen voluntarily to resign with full knowledge of the consequences. Even then, the bar is only for six months. (3) As far as voters in the relevant constituency are concerned, the by‑election is held in any event and their choice of candidate is unrestricted (except for the resigning member). E. CONCLUSION 62. For the above reasons, I would dismiss this appeal. 63. I would also make an order nisi as to costs that the applicant should pay to the respondent the costs of the appeal, to be taxed if not agreed and that the applicant’s own costs be taxed in accordance with the Legal Aid Regulations. If any party wishes to have a different order for costs, written submissions should be served on the other party and lodged with the Court within fourteen days of the handing down of this judgment, with liberty on the other party to lodge written submissions in reply within fourteen days thereafter. In the absence of such written submissions, the order nisi will stand absolute at the expiry of the time limited for these submissions. Mr Justice Ribeiro PJ: 64. I agree with the judgment of the Chief Justice. Mr Justice Tang PJ: 65. I agree with the judgment of the Chief Justice. Mr Justice Fok PJ: 66. I agree with the judgment of the Chief Justice. Mr Justice Robert French NPJ: 67. I agree with the judgment of the Chief Justice. Chief Justice Ma: 68. For the above reasons, the appeal is dismissed. There will also be an order nisi as to costs as set out in para. 63 above. Mr Hectar Pun SC and Mr Newman Lam, instructed by Ho Tse Wai & Partners, assigned by the Director of Legal Aid, for the Appellant Lord Pannick QC, Mr Johnny Mok SC and Mr Jin Pao, instructed by the Department of Justice, for the Respondent [1] Cap. 542. [2] He is the appellant in this appeal. Mr Kwok Cheuk Kin is a registered voter entitled to vote at general elections and by‑elections in Hong Kong. He was represented in these proceedings by Mr Hectar Pun SC and Mr Newman Lam. [3] As to this test, see further below at Section C below. [4] He is the respondent in this appeal. The Secretary for Constitutional and Mainland Affairs was responsible for handling and dealing with the relevant provision in the present case. The Secretary was represented in this appeal by Lord Pannick QC, Mr Johnny Mok SC and Mr Jin Pao. [5] The Legislative Council consists of members returned by geographical constituencies and functional constituencies (35 members each):- see Article 68 of the Basic Law and Annex II thereto (“Method for the Formation of the Legislative Council of the Hong Kong Special Administrative Region and Its Voting Procedures”, as amended in 2010). See also Part 3 of the LCO as to geographical and functional constituencies. [6] Members of the LSD and the Civic Party. [7] Where a member of the Legislative Council resigns, a vacancy arises and the Electoral Affairs Commission must arrange for a by‑election to be held: see ss 14, 15, 35 and 36 of the LCO. [8] It was in fact a record low, only 17.19% (about 580,000 electors). This was about a third of those who had voted at the previous three elections. [9] In this judgment, the terms “mischief” and “abuse” are referred to. These are not intended to be tendentious nor are they used as any form of criticism but are used as they were the terms which appeared in the various documents before us. [10] Paras. 1.03 and 5.03 of the Consultation Paper. [11] The Consultation Report on Arrangements for Filling Vacancies in the Legislative Council dated January 2012. [12] In para. 3.17 of the Report, it was said that 31,120 written submissions were received with the majority (70%) considering that the said mischief was a loophole that needed to be plugged. [13] In para. 6.10. [14] The opinion was provided by Lord Pannick QC. [15] At para. 15. [16] This is a Brief provided to the Legislative Council to explain and justify the need for legislation or amendments to legislation. [17] At para. 8(a). [18] At para. 11(d). [19] Leung Kwok Hung v President of the Legislative Council (Filibuster) [2012] 3 HKLRD 470. This was eventually appealed to this Court: Leung Kwok Hung v President of the Legislative Council (No. 1) (2014) 17 HKCFAR 689. [20] At para. 6. [21] In a judgment dated 5 March 2014. [22] In a judgment dated 22 October 2015 (Cheung CJHC, Lam VP and Poon JA). [23] At para. 6. [24] Ma CJ, Tang and Fok PJJ. [25] Pursuant to s 22(1)(b) of the Hong Kong Court of Final Appeal Ordinance Cap. 484. [26] At para. 3 above. [27] The term “without unreasonable restrictions” is used in that article. Where a right is unqualified in the Basic Law but the corresponding right is qualified in the Bill of Rights, the right is not absolute: see Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229, at paras. 16-21; Chan Kin Sum v Secretary for Justice [2009] 2 HKLRD 166, at paras. 55-62. See also in this context Gurung Kesh Bahadur v Director of Immigration (2002) 5 HKCFAR 480, at paras. 26-28. [28] See Section C below. [29] At paras. 63-78. [30] See Proportionality: Constitutional Rights and their Limitations (Aharon Barak) (2012) (Cambridge University Press) at Pg. 371-378. [31] See s 4(1). Articles 68 and 69 of the Basic Law are both contained in Chapter IV Section 3 of the Basic Law. [32] See ss 13(3), 14 and 15(1)(a) of the LCO. [33] Section 15(1) of the LCO. [34] Under s 35. [35] See s 36(1)(a). [36] See s 16. [37] Such as for judicial officers and prescribed public officers are disqualified as officers of the Legislative Council etc. [38] See Section D.3 below. [39] (2016) 19 HKCFAR 372 at paras. 43-44 (in the judgment of Ribeiro PJ with whose judgment the other members of the Court agreed). [40] Usually the applicant in judicial review proceedings. [41] See Catholic Diocese of Hong Kong v Secretary for Justice (2011) 14 HKCFAR 754 at para. 65. In exceptional cases, however, a proportionality exercise is performed where a constitutional challenge is made against measures said to be inconsistent with the Basic Law even though no individual right is engaged: see Incorporated Owners of Po Hang Building v Sam Woo Marine Works Ltd., FACV 10/2016, at para. 12; Mok Charles v Tam Wai Ho (2010) 13 HKCFAR 762, at paras. 20, 21, 26 and 34. [42] Such as Article 30 of the Basic Law where the freedom and privacy of communication may not be infringed except where the relevant authority may inspect communications “in accordance with legal procedures to meet the needs of public security or of investigation into criminal offences.” [43] See para. 22 above. [44] See Hysan at para. 47. [45] At paras. 134-135. [46] Although in the present case, both steps are disputed. Indeed, as will be seen, the first step (legitimate aim) is crucial. [47] See Section G of the judgment. [48] At paras. 136-141. [49] The phrase “sliding scale” has also been used: see Pham v Secretary of State for the Home Department [2015] 1 WLR 1591, at para. 106 per Lord Sumption JSC (the passage is quoted in Hysan at para. 63). [50] Hysan at para. 140. [51] See para. 29 above setting out para. 43 of Hysan. [52] See Hysan at paras. 108-113; Fok Chun Wa v Hospital Authority (2012) 15 HKCFAR 409, at paras. 76-81. [53] See, for example, para. 37 above. [54] At para. 81. [55] See Hysan at paras. 99 and 118; Fok Chun Wa at para. 64. [56] At para. 116. [57] [2015] AC 945 at para. 34. [58] See Hysan at para. 117. [59] See Leung Chun Ying v Ho Chun Yan Albert (2013) 16 HKCFAR 735 at para. 45. [60] Apart from Lord Carlile (see para. 41 fn 57 above), see also A v Secretary of State for the Home Department [2005] 2 AC 68, at para. 29 and R (Countryside Alliance) v Attorney General [2008] 1 AC 719, at para. 45. [61] Such as Zdanoka v Latvia (2006) 45 EHRR 478, at para. 115(c) (where the political evolution of the country concerned was referred to); Sukhovetskyy v Ukraine (2007) 44 EHRR 1185, at para. 67 (where the “delicate balance between conflicting [political] interests” was referred to). [62] See Huang v Secretary of State for the Home Department [2007] 2 AC 167, at para. 17; see also Sukhovetskyy at para. 65. [63] [2002] 3 SCR 519. [64] Section 51(e) of the Canada Elections Act. [65] As breaching the right to vote (Article 3 of the Canadian Charter of Rights and Freedoms) and the right to equality (Article 15(1) of the Charter). [66] At para. 13. See also para. 15 where the Chief Justice also said, “While a posture of judicial deference to legislative decisions about social policy may be appropriate in some cases, the legislation at issue does not fall into this category.” [67] At paras. 22-26. [68] At paras. 8 and 9. [69] The term “deference”, intended to be the same concept as margin of appreciation, is not, however, an appropriate term to use: see Hysan at para. 99; see also Mok Charles at para. 79. [70] At para. 56. [71] See Sections F.2 and F.3 of the judgment. [72] See Hysan at para. 78. See also para. 79. [73] See para. 8 above. [74] See para. 17 above. [75] This was said to be “strong”: see para. 16 above. [76] It will be recalled that the May 2010 by‑elections involved expenditure of about $126 m: see para. 9 above. [77] See para. 25 above. [78] See paras. 15 and 16 above. [79] See paras. 57-63 of the judgment of Au J; para. 52 of the judgment of the Court of Appeal. [80] See para. 50(4) above. [81] See Section A.3 and para. 50 above. [82] See para. 13 above. Whether or not the other options were feasible or acceptable is not necessary to determine. [83] [2008] 1 AC 1312, at para. 33. [84] At para. 71. [85] See paras. 26 and 27 above. [86] See Section C.4 above. [87] Chan Kin Sum at para. 164 (para. 22 fn 27 above). [88] See Hysan at para. 73. 
------------------------------ Chief Justice Ma, Mr Justice Ribeiro PJ, Mr Justice Tang PJ, Mr Justice Bokhary NPJ, Lord Collins of Mapesbury NPJ I Introduction 1. This is an appeal from a judgment of the Court of Appeal refusing a stay of matrimonial proceedings in Hong Kong. The appellant husband (H) and the respondent wife (W) are German nationals. W has lived in Hong Kong since 1997 and is a permanent resident. They were married in Hong Kong and during their short marriage they lived there. W commenced divorce proceedings in Hong Kong. H commenced divorce proceedings in Germany and sought a stay of the Hong Kong proceedings. Prior to the marriage they executed an ante-nuptial agreement under German law, and they also executed a post-nuptial, or separation, agreement before the divorce proceeding. 2. Differing from Poon J, the Court of Appeal (Cheung, Fok and Lam JJA) decided that a stay should not be granted. The reasoning of the Court of Appeal will be outlined below. At this stage it is sufficient to mention that the principal reason why it decided to interfere with the judge’s decision, in a case which essentially involved the exercise of a discretion in accordance with well-settled principles of forum non conveniens, was that it considered that Poon J’s exercise of the discretion was plainly wrong because, in finding at the first stage of the enquiry that Germany was clearly and distinctly the more natural and appropriate forum, the judge had failed to give sufficient weight to the facts that W had commenced divorce proceedings in Hong Kong as of right and that the connection of the parties with Hong Kong was overwhelming. 3. The Court of Appeal exercised the discretion afresh, and in the normal course, where it had done so properly, such a case would not have been a suitable one for this court. But what takes it out of the ordinary is that, as mentioned above, the parties entered into an agreement prior to their marriage varying their matrimonial property rights under German law, and when they were about to divorce the parties entered into a separation agreement restricting W’s rights to claim maintenance. 4. Consequently the appeal has a much wider significance, since it involves the potential impact of these agreements in Hong Kong should the proceedings continue here. Such agreements were once considered at common law to be contrary to public policy because, not only were they seen somehow to affect the institution of marriage itself, they ousted the jurisdiction of the court to grant ancillary relief, but they have been the subject of a slow evolution in English law, and also reflected in Hong Kong law, to the point where, although not regarded as definitively binding, they have substantial legal effects. The two crucial decisions were, first, Edgar v Edgar [1980] 1 WLR 1410 (CA), on the effect of post-nuptial agreements or separation agreements (to which we will generally refer as “separation agreements”), which was applied in Hong Kong by the Court of Appeal in L v C [2007] 3 HKLRD 819, and, second, Radmacher v Granatino [2010] UKSC 42, [2011] 1 A.C. 534 in which the UK Supreme Court gave qualified effect to ante- or pre-nuptial agreements (to which we will refer as “ante-nuptial agreements”). This is an opportune occasion to consider whether Radmacher v Granatino represents the law in Hong Kong. If it does not, this would obviously be an extremely important factor in considering whether or not the present proceedings ought to be stayed. II The marriage and its breakdown 5. As is so often the case in bitter matrimonial disputes, there are almost no uncontested facts, and there are angry disputes in the affidavits on questions (concerning such matters as the size of W’s engagement ring, and the use by the couple of first class air travel) which are peripheral to the issues which arise on the present appeal. There have, of course, been no findings of fact, and the following account must be read in that light. 6. H is German. He is now 50. He is a successful specialist in luxury hotel interior design. His business is based in Cologne, Germany, and has grown internationally over the years. He now has a business presence in several parts of the world including Hong Kong, which he regards as a gateway to enter into the mainland and the Asia Pacific region. In about 2006, he set up Company A, a hotel interior design company, in Hong Kong. Company A’s business is funded by loans from HSBC, and he has paid income tax in Hong Kong. 7. W is also German. She is 46. She has been living in Hong Kong since 1997 and is a permanent resident. Prior to the events giving rise to the present proceedings, she worked in Hong Kong as a general manager for a German-based Hong Kong company. 8. H and W first met in Hong Kong in 2005 when H was on a business trip here. He was then married to his first wife, whom he married in 1997 and from whom he separated in 2006. Later that year H and W began an intimate relationship. In early 2007 H proposed marriage, and they were married at the Peninsula Hotel, Hong Kong, on February 15, 2008. The Hong Kong marriage certificate was used also to obtain registration of the marriage in Germany by April 3, 2008 at the Family Court of Berlin. 9. During their married life, the parties lived together in Hong Kong, although H (often accompanied by W) spent much of the time outside Hong Kong in the pursuit of business and leisure interests, and they frequently visited Germany. 10. H owns, or owned, a castle in Germany (“the castle”) initially through a company, Company C. He had acquired it for €2 million from his former wife as part of their ante-nuptial agreement or their separation agreement – it is not clear from the evidence which is the relevant agreement. H allowed W’s elderly retired parents to live in the guest house in the castle. H claims that as a result of the financial crisis in 2008 he was forced to sell the castle to discharge the loan used to purchase the castle from his former wife. H says that the purchaser was an independent wealth management company, which, on H’s request, granted a lease of the guest house to W’s parents (from which they have moved).W says that H is still the owner. 11. The marriage was a short one, and by 2010 it had broken down. W presented a divorce petition in Hong Kong in October 2010. In the following month H applied for a stay of the Hong Kong proceedings on the basis that the German courts were the forum conveniens. H also commenced divorce proceedings in December 2010 in the Family Court of Berlin. Poon J granted the stay by a decision of November 10, 2011, and an appeal was allowed by the Court of Appeal on February 8, 2013. H now appeals. III The Agreements 12. H and W entered into two agreements. The first was an ante-nuptial agreement executed on December 17, 2007 (“the Ante-nuptial Agreement”). The second was a separation agreement executed on September 9, 2010 (“the Separation Agreement”). The Ante-nuptial Agreement 13. H’s evidence of the genesis of the Ante-nuptial Agreement is as follows. W knew that he had entered into an ante-nuptial agreement with his first wife. H told W that as they had already accumulated wealth before meeting each other, their finances should remain independent and separate from each other. W agreed, and suggested that they should sign an ante-nuptial agreement. W knew that H had to carry a lot of business loans after his divorce with his ex-wife, including the €2 million he took out to purchase the castle from his former wife. If there were an ante-nuptial agreement he would be free from any claim against her business and assets, and she would be free from any potential claims from his creditors. So he agreed with her suggestion, even though he was concerned that the process would damage their relationship. 14. H then instructed his lawyers to draft an ante-nuptial agreement. On December 17, 2007, H and W flew back to Germany to execute the Ante-nuptial Agreement before a notary, who explained the contents and legal effect to them beforehand. H says that both parties intended their marriage to be governed by German law, and to have all matrimonial matters dealt with by the German courts. 15. W's evidence is substantially different. She accepts that H told her about the €2 million he had agreed to pay to his ex-wife for the castle, but not that he raised a bank loan to fund it. She denies that it was she who suggested that they should have an Ante-nuptial Agreement. The idea came from H who said that it was for her protection and for the purpose of keeping them safe, just in case if he had to go bankrupt, or had to let his companies go bankrupt. She received no independent legal advice before signing the Agreement before the notary, who was a long standing personal friend of H’s. She says that they did not discuss what law should govern their marriage, or what courts should deal with matrimonial proceedings. 16. The Ante-nuptial Agreement executed on December 17, 2007 stated that the parties (whose residential addresses were stated to be Hong Kong ones) declared: (1) they intended to be married in Hong Kong on February 15, 2008; (2) the notary had pointed out to them that in his opinion the proprietary effects of the marriage were governed by German law since they were both German citizens; (3) purely as a matter of precaution, they selected German law, in accordance with Article 15 of the Introductory Act to the German Civil Code, to govern the proprietary effects of the marriage; and (4) they had been informed by the notary of the consequences of the agreement, in particular about the meaning of the modification of the statutory matrimonial property regime. 17. The Agreement provided that they agreed to retain the German matrimonial property regime for the common ownership of gains accrued in marriage, subject to modification. They had both acquired substantial assets, in particular shares in companies (namely W’s Hong Kong company, and several companies owned by H, including Company A and Company C), and they assumed they would receive substantial assets from parents or third parties by way of gift, inheritance or succession. Accordingly, the companies which they owned (and debts associated therewith) and any such gifts etc would be excluded from computation of the parties’ assets in the event of the termination of the marriage (except on death). The present value of the assets was declared to be only €1 million, perhaps for the purpose of the calculation of ad valorem fees. 18. The Agreement was therefore essentially concerned with the exclusion from the matrimonial property regime of business assets and assets to be derived from gifts and inheritance, and not with the position on divorce. The Separation Agreement 19. H’s evidence is that the relationship deteriorated from February 2010, and that after a public incident in London in August 2010, he decided to divorce her, and told her of his decision on September 4, 2010. W had no objection and they had detailed discussions about the terms of the divorce on the following day. H then asked his lawyer to draft a separation agreement based on their agreement. On September 7, 2010, the notary emailed a copy of a draft Separation Agreement to H’s office in Hong Kong. H’s secretary printed it out and gave it to H and W, who then discussed it. W agreed to various handwritten amendments. In the same evening, H and W left for Germany separately. H denies that W signed the Agreement under duress. 20. W’s evidence is that there was no sign or indication from H that he had any problem with their marriage. Then on September 4, 2010, while she and H were discussing some business at the office, H suddenly told her that they were to separate and that she had to sign a separation agreement. He told her later that day that he had instructed his lawyers in Germany that she had to sign a separation agreement when she was there. He asked her to fly back to Germany immediately to inform her parents of their separation and if she were not to sign the agreement, she would be left with nothing. H then presented to her orally a list of terms to be included in the agreement. W denied that she saw the draft. 21. On September 9, 2010, she was collected from the castle by a tax adviser, a friend of H’s, who drove her to the notary’s office. It was during the journey that the tax adviser showed her, for the first time, a copy of the Separation Agreement. W declined to look at it, as she was already very upset and could not read while riding in a car. When they arrived at the notary’s office, she was presented with the original document and was told to sign. She had not received any financial disclosure or any legal advice nor was there any time to consider the contents away from H’s representatives. She felt pressured to sign, and her signature was procured by H through his representatives while she was under duress. 22. The Separation Agreement is dated September 9, 2010. The residential addresses of both parties were again Hong Kong ones. The Agreement provided that the Ante-Nuptial Agreement was rescinded, and that the assets in their respective possession should be retained in sole ownership. The Agreement made provision (inter alia) for W to stay in the matrimonial home until September 30, 2012, and for W to continue to receive her existing salary until September 30, 2012; for W to be able to use the castle as a guest until the end of 2011; for H to pay €500 per month for 2 years for the accommodation of a horse which was to be transferred to W, and for H to pay W’s private health insurance in Germany of €1200 per month for 2 years; H agreed to pay W’s parents an allowance of €20,000 should Company C terminate the tenancy which enabled them to stay at the castle. 23. For present purposes the most important provisions are those by which claims for maintenance were renounced. There are competing translations of the Agreement, but the following translation seems to represent its sense: “11. The parties concerned are in agreement that mutual support will not be due upon implementation of the present Agreement, since both will be in the position of supporting themselves. In the event of a legally effective divorce from the marriage, the husband and wife will waive mutual support, even in the case of emergency need, and adopt this waiver mutually. 12. The parties concerned are in agreement that a maintenance settlement should not be implemented. Claims for a maintenance settlement are mutually waived. With this provision it has been taken into account that both parties, even if indeed they have only earned negligible accrued pension rights, are each capable in the future of themselves looking after their own security in old age. … 15. All claims for equitable division of community property as well as possible effects of divorce shall be governed by the foregoing provisions. No mutual claims shall exist after implementation of the foregoing agreement. 16. Since all agreements to be made between us shall be governed by this contract, in the event of the initiation of a divorce proceeding only one of us will be represented by counsel. The costs arising thereupon shall be borne by the husband alone …” German law 24. There is substantial written evidence from four experts (two instructed by each party) on German law. Some of the opinions go far beyond the proper scope of an opinion on foreign law, and express views on the merits of the case. Nothing in this judgment is to be taken as a finding on German law in relation to any contentious or unclear matter. 25. This point will be expanded below, but at this point it is only necessary to say that the effect of the Ante-nuptial Agreement was to modify the German matrimonial property regime, which is not one of community of property, but community of accrued gains, i.e. separation of property with a claim for participation in the gains which accrued during the marriage. The parties can opt out of the matrimonial regime: BGB (German Civil Code), section 1408(1). According to the husband’s expert and one of the wife’s experts (but not the other) there is no need for independent legal advice in relation to marital agreements. There is evidence that conclusion of the agreement requires a notarial deed, and the notary is an independent holder of public office, who is supposed to be a neutral adviser owing a duty to both parties to explain the consequences of the agreement. 26. Three of the experts have given evidence on questions relating to the validity of the Agreements. Dr Palm and Dr Scherpe, instructed on behalf of H and W respectively, are of the view that there was no duty of disclosure prior to the Ante-nuptial Agreement, although Dr Pilati, also instructed on behalf of W, takes a contrary view. Dr Scherpe considers that there is no requirement for W to have separate advice where there is a notary, whose position is independent. He also expresses the view that an Ante-nuptial Agreement may be challenged on the ground that it is so one-sided as to be contrary to public policy, or on the basis that there was undue influence, fraud etc. Dr Pilati considers that an Ante-nuptial Agreement may be challenged on the basis of mistake if W was labouring under the mistake that (as she was told by H) the Agreement was necessary for her own protection. 27. On the validity of the Separation Agreement, Dr Scherpe considers that if W entered into the Ante-nuptial Agreement without disclosure or knowledge of the assets, then the Separation Agreement might be open to challenge. It might also be set aside on the ground H is found to have been abusing his dominant position. Dr Pilati also expresses the view that the Separation Agreement can be challenged if W did not receive independent advice, she was in a state of shock and subject to unequal bargaining power, and proper disclosure was not made. IV Radmacher v Granatino 28. There was extensive discussion in the courts below and in the arguments of the parties of the application of the decision of the UK Supreme Court in Radmacher v Granatino [2010] UKSC 42, [2011] 1 A.C. 534 concerning the effect of ante-nuptial agreements and, as indicated above, this is an opportune time for this court to state the law in Hong Kong. But a word of qualification is necessary in this respect, to which it will be necessary to return. Some ante-nuptial agreements, as frequently happens in the United States, are designed primarily for the consequences of divorce. This was partly so in Radmacher v Granatino, where the wife was a German heiress whose father had threatened that she would not receive a further inheritance unless she entered into an agreement with the French husband, providing for a mutual waiver of claims for maintenance on divorce, as well as for separation of property on marriage.But, as will appear below, that is not the typical ante-nuptial agreement in civil law countries. Separation or post-nuptial agreements 29. Separation or post-nuptial agreements were the subject of the leading English decision in Edgar v Edgar [1980] 1 WLR 1410 which decided that, although separation agreements did not override the powers of the court to grant ancillary relief, they carried considerable weight in relation to the exercise of the court's discretion when granting such relief. Such an agreement should be upheld unless there were vitiating factors or a compelling case of unforeseeable circumstances. 30. In Hong Kong the Court of Appeal conducted an extensive review of the English cases in L v C [2007] 3 HKLRD 819 and confirmed that when parties who are sui juris freely enter upon a bargain for the division of matrimonial assets then, in the absence of unfair or unconscionable circumstances surrounding the conclusion of the agreement and material and drastic unforeseen circumstances arising thereafter such as to cause manifest prejudice to one of the parties, the courts will hold the parties to their bargain. Determining whether injustice would be done under the agreements involved more than simply ascertaining whether there was any disparity in the value of the assets. The court would only allow one of the parties to depart from an agreement if that party demonstrated good and substantial grounds for doing so. Ante-nuptial agreements 31. The evolution in relation to ante-nuptial agreements was considerably later. In MacLeod v MacLeod [2008] UKPC 64, [2010] 1 AC 298 the Privy Council (on appeal from the Isle of Man) held that it was not open to the Privy Council to reverse the long standing rule that ante-nuptial agreements were contrary to public policy and thus not valid and binding in the contractual sense, since it was more appropriate that any such policy change should be made by legislation rather than by judicial development: [31], [35]. 32. In Radmacher v Granatino [2010] UKSC 42, [2011] 1 A.C. 534 a wealthy German heiress entered into an ante-nuptial agreement with her prospective husband, who was French. The agreement was subject to German law and provided (inter alia) for separation of property and also for a waiver of claims of maintenance on termination of the marriage. When the marriage broke down, the husband, despite the terms of the agreement, brought a claim for ancillary relief, seeking an order for a lump sum and periodical payments. The UK Supreme Court decided that, although it was the court and not any prior agreement between the parties which would determine the appropriate ancillary relief when a marriage came to an end, the rule that agreements providing for the future separation of the parties to a marriage were contrary to public policy was obsolete and no longer applied; the court should give weight to an agreement, made between a couple prior to and in contemplation of their marriage, as to the manner in which their financial affairs should be regulated in the event of their separation in circumstances where it was fair to do so; in appropriate circumstances, the court could hold the parties to the agreement even when the result would be different from that which the court would otherwise have ordered; on an application for ancillary relief the court should apply the same principles when considering ante-nuptial agreements as it applied to post-nuptial agreements. 33. In particular, an agreement would carry full weight only if each party had entered into it of his or her own free will, without undue influence or pressure, having all the information material to his or her decision to enter into the agreement and intending that it should be effective to govern the financial consequences of the marriage coming to an end; and the court should give effect to an agreement which was freely entered into by each party with a full appreciation of its implications unless in the circumstances prevailing it would not be fair to hold the parties to the agreement. Enforcement of the agreement could be rendered unfair by the occurrence of contingencies unforeseen at the time of the agreement or where, in the circumstances prevailing at the time of separation, one partner would be left in a predicament of real need while the other enjoyed a sufficiency. 34. The particular matters which were stressed by the Supreme Court were these. The court when considering the grant of ancillary relief was not obliged to give effect to nuptial agreements—whether they were ante-nuptial or post-nuptial. The parties could not, by agreement, oust the jurisdiction of the court. The court must, however, give appropriate weight to such an agreement. But it was the court, and not any prior agreement between the parties, that would determine the appropriate ancillary relief when a marriage came to an end, for that principle was embodied in the legislation. [2], [7]. 35. The Supreme Court said: “68 If an ante-nuptial agreement, or indeed a post-nuptial agreement, is to carry full weight, both the husband and wife must enter into it of their own free will, without undue influence or pressure, and informed of its implications. … 69 … the Court of Appeal was correct in principle to ask whether there was any material lack of disclosure, information or advice. Sound legal advice is obviously desirable, for this will ensure that a party understands the implications of the agreement, and full disclosure of any assets owned by the other party may be necessary to ensure this. But if it is clear that a party is fully aware of the implications of an ante-nuptial agreement and indifferent to detailed particulars of the other party's assets, there is no need to accord the agreement reduced weight because he or she is unaware of those particulars. What is important is that each party should have all the information that is material to his or her decision, and that each party should intend that the agreement should govern the financial consequences of the marriage coming to an end. … 71 … The first question will be whether any of the standard vitiating factors: duress, fraud or misrepresentation, is present. Even if the agreement does not have contractual force, those factors will negate any effect the agreement might otherwise have. But unconscionable conduct such as undue pressure (falling short of duress) will also be likely to eliminate the weight to be attached to the agreement, and other unworthy conduct, such as exploitation of a dominant position to secure an unfair advantage, would reduce or eliminate it. 72 The court may take into account a party's emotional state, and what pressures he or she was under to agree. But that again cannot be considered in isolation from what would have happened had he or she not been under those pressures. The circumstances of the parties at the time of the agreement will be relevant. Those will include such matters as their age and maturity, whether either or both had been married or been in long-term relationships before. For such couples their experience of previous relationships may explain the terms of the agreement, and may also show what they foresaw when they entered into the agreement. What may not be easily foreseeable for less mature couples may well be in contemplation of more mature couples. Another important factor may be whether the marriage would have gone ahead without an agreement, or without the terms which had been agreed. This may cut either way. 73 If the terms of the agreement are unfair from the start, this will reduce its weight, although this question will be subsumed in practice in the question of whether the agreement operates unfairly having regard to the circumstances prevailing at the time of the breakdown of the marriage.” The foreign element 36. There are several distinct aspects to the foreign element. One is what law governs the matrimonial property rights of the parties. The relevant conflict of laws rules in Hong Kong (which are the same as those in England) are that the rights of married persons in each other’s movable property are governed by the law of the matrimonial domicile (which for this purpose is almost certainly Germany), and the law governing the effect of the Ante-nuptial Agreement is German law, which was expressly chosen to govern it: see Dicey, Morris and Collins, Conflict of Laws(15th ed. 2012), Rules 165 and 166; Johnston, Conflict of Laws in Hong Kong (2nd ed 2012), para 7.099. 37. Radmacher v Granatino was a case (as is the present case) in which the ante-nuptial agreement was expressed to be governed by German law. The Supreme Court said (at [74]) that foreign elements may bear on the important question of whether or not the parties intended their agreement to be effective: “In the case of agreements made in recent times and, a fortiori, any agreement made after this judgment, the question of whether the parties intended their agreement to take effect is unlikely to be in issue, so foreign law will not need to be considered in relation to that question.” 38. The second, and distinct, question is what law governs the availability of financial relief. In Hong Kong, as in England, when the court exercises its jurisdiction to make an order for ancillary relief under the Matrimonial Proceedings and Property Ordinance (“MPPO”), or the Matrimonial Causes Act 1973, it will normally apply Hong Kong law, or English law, as the case may be, irrespective of the domicile of the parties, or any foreign connection: Dicey, Morris & Collins, Conflict of Laws (15th ed 2012), vol 2, rule 99(9) and e.g. C v C (Ancillary Relief: Nuptial Settlement) [2005] Fam 250, [31]. Consequently, the issues in Radmacher v Granatino were governed exclusively by English law, and the relevance of German law and the German choice of law clause was that they clearly demonstrated the intention of the parties that the ante-nuptial agreement should, if possible, be binding on them: [108]. The application of Radmacher v Granatino in Hong Kong 39. There have been signs of approval of Radmacher v Granatino in this court in LKW v DD [2010] HKEC 1727, (2010) 13 HKCFAR 537 (per Ribeiro PJ at [53], [105], obiter since the appeal did not concern an ante-nuptial agreement). In the view of this court, the principles enunciated in Radmacher v Granatino should also be regarded as the law in Hong Kong. In common with the UK Supreme Court, we see no reason for distinguishing between ante-nuptial agreements and separation agreements. 40. As we have said, the Hong Kong Court of Appeal has already accepted in L v C [2007] 3 HKLRD 819 that the old rule that agreements providing for future separation are contrary to public policy is obsolete, and we endorse its judgment. We agree with the UK Supreme Court that this should not be restricted to separation agreements. None of the supposed distinctions between them can any longer be supported, although we accept that there may be circumstances where it is appropriate to distinguish between an ante-nuptial and a separation agreement. As the UK Supreme Court said (at [61]) the circumstances surrounding the agreement may be very different dependent on the stage of the couple's life together at which it is concluded, but it is not right to proceed on the premise that there will always be a significant difference between an ante-nuptial agreement and a separation agreement. V Matrimonial property regimes 41. It is now necessary to revert to the word of qualification referred to above, when it was pointed out that the agreement in Radmacher v Granatino went beyond the normal matrimonial property agreement by providing that the parties were to have no claims against each other on divorce. In the present case the Ante-nuptial Agreement, like many such agreements entered into in civil law countries in Europe, was concerned only with the effect of the marriage on property rights. European countries in the civil law tradition operate matrimonial property regimes, which take a number of different forms. Some involve immediate community of property, others involve deferred community of property where the property is pooled on death, bankruptcy or divorce. Some involve total community, where all property is jointly owned. Others involve a community of acquests, where property acquired before marriage, and property gained gratuitously (e.g. by inheritance or gift) is excluded. But a common feature of these systems is that the parties may contract out of them: see Law Commission of England and Wales, Marital Property Agreements: A Consultation Paper (Consultation Paper No 198, 2011), paras 4.6 et seq. Contractual matrimonial property regimes are often entered into, not in contemplation of divorce, but to regulate the financial position of the parties during and after the marriage, sometimes to protect one of the spouses from the insolvency of the other. 42. Under German law the default matrimonial property regime is an accruals system, that of the Zugewinngemeinschaft, a community of acquired gains or a system of separation of property with a claim for participation in the gains of the spouses accrued during the marriage (Zugewinnausgleich): BGB, sections 1363 et seq. In the present case the Ante-nuptial Agreement excluded participation in the gains of the major assets of the parties at the time of the marriage and in assets to be acquired by gift or inheritance after marriage. It did not purport to restrict claims for financial relief on divorce. 43. Under German law, the matrimonial property regime and financial provision on divorce are entirely separate matters. The German rules on divorce provide for maintenance provision: (a) for the care of joint children; (b) if sufficient income cannot be expected due to age, or due to illness, or unemployment, or education or re-education requirements. The yardstick is standard of living during marriage (with a statutory discretion to limit it to reasonable requirements). The parties may, by an agreement made before the marriage or in contemplation of divorce, modify the default rules on maintenance (as in Radmacher v Granatino itself, above), but the Ante-nuptial Agreement in the present case dealt only with the matrimonial property regime. 44. It is not yet settled what effect matrimonial property regime agreements in the strict sense have in the case of matrimonial proceedings in Hong Kong (or England). 45. Comparable agreements in France providing for séparation de biens were involved in the stay cases, de Dampierre v de Dampierre [1988] AC 92 (which appears only from the transcript of the decision in the Court of Appeal) and Louvet v Louvet [1990] 1 HKLR 670 (see at 672), and several other stay cases discussed below have involved similar agreements under other laws. 46. Since Radmacher v Granatino matrimonial property agreements have been considered in several cases at first instance, but the position cannot be regarded as settled. 47. In Z v Z [2011] EWHC 2878 (Fam) such an agreement was treated by Moor J as an exclusion of the sharing principle in White v White [2001] AC 596 (adopted in Hong Kong in LKW v DD (2010) 13 HKCFAR 582). Moor J gave effect to what he regarded as the agreement to exclude sharing and granted ancillary relief on a reasonable needs only basis in a case which would otherwise “undoubtedly be a case for equal division of the assets” (at [31]). The decision is questioned by Cretney and Probert, Family Law (8th ed Probert, 2012), para 8-005, who say: “This inevitably raises questions about the very basis of the ‘sharing’ principle, and why any requirement to share, as opposed to meeting the other’s needs, should be capable of being displaced by agreement. In other words, is sharing a fundamental element of marriage, to be required whatever the individual wishes of the spouses, or does it rest on the inferred intentions of the spouses? If the former, one might ask why we have a system of separate property within marriage; if the latter, we perhaps need more empirical data about the actual intentions of married couples. And if, as the Law Commission have claimed, the prospect of sharing assets on divorce ‘may indeed be a serious disincentive to marriage for some’ [citing Law Commission of England and Wales, Marital Property Agreements (Consultation Paper No 198, 2011), para 5.23], then is the solution to move away from sharing, to allow those who do not want to share their wealth to marry but opt out of sharing, or simply to leave this group to live together outside marriage? Any of these approaches could, of course, be described as promoting marriage – and any of them could equally be viewed as devaluing marriage, depending on whether it is the obligations of marriage or the numbers marrying that is regarded as more important.” 48. In B v S [2012] EWHC 265 (Fam) [2012] 2 FLR 502 Mostyn J said (at [5]), in a case involving an agreement varying the Spanish matrimonial property regime, that there is a marked difference between a negotiated ante-nuptial agreement which specifically contemplates divorce and which seeks to restrict or influence the exercise of discretion to which the law gives access, and an agreement made in a civil law jurisdiction which adopts a particular marital property regime. On the facts Mostyn J found that the marital property agreement was only to be effective in England if the parties intended the agreement to have effect wherever they might be divorced and particularly were they to be divorced in a jurisdiction that operated a system of discretionary equitable distribution. See also GS v L [2011] EWHC 1759 (Fam) (Spanish law agreement between Spanish wife and German-Polish husband); [2013] 1 FLR 300; V v V [2011] EWHC 3230 (Fam) (Swedish law agreement between Italian husband and Swedish wife); AH v PH [2013] EWHC 3873 (Fam) (Scandinavian couple); Law Commission of England and Wales, Marital Property Agreements (Consultation Paper No 198, 2011); Matrimonial Property, Needs and Agreements: A Supplementary Consultation Paper (Consultation Paper No 208, 2012), Part 6; Matrimonial Property, Needs and Agreements (Law Com No 343, 2014), Chap 8. 49. The application of Radmacher v Granatinoto matrimonial property agreements of the type involved in this case (which the Separation Agreement purported to supersede) was not argued on this appeal. The material referred to above shows that this is a controversial area which is not free from difficulty, and nothing in this judgment is intended to pre-judge any question which may arise on this aspect in the substantive divorce proceedings, and in particular whether the application of Radmacher v Granatinorequires any adjustment or qualification in such cases. VI Forum non conveniens: principles 50. It is now well established in Hong Kong that the general principles of forum non conveniens apply to the stay of matrimonial proceedings: Johnston, Conflict of Laws in Hong Kong (2nd ed 2012), para 7.104. 51. We adopt the re-statement of the principles in matrimonial proceedings by the Court of Appeal (Cheung JA and Tang JA (as he then was)) in DGC v SLC (née C) [2005] 3 HKC 293, 297-298, applying Spiliada Maritime Corporation v. Cansulex Limited [1987] 1 AC 460, 477 and Louvet v. Louvet [1990] 1 HKLR 670, 674-675: “1. The single question to be decided is whether there is some other available forum, having competent jurisdiction, which is the appropriate forum for the trial of an action i.e. in which the action may be tried more suitably for the interests of all the parties and the ends of justice? 2. In order to answer this question, the applicant for the stay has to establish that first, Hong Kong is not the natural or appropriate forum (‘appropriate’ in this context means the forum has the most real and substantial connection with the action) and second, there is another available forum which is clearly or distinctly more appropriate than Hong Kong. Failure by the applicant to establish these two matters at this stage is fatal. 3. If the applicant is able to establish both of these two matters, then the plaintiff in the Hong Kong proceedings has to show that he will be deprived of a legitimate personal or juridical advantage if the action is tried in a forum other than Hong Kong. 4. If the plaintiff is able to establish this, the court will have to balance the advantages of the alternative forum with the disadvantages that the plaintiff may suffer. Deprivation of one or more personal advantages will not necessarily be fatal to the applicant for the stay if he is able to establish to the court's satisfaction that substantial justice will be done in the available appropriate forum.” 52. The Court of Appeal in that case (as in the present case) emphasised that the husband was entitled to sue in Hong Kong as of right. Where jurisdiction is founded in the Hong Kong court as of right (as in divorce proceedings like the present case), the party seeking the stay has to establish that there is another available forum which is clearly or distinctly more appropriate than the Hong Kong forum. This derives from what Lord Goff said in Spiliada (at 477), which has been regularly applied in Hong Kong: e.g. The Kapitan Shvetsov [1997] HKLRD 374 at 377; The Peng Yan [2009] 1 HKLRD 144, at [22]. 53. The existence of an ante-nuptial or post-nuptial agreement (particularly one governed by foreign law) is plainly a factor in the exercise of the discretion to stay on the ground of forum non conveniens, and there have been several decisions in this area involving such agreements. The facts in one case are not, of course, a guide to the exercise of discretion in another case, and they are presented here as merely illustrative. In England the decisions arise in the context of a statutory discretion to stay on “the balance of fairness (including convenience)” (Domicile and Matrimonial Proceedings Act 1973, Sched 1, para 9), which was held in de Dampierre v de Dampierre [1988] AC 92 to engage the same principles as those in the Spiliada case. In de Dampierre v de Dampierre itself the agreement for séparation de biens played no express part in the House of Lords’ reasoning that France was the appropriate forum for the divorce proceedings. In the Court of Appeal (whose decision was reversed) Dillon LJ thought it had no bearing on the appropriate forum. That approach was followed at first instance and on appeal in Louvet v. Louvet [1990] 1 HKLR 670 (see at 681). 54. In R v R (Divorce: Stay of Proceedings) [1994] 2 FLR 1036, where there was a Swedish separation of property contract, Ewbank J held that justice demanded that a stay of English proceedings be refused because the Swedish court could do no more than apply Swedish law and enforce the marriage contract, whereas the English court could grant a property adjustment, a lump sum and periodical payments. In S v S (Divorce: Staying Proceedings) [1997] 2 FLR 100 Wilson J granted a stay of English proceedings in a case where the parties had entered into an ante-nuptial agreement (in the negotiation of which each of the parties was represented by a distinguished New York lawyer) which provided for the financial aspects of divorce, and which was governed by New York law, and provided that it could only be enforced in New York. In C v C (Divorce: Stay of English Proceedings) [2001] 1 FLR 624 Johnson J placed decisive importance on the French ante-nuptial agreement (apparently for séparation de biens) in granting a stay of English proceedings. In Ella v Ella [2007] 2 FLR 35 the parties had entered into an ante-nuptial agreement, governed by Israeli law, providing for separation of property and for Israeli jurisdiction (see at [38]), and it was treated as a major factor in granting a stay, even though the wife contested its validity: but the wife’s lawyers had taken steps in the Israeli proceedings which were virtually a submission. In Hong Kong, in L v H, unreported, November 27, 2007 Rogers VP, refusing the husband’s leave to appeal from Judge Chan’s refusal to stay proceedings in favour of Germany, said, at [10], that the terms of an ante-nuptial agreement, if adhered to, would involve a grave injustice to the wife. VII Financial relief in Hong Kong after divorce outside Hong Kong 55. The possibility of financial relief in Hong Kong after a divorce in Germany is relevant because Poon J took the view that the balance of fairness between the parties would be achieved by staying the Hong Kong proceedings without prejudice to W’s right to make an application under the provisions in Part IIA of the MPPO, which came into effect on March 1, 2011. 56. Part IIA is based on, and is in material respects identical to, Part III of the English Matrimonial and Family Proceedings Act 1984 (the relevant Scottish provisions being different), which was introduced as a result of concern at the hardship to wives and children caused by the effect of the combination of the liberality of the rules for recognition of foreign divorces and the restrictive approach of some foreign jurisdictions to financial provision: Agbaje v Agbaje [2010] UKSC 13, [2010] 1 AC 628, at [4]. 57. The conditions for the application of Part IIA are stringent. The applicant must obtain leave to bring an application. The court has to consider whether Hong Kong is an appropriate venue, and then the applicant must persuade the court to make an order. Part IIA contains a filter mechanism in section 29AC, which provides that no application for an order may be made unless the court considers that there is substantial ground for the making of an order for financial relief. Section 29AE sets out the jurisdictional criteria for entertaining an application for an order for financial relief under Part IIA, which (a) the domicile of either of the parties in Hong Kong on the date of the application for leave, or the date on which the foreign divorce took effect; (b) the habitual residence of either of the parties for 3 years before such dates; and (c) a substantial connection of either of the parties with Hong Kong on such dates. 58. Before making the order, the court must consider whether it would be appropriate for the court to make the order, and must have regard to a number of factors, including the connection of the parties with Hong Kong and any place outside Hong Kong: section 29AF, which is headed “Duty of court to consider whether Hong Kong is appropriate venue for application.” 59. So far as material, the factors are these: “(a) the connection that the parties to the marriage have with Hong Kong; (b) the connection that those parties have with the place where the marriage was dissolved …; (c) the connection that those parties have with any other place outside Hong Kong; (d) any financial benefit that the applicant … has received, or is likely to receive, in consequence of the divorce … by virtue of any agreement or the operation of the law of a place outside Hong Kong; (e) if an order has been made by a competent authority outside Hong Kong requiring the other party to the marriage to make any payment or transfer any property to, or for the benefit of the applicant ….- (i) the financial relief given by the order; and (ii) the extent to which the order has been complied with or is likely to be complied with; (f) any right that the applicant has, or has had, to apply for financial relief from the other party to the marriage under the law of any place outside Hong Kong and, if the applicant has not exercised that right, the reason for that; (g) the availability in Hong Kong of any property in respect of which an order for financial relief in favour of the applicant may be made; (h) the extent to which any order for financial relief is likely to be enforceable; (i) the length of time that has elapsed since the date of the divorce ….” 60. The English equivalent of Part IIA was considered in detail by the UK Supreme Court in Agbaje v Agbaje, ante. In particular the UK Supreme Court, in a judgment of the court delivered by Lord Collins, said (at [50]) (adapted for the MPPO): “Many of the factors in section [29AF] have much in common with those which would be relevant in a forum non conveniens enquiry, but they are not directed to the question of which of two jurisdictions is appropriate. They are directed to the question whether it would be appropriate (which is the meaning of the word conveniens in forum conveniens) for an order to be made by a court in [Hong Kong] when ex hypothesi there have already been proceedings in a foreign country (including proceedings in which financial provision has been made). Little assistance can therefore be obtained from the stay cases (and still less from the anti-suit injunction cases) in the Part [IIA] exercise. The task for the judge under Part [IIA] is to determine whether it would be appropriate for an order to be made in [Hong Kong], taking account in particular of the factors in section [29AF], notwithstanding that the divorce proceedings were in a foreign country which may well have been the more appropriate forum for the divorce.” 61. As regards the basis for an order, the UK Supreme Court held that hardship was not a pre-condition of the exercise of the power, and there was no rule that the court would only make an award to the minimum extent necessary to remedy injustice ([60]-[63]); but it was not the intention of the legislation to allow a simple “top-up” of the foreign award so as to equate with a forum award ([65]); nor was it the purpose of the legislation to allow a spouse to make an application in order to take advantage of the more generous approach in England (and Hong Kong) in “big-money” cases ([72]). 62. The possibility of an application for financial relief after a foreign divorce has been taken into account in two stay cases involving ante-nuptial agreements, both of which were relied upon by Poon J. 63. In S v S (Divorce: Staying Proceedings) [1997] 1 WLR 1200 the wife was Swedish. The husband was a citizen of Austria, Turkey and Israel, and was enormously wealthy. After their marriage the parties lived together in New York and London, but the wife was based primarily in London. Prior to the marriage, the parties had entered into an ante-nuptial agreement, in which all rights to financial provision on divorce were renounced. Each of the parties was represented by a prominent New York lawyer. The agreement was governed by New York law and in substance provided for the exclusive jurisdiction of the New York courts. The husband was granted a stay of the wife’s English divorce proceedings. In the course of his judgment, Wilson J said (at 1202): “Were the English proceedings to be stayed, she would probably be confined to such financial claims, if any, as she could mount under New York law. She would probably be so confined because, although there is power under Part III of the Matrimonial and Family Proceedings Act 1984 to entertain claims for ancillary relief following an overseas divorce, I have doubts whether, had her suit been stayed, she would be able to satisfy the threshold requirement as to appropriate venue set by section 16 of the Act of 1984.” 64. In Ella v Ella [2007] EWCA Civ 99, [2007] 2 FLR 35 a couple of Israeli nationality entered into an ante-nuptial agreement immediately prior to their marriage in Israel. The agreement provided for separation of property with future assets belonging exclusively to the spouse creating them – the wife was not independently advised and the agreement was drawn up by a notary who had acted for the husband for some time: at [4]. The family home was in London. The wife petitioned for divorce in London, and the husband commenced proceedings in the rabbinical court in Israel (where divorce is administered by the religious courts). The wife’s lawyer took steps in the Israeli proceedings which amounted virtually to a submission. She then appointed new lawyers to contest the jurisdiction of the rabbinical court, and the husband sought a stay of the English divorce proceedings. Macur J granted a stay, and said that she was reassured by the fact that if the wife did not receive substantial justice in Israel she could make an application under Part III of the 1984 Act. The Court of Appeal upheld the decision. Thorpe LJ said (at [28]) that if the rabbinical court in Tel Aviv were to impose on the wife the terms of the pre-nuptial agreement with fullest vigour then the likelihood was that she would bring an application for ancillary relief in London under Part III of the 1984 Act. He did not express a view on the prospects of success on an application, but Charles J (at [56]) considered that, if the husband succeeded in Israel, the wife’s prospects of getting permission under Part III of the 1984 Act in England were good. VIII Conclusions 65. The application for a stay, of course, involves the exercise of a discretion and the Court of Appeal may only interfere with exercise of the judge’s discretion in accordance with well-settled principles which it is unnecessary to repeat: Hadmor Productions Ltd v Hamilton [1983] 1 AC 191, which has been applied frequently in Hong Kong (e.g. Tsit Wing (Hong Kong) Co Ltd v TWG Tea Co Pte Ltd [2013] 2 HKLRD 505 (CA)). Also, if the Court of Appeal has good grounds for interfering with the exercise of the judge’s discretion, and exercises the discretion afresh, this court will only interfere with the exercise of that discretion in accordance with the same principles. The decision of Poon J 66. In granting the stay, Poon J decided that Germany was clearly and distinctly the more natural and appropriate forum, in the light of the following matters in particular: both parties were German nationals with family ties in Germany; H’s business was essentially based in Cologne, Germany, although he had an international business presence at different places including Hong Kong; H had already commenced divorce proceedings in Germany; the Ante-nuptial Agreement and the Separation Agreement were executed in Germany, and the Ante-nuptial Agreement was expressly, and the Separation Agreement was impliedly, governed by German law; it was arguable that the parties had also, by necessary implication, designated Germany as the forum for their divorce; in resolving the question of validity of the Agreements, Germany was distinctly the more natural and appropriate forum, because most of the vitiating events relied on by W took place in Germany, and the notary and the tax adviser were allegedly involved, and were material witnesses in Germany; and German law was best dealt with by a German court. 67. As regards juridical advantage, W would not be disadvantaged in having to dispute the validity of the Agreements in a German court, which she was entitled to do. But if the Agreements were held to be valid in Germany, she would suffer “serious juridical disadvantage” because the German court would simply apply them, thus severely restricting her entitlement to financial relief. We would point out here that this view was consistent with the opinion of Dr Scherpe that the courts in Germany do not adopt what he calls a “holistic approach”; in his view the German courts aim more for certainty and there is not much scope for judicial discretion. 68. If the German court were to find that the Agreements were valid and simply apply them, W could still come back to Hong Kong to make an application under Part IIA of the MPPO for additional financial relief. Accordingly, the balance of fairness was achieved by staying the proceedings in favour of Germany without prejudice to W's right to make an application under Part IIA of the MPPO after the conclusion of the German proceedings. Court of Appeal 69. In the Court of Appeal Cheung JA emphasised that because W had brought the divorce proceedings in Hong Kong as of right, the burden was on H to show that not only Hong Kong was not the appropriate forum, but that Germany was distinctly more appropriate than Hong Kong. 70. Apart from the Agreements, the real and substantial connection of the parties with Hong Kong was overwhelming. There were three matters for determination by the Hong Kong courts: the divorce; the applicability of the Agreements (both in respect of the issue whether they had been vitiated by duress and also the weight to be given to them); and if the Agreements were not to be given effect, what should be the financial relief to be given to the parties, in the light of the circumstances of the case including the seven factors identified in section 7 of the MPPO. 71. The Agreements did not provide for exclusive jurisdiction in the German courts. Expert evidence on German law as regards the validity of the Agreements had already been filed. The German notary and tax adviser might have to testify in Hong Kong (so far there were no witness statements from them), but this was a matter affecting convenience or expense only. If the case were to be tried in Germany, then the parties and a Hong Kong witness (an employee of H) would have to travel there to attend court. 72. As regards juridical advantage, if the Agreements were found to be valid in the German court, there would be no basis for merely giving them appropriate weight in the light of fairness. Under German law, there was apparently no requirement for full financial disclosure or access to independent legal advice, which was of particular importance in the context of H’s admittedly close and life-long relationship with the notary. By staying the Hong Kong proceedings, the Hong Kong court would already have considered Germany to be the appropriate forum where substantial justice could be given to W, and W might not be able to satisfy the “substantial ground” requirement in the leave application under Part IIA of the MPPO, which could not be a proper factor to be considered in an application for stay. Overall conclusion 73. In the view of this court, the Court of Appeal was entitled to interfere with the exercise of discretion by the judge for these reasons. 74. While the judge referred to the principle that in stay cases, where jurisdiction is founded as of right, the applicant must show that the foreign forum is distinctly more appropriate than Hong Kong, he failed to give appropriate weight to the factors connecting the parties, the marriage, and the matrimonial home with Hong Kong, and gave inappropriate weight to their nationality, and to the Ante-Nuptial Agreement and the Separation Agreement (which, as observed earlier, stated the residence of the parties to be in Hong Kong). The principal issues in the Hong Kong proceedings would relate to the matters to be considered under MPPO, section 7, in determining the financial relief to be granted to W, and the weight to be given to the Agreements. 75. It is true that the Agreements are governed by German law. However, like the agreement in Radmacher v Granatino (see NG v KR (Pre-nuptial Contract) [2008] EWHC 1532 (Fam), [2009] 1 FLR 1478, [43]), the Ante-nuptial Agreement in this case says nothing about jurisdiction. The Separation Agreement contains no exclusive submission to the German courts, although it plainly contemplates divorce proceedings in Germany by providing that one lawyer is to act for both parties. It is also true that proceedings would be much more expensive in Hong Kong, although W can afford that less than H. But W contests the validity of the Agreements, and the evidence seems to show that their validity can be challenged on a number of grounds which are available under German law, and which have already been the subject of extensive written evidence in these proceedings. 76. We also agree with the Court of Appeal that if the judge had not erred at the first stage, he was correct in concluding that W would suffer a juridical disadvantage if the stay were granted. If the German court held that the Agreements were valid, they would be applied without the discretion inherent in the Radmacher v Granatino approach. The judge found that W would suffer a serious juridical disadvantage but we agree with the Court of Appeal that he erred at the third stage in finding that the balance of fairness would be achieved by staying the proceedings in favour of Germany without prejudice to W’s right to make an application under Part IIA of the MPPO. He placed too much reliance on statements by the English Court of Appeal in Ella v Ella [2007] EWCA Civ 99, [2007] 2 FLR 35 that if the Israeli court gave full effect to the ante-nuptial agreement, the wife could apply in England under Part III of the 1984 Act. That appeal was decided before Agbaje v Agbaje and Radmacher v Granatino, and it must be doubtful whether the decision could be supported on that ground. More relevant is the statement by Wilson J in S v S (Divorce: Staying Proceedings) [1997] 1 WLR 1200, 1202, that, where proceedings are stayed on the ground of forum non conveniens, there is room for doubt whether the threshold requirements for appropriate venue under Part III would be met. That is so even though the UK Supreme Court recognised in Agbaje v Agbaje that the Part III requirements were not the same as the forum convenienstest. 77. Since in our judgment the Court of Appeal was justified in interfering with the exercise of discretion, the sole remaining question is whether the exercise of discretion by the Court of Appeal is flawed and open to challenge on familiar grounds. In our view it is not. The Court of Appeal correctly held that H had not shown that Germany was clearly or distinctly more appropriate than Hong Kong. 78. We therefore dismiss the appeal. We would only add that W should not necessarily think that the very considerable costs she is bound to incur in these proceedings will be well spent. As was indicated in the oral argument, victory on this appeal might well be pyrrhic. 79. As to costs, we make an order nisi that the respondent (W) should have the costs of this appeal, such costs to be taxed if not agreed. If either party wishes to have a different order for costs, written submissions should be served on the other party and lodged with the Registrar of the Court within 14 days of the handing down of this judgment, with liberty on the other party to serve and lodge written submissions within 14 days thereafter. In the absence of such written submissions, the order nisi will stand absolute at the expiry of the time limited for these submissions. Mr David Pilbrow SC and Ms Frances Irving instructed by Joseph Chu, Lo & Lau, for the Appellant Mr Russell Coleman SC and Mr Richard Todd QC instructed by Withers, for the Respondent 
Chief Justice Cheung: 1. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Mr Justice Ribeiro PJ and Mr Justice Fok PJ: 2. This judgment concerns a contested application for the retrospective validation of the disposition of a company’s property made after commencement of the company’s winding up. A. The Joint Venture Agreement and the Company’s financial difficulties 3. Pursuant to a Joint Venture Agreement dated 21 November 2013 (“the JVA”) Hsin Chong Construction Company Limited (“the Company”) and Build King Construction Limited (“Build King”) tendered for and, on 22 June 2016, were awarded a Government contract to design and build a large police facility in Kowloon with the Company taking a 65% interest and Build King the remaining 35% in their unincorporated joint venture. 4. In 2017 and 2018, it became increasingly clear that the Company was in dire financial straits. Among other indications, trading in the Company’s listed shares was suspended in March 2017, defaults occurred in meeting its debt obligations, its auditors qualified its accounts on the going concern basis and, on 17 July 2018, the West Kowloon Cultural District Authority, the employer under a separate contract for the Company to construct the M+ Museum, determined that contract on the ground of the Company’s insolvency. On 27 August 2018, a creditor issued a petition to wind up the Company.[1] 5. Clause 17 of the JVA provides a mechanism whereby one party may exclude the other from the joint venture on the ground of that other’s insolvency. It enables the “Continuing Party” to exclude the “Defaulting Party” from “further participation in the management of the Joint Venture and the Contract and the profits arising therefrom” and permits the Continuing Party “to take over the benefits of the Defaulting Party in the Joint Venture (but without releasing the Defaulting Party from its obligation to bear its proportionate share of any loss resulting or to result from the Contract) ...”. 6. If the option to exclude is exercised, Clause 17 provides for an accounting exercise to be performed at the completion of the project to calculate the amount of any accrued profit the Defaulting Party might be entitled to up to the date of exclusion, less the Defaulting Party’s share of any losses, pre- and post-exclusion, and any expenditure or loss incurred by the Continuing Party due to the Defaulting Party’s default. Clause 17.5 accordingly provides: “Upon completion or termination of the Contract, receipt of all amounts due to be paid by the Client pursuant to the terms thereof and the ascertainment of all liabilities of the Continuing Party under or arising out of or in connection with the Contract or the construction of the Works, the Continuing Party shall, subject as hereinafter provided, account to the Defaulting Party who shall be entitled to receive an amount equal to the sum provided by him towards the Working Capital together with his proportionate share (as adjusted in accordance with the terms and conditions hereof as the case may be) of any profits or other entitlements earned and received by the Joint Venture in respect of the Contract but calculated up to the date when the Defaulting Party was excluded from the Joint Venture less:- (a) his share of any losses arising from the Contract calculated in accordance with the Proportions whether before or after the date of exclusion; and (b) all costs, expenses, losses and damages incurred by the Continuing Party directly or indirectly as a result of the default of the Defaulting Party. 7. The final account might show a sum due to the Continuing Party and Clause 17.6 accordingly stipulates: “In the event that the share of the costs, expenses, losses and damages chargeable to the Defaulting Party exceeds its entitlements, the Defaulting Party, or (as the case may be) its successors, receivers or other legal representatives, shall promptly pay the excess to the Continuing Party as a primary obligation and as a debt.” 8. It will be convenient to refer to the benefits contingently claimable by the Defaulting Party in the event of Clause 17 being triggered as “the residual rights”. 9. On 13 December 2018, having taken the view that the Company was insolvent, Build King did indeed invoke Clause 17 constituting itself the “Continuing Party” and the Company the “Defaulting Party” for the purposes of that clause. At that stage, the design phase was close to completion and construction works on the structure were well advanced, while work on the architectural finishes and building services was just commencing. B. The Supplemental Agreement 10. Negotiations ensued. They resulted in Build King and the Company entering into a Supplemental Agreement (“the SA”) dated 17 December 2018 whereby Build King agreed to acquire the Company’s residual rights together with its interests in materials, plant and equipment, deposits and partial payments which it had contributed to the joint venture. The consideration agreed to be paid by Build King for such acquisition was the sum of $53.6 million to be paid in two instalments of $20 million and $33.6 million respectively. It is common ground and has been accepted below that this represented a fair, and indeed, favourable valuation of the Company’s rights and interests being acquired since it included a significant amount for projected post-exclusion profits to which the Company was not entitled under the JVA. It also eliminated certain contingencies and removed the need to await a final accounting at the project’s completion. 11. However, crucially, by its clause 5(d) the SA specified that the $53.6 million “will be paid in [sic] the following designated account requested by [the Company]: Bank of East Asia 136-68-00013-1 savings account, Cogent Spring Limited [‘Cogent Spring’].” Clause 5(e) went on to stipulate: “[The Company] confirms that [Cogent Spring] is a wholly owned subsidiary of Hsin Chong Group Holdings Limited and that Build King’s payment to [Cogent Spring] will be understood as and/or equivalent to the effect that Build King has discharged its liability in connection with or arising from this SA.” 12. Initially, as indicated in a resolution by the Company’s board dated 13 December 2018, the intention was for the Company itself to use the amounts received from Build King “to settle the outstanding payment of MPF [Mandatory Provident Fund] and staff wages/salaries”, without any mention of Cogent Spring or any other third party recipient. However, as the trial judge found: “On 14 December 2018, the Company requested that payment be made to [Cogent Spring] ... as the Company’s bank accounts were frozen because of the petition and outstanding MPF contributions and employees’ wages could not be paid.”[2] 13. That proposal raised concerns on the part of certain directors on both sides of the transaction. The reaction of Mr Desmond Chang Kam Chuen, a Build King director, was to send an e-mail dated 14 December 2018 to Mr Wilfred Wu (then a director of the Company) stating: “I just noted that you request us to pay the first $20 million to another company named Cogent Spring. I’m sorry that we can’t entertain this request; instead, we will provide you a bank draft payable to [the Company], the party to the agreement.” 14. And Mr James Lee Kok Foo, who was one of the Company’s directors until he stepped down from the board on 14 December 2018, protested when he saw draft board minutes purporting to approve such payment to Cogent Spring and in his e-mail to the Company’s directors dated 16 December 2018, he stated: “I remind all to check the lawfulness of requesting Build King to make payments to [Cogent Spring] while [the Company’s] accounts have been frozen by concerned banks. This may be viewed as fraud rendering legal liability to directors so involved.” 15. Nonetheless, Build King proceeded to pay the first instalment to Cogent Spring instead of the Company, evidently to circumvent the problem of the latter’s frozen bank accounts with a view to paying outstanding salaries and MPF contributions. On 17 December 2018, the sum of $20 million was transferred into Cogent Spring’s account with the Bank of East Asia, as specified in the SA. The funds were then dissipated in making a variety of payroll and MPF payments, as well as towards meeting legal costs and miscellaneous utility and other expenses, involving not only the Company but other entities in the Group. C. Section 182 and the application for validation 16. Section 182 of the Companies (Winding Up and Miscellaneous Provisions) Ordinance[3] (“CWUMPO”) materially provides as follows: “In a winding up by the court, any disposition of the property of the company, including things in action, ... made after the commencement of the winding up, shall, unless the court otherwise orders, be void.” 17. As we have seen, the SA was executed on 17 December 2018 and the payment of $20 million to Cogent Spring was effected on the same date. As this was after presentation of the petition (on 27 August 2018) which is when the winding up is deemed to have commenced,[4] the parties faced the possibility that the transaction might constitute a disposition of the Company’s property which was void unless the court otherwise ordered. 18. Build King might have, but did not, seek in advance the court’s approval of the SA and the payment. Instead, it sought the court’s retrospective validation by issuing a summons filed on 18 January 2019 seeking an order confirming that the SA should not be avoided by section 182.[5] In taking that course, Build King took the risk of the court refusing to make the order.[6] Since section 182 deems dispositions of company property occurring after commencement of the winding up to be void unless the court otherwise orders, if entry into and performance of the SA were held to constitute a disposition of the Company’s property (as the Company’s provisional liquidators contend), the transaction would be presumptively void and the burden would fall on Build King to show why the court ought to make a validating order.[7] D. The decisions in the Courts below D.1 At first instance 19. Deputy High Court Judge Le Pichon held in favour of Build King and ordered that the SA and any disposition of property thereunder should not be avoided by virtue of section 182.[8] 20. Her Ladyship found (as pointed out above) that the Company had requested “that payment be made to [Cogent Spring] ... as the Company’s bank accounts were frozen because of the petition and outstanding MPF contributions and employees’ wages could not be paid.”[9] 21. She also found that: “... at the very least, [Build King] had an inkling that part of the proceeds would be applied by the Company in settling outstanding MPF contributions and employees’ wages that would contravene section 182 (absent any prior validation order) and that [Build King] could be said to be facilitating a possible contravention by making payment to Cogent Spring instead of the Company directly.”[10] 22. In deciding in favour of validation, the learned Judge held as follows:[11] “BK’s position was that that was a matter internal to the Company and not for a third party purchaser who is making a payment into the Group. BK made payment to the Company’s nominee to discharge its obligations as purchaser. Any disposition made subsequently of the proceeds would not have been made by BK but by the Company.” (§93) “How should the court exercise its discretion? In so far as the proceeds (or part thereof) have been applied in contravention of section 182 and so prejudiced the unsecured creditors, it would have been because of misapplication of the purchase monies by the Company and/or its directorsand not BK. In any event, BK was and is not in a position to control or direct the Company’s application of the proceeds.” (§94) “For BK, the acquisition of the Company’s residual rights in the JV was a commercial transaction negotiated at arm’s length. No ulterior purpose or agenda in BK making payment into the designated account can be discerned. It is not the PLs’ case that BK breached any duty or obligation whether to the unsecured creditors or anyone else. The only reason not to validate would be to punish BK for somehow facilitating a potential breach by the Company. But how would that benefit the unsecured creditors?” (§95) “For a start, if the transaction is rendered void, the $53.6 million will have to be repaid to BK. The Company will have to await the completion of the project and the final accounting before it could receive the accrued profits up to the date of its exclusion. It has no right to any share of post-exclusion profits and its exposure to the possibility of having to carry its share of the loss on the Project will continue until its termination or completion.” (§96) “In those circumstances, I can see no good reason for the court not to exercise its discretion to validate the transaction.” (§97) 23. Her Ladyship’s focus was thus on Build King’s payment which she regarded as a payment to the Company to discharge its obligations as purchaser under the SA and, for that reason, not a disposition of the Company’s property; that any subsequent misapplication of the funds was internal to the Company or its directors and did not involve Build King; and that Build King had not breached any duty to the unsecured creditors or otherwise. She was also concerned that the consequence of refusing validation would involve the Company having to “repay” $53.6 million to Build King. D.2 In the Court of Appeal and submissions on the present appeal 24. The Court of Appeal held that there were no valid grounds to interfere with the judge’s exercise of discretion.[12] It referred to clause 5(e) of the SA[13] which it took to mean that Cogent Spring “was designated as the Company’s agent to receive the consideration from BK on the Company’s behalf”,[14] and held as follows:[15] “The judge has fully considered the circumstances in which $20 million was paid to Cogent Spring. Insofar as part of the proceeds had been applied in contravention of section 182 and so prejudiced the unsecured creditors, it would have been because of the subsequent misapplication of the proceeds by the Company and/or its directors and not BK. BK was not in a position to control or direct the Company’s application of the proceeds. The judge could discern no ulterior purpose or agenda in BK making payment into the designated account.” (§60) “There is no evidence that BK had actual or constructive notice of any acts of impropriety of the payments out from the account of Cogent Spring caused to be made by the Company. When BK raised concern about the request of the Company to pay $20 million to Cogent Spring, this was before BK was informed by the Company that the reason for the request was because the Company’s accounts were frozen and hence it was not possible for the Company to apply the funds to pay its employees.” (§61) 25. We return to the aforesaid decisions in our analysis of the transaction below. We pause however to note that the Court of Appeal’s suggestion in (§61) that Build King’s concern about the request to pay Cogent Spring was raised before Build King was aware that this was necessitated by the fact that the Company’s accounts were frozen because of the petition appears to run contrary to the trial judge’s findings set out above.[16] 26. The submissions made by Mr Charles Manzoni SC[17] reflected the approach adopted below. He placed much emphasis on the proposition that, because of what had been contractually agreed between the Company and Build King, payment by Build King to Cogent Spring in accordance with clause 5 of the SA was deemed to be payment to the Company and that only subsequent disbursements of the monies could constitute dispositions caught by section 182. E. The principles regarding section 182 validation orders 27. Section 182 is modelled on virtually identical provisions which exist in other common law jurisdictions, especially in England and Wales[18] and in Australia, so that reported decisions in those jurisdictions provide valuable guidance. 28. The minimalist language of section 182 does not specify the criteria or circumstances in which the court should or should not “otherwise order” so as to save a disposition from being void. However, construing the section purposively, as Buckley LJ put it in Re Gray’s Inn Construction Co Ltd, the section should be understood as giving effect to: “... a basic concept of our law governing the liquidation of insolvent estates, whether in bankruptcy or under the Companies Acts, that the free assets of the insolvent at the commencement of the liquidation shall be distributed rateably amongst the insolvent's unsecured creditors as at that date.” [19] 29. His Lordship noted that, consistently with that basic concept: “There may be occasions ... when it would be beneficial, not only for the company but also for its unsecured creditors, that the company should be enabled to dispose of some of its property during the period after the petition has been presented but before a winding up order has been made.”[20] 30. Many examples of such beneficial dispositions can be found in the reported cases. These include payments made in good faith in the ordinary course of business to keep the company afloat with a view to its possible longer-term survival or its being sold as a going concern and thus benefiting the company and its creditors more than would a break up realisation of the company’s assets;[21] transactions entered into at full value after presentation of the petition which add to or prevent reduction of the company’s assets;[22] transactions seizing an opportunity speedily to dispose of some piece of property at an exceptionally good price or incurring expenditure necessary to enable the company to complete a profitable contract;[23] and so forth.[24] 31. The interests of the general body of creditors are determinative in deciding whether a disposition of the company’s property after commencement of the winding up should be validated: the court must do its best to ensure that their interests are not prejudiced.[25] Validation orders may be made where the applicant shows that the disposition is likely to be or actually has been for the benefit of the unsecured creditors.[26] Conversely, an order is likely to be refused if such benefit cannot be shown. This would obviously a fortiori be the case where the disposition is seen to involve a misapplication of the company’s property with a view to preferring the rights of certain creditors at the expense of the others, or to reducing overall the assets available for distribution to the general body of creditors.[27] F. The transaction analysed 32. The concern of section 182 is with the disposition of a company’s property made after commencement of the winding up. If a transaction amounts to such a disposition, it is void unless the court orders otherwise. And as we have seen, in deciding whether to validate the disposition, the court regards the interests of the general body of creditors as of central importance. 33. It is therefore important to identify correctly the Company’s property and the disposition in question. Here, the Company’s residual rights plus incidental interests under the JVA constituted the initial property concerned. The Company agreed to sell them to Build King for the consideration of $53.6 million, payable in two instalments. So the residual and incidental rights were converted into a contractual chose in action consisting of a right to payment of that consideration which was the Company’s property. By entering into the SA with its clause 5, the Company and Build King agreed that that payment should not be made into the Company’s coffers (to circumvent the freezing of its bank accounts) but should instead be made to Cogent Spring, a different entity within the Group. In performing that agreement, a disposition of a chose in action with the value of $20 million, the first instalment, took place. Section 182 became applicable and rendered the SA and payment made thereunder void unless the Court otherwise orders. 34. With respect, both the Judge and the Court of Appeal focussed on the wrong property and the wrong “disposition”. They failed to recognise that the relevant property was the right to payment of $53.6 million and its first instalment as the consideration for sale of the Company’s residual and incidental JVA rights. They focussed on Build King’s payment of the monies which they described as merely Build King’s performance of its payment obligation under the SA, contractually acknowledged by the Company to constitute discharge of that obligation and, for that reason, not a relevant disposition. They considered that only the subsequent disbursements might constitute a misapplication of the purchase proceeds by “the Company and/or its directors”, over which Build King had no control and which involved no breach of duty on Build King’s part. 35. There are major flaws in the aforesaid approach. In the first place, it is a non sequitur to suggest that because payment is made to a third party in accordance with a contractual provision which deems such payment a discharge of the payor’s obligation, this somehow prevents that payment from being a section 182 disposition. If as a result of that payment, the Company’s property is transferred or dissipated so that the interests of the general body of creditors are prejudiced, it matters not that the transfer or dissipation is wrapped in contractual clothing. It is still a disposition which attracts section 182 and its prejudicial effect on the unsecured creditors prevents the transaction from being validated. The great majority of dispositions for which validation has been refused in the reported cases will have been made pursuant to contractual or other legal arrangements.[28] 36. The question of whether a disposition has taken place is one of substance and not form. As was explained by HH Judge Paul Matthews (sitting as an additional judge of the High Court) in Officeserve Technologies Ltd (in Liquidation) v Anthony-Mike,[29] whether a disposition occurs is judged by what happens to the value of a company’s asset as a result of the transaction in question: “In considering what is and what is not a disposition for the purposes of s.127, it is necessary not to be constrained by what, in formal terms, may be the transfer of one interest in property, wholly and separately, to another person. The mischief against which the section is directed is clear. The destruction, or at least the reduction in value, of a property right belonging to the company, causing an immediate and equivalent accrual in value to another person, is well within that mischief.”[30] 37. His Lordship added: “In my judgment, it is sufficient that identifiable property by some act having legal consequences (so excluding mere effluxion of time) ceases to be in the ownership of the company, so that it is no longer available to the liquidator of the company for the statutory purposes, and the value accrues to some other person (so excluding consumption or waste), even though that other person cannot necessarily be said to become the owner of the same property.”[31] 38. In the present case, it is clear that the value of the first instalment of $20 million never accrued to (and was never intended to accrue to) the Company but went entirely to Cogent Spring to be dissipated in favour of various third parties to the prejudice of the Company’s unsecured creditors and the pari passu principle. Thus, invocation of a principle of agency law that payment to an agent conferred with authority to receive money on behalf of his principal is deemed receipt by the principal is of no assistance to Build King here. Such a doctrinal deeming cannot obscure the fact that, in terms of substance and value, none of the $20 million reached the Company to be available for distribution to the general body of creditors. 39. Moreover, on the evidence and findings of the trial judge, the authority conferred on Cogent Spring was not to “receive payment on behalf of the Company” but to receive it to make intended disbursements to recipients other than the Company, circumventing the frozen bank accounts. Furthermore, the suggestion that any void disposition “only would have been because of misapplication of the purchase monies by the Company and/or its directors and not Build King” cannot be accepted since the Company and/or its directors never received Build King’s payment and did not have any of those funds to misapply. 40. Secondly, the judgments below erroneously emphasise that Build King had no ulterior purpose and breached no duty in making the payment to Cogent Spring as matters favouring the making of a validation order. As indicated above, the concern of section 182 is to preserve the Company’s property for proper distribution under the statute, applying to dispositions made after commencement of the winding up. The section does not concern itself with the parties to the transaction and does not require it to be shown that such parties were involved in any breaches of duty before the disposition is rendered presumptively void. Incidentally, it will be recalled that the trial judge found that Build King “had an inkling” – that is, “knew” – “that part of the proceeds would be applied by the Company in settling outstanding MPF contributions and employees’ wages that would contravene section 182 (absent any prior validation order) and that [Build King] could be said to be facilitating a possible contravention by making payment to Cogent Spring instead of the Company directly.”[32] Whether or not such an intended circumvention of section 182 constitutes a breach of duty by someone or other is not presently in issue. However, such knowledge certainly does not militate in favour of a validation order. 41. Finally, the trial judge’s suggestion that “if the transaction is rendered void, the $53.6 million will have to be repaid to”[33] Build King was incorrect. No part of those funds was ever paid to the Company. Indeed, the second instalment of $33.6 million was not paid out by Build King at all. So there was no question of any repayment by the Company to Build King. The effect of holding the SA void would be that the Company could revert to a claim against Build King for the value of its residual rights under the JVA as determined on a final accounting.[34] It is uncertain how that amount would compare to the sum of $53.6 million agreed to under the SA. G. Disposal of the appeal 42. For the aforesaid reasons, we would allow the appeal, set aside the validation orders made below and declare the SA and dispositions made thereunder to be void by reason of section 182. 43. We would make an order nisi that Build King should pay the costs here and below, giving the parties liberty if so advised, to lodge submissions in writing concerning costs within 14 days of the date of this judgment, in default of which the order nisi is to stand as an order absolute without further order. Mr Justice Tang NPJ: 44. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Mr Justice Gummow NPJ: 45. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Chief Justice Cheung: 46. The appeal is unanimously allowed and we make the orders referred to in paragraphs 42 and 43 above. Ms Audrey Eu SC, Mr John Hui and Mr Anson Wong Yu Yat, instructed by Wilkinson & Grist, for the Appellant Mr Charles Manzoni SC and Mr John Leung, instructed by Hogan Lovells, for the Respondent The Official Receiver, attendance dispensed with [1] After settlement was reached with the original petitioner, another creditor was substituted as petitioner, supported by others. A number of debt recovery actions were also commenced against the Company. [2] Deputy High Court Judge Le Pichon [2019] HKCFI 1531 at §87. [3] Cap 32. [4] Under CWUMPO section 184(2). [5] The Summons also sought validation of Build King’s exercise of its right to exclude the Company from the joint venture under Clause 17 of the JVA. The validity of such exclusion is no longer in issue and does not require discussion in this judgment. [6] As recognised in In re Gray’s Inn Construction Co Ltd [1980] 1 WLR 711 at 718D. [7] Jardio Holdings Pty Ltd v Dorcon Construction Pty Ltd (1984) 2 ACLC 574 at 579; Re Leric International Ltd [2009] 2 HKLRD 238 at §28. [8] [2019] HKCFI 1531 (13 June 2019). [9] Ibid at §87. [10] Ibid at §92. [11] BK and PLs being her abbreviations for Build King and the provisional liquidations of the Company. [12] [2019] HKCA 1305 (4 December 2019) at §58, per Kwan VP, with whom Cheung and Yuen JJA agreed. [13] Set out above at §11. [14] [2019] HKCA 1305 at §59. [15] Footnote references have been omitted. [16] At §§12 and 13. [17] Appearing with Mr John Leung on behalf of Build King. [18] In recent times, section 227 of the Companies Act 1948 as re-enacted by the Companies Act 1985, section 522 and the Insolvency Act 1986, section 127. Those provisions derive from section 153 of the Companies Act 1862, and subsequent enactments in the Acts of 1908 and 1929. See Re S A & D Wright Ltd [1992] BCC 503 at 503. [19] [1980] 1 WLR 711 at 717D; Goldlion Properties Ltd v Regent National Enterprises Ltd (2009) 12 HKCFAR 512 at §117; Express Electrical Distributors Ltd v Beavis [2016] 1 WLR 4783 at §20. [20] [1980] 1 WLR 711 at 717E. [21] Eg, Re Stean’s (Bournemouth) Ltd [1950] 1 All ER 21 at 24; In re Clifton Place Garage Ltd [1970] Ch 477 at 493-494; In re J Leslie Engineers Co Ltd (In Liquidation) [1976] 1 WLR 292 at 301, 304; In re Gray’s Inn Construction Co Ltd [1980] 1 WLR 711 at 717F-G; Re Luen Cheong Tai Construction Co Ltd [2004] 1 HKLRD 735 at §12; Express Electrical Distributors Ltd v Beavis [2016] 1 WLR 4783 at §21. [22] In re Gray’s Inn Construction Co Ltd [1980] 1 WLR 711 at 719B-E; Express Electrical Distributors Ltd v Beavis [2016] 1 WLR 4783 at §§43-44. [23] In re Gray’s Inn Construction Co Ltd [1980] 1 WLR 711 at 717F; Express Electrical Distributors Ltd v Beavis [2016] 1 WLR 4783 at §21. [24] While the point does not arise in the present case, we note and respectfully agree with the comments of Sales LJ in Express Electrical Distributors Ltd v Beavis [2016] 1 WLR 4783 at §§33-40 qualifying an aspect of Buckley LJ’s judgment in In re Gray’s Inn Construction Co Ltd [1980] 1 WLR 711 at 718F-G involving the “bald proposition” that “a disposition carried out in good faith in the ordinary course of business at a time when the parties are unaware that a petition has been presented may, it seems, normally be validated by the court”. [25] In re Gray’s Inn Construction Co Ltd [1980] 1 WLR 711 at 717G; Re S A & D Wright Ltd [1992] BCC 503 at 504-505 (applying In re Gray’s Inn Construction); Express Electrical Distributors Ltd v Beavis [2016] 1 WLR 4783 at §20. [26] As Sales LJ pointed out in Express Electrical Distributors Ltd v Beavis [2016] 1 WLR 4783 at §25: “In a case where a retrospective validation order is sought, as distinct from a prospective order, the range of evidence available is likely to be different. In a case where a retrospective order is sought it may have become clear whether a particular transaction or the carrying on of the company’s general business in fact turned out to be for the benefit of the general body of creditors or not, whereas in a case where a prospective order is sought the court will have to make an assessment on the basis of such evidence as is available of what is likely to transpire in the future.” [27] Re Stean’s (Bournemouth) Ltd [1950] 1 All ER 21 at 24. [28] See eg, B Mullan & Sons (Contractors) Ltd v Ross and another (1996) 54 Con LR 163 at 185 on the applicability of the equivalent section to a proposed payment by an employer directly to a sub-contractor pursuant to the construction contract after winding up proceedings had been started against the contractor, validation being refused. [29] [2017] EWHC 1920 (Ch), [2017] BCC 574. [30] [2017] BCC 574 at §90 (emphasis in original). Section 127 referred to is in the Insolvency Act 1986 which is materially identical to section 182. [31] Ibid at §99 (emphasis in original). [32] [2019] HKCFI 1531 at §92. [33] Ibid at §96. [34] As held, for instance, in Re AGI Logistics (Hong Kong) Ltd (in liq) [2015] 4 HKLRD 300, a tax refund which was an asset of a parent company in liquidation, wrongly paid by the Commissioner of Inland Revenue to a subsidiary rather than the parent and so held void under section 182, had to be paid again to the liquidators since the Company’s statutory right to the refund had not been met. 
Mr Justice Tang PJ (Judgment of the Court): 1. These proceedings are about the beneficial ownership of 200 million shares in Mayer Holdings Ltd, a listed company in Hong Kong (“Mayer HK”). The shares were represented by two share certificates, No 70 and 71, each for 100 million shares registered in the name of Mayer Corporation Development International Ltd (“Mayer BVI”). The share certificates together with instruments of transfer signed in blank were held by Alliance Financial Intelligence Ltd (“AFIL”) under a Custodian Agreement dated 19 June 2009 (“the Custodian Agreement) signed by Alex Ku, its managing director, and on behalf of Mayer BVI by Mr Lai Yueh Hsing, its sole director. 2. The rival claimants were Mayer BVI (the appellants), and Aspial Investment Ltd (“Aspial”) and Bumper East Ltd (“Bumper”)(who were two of the respondents before us). After a 4 day trial, Reyes J decided in favour of Aspial and Bumper. His decision was upheld by the Court of Appeal after a 3 day hearing. Despite the concurrent findings, Mayer BVI was able to appeal to us because the value of these shares exceeded $1 million, accordingly, it was entitled to appeal as of right under s 22(1)(a) of the Hong Kong Court of Final Appeal Ordinance.[1] 3. It is our settled practice not to review concurrent findings of fact save in rare and exceptional circumstances. Nor would we embark on such a review, unless we were persuaded that it would be purposeful to do so.[2] We heard arguments on whether it would be purposeful to embark on such a review and decided that there was no reason to permit such a review. Accordingly, we dismissed the appeal with reasons to be handed down later. These are our reasons. 4. The background facts are complicated and were set out in some detail in the judgments below.[3] We will be deal with them as briefly as possible. 5. Aspial and Bumper were the plaintiffs in HCA 238/2012 and Mayer BVI the defendant. Aspial and Bumper’s case was that they purchased the shares from Capital Wealth Finance Company Ltd (“Capital Wealth”), of which Mr Lam Chin Chun (“Lam”) was the CEO. Capital Wealth was authorised to sell these shares by Mayer BVI. 6. Mayer BVI was the plaintiff in HCCL 3/2012[4] against AFIL, Charles Chan Wai Dune[5] (“Charles Chan”), Lam, Bumper and Aspial[6], alleging that AFIL was guilty of a breach of its fiduciary duties owed to Mayer BVI in respect of the certificates which Mayer BVI had deposited with AFIL under the Custodian Agreement, that Charles Chan and Lam dishonestly assisted AFIL in its breaches of fiduciary duty, and that Aspial and Bumper received the share certificates with notice of AFIL’s breaches of fiduciary duty. According to Lai, the share certificates were deposited with AFIL for one year to take advantage of market movements, but no sale should be made without written authorization from Mayer BVI. 7. According to Lam, he on behalf of Capital Wealth made an oral agreement with Lai who was acting for himself as well as on behalf of Mayer BVI to sell within one year of 19 June 2009, 100 million[7] and 300 million shares respectively in Mayer HK for $100 million, and that Capital Wealth was entitled to keep any excess over $100 million as its profit. Under that agreement Mayer BVI would deposit 200 million shares with AFIL for one year and Capital Wealth would deposit $50 million with AFIL. The deposit of the 200 million shares was evidenced by the Custodian Agreement. The deposit of the $50 million was evidenced by a letter dated 3 June 2009 signed by Ku on behalf of AFIL, addressed to Capital Wealth acknowledging receipt of $50 million “being 承諾金 (in translation ‘promise money’) to be paid to [Mayer BVI] and/or its representative.” Although Capital Wealth’s payment of $50 million to AFIL on 3 June 2009 was not disputed, the authenticity of the letter of 3 June 2009 was very much so, evidently because it described the $50 million as “(promise money) to be paid to (Mayer BVI) …” The authenticity of the Custodian Agreement was common ground as well as the fact that Mayer BVI had procured the issue of the 2 certificates so that they could be deposited with AFIL. 8. Capital Wealth’s case was that $100 million had been paid. As to $55 million[8] upon Lam selling 100 million shares in the market, this sale was common ground and payment admitted. Capital Wealth’s case was that after the sale of 100 million shares and payment of $55 million, there was a further oral agreement made on 24 June 2009[9] between Lam and Lai (the 2nd oral agreement) because “Mayer BVI and Lai were in need of funds”[10] under which Lai/Mayer BVI would be paid the balance of the $45 million, and the 200 million shares held by AFIL would be released to Capital Wealth to sell as it saw fit without any time limit. The shares were in due course sold to Aspial and Bumper. 9. According to Lam, as a result of the 2nd oral agreement Capital Wealth authorized AFIL to pay $45 million to Mayer BVI out of the $50 million promise money paid to AFIL by Capital Wealth. However, only $42.5 million was paid because AFIL was entitled to a fee of $2.5 million being 2.5% of $100 million. The promise money of $50 million was disposed of as follows: $42.5 million to Mayer BVI, $5 million returned to Capital Wealth and $2.5 million retained by AFIL. The disposition of the $50 million was not disputed butMayer BVI’s case was that the payment of $42.5 million had nothing to do with Mayer BVI nor was Lai involved with such payment. According to Lai, in March 2009, his long term friend, a Mr Wang Ing Jye (“Wang”) of Taiwan, was approached by 7 Taiwanese investors who together owned 99 million Mayer HK shares to help them sell their shares in Hong Kong. But because Wang did not know any stock broker in Hong Kong, Lai’s personal assistant, Flora Kao (“Kao”), who also worked for Wang part time, recommended Charles Chan to Wang whom Wang said he met in March 2009. 10. This is what Reyes J said of Wang’s evidence: “ 57. … Charles Chan had (Wang says) responded that it would not be difficult to fetch at least $1 per share, but to do so might take time. 58. Charles Chan is thus said to have suggested to Wang in March 2009 that Wang deposit the shares with CCIF for 1 year. In return, Charles Chan would arrange an initial deposit of $42.5 million to be paid to Wang or other nominees of the investors owning the 99 million shares. 59. When the 99 million shares were handed over on 19 June 2009, Ku is said to have produced another Custodian Agreement…” 11. Wang’s evidence was that it was pursuant to this agreement that on 24 June 2009 $42.5 million[11] were paid by AFIL to persons whose names appeared on a list produced by Kao, acting on his behalf. On the other hand, Ku produced a copy of a letter dated 24 June 2009 from AFIL signed by him and addressed to Mayer BVI which stated: “as instructed by your company’s representative, Ms Flora Kao, we have distributed (Promise Money) of HK$50,000,000 received from (Capital Wealth) in accordance with the endorsed payment schedule presented by her.” Ku’s evidence was that the original was given to Kao, which Kao denied. 12. Both Charles Chan and Ku denied ever having met Wang. Ku also denied that he had signed the Custodian Agreement covering 99 million shares produced by Wang at trial. 13. Charles Chan’s evidence, accepted by Reyes J, was that he was approached by Lai on 24 June 2009 about the payment out from the promise money. To cut a long story short, later on the same day, both Lam and Lai confirmed to him the content of the 2nd oral agreement. Charles Chan also spoke to Ku about the release of the promise money and Ku told him that he saw no problem. 14. Reyes J gave detailed reasons why he rejected Wang’s evidence as unreliable. Reyes J preferred the evidence of Lam, Charles Chan and Ku. The judge said: “164. [Kao and Wang’s] explanation [on how they dealt with the $42.5 million] is incredible. ...... 167. For those reasons, Wang’s evidence must be rejected as unreliable. ...... 169. … In light of my assessment of Wang’s evidence, I cannot regard that [the disputed Custodian] Agreement as authentic.” 15. At paras 170 and 171, Reyes said he accepted Ku’s evidence and rejected the disputed Custodian Agreement as unreliable. He went on to say at para 171: “On the balance of probability, the document is a fake.” 16. Mr Benjamin Yu SC, for Mayer BVI, submitted that Reyes J should not have found the disputed Custodian Agreement to be a fake on a balance of probabilities. He said the judge had overlooked Re H[12]. Mr Yu said that was why we should review the concurrent findings. In support, Mr Yu also pointed to Re D [2008] 1 WLR 1499 and said Reyes J should have subjected the question of the authenticity of the disputed Custodial Agreement to a “heightened examination” but as Lord Carswell went on to explain in Re D the seriousness of the allegation to be proved “[does] not require a different standard of proof or a specially cogent standard of evidence, merely appropriately careful consideration by the tribunal before it is satisfied of the matter which has to be established.” 17. Neither Re H nor Re D assisted Mr Yu. The burden was on Capital Wealth to prove the 1st and 2nd oral agreements[13], on a balance of probabilities. Insofar as Mayer BVI relied on the alleged 99 million shares agreement, it also had to establish it on a balance of probabilities. The fact that, in the process, Wang produced a document the authenticity of which was disputed made no difference.[14] The burden was on Mayer BVI to prove that the document was authentic, again, on a balance of probabilities. That Mayer BVI failed to do. The judge accepted the evidence of Ku who denied having signed the disputed Custodial Agreement. Reyes J accepted the evidence of Charles Chan on the events of 24 June 2009 which were irreconcilable with the existence of a separate agreement with Wang. Moreover, he regarded the evidence of Wang as unreliable and incredible. After a careful examination of the totality of the evidence, Reyes J was satisfied that both the 1st and 2nd agreements were made. The learned judge’s decision was affirmed by the Court of Appeal, where Barma JA said, with the agreement of the other members of the Court: “57. In my view, none of the criticisms directed at the judgment could justify this court in overturning the judge’s factual findings and the conclusions which he reached. On the contrary, for the reasons explained by the judge, there was ample evidence on the basis of which he was justified in making those findings and disposing of these proceedings as he did. The appeal must therefore be dismissed.” 18. The Re H argument is totally unmeritorious, and raised for the first time by Mr Yu before us. It focused on an isolated part of the case and an isolated part of the judgment of Reyes J without any regard of the wider, and more crucial, issues dealt with by the judge. This is as strong a case on unassailable concurrent findings as one could imagine. 19. The second basis upon which Mr Yu submitted we should review the concurrent findings was the allegation that Lai did not have authority to commit Mayer BVI to the oral agreements. This occupied 20 pages of Mr Yu’s written case. Many of the submissions are poorly disguised submissions on why Reyes J was wrong to have found in favour of the 2nd oral agreements. What are left can be disposed of shortly. 20. First, that Lai had no actual, ostensible or apparent authority to bind Mayer BVI to the oral agreements. Lai’s lack of authority was not pleaded. Nor was it an issue at trial. 21. Then, that the 2nd oral agreement, as pleaded, showed that it was made by Lai in breach of his fiduciary duty to Mayer BVI. Mr Yu submitted that it mattered not that neither Lai’s breach nor Lam’s knowledge of such breach was unpleaded. He relied on Lysaght Bros & Co Ltd v Falk (1905) 2 CLR 421. In Lysaght, the plaintiff sued the defendant in the Supreme Court of New South Wales on a contract made with the defendants’ agent. The issue was whether under a plea of non-assumpsit, the principal might give evidence of all circumstances tending to show that the agent, in making the contract, was acting without authority, to the knowledge of the plaintiff, even though that evidence might also show that there was fraudulent collision between the plaintiff and the agent in making the contract. The principal was not allowed to adduce such evidence because the Supreme Court took the view that, under the pre-judicature pleading rules as they then applied in New South Wales, a specific or “special” plea of fraud was required in order to raise that issue. For reasons irrelevant to this appeal, the plaintiff’s applications for leave to amend were unsuccessful. The plaintiff’s appeal to the High Court of Australia was successful and a new trial was ordered. 22. Lysaght was concerned “with a specific rule of pleading” per Griffith CJ at 434 and as Barton J explained at 436 “whether the evidence was not admissible under the plea of non-assumpsit”, and does not help Mr Yu. Time has moved on, our current rule, O18 r8 (1) requires a party to “plead specifically any matter, … (a) which he alleges makes any claim or defence of the opposite party not maintainable; or (b) which, if not pleaded, might take the opposite party by surprise”. The role of pleadings post the Civil Justice Reform has been highlighted by this Court on a number of occasions.[15] 23. We do not agree that a defence based on self dealing by an agent with someone who has knowledge (constructive or actual) of it, need not be pleaded. Moreover, pleaded or not, it was not raised as an issue at trial. So anyway Lysaght cannot help Mr Yu. 24. Mr Yu also submitted that neither Flywin[16] nor lack of pleading mattered[17] because Mayer BVI had an unanswerable case. This is an impossible argument. The issues are fact sensitive. 25. For these reasons, the appeal was dismissed. Costs 26. We make an order nisi against the appellant, such costs to be paid on an indemnity basis. It may assist if we indicate, on a provisional basis, why we consider costs on an indemnity basis appropriate. The appeal was brought as of right because the value of the shares exceeded $1 million. It was plainly hopeless. Had leave been required, it would have been refused. This appeal represents yet again the worst excesses caused by the as of right route of appeal under s 22(1)(a) of the Hong Kong Court of Final Appeal Ordinance. This Court has in the past commented on the anachronistic nature of this provision and the need for its repeal.[18] The present appeal was set down for 2 days. The parties’ written cases amounted in total to over 170 pages. The authorities numbered 75. Ten counsel were briefed on appeal. The following passage from the judgment of the Chief Justice in Kwok Chin Wing v 21 Holdings Limited is apt: “Cases such as the present appeal are inherently wasteful of resources which the Courts have in limited supply. Following the Civil Justice Reform introduced in 2009, this kind of wastage ought no longer be tolerated by the public. It is unfair to the successful parties in a litigation, it is unfair on other litigants who have deserving cases before the Courts and it is ultimately unfair to the community.” 27. If any party wishes to have a different order for costs, written submissions should be served on the other party or parties and lodged with the Registrar of the Court within 14 days of the handing down of this judgment, with liberty on the other party or parties to lodge written submissions within 14 days thereafter. In the absence of such written submission, the order nisi will stand absolute at the expiry of the time limited for these submissions. Mr Benjamin Yu SC, Mr Bernard Man & Mr Justin Ho, instructed by K & L Gates, for the defendant in HCA 238/2012 and the plaintiff in HCCL 3/2012 / appellant Mr Simon Chiu & Mr Timothy Wong, instructed by Hastings & Co, for the 1st & 2nddefendants in HCCL 3/2012 / 1st & 2ndrespondents Mr Edward Chan SC & Mr Law Man Chung, instructed by Fred Kan & Co, for the 3rddefendant in HCCL 3/2012 / 3rdrespondent Mr John Litton, Mr Fredrick H F Chan & Mr Eric Chow, instructed by Samuel LC Yang & Co, for the 1st & 2ndplaintiffs in HCA 238/2012 and the 4th & 5thdefendants in HCCL 3/2012 / 4th & 5threspondents [1] Cap 484. [2] Z v X v C,FACV No 11, 12/2013 (unreported, dated 23 May 2014). [3] In paras 22-24, Barma JA set out the respective cases of the parties. He said at para 21 that the parties’ respective cases on the beneficial ownership of the 200 million shares “were irreconcilably different.” According to Reyes J: “8. … The accounts are contradictory and cannot be reconciled.” [4] HCA 238/2012 and HCCL 3/2012 were consolidated by Reyes J by order dated 22 March 2012. [5] Charles Chan Wai Dunewas the founder of CCIF CPA Ltd who were Mayer HK’s auditors. CCIF changed its name to Crowe Horwath (HK) CPA Ltd which remained as Mayer HK’s auditors until 2011. [6] These defendants in HCCL 3/2012 were referred to below as the Charles Chan faction. However, the main protagonists were in fact Capital Wealth represented by Lam, and Mayer BVI represented by Lai. [7] However, eventually only 99 million shares were provided. According to Lam, Lai told him that he was the beneficial owner of 99 million shares and Mayer BVI 300 million. [8] In fact, $55,009,344.21 was paid. [9] Reyes J said: “69. There are diametrically opposed versions of what happened on 24 June 2009.” The learned judge dealt with these versions at paras 69-88 of his judgment. [10] Barma JA at para 22(5). [11] It should be noted that these proceedings were not directly about the 99 million shares. However, Wang’s allegation about the 99 million shares, if true, would provide an explanation for the payment of the $42.5m. [12] [1996] AC 563, where Lord Nicholls said at 586D-F: “When assessing the probabilities the court will have in mind as a factor, to whatever extent is appropriate in the particular case, that the more serious the allegation the less likely it is that the event occurred and, hence, the stronger should be the evidence before the court concludes that the allegation is established on the balance of probability.” [13] Issues 3 & 4 at trial respectively. Reyes J held both oral agreements were made. The 2nd agreement is irreconcilable with the allegation that the $42.5 million was paid under a separate agreement with Wang. [14] The authenticity of AFIL’s letters of 3 June 2009 and 24 June 2009, paras 7 & 11 above, was also disputed. [15] See Kwok Chin Wing v 21 Holdings Limited and another, FACV 9/2012 (unreported, 30 September 2013), Sinoearn International Limited v Hyundai-CCECC Joint Venture (a firm), FACV 22/2012 (unreported, 30 September 2013). [16] Flywin Co Ltd v Strong & Associates Ltd (2002) 5 HKCFAR 356. [17] Because there was no reasonable possibility that the state of evidence relevant to the point would have been materially more favourable to the other side. [18] See Kwok Chin Wing v 21 Holdings Limited. 
Chief Justice Ma: 1. For the reasons contained in the judgment of Lord Collins of Mapesbury NPJ, I agree that these appeals should be dismissed. Like Lord Collins, I believe that the outcome can be determined by reference to applicable conflict of laws principles (for we are here concerned with a foreign illegality). It is not necessary to discuss the matter in the context of a domestic illegality. As Lord Collins points out, the lower courts determined the case on the basis of the principle contained in Tinsley v Milligan [1994] 1 AC 340 and also applied the proportionality test contained in ParkingEye Ltd v Somerfield Stores Ltd [2013] QB 840. Since the decisions of the lower courts, there have been important developments in the law in this area and reference is made below to the decisions of the UK Supreme Court in Hounga v Allen [2014] 1 WLR 2889, Les Laboratoires Servier v Apotex Inc[2015] AC 430 and Bilta (UK) Ltd v Nazir (No 2) [2015] 2 WLR 1168. It is not necessary, as I have said, to discuss illegality in the domestic context and I would accordingly leave open any detailed discussion of the applicable principles for a case in which the point arises. I would like, however, to make this point. Both the lower courts applied a proportionality test, this being the test advocated by both parties based on their reading of ParkingEye. I am not in favour of applying this test as the applicable test for illegality since it appears to suggest that some kind of judicial discretion to be exercised. It seems to me that the question of illegality must be based on firmer principle and policy, although I accept that the facts in any given case will inevitably differ. However, as I have indicated, it is not necessary to go into this aspect in the present appeals. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Collins of Mapesbury NPJ and the observations made by the Chief Justice. Mr Justice Tang PJ: 3. I agree with the judgment of Lord Collins of Mapesbury NPJ and the observations made by the Chief Justice. Mr Justice Chan NPJ: 4. I agree with the judgment of Lord Collins of Mapesbury NPJ and the observations made by the Chief Justice. Lord Collins of Mapesbury NPJ: I Introduction 5. These appeals from the judgment of the Court of Appeal arise out of two consolidated actions by Ryder Industries Limited, formerly known as Saitek Limited (“Saitek”) against Timely Electronics Company Limited (“Timely”), both of which are Hong Kong companies, and against Mr Chan Shui Woo (“Mr Chan”), the majority shareholder and director of Timely, as guarantor. At the conclusion of the hearing the Court indicated that the appeals would be dismissed for reasons to be given later. 6. The claims are for about HK$6.5 million claimed to be due under a series of agreements between 2005 and 2007 in connection with a form of joint venture for the manufacture of mobile phones in China. Saitek was to make available factory facilities and part of the equipment necessary for the manufacture of the phones, and Timely was to introduce customers, provide also a part of the machinery to enable the manufacturing to take place, as well as provide management and supervision in the manufacturing process. 7. Mr Recorder Anthony Houghton SC (“the judge”) gave judgment[1] for Saitek, and an appeal to the Court of Appeal was dismissed.[2] 8. The appeals raise the question of the enforceability in Hong Kong of a contract governed by Hong Kong law when it has been performed in the PRC partly in breach of PRC law. The relevant contracts are the agreements referred to in paragraph 6 above. It was accepted by the parties that Mr Chan’s liability as guarantor stood or fell with the validity of the agreements. II The background: Co-operation Agreement and Supplemental Agreements 9. In 2002 Saitek established an operating division doing business in China through a “Commission Processing Enterprise” (CPE) in conjunction with a local authority in Shenzhen, under the name Saitek Baoan Shanghe Saitek Electronics and Plastics Factory (“Saitek CPE”). Saitek also established in the PRC a Wholly Foreign Owned Enterprise (“Saitek WFOE”). 10. Timely also had an associated wholly-owned company incorporated in the PRC which is a Wholly Foreign Owed Enterprise (“Timely WFOE”), and which operated a factory at Kung Ming Property Development Main Company, Baoan, Shenzhen (“the Kung Ming Factory”). 11. Saitek CPE had substantial spare capacity at its factory. In 2005 a former employee of Saitek introduced Timely to Saitek, with a view to the introduction by Timely of customers seeking a manufacturing facility for mobile phones in China at Saitek CPE. Saitek CPE did not have all of the equipment necessary to carry out the manufacturing function and was not willing to invest in the purchase of the remaining necessary equipment. What was envisaged was that Timely would supply, in addition to customers, the machinery not then available to Saitek CPE to enable the manufacturing to take place, and Timely would also provide management and supervision of the manufacturing process. 12. These discussions resulted in a written memorandum of understanding and, subsequently, in a written agreement (headed Agreement of Co-operation between Saitek and Timely) between Saitek and Timely in late October 2005. It was common ground that the agreement is governed by Hong Kong law. 13. The agreement recorded that Saitek and Timely had a mutual interest in the manufacturing of mobile phones and that Saitek would provide resources to support the manufacturing process in return for a share of revenue. Saitek was to be responsible for customs declarations and for logistics/freight costs of importing the components from Hong Kong to Saitek CPE’s factory, and the freight costs to export the finished goods to Hong Kong. Timely was to be responsible for the transportation costs between the Saitek CPE and Timely factories, and for payment collection from customers. Timely was to pay Saitek a monthly rental for the premises, and a monthly depreciation charge for the equipment, and also utility charges and staff wages. Saitek was to account for a share of the revenue received from customers. 14. The parties operated a periodic mutual account, which by mid-2006 was considerably in favour of Saitek, and the sums due had been accruing and unpaid for some time. 15. As a result, in September 2006 Saitek and Timely entered into a first supplementary agreement, which provided that Timely would pay all overdue debts to Saitek before September 2007, and would pay interest on overdue balances, at 1% above base lending rate. A second supplementary agreement was made in March 2007, under which Timely and Saitek agreed (inter alia) that the net outstanding payable to February 2007 was HK$5,615,394.97, and that Timely would pay the debts before December 31, 2007. Mr Chan personally guaranteed that all outstanding debts would be paid by then. 16. When proceedings were commenced by Saitek, Timely and Mr Chan resisted payment on the ground that enforcement of the agreements was barred as a result of acts of performance of the co-operation agreement which were illegal under PRC law. 17. The judge found that two of the four alleged illegalities had occurred, but that they did not affect the enforceability of the contracts, and his decision was affirmed by the Court of Appeal. Timely and Mr Chan now appeal to this Court. III Alleged illegalities under PRC law 18. Four illegalities were alleged. The first was that the arrangement was designed to evade PRC law, which provided that Timely, a foreign company for these purposes, could only lawfully carry on an independent business through a mainland body. The judge rejected the allegation on the basis that, although the agreement envisaged the payment of something akin to rent by Timely to Saitek in respect of the factory premises for use by Timely, it had not created a lease of these premises; it did not envisage Timely operating a separate or independent business under the agreement, which was a co-operation agreement. 19. The second allegation related to the fact that semi-finished processed goods were transported from the Saitek CPE premises to the Kung Ming Factory of Timely WFOE for testing before being returned to the Saitek CPE premises. Under the applicable law, materials (such as those used for processing the goods under the agreement between Timely and Saitek) imported to China for processing were imported duty free and should have been kept as bonded goods under constant supervision before being re-exported from China. Transporting goods between the factories meant technically that the supervision was broken. Written authorisation could have been obtained from the customs authorities, but obtaining it was not practicable within the timeframe required. Accordingly this requirement was said to be frequently honoured in its breach. The expert witnesses were agreed that such conduct would be in breach of the customs law of the PRC, but disagreed as to how seriously the authorities would view a breach. In the event, following a raid by customs authorities on Timely’s Kung Min factory in 2007 goods which had been seized and impounded were released on payment of a substantial “administration fee”, described by the judge as “inferentially illicit”. The judge noted that Timely did not seek to rely on this alleged illegality as, by itself, constituting a defence to the claims of Saitek, but as a factor in assessing the extent to which the agreement was tainted by illegality. This was relevant to an assessment of the fourth allegation of illegality. 20. The third allegation of illegality was that, in breach of PRC law, Saitek CPE was said effectively to be fulfilling orders for mainland customers (which subsequently expanded to an allegation that use of machinery by Saitek CPE for processing of domestic orders would also be in breach of bonded equipment regulations). The judge rejected this allegation on the basis that the work for mainland customers was done, not by Saitek CPE, but by Timely, and there was no misuse of the bonded equipment. 21. The fourth ground, and the only ground of illegality which the judge accepted, was that Saitek CPE used materials imported duty-free by Saitek WFOE, another operation of Saitek in Shenzhen, for the production of mobile phones. If imported by Saitek WFOE duty-free, the materials should have remained under the control of Saitek WFOE, and it was undisputed that outsourcing of bonded materials from a WFOE to a CPE would be a breach of Article 23 of the Measures of the Customs of the People’s Republic of China for the Supervision of Goods for Processing Trade, which provides for outsourcing by an operating enterprise to be subject to the approval of the customs office. In the absence of approval, outsourcing is not permitted. Failure to comply attracts a sanction contained in the Regulation of the People’s Republic of China on the Implementation of Customs Administrative Punishment, the level of punishment depending upon the severity of the offence, and ranging from a reprimand or criticism to a fine and confiscation of gains. This ground of illegality can therefore be seen to be linked to the second ground. IV The decisions of the judge and the Court of Appeal The judge’s decision 22. The judge noted that the evidence showed that there was a wide range of potential penalties, but that the evidence as to the seriousness with which any such breach would be viewed was very limited. His conclusion was that, so far as the evidence went for the fourth ground of illegality, although a substantial amount of goods were involved, he was not persuaded that the mode of performance would be considered a very serious contravention of the law. 23. The applicable legal principles were, according to the judge, not in dispute before him. He said that illegality was an aspect of public policy, and the courts would not, in the ordinary course of events, enforce a contract which was illegal under domestic law, and that a contract to be performed in a foreign jurisdiction, the performance of which would be illegal in that place of performance, would not be enforced (citing Regazzoni v KC Sethia (1944) Ltd [1958] AC 301 and Ralli Brothers v Compania Naviera Sota y Aznar [1920] 2 KB 287). 24. The essence of his conclusion was based on the reasoning of the Court of Appeal in England in ParkingEye Ltd v Somerfield Stores Ltd [2012] EWCA Civ 1338, [2013] QB 840, from which he drew the following points, inter alia: a contract which was not formed for an illegal purpose and which was performed over a period of time might be susceptible to some illegality arising in its performance; whether such illegality tainted the whole contract such that it would not be enforced by the court required consideration of the proportionality of not enforcing the contract and the furtherance of the policy objectives underlying the illegality defence; and the necessity or otherwise for an illegal mode of performance to be adopted, and the question whether illegal performance was the object of the contract are relevant factors, as is the question as to whether the plaintiff has to plead or rely on any illegality as a basis for the claim. 25. On that basis the judge, having found that there was illegality in the performance of the agreement, primarily on the part of Timely, in the arrangements described as the second illegality, and illegality in performance on behalf of Saitek in regard to the fourth illegality, decided that he should not, as a matter of policy, decline relief to Saitek. There had been some illegal conduct, for which the parties largely shared responsibility. It was not conduct which could be described as iniquitous, and it had not resulted in actual criminal or other enforcement proceedings in the PRC. There was no suggestion of any evasion of taxes or duties, and the contraventions were, in a sense, administrative. Saitek did not need to rely on the illegalities as a basis for the claim (inferentially a reference to the principle in Tinsley v Milligan [1994] 1 AC 340) and it would be disproportionate to decline to enforce the payment obligation under the agreement, particularly where it had otherwise been performed. Court of Appeal 26. On the appeal by Timely, Saitek contended in its respondent’s notice that the judge should have found that there are only two situations in which a contract governed by Hong Kong law may be held unenforceable by reason of a foreign illegality: (i) the contract could not be performed in accordance with its terms without the commission of an illegal act, and/or (ii) the contract was entered into for the common purpose of doing an illegal act under the foreign law of the place of performance. But the Court of Appeal decided that it was unnecessary to decide the point, and affirmed the decision on the basis of what it described as the well established principle in Tinsley v Milligan that if a claimant seeking to enforce a contract does not need to rely on his illegal performance then the contract is enforceable. 27. On the facts, the Court of Appeal considered that the judge was plainly right to find that Saitek did not need to rely on the fourth illegality. Timely’s own evidence was that the entire outstanding balance in the running account, the subject matter of Saitek’s claims, was exclusively referable to transactions within the third alleged illegality (which, as the judge had found, involved no illegality); and Timely’s evidence supported Saitek’s case that no reliance had been placed on the fourth illegality for its claims based on the running account. Even if the fourth illegality could be relied on as a ground for not enforcing the agreement, it would not take Timely’s defence any further. By applying the proportionality test as propounded in ParkingEye Ltd v Somerfield Stores Ltd, the Court of Appeal agreed that in the overall circumstances of this case it would be wholly disproportionate to decline to enforce the agreement. V Discussion The illegality defence 28. As noted, the Court of Appeal in this case applied a combination of, first, the principle in Tinsley v Milligan [1994] 1 AC 340 that a claimant to an interest in property which had been acquired in the course of an illegal transaction was not barred from recovery if the claimant was not forced to plead or rely on the illegality, and, second, the proportionality test in ParkingEye Ltd v Somerfield Stores Ltd [2012] EWCA Civ 1338, [2013] QB 840. 29. It is notorious that the common law in England on the scope of application of the illegality defence is difficult and uncertain, and the position has not been made easier by recent, conflicting, decisions of the UK Supreme Court. 30. Shortly before, and soon after, the decision of the Court of Appeal in this case, there were three decisions of the UK Supreme Court on the illegality defence: Hounga v Allen [2014] UKSC 47, [2014] 1 WLR 2889 (claim for unlawful discrimination); Les Laboratoires Servier v Apotex Inc [2014] UKSC 55, [2015] AC 430 (claim for damages on a cross-undertaking) and Bilta (UK) v Nazir [2015] UKSC 23, [2015] 2 WLR 1168 (claim by a liquidator against directors arising out of a VAT fraud). 31. In Hounga v Allenthe claimant was an illegal immigrant from Nigeria. It was held that she was entitled to compensation for the physical abuse inflicted on her by her employers, notwithstanding that an illegal contract of employment formed a material part of her complaint. Lord Wilson said (at [42]; Lady Hale and Lord Kerr agreed) that it was necessary to ask, first, what was the aspect of public policy which founded the defence, and second, whether there was another aspect of public policy to which application of the defence would run counter. It was held that the public policy against trafficking and in favour of the protection of its victims far outweighed the policy behind the illegality defence: [44]-[52]. Lord Hughes (with whom Lord Carnwath agreed) adopted the test of proportionality from Saunders v Edwards [1987] 1 WLR 1116 (at 1134, per Bingham LJ[3]) and concluded that there was an insufficiently close connection between the immigration offence and the claims for discrimination, since the former merely provided the setting or context for the claim (at [59], [67]). 32. The decision in Hounga v Allen was handed down after argument had taken place before an entirely different constitution of the court in Les Laboratoires Servier v Apotex Inc but before judgment was given in the latter case.In that case the claimant’s defence on the claim for damages on a cross-undertaking was that the defendant would have made its profits through illegal conduct. On the facts, the Supreme Court held that there would have been no relevant illegality, but the majority rejected the notion that the court was entitled to make a value judgment about the seriousness of the illegality and the impact on the parties of allowing the illegality defence: [21], per Lord Sumption; while Lord Toulson, on the other hand, considered that Hounga v Allen was authority for a more policy-based approach: [61]-[62]. 33. In Bilta (UK) v Nazirthe UK Supreme Court returned to the question. A panel of seven sat on this appeal, only one of whom (Lord Carnwath) had been on the panel in Hounga v Allen. In Bilta (UK) v Nazirthe Court held that the claim by liquidators of a company against directors for losses caused by a VAT fraud was not barred by the illegality defence because the directors’ conduct was not attributable to the company. Lord Sumption (at [62]) repeated the view expressed by him in Les Laboratoires Servier v Apotex Inc that the illegality defence is based on a rule of law on which the court is required to act in every case to which it applies, and is not a discretionary power on which the court is merely entitled to act, nor is it dependent on a judicial value judgment about the balance of the equities in each case. On the other hand, Lord Toulson and Lord Hodge repeated, in a joint judgment, Lord Toulson’s view that the proper approach was the balancing of policy interests enunciated in Hounga v Allen; and they also cast doubt on the utility of the reliance test in Tinsley v Milligan [1994] 1 AC 340 in the light of criticisms by the Law Commission of England and Wales, by Lord Phillips in Stone & Rolls Ltd v Moore Stephens [2009] UKHL 39, 1 [2009] AC 1391, at [21]-[25] and Lord Wilson in Hounga v Allen at [30] and by McHugh J in the High Court of Australia in Nelson v Nelson (1995) 184 CLR 538, at [27]: see [2015] 2 WLR 1168 at [171]-[173]. 34. Lord Neuberger thought it was not appropriate for the Court to address this difficult and controversial issue without full argument (at [17]) and Lord Mance agreed (at [34]). 35. Faced with these differences of view on the effect and correctness of the leading decisions and an apparent conflict of principle between two differently constituted panels, it is not surprising that the President of the UK Supreme Court, Lord Neuberger, said in Bilta (UK) v Nazir(at [15]) that the proper approach to the defence of illegality needed to be addressed by the UK Supreme Court as soon as appropriately possible in a panel of seven or nine justices. I also agree with the observations of the Chief Justice in relation to the position in Hong Kong. Illegality in the conflict of laws 36. But on this appeal the fundamental question of the scope and content of the illegality defence under the common law of Hong Kong does not fall for decision because, in my judgment, the judge and the Court of Appeal were wrong to have treated the case as if it were purely an internal Hong Kong case, and should have addressed the issues on the basis of well established rules of the conflict of laws, as the Court of Appeal had been invited, and declined, to do. 37. The HKSAR and the PRC are parts of one country, but for the purposes of the conflict of laws they are separate law districts: First Laser Ltd v Fujian Enterprises (Holdings) Co Ltd (2012) 15 HKCFAR 569, at [43]. 38. It was accepted by the parties that the co-operation agreement, and the supplementary agreements, were governed by Hong Kong law. Most of the performance (other than payment) was to take place on the mainland. 39. The basic principles (which have to be read in the light of the fact that Hong Kong and the PRC are separate law districts rather than different countries) are helpfully summarised in Johnston, Conflict of Laws in Hong Kong, 2nd ed, 2012, at para 5.012 (footnotes omitted): The following principles appear to represent Hong Kong law. The underlying rationale is international comity coupled with Hong Kong public policy. First, if the contract is unenforceable under its proper law (whether chosen by the parties or otherwise), then it will not be enforced by the Hong Kong court. The importance of this principle is that it applies to limit the enforceability of the contract regardless of the place of required, intended or actual performance. Moreover, it is irrelevant whether the bar on enforcement is a foreign penal law of the sort which will not be directly enforced by a Hong Kong court. Secondly, if the performance of the contract requires or necessarily involves conduct which is illegal under the laws of the place where it is required to be performed, then it will not be given effect regardless of its proper law. Thirdly, the contract will not be given effect regardless of its proper law “if the real object and intention of the parties [at the time of concluding the contract] necessitates them joining in an endeavour to perform in a foreign and friendly country some act which is illegal by the law of such country notwithstanding the fact that there may be, in a certain event, alternative modes or places of performing which permit the contract to be performed legally” [citing Foster v Driscoll [1929] 1 KB 470, 521, per Sankey LJ]. Fourthly, violation of foreign laws in the actual performance of a contract may, even though not required or initially intended, lead to the unenforceability of the contract before a Hong Kong court, regardless of its proper law. It has recently been stated in England at first instance that a contract will not be enforced if it has been “performed in such a way that one party (or both parties) commits a legal wrong”. It is, however, respectfully suggested that this is to state the principle too rigidly, and that a more flexible approach having regard to the seriousness of the foreign illegality is required to determine whether public policy and comity really require enforcement of the contract to be denied in such a case. Fifthly, the above four principles apply irrespective of whether the illegality under foreign law existed at the time of contracting or arose subsequently. Ralli Brothers v Compania Naviera Sota y Aznar [1920] 2 KB 287 40. In the Ralli Brotherscase the issue was whether Spanish owners could require English charterers to pay the full amount of the unpaid balance of freight for a cargo of jute from Calcutta to Barcelona, when, as a result of changes in exchange rates, the freight exceeded the maximum permitted by Spanish law on pain of penalties. The question arose in an arbitration in England after the Spanish courts had ordered the owners to discharge the cargo against the deposit by the buyers of the freight allowed by Spanish law. The relevant question for the opinion of the court on the case stated was essentially whether payment by the buyers of the permitted amount into court discharged the charterers. It was held that the effect of illegality under Spanish law was that payment of the excess could not be enforced against the charterers even though the charter was governed by English law. 41. All three members of the Court of Appeal referred with approval to the proposition in Dicey, Conflict of Laws, 2nd ed 1908, p. 553, that: “A contract … is, in general, invalid in so far as … the performance of it is unlawful by the law of the country where the contract is to be performed” – “the Dicey Rule” – and Ralli Bros has ever since 1920 been treated as authority for that proposition. In De Beéche v. South American Stores (Gath & Chaves) Ltd [1935] AC 148, 156 it was said (per Viscount Sankey): “it cannot be controverted that the law of this country will not compel the fulfilment of an obligation whose performance involves the doing in a foreign country of something which the supervenient law of that country has rendered it illegal to do.” The same (or a closely similar) principle applies to existing illegality: Toprak Mahsulleri v Finagrain [1979] 2 Lloyd’s Rep 98, 107, per Robert Goff J, approved [1979] 2 Lloyd’s Rep 112, 117 (CA). 42. The underlying basis of the Ralli Brothers decision has been a matter of controversy for most of the 95 years since it was decided. It has been cited in many cases. But it has been applied in very few commercial decisions to hold that performance of a contract or a particular obligation was discharged. Examples include Harrison, Sons & Co Ltd v Jules Cavroy (1922) 12 Ll LR 390; Kursell v Timber Operators and Contractors Ltd [1927] 1 KB 298; Zivnostenska Banka v Frankman [1950] AC 57, 71; Nile Co for the Export of Agricultural Crops v H & JM Bennett [1986] 1 Lloyd’s Rep 555. But in the vast majority of the many reported cases in which reliance has been placed upon the decision, it has not been applied, most often either because performance was not found to be illegal, or because performance was not required at the place of the illegality (applying Kleinwort, Sons and Co v Ungarische Baumwolle Industrie AG [1939] 2 KB 678). 43. Although the Dicey Rule has sometimes been treated as if it were a statutory provision or a strict rule of private international law, it has long been recognised that it forms part of the law of contract. F A Mann, Proper Law and Illegality in Private International Law (1937) 18 BYIL 97 argued (at 111) that the decision in Ralli Brothersturned on the doctrine of impossibility of performance in English law. That continues to be the prevailing view: “… Ralli Bros established a principle of the domestic law of contract relating to discharge by supervening illegality and does not establish a rule of the conflict of laws”: Chitty on Contracts, 32nd ed, 2015, para 30-360; see also Dicey, Morris & Collins, Conflict of Laws, 15th ed, 2012, paras 32-097 et seq. In practice the distinction would only matter if the governing law of the contract were not the law of the forum. Foster v Driscoll [1929] 1 KB 470 44. In this famous case, Sir Harry Foster, the Conservative and Unionist Member of Parliament for Portsmouth Central, entered into an agreement with Driscoll and Miller (shipbrokers), Lindsay (a distiller in Scotland), and Attfield (a retired schoolteacher). The agreement provided that Lindsay would sell 7000/7500 cases of Scotch whisky fob Leith or Glasgow; the syndicate would take delivery of the goods on board a regular line of steamships; and Attfield would lend the syndicate £2500 to buy the steamship Wearhome. The plan was that Attfield would see to the disposal of the whisky in the United States if that were possible, or if not, in Canada, or on the high seas at some point sufficiently near the territory of the United States to facilitate a sale in violation of the United States prohibition laws. The whisky was never shipped, and the financial transaction unravelled, leading to three actions, which raised the question (inter alia) whether Foster was liable to Lindsay on bills drawn by him for the purchase of the whisky. 45. The Court of Appeal decided by a majority that the object to be attained by this agreement was a breach of international comity and therefore that the agreement was contrary to public policy and void. Lawrence LJ said (at 510): …a partnership formed for the main purpose of deriving profit from the commission of a criminal offence in a foreign and friendly country is illegal, even although the parties have not succeeded in carrying out their enterprise, and no such criminal offence has in fact been committed; and none the less so because the parties may have contemplated that if they could not successfully arrange to commit the offence themselves they would instigate or aid and abet some other person to commit it. The ground upon which I rest my judgment that such a partnership is illegal is that its recognition by our Courts would furnish a just cause for complaint by the United States Government against our Government (of which the partners are subjects), and would be contrary to our obligation of international comity as now understood and recognized, and therefore would offend against our notions of public morality… There only remains to be considered the statement in Dicey’s Conflict of Laws, 4th ed., p. 620, that an English contract will only be held invalid on account of illegality if it actually necessitates the performance in a foreign and friendly country of some act which is illegal by the law of such country. … I am of opinion that, in view of the main object of the contract of partnership between the parties in the present case, it is not saved from illegality merely because the partners may have contemplated the event of not being able themselves to import the whisky into the United States, and may have considered the possibility of having to deliver the whisky to the illicit buyers on the high seas or at such other convenient place as might be arranged between themselves and their buyer consistently with their being able to obtain the high price ruling on the illicit market in which they intended to sell the whisky. 46. Sankey LJ said (at 518-519): …the mere fact that a vendor of goods knows that the purchaser proposes to run them into a country where they are prohibited by some revenue law is not sufficient to render the contract of sale illegal, but if beyond mere knowledge the vendor actively engages in an adventure to get the goods into such country, the Court will not assist the parties to the adventure by entertaining or settling any dispute between the parties arising out of the contract … … [T]he Courts of this country are not bound to entertain such actions in view of the obligations of international comity... 47. The principle of the case is that “an English contract should and will be held invalid on account of illegality if the real object and intention of the parties necessitates them joining in an endeavour to perform in a foreign and friendly country some act which is illegal by the law of such country notwithstanding the fact that there may be, in a certain event, alternative modes or places of performing which permit the contract to be performed legally” (at 521-522, per Sankey LJ). Regazzoni v KC Sethia (1944) Ltd [1958] AC 301 48. The decision in Foster v Driscoll was endorsed by the House of Lords in Regazzoni v Sethia. An Indian company, the sellers, agreed to sell to Regazzoni, resident in Switzerland, cif Genoa, 500,000 jute bags. Both parties knew that it was impossible to obtain so large a quantity of the type of jute for early shipment from any source other than India. Both parties knew that the ultimate destination was South Africa, and that Indian law prohibited the export of designated goods from India to South Africa, directly or indirectly, and that the prohibition included the goods which were the subject of the contract. The sellers intended, and Regazzoni knew that they intended, to evade the Indian prohibition by finding a shipper in India who would not ask inconvenient questions about the destination of the goods and who would be able to get the goods out of India. English law was the governing law of the contract. The sellers repudiated the contract. 49. Foster v Driscoll was applied and the contract was held to be unenforceable. Viscount Simonds said (at 323) that it had been conclusively found that the common intention of the parties was to violate the law of India, and it was of no consequence that the documents did not disclose their intention. Lord Reid also said that the crucial fact was that both parties knew that the contract could not be performed without the suppliers procuring a breach of the law of India within the territory of that country. He said (at 318-319): It is … nothing else than comity which has influenced our courts to refuse as a matter of public policy to enforce, or to award damage for the breach of, a contract which involves the violation of foreign law on foreign soil, and it is the limits of this principle that we have to examine. … Just as public policy avoids contracts which offend against our own law, so it will avoid at least some contracts which violate the laws of a foreign state, and it will do so because public policy demands that deference to international comity.” 50. Because these were international cases, the decisions speak of international comity, but the principle applies when the case concerns, as here, separate law districts rather than separate countries. The principle is one of public policy, and these decisions have to be read in the light of the foreign legislation which was involved. In Foster v Driscoll it was the prohibition laws mandated under the 18th amendment to the US Constitution. In Regazzoni v Sethia it was Indian sanctions against South Africa. Plainly, it does not apply to every breach of foreign law. 51. It is not a pre-condition to the application of the Foster v Driscoll/Regazzoni v Sethia principle that the plaintiff is relying on its own illegality. In Foster v Driscoll the seller of the whisky was suing on the bill drawn to pay for the whisky. In Regazzoni v Sethia the buyer was suing the seller for repudiation of the contract to sell the jute. Neither had to rely on the illegality abroad, but neither was allowed to recover. 52. It is possible that the line between foreign illegality and domestic illegality has been blurred in two recent cases on enforcement of cross-undertakings in damages following the grant of injunction ultimately held to have been wrongly granted. In each of the cases it was argued, unsuccessfully, that the defendant was not entitled to damages on the cross-undertaking because its profits would have accrued through illegal conduct abroad. 53. First, in Lilly Icos LLC v 8PM Chemists Ltd [2009] EWHC 1905 (Ch), [2010] FSR 95the claimantscontended that the defendants were not entitled to recover profits for lost sales of generic pharmaceuticals to Canadian internet pharmacies, because the drugs would have been imported into the United States in breach of United States law. Applying dicta of Robert Walker LJ in Ispahani v Bank Melli Iran [1998] Lloyd’s Rep Banking 133, 136-137, Arnold J accepted that where a contract was governed by English law, the relevance of illegal performance under foreign law was limited to the application of the Ralli Brothers and Foster v Driscoll/Regazzoni v Sethia lines of authority: [262]-[264]. But the case was not one of contract and the judge held (at [287]) that (a) the court would not award compensation under a cross-undertaking for the loss sustained by an unlawful business or where the beneficiary of the cross-undertaking has to rely to a substantial extent upon his own illegality in order to establish the loss, and (b) as a matter of international comity it did not matter whether the acts in question were unlawful under English law or foreign law. Arnold J rejected the claimants’ contention on the basis that the defendants did not have to rely on their own illegality in order to establish their loss, since their sales to the Canadian internet suppliers was not illegal, and it was the claimants who were relying on the illegal acts of importation into the United States by others. Consequently, the judge did not have to address the question whether the whole scheme might have fallen within an extended version of the principle in Regazzoni v Sethia. 54. Second, in Les Laboratoires Servier v Apotex Inc,as indicated above (para 32),the claimant resisted a claim for damages on the cross-undertaking on the basis that the defendant’s lost profits would have accrued from sales in England of products manufactured abroad and infringing foreign patents. Arnold J applied his decision in the Lilly Icos case above, and held that the defendant failed because its claim was founded on its own illegality: [2011] EWHC 730 (Pat), [2011] RPC 574. It was conceded in the Court of Appeal that the illegality defence could apply where the source of the profits from sales in England was illegality under foreign law ([2012] EWCA Civ 593, [2013] Bus LR 80, at [69]), but the point was not discussed in the Supreme Court, which decided that patent infringement was not an illegal act for the purposes of the illegality defence: [2014] UKSC 55, [2015] AC 430. In that case neither the Court of Appeal nor the UK Supreme Court articulated the reason why illegality under foreign law was relevant. 55. No principle can be derived from that case (or the Lilly Icos case) which is relevant to the present case, or which suggests that purely domestic rules of illegality can be applied to the consequences of the illegal performance of a contract in a foreign country. 56. It has been suggested (obiter) that a contract which is valid by the governing law of the forum, English law, or in this case, Hong Kong law, may be refused enforcement if it has been “performed in such a way that one party (or both parties) commits a legal wrong”: Barros Mattos Jnr v MacDaniels Ltd [2004] EWHC 1188, [2005] 1 WLR 247, [30] (Laddie J). But, as Johnston, Conflict of Laws in Hong Kong, para 5-012, text at note 80, rightly points out, this obiter suggestion states the principle much too widely. Thus in Re O’Connor’s Bill of Costs [1993] 1 Qd 423 a Queensland solicitor was entitled to recover on his bill of costs although he had done some of the work in New South Wales where he was not entitled to practise, and where therefore his work was illegal (and must have been contrary to an implied term of the retainer). But he was entitled to recover because the contract did not require an illegal mode of performance and he did not seek to enforce any illegal mode of performance by the client. 57. There may nevertheless be cases in which a sufficiently serious breach of foreign law which reflects important policies of the foreign state or separate law district may be such that it would be contrary to public policy to enforce a contract. But there is no basis in authority or principle for holding that every breach of foreign law would come into this category. In Euro-Diam Ltd v Bathurst [1990] 1 QB 1 diamond dealers exported diamonds to the Federal Republic of Germany and when the diamonds were stolen from the warehouse insurers refused to pay on the ground that the dealers had misrepresented their value in an invoice in order to reduce VAT payable in Germany. This was a criminal offence in Germany. The Court of Appeal decided that it was not contrary to public policy to enforce the insurance contract because (among other reasons) the false invoice did not involve any deception of the insurers and the dealers were not relying on the invoice in their action against the insurers. Aspects of this decision must be treated with considerable reserve because its “public conscience” discretionary approach to illegality was disapproved in Tinsley v Milligan [1994] 1 AC 340, at 360-361, and in Apotex[14]-[15] and Bilta at [61], although it may be consistent with Hounga v Allen. 58. I would therefore reject the submission made on behalf of Timely that comity requires the Hong Kong court to treat the contract as unenforceable because of incidental breaches under PRC law in its performance. 59. It follows that, when these principles are applied to the facts of this case, there is no basis for denying relief. First, there is no suggestion that performance of the contract in accordance with its terms was prohibited by PRC law. Second, there was no finding that the parties had agreed to a scheme whereby PRC law would be contravened. The judge refused to allow Timely to broaden the scope of evidence to encompass assertions that the agreement had been formulated with an illegal objective in mind: at [24]. Third, there was no finding that Saitek always intended to commit what is described as the fourth illegality. But, even if it had so intended, it would be extraordinary if it could be regarded as contrary to public policy in Hong Kong to enforce a contract because of breaches in the PRC which the judge found (a) not to be a very serious contravention of the law; (b) not to be conduct which could be described as iniquitous; (c) not to have resulted in actual criminal or enforcement proceedings in the PRC; (d) to have been mere administrative contraventions (at [70]-[72]). There is no principle of law or public policy which would lead to such a conclusion, which would be contrary to commonsense and justice. 60. I would therefore dismiss both appeals. 61. As to costs, I would make an order nisi that the appellants pay the costs of these appeals, such costs to be taxed if not agreed. Should any party wish to have a different order for costs, written submissions should be served on the other party or parties and lodged with the court within 14 days of the handing down of this judgment, with liberty on the other party or parties to lodge written submissions within 14 days thereafter. In the absence of such written submissions, the order nisi will stand absolute at the expiry of the time limited for these submissions. Chief Justice Ma: 62. For the above reasons, the appeals are dismissed. The order as to costs is as set out in para 61 above. Mr Edward Chan SC, Mr Simon Chiu and Mr Albert Chan, instructed by Allen Chan & Co., for the Appellants Mr Richard Zimmern and Mr Jason Yu, instructed by Munros, for the Respondent [1] HCA 2358/2007 & HCA 109/2009 (July 11, 2013). [2] Lam VP, Barma JA and Poon J, CACV 164 & 165/2013 (September 22, 2014). [3] Which seems to have been disapproved in Tinsley v Milligan [1994] 1 AC 340, at 359-361. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: 3. This appeal concerns a charge to profits tax in relation to a payment received by the appellant upon the disposal of a property that had been owned by it as a long-term capital asset. The correctness of that charge depended on the question of whether the appellant changed its intention concerning its ownership of the property so that, when disposing of it, it was carrying on a trade or business. A. Introduction A.1 Background facts 4. The background facts were substantially agreed between the parties and the following summary, taken from the recitation of those agreed facts in the judgment of the Court of Appeal,[1] will suffice for present purposes. 5. The appellant, Perfekta Enterprises Limited, was incorporated in Hong Kong in 1965 and carried on the business of toy manufacturing. In 1969, it acquired most of a building in Kwun Tong, called Vanda Industrial Building, which it used as its manufacturing base in Hong Kong. In 1977, it acquired the remainder of the building and continued to manufacture toys there. This building will be referred to in this judgment as “the Property” and the site on which it was situated as “the Lot”. 6. The appellant’s manufacturing base shifted from Hong Kong to the Mainland from the late 1970s. In 1978, the appellant established a toy factory in Guangzhou, which relocated to Shenzhen in 1985. In 1987, the appellant ceased its manufacturing operations at a factory it owned in Tsuen Wan in Hong Kong and disposed of that factory. 7. From 1991, a series of applications was made in relation to the Lot to enhance its value. In 1991, an application was made to the Town Planning Board for permission to develop a composite industrial and office building on the Lot. This was rejected. A second application to the Town Planning Board, in June 1992, of the same nature was approved. In October 1992, an application was made to the District Lands Office, Kowloon East, for modification of the lease conditions for the redevelopment of the Lot by the construction of a composite industrial and office building. In 1993, a third application to the Town Planning Board in respect of a revised design of the proposed composite building was approved. General building plans for such a building submitted to the Buildings Department in 1993 were initially disapproved but, on re-submission, were approved. In 1994, the District Lands Office indicated a preparedness to recommend a modification of the lease to allow the development of the Lot by way of surrender and re-grant, subject to the payment of a premium of HK$61,420,000. 8. On 21 April 1994, at a meeting of the board of directors of the appellant, a proposal from Cheung Kong (Holdings) Limited (“Cheung Kong”) for the redevelopment of the Property was discussed. The material paragraph of the minutes (“the Minutes”) recorded as follows: “Re: 25 Chong Yip Street, Kwun Tong, Redevelopment plan The Chairman reported that discussions had taken place with representatives of Cheung Kong (Holdings) Limited who had approached [the appellant] with a suggestion of a joint development of the Company’s industrial premises at 25 Chong Yip Street, Kwun Tong. As the premises were acquired 24 years ago (since May 1969) and were in need of upgrading it was recommended that the duscussions [sic] with Cheung Kong proceed. Consideration would have to be given to the leasing of alternative premises for the Company’s manufacturing operation during the period of development if such were to proceed. The proposal from Cheung Kong (Holdings) Limited envisaged the sale to third parties of the newly developed industrial & office building as the manufacturing business of [the appellant] was seen to be a ‘sunset industry’ in Hong Kong and with more production being carried out in PRC, the redevelopment of the site and subsequent sale would be an appropriate method for [the appellant] to realise its long term asset. It was agreed that any joint development program would have to provide for [the appellant] to have an entitlement to take up sufficient space for its own manufacturing requirements in the future. It was decided that for internal purposes any such joint development should be carried out in an entity separate from [the appellant] and that consideration be given to a sale of the property to a wholly owned subsidiary which would subsequently enter into a development venture with Cheung Kong.” 9. On 23 April 1994, the Lot was valued at HK$418,000,000, reflecting its development potential, but without taking into account any premium payable. 10. In May 1994, Cheung Kong proposed to the appellant a joint venture arrangement for the redevelopment of the Property. This led to the execution, by the appellant, Cheung Kong and Great Poka Limited (a subsidiary of Cheung Kong) (“Great Poka”) of a Redevelopment Agreement relating to the Property dated 30 July 1994 (“the Redevelopment Agreement”). The material terms of that Redevelopment Agreement are set out below in Section A.2. Pursuant to clause 3.02 of the Redevelopment Agreement (see below), Great Poka paid a sum of HK$165,104,100, described as an “Initial Payment”, to the appellant as consideration for the right to redevelop the Lot in accordance with its terms. As will presently be seen, under this agreement, the actual joint venture parties were to be Cheung Kong, Great Poka and the appellant’s wholly-owned subsidiary company (then as yet not formed, so it was known simply as “Newco”). 11. In August 1994, the appellant and the Government entered into an Agreement and Conditions of Exchange in respect of the Lot. 12. By an Assignment dated 14 November 1994, the appellant assigned the Lot to its wholly owned subsidiary, Prodes Company Limited (“Prodes”) for a consideration of HK$314,315,900. On 24 November 1994, Prodes, Great Poka and Cheung Kong entered into the New Agreement (“the New Agreement”) (a draft of which had been annexed to the Redevelopment Agreement) for the carrying out of the redevelopment joint venture. The profits of the joint venture, after deduction of expenses, were to be shared equally between Prodes, on the one hand, and Great Poka, on the other. In December 1994, vacant possession of the Lot was given to Great Poka and, in the same month, the appellant’s manufacturing operations at the Property ceased. 13. On 10 February 1999, Prodes, Great Poka and Winrise Limited (“Winrise”) (another subsidiary of Cheung Kong) entered into an Agreement for Sale and Purchase of the Lot, whereby Prodes sold the Lot to Winrise for a consideration of HK$332,661,000. On the same date, those parties also entered into an Agreement relating to the Manner of Payment of the Purchase Price whereby it was agreed that, of the purchase consideration, HK$315,210,899.27 would be paid to Great Poka by way of reimbursement to it of the land premium paid and construction costs incurred in the redevelopment of the Lot. 14. On 11 February 1999, the board of directors of the appellant ratified and approved an Assignment between the appellant and Prodes whereby Prodes assigned its rights under the New Agreement to the appellant and the appellant released and discharged Prodes from its liabilities owing to the appellant. 15. In August 2007, following the redevelopment of the Lot and the sale of the units in the redeveloped building, the appellant received 50% of the balance of the net proceeds in the sum of HK$386,223.21. A.2 The Redevelopment Agreement 16. Material terms of the Redevelopment Agreement included the following: (1) Recital (6): “Subsequent to surrender and regrant of the Lot but prior to redevelopment, [the appellant] intends to transfer the registered and beneficial ownership of the Lot to its wholly owned subsidiary (‘Newco’).” Prodes became the subsidiary known as Newco. (2) Clause 3.02: “[Great Poka] shall pay to [the appellant] as consideration for [the appellant] granting to [Great Poka] the right to redevelop the Lot in accordance with the terms of this Agreement an Initial Payment totalling HK$165,104,100 …”. (3) Clause 5.01: “Within four (4) months after the date of issue of the Conditions [the appellant] shall at its sole cost and expense transfer the registered and beneficial ownership of the Lot to Newco subject to Newco executing the New Agreement referred to in Clause 5.02 below.” (4) Clause 5.02: “[The appellant] hereby agrees and undertakes with [Great Poka] to procure that Newco shall simultaneous [sic] with but immediately after the execution of an assignment of the Lot in its favour enter into a new agreement (‘the New Agreement’) with [Great Poka] and [Cheung Kong] in the form as set out in Appendix II. [Great Poka] and [Cheung Kong] agree and undertake with [the appellant] that they shall enter into the New Agreement with Newco. Should Newco fail to execute the New Agreement as contemplated in this Clause, the assignment of the Lot to it shall not take effect and shall become null and void and [the appellant] shall be deemed to have replaced Newco in its position and continue as the registered and beneficial owner of the Lot under the New Agreement which shall be deemed to have been executed and shall come into operation immediately upon Newco’s failure to execute the New Agreement. …”. The New Agreement referred to in this clause was the agreement that was eventually executed by Prodes, Great Poka and Cheung Kong on 24 November 1994 (referred to earlier at [12] above). (5) Clause 6.01: “This Agreement shall be terminated in the event that:- (i) Newco refuses or fails for whatever reason to execute the New Agreement; …” (6) Clause 6.02: “Where this Agreement is terminated by reason of sub-clause (i) above [the appellant] shall forthwith refund to [Great Poka] (a) all moneys paid to [the appellant] by [Great Poka] as Initial Payment, …”. (7) Clause 8.01: “This Agreement is in the nature of a joint venture and sale and purchase of interest in property. Nothing herein contained shall be deemed to constitute a partnership between [the appellant] on the one part and [Great Poka] and [Cheung Kong] on the other part. …”. B. Procedural history B.1 Determination 17. By a Determination dated 19 May 2011, the Deputy Commissioner of Inland Revenue agreed with the assessor’s view that the Initial Payment to the appellant of HK$165,104,100 should be assessable to profits tax and accordingly increased the appellant’s Additional Profits Tax Assessment for the year of assessment 1994/95 by the net sum of HK$162,624,798 (being the amount of the Initial Payment less professional and legal fees and agreed adjustments), on which the additional tax payable was assessed at HK$26,833,092. B.2 Board of Review 18. The appellant appealed to the Board of Review[2] contending that the Initial Payment was a capital receipt on which profits tax was not chargeable. By a majority, the Board of Review allowed the appellant’s appeal, but on an unusual basis not contended for by either party before it. The Board majority[3] found that, although there was a change of intention on the part of the appellant, there was no relevant change of intention to one of trading and remitted the assessment to the respondent for revision or annulment. The basis of this decision, however, was based on a “reinvestment theory” that posited that the appellant sold the Property and used part of the proceeds to invest in a joint venture to be carried out by Prodes. The Board minority[4] found that there was a change of intention on the part of the appellant to trade as at the date of the Redevelopment Agreement.[5] B.3 Case Stated Appeal to Court of First Instance 19. The respondent appealed by way of Case Stated and questions arising from the decision of the Board of Review submitted by it and also by the appellant were put to the Court of First Instance for determination.[6] By his judgment dated 27 April 2017 (“the CFI Judgment”), Chung J held that the Board majority decision to remit the assessment to the respondent should be set aside, as there was no evidential basis for the “reinvestment theory”.[7] He rejected the appellant’s argument that the appellant did not change its intention to trade since it was Prodes, its subsidiary, that participated in the joint venture and therefore carried out the redevelopment project. B.4 Court of Appeal 20. The appellant appealed to the Court of Appeal,[8] materially contending that Chung J was wrong in holding that the Board minority did not err in finding that it had changed its intention to trade in respect of the Lot. By a majority,[9] the Court of Appeal rejected the appellant’s ground of appeal in relation to there being no change of intention on its part to trade in relation to the Lot. Godfrey Lam J, dissenting in relation to this ground of appeal, held that there was no change of intention on the part of the appellant to trade and would therefore have allowed the appeal and annulled the assessment. In the event, by reason of the Court of Appeal’s unanimity in respect of a separate ground of appeal advanced by the appellant,[10] the appeal was allowed to the extent that the appellant was permitted to amend its ground of appeal against the assessment and the question of the valuation of the Lot for the purpose of assessing the taxable profit and the amount of tax thereon was remitted to the Board for determination. B.5 Leave to appeal to this Court 21. The parties’ respective applications for leave to appeal to this Court were dismissed by the Court of Appeal by its Decision dated 21 August 2018.[11] Their renewed applications to this Court for leave to appeal were heard by the Appeal Committee on 13 November 2018.[12] Leave to appeal was only granted to the appellant on the “question of law” basis and also on the “or otherwise” basis under the Court’s statute.[13] 22. The question of law for which leave to appeal was granted arose because of a dispute between the parties as to the status of the Board minority’s decision in the light of the successful appeal against the Board majority’s decision and was in the following terms: “In the event of a successful case stated/appeal vitiating a majority decision of a board of review/tribunal (‘the Majority Decision’), can the Court substitute the finding in the minority decision (‘the Minority Decision’) for that of the board of review/tribunal, with the consequence that any challenge to the Minority Decision is a challenge to a conclusion from primary facts which can only succeed if it is demonstrated to be unreasonable, illogical or plainly wrong? Or should the matter be remitted to the board of review/tribunal with the opinion of the court for proper findings to be made free from the legal errors identified, unless it can be shown that the conclusion of the Minority Decision is the true and only reasonable conclusion?”[14] 23. Leave to appeal on the “or otherwise” basis was granted to the appellant to contend that “the majority of the Court of Appeal erred in concluding that the Taxpayer changed its intention as to the basis on which it held its property at Chong Yip Street, Kwun Tong, so that it disposed of the said property as a trading asset in the nature of a trade rather than selling the same as a capital asset.”[15] C. The ambit of this appeal C.1 The relevant tax principles 24. It was not in dispute that profits tax would not be chargeable on the Initial Payment if it was a profit arising from the sale of a capital asset and would only be chargeable if it was derived by the appellant from its carrying on of “a trade, profession or business” in Hong Kong: Inland Revenue Ordinance (Cap.112) (“the IRO”), s.14(1). The term “trade” includes every trade and manufacture, and every adventure and concern in the nature of trade: ibid. s.2(1). 25. Since it was common ground that the Property had been held by the appellant as a long-term capital asset prior to its disposal, in order for the Initial Payment to be taxable under s.14(1), it would be necessary to find that there was a change of intention on the part of the appellant such that its intention was to dispose of the Property as part of a trade or business: Church Body of Hong Kong Sheng Kung Hui v Commissioner of Inland Revenue.[16] 26. Disposal of land at an enhanced value would not necessarily indicate an intention to trade by its owner, nor would the expenditure of money on the property in order to enhance its sale price necessarily lead to the conclusion that the landowner was engaging in an adventure in the nature of trade.[17] 27. The question of whether there is an intention to trade is a question of fact and: “… in determining whether an activity amounts to trading, the fact-finding tribunal must consider all the circumstances involved in the activity. It will then have to make a ‘value judgment’ as to whether this constitutes trading and whether the requisite intention to trade can be inferred. Regardless of what is claimed to be the intention subjectively, the question falls to be determined objectively having regard to all the surrounding circumstances.”[18] As that passage shows, in determining an intention to trade, it is important to identify the activity that is said to amount to trading and, in practical terms, to ask the question: “What trading or business venture has the taxpayer embarked upon?” C.2 The approach of an appellate court 28. There was no dispute between the parties that the approach of an appellate court on an appeal on a point of law only is limited by the three propositions identified in Kwong Mile Services Ltd v Commissioner of Inland Revenue.[19] 29. In his judgment in that case, at [37], Bokhary PJ stated: “In an appeal on law only the appellate court must bear in mind what scope the circumstances provide for reasonable minds to differ as to the conclusion to be drawn from the primary facts found. If the fact-finding tribunal’s conclusion is a reasonable one, the appellate court cannot disturb that conclusion even if its own preference is for a contrary conclusion. But if the appellate court regards the contrary conclusion as the true and only reasonable one, the appellate court is duty-bound to substitute the contrary conclusion for the one reached by the fact-finding tribunal. The correct approach for the appellate court is composed essentially of the foregoing three propositions. These propositions complement each other, although the understandable tendency is for those attacking the fact-finding tribunal’s conclusion to stress the third one while those defending that conclusion stress the first two.” 30. It follows that, even if the Board of Review’s decision was vitiated by an error of law, it was not open to Chung J or the Court of Appeal (nor indeed is it open to this Court) to substitute a different conclusion to that drawn by the Board on the primary facts found unless the contrary conclusion is the true and only reasonable conclusion on those primary facts. Instead, the proper course would be to remit the matter to the Board of Review with the court’s opinion for it to make findings on the relevant question (here, the question of whether there was a change of intention on the part of the appellant in respect of the Lot). Both parties recognised this proposition in making their respective submissions on this appeal and, as will be seen, each advocated that their conclusion on the primary facts was the true and only reasonable conclusion that could be drawn. C.3 The parties’ respective contentions 31. The appellant’s case on appeal is straightforward. It being common ground that the Lot was held by the appellant as a long-term capital asset, it was for the respondent to show that there was a change of intention on its part to trade the Lot. The appellant maintains that it had no intention to trade the Lot and that any intention to trade was that of Prodes, a separate legal entity to the appellant. This, it was contended, is embodied in the Minutes, the Redevelopment Agreement and the New Agreement. Since there was no change of intention to trade on the part of the appellant, the Initial Payment received by it was not chargeable to profits tax. 32. For the respondent’s part, it was contended that the true and only reasonable conclusion on the facts was that the appellant did change its intention from capital holding to trading via its subsidiary, Prodes, as held by the Board minority. The contention was that the appellant was embarking on a venture in the nature of a trade in respect of the Lot as a property developer. As will be discussed in further detail later in this judgment (at Section D.3), at the hearing of this appeal it became apparent that the respondent was advancing an alternative argument, not dealt with by the courts below, concerning the nature of the trading venture on which the appellant embarked. D. Did the appellant change its intention in respect of the Lot? D.1 The conclusions in the courts below 33. The Board minority’s finding – consistent with what Chung J and the majority of the Court of Appeal considered to be the correct conclusion – was that the appellant changed its intention to one of trading the Lot “on 30 July 1994 when the appellant entered into the Redevelopment Agreement”.[20] The crux of the Board minority’s reasoning leading to that finding was as follows: “121. The nature of a joint venture involves a commercial, business or trade purpose. Cheung Kong, Great Poka and the appellant entered into the Redevelopment Agreement to engage in trade with a view to making a profit. I do not accept that Cheung Kong and Great Poka joined hands with the appellant to merely enhance the old property for the benefit of the appellant. 122. In my Decision, by the time the appellant had entered into the Redevelopment Agreement, the activities had gone beyond mere enhancement for the purpose of realising the old property for its maximum profit. The Redevelopment Agreement was in express terms binding on the appellant. The appellant was then engaged in trade.”[21] 34. To the argument that it was Prodes, a separate legal entity to the appellant, that engaged in the nature of a trade in respect of the Lot, the Board minority held that the facts showed that Prodes “was the appellant’s alter ego”.[22] 35. In the Court of First Instance, Chung J rejected the appellant’s argument that it did not change its intention and that the redevelopment was carried out by its subsidiary Prodes for three reasons.[23] First, he held that “the use of another legal entity to carry out the redevelopment was meant to be an ‘internal’ arrangement (hence the phrase ‘for internal purposes’)”,[24] referring to the Minutes. Secondly, he held that the appellant’s obligations under the Redevelopment Agreement did not end with the transfer of the Lot to Prodes because, in the event Prodes failed to enter into the New Agreement, clause 5.02 of the Redevelopment Agreement would deem the appellant to have replaced Prodes and, at Great Poka’s request, the appellant would be obliged to execute the New Agreement. Thirdly, he held that clause 8.01 of the Redevelopment Agreement made it clear that the joint venture was between, on the one part, the appellant and, on the other, Great Poka and Cheung Kong. 36. In the Court of Appeal, Cheung JA (with whom McWalters JA agreed) rejected the appellant’s first ground of appeal, which was that Chung J was wrong in holding that the Board minority did not err in finding that the appellant had changed its intention to trade in respect of the Lot.[25] His reasoning was similar to that of Chung J in the Court of First Instance. Thus, he held that: “By the terms of the Redevelopment Agreement together with the receipt of the initial payment, the Board cannot be faulted from forming the view that the only true and reasonable inference is that taxpayer had engaged in trade.”[26] Cheung JA also rejected the appellant’s argument based on the separate legal personality of Prodes. He held that “the use of Prodes was only a method or mechanics of implementing the [appellant’s] intention to trade.”[27] D.2 The intention to trade was that of Prodes, not the appellant 37. For the following reasons, I respectfully disagree with the conclusion reached by the courts below insofar as they held that the appellant changed its intention from disposing of the Lot as a capital asset to one of trading the same. 38. As already noted, it was common ground that the Lot was held by the appellant as a long-term capital asset. The steps taken by it, from 1991 onwards to enhance the value of the Lot by obtaining planning permission, government consent for a variation of the lease and approval of the building plans, are steps that are entirely consistent with the appellant disposing of the Lot as a capital asset for the best price obtainable and do not necessarily evidence its intention to enter into a venture in the nature of a trade.[28] 39. The Minutes show clearly that, on 21 April 1994, the appellant decided to dispose of the Lot by way of a joint venture with Cheung Kong. However, the Minutes also show clearly that an entity “separate” to the appellant was to be used for this purpose and that it was the appellant’s intention to sell the Lot to a subsidiary, which would then enter into the development venture with Cheung Kong. On their face, the Minutes support the appellant’s contention that it did not change its intention with regard to the Lot but rather that it was going to dispose of the Lot by way of transfer to a subsidiary, which would then trade the Lot by redeveloping it with the assistance of Cheung Kong. 40. The fact that a subsidiary of the appellant was to be used for the purpose of the redevelopment of the Lot is important. The appellant and Prodes were two separate legal entities and “the court is not free to disregard the principle of Salomon v. A. Salomon & Co. Ltd. [1897] A.C. 22”,[29] save in limited circumstances.[30] In the present context, one such circumstance might have been where the respondent was able to rely on section 61 or section 61A of the IRO,[31] which deal respectively with “artificial or fictitious” transactions and transactions designed to avoid liability for tax, but there is no suggestion in the present case that those provisions apply and the respondent has not sought to invoke them. Otherwise, as Lord Millett NPJ held in ING Baring Securities (Hong Kong) Ltd v Commissioner of Inland Revenue:[32] “… for tax purposes in this jurisdiction a business which is carried on in Hong Kong is the business of the company which carries it on and not of the group of which it is a member; the profits which are potentially chargeable to tax are the profits of the business of the company which carries it on; and the source of those profits must be attributed to the operations of the company which produced them and not to the operations of other members of the group.” 41. In the circumstances, and with respect, the Board minority erred in holding that Prodes was the “alter ego” of the appellant. Chung J’s reference to the phrase “for internal purposes” in the Minutes cannot change the fact that, absent some reason to ignore the separate corporate personality of Prodes, the operations of the appellant’s subsidiary were not those of the appellant itself. Similarly, Cheung JA’s conclusion that Prodes was “only a method or mechanics of implementing the [appellant’s] intention to trade” [33] wrongly treats the appellant’s subsidiary as a mere nominee or alter ego of the appellant. In my view, the courts below wrongly overlooked the fact that, interposed as it was into the redevelopment, Prodes was a separate legal entity embarking on its own account on a trading venture to redevelop the Lot and that, since it was doing so, it would be redundant for the appellant to engage in that venture. 42. That it was the appellant’s intention that any redevelopment of the Lot be undertaken by its subsidiary, rather than by itself, was underscored by the terms of the Redevelopment Agreement entered into on 30 July 1994 (set out in Section A.2 above): (1) That agreement recited that, subsequent to the surrender and re-grant of the Lot but “prior to redevelopment”, the appellant “intends” to transfer ownership of the Lot “to its wholly owned subsidiary”, defined as “Newco”. (2) By clause 3.02, the appellant granted the right to redevelop the Lot to Great Poka in consideration of the Initial Payment. (3) Within four months of the grant of the new lease, the appellant was to transfer the registered and beneficial ownership of the Lot to Newco (clause 5.01). (4) By clause 5.02, the appellant undertook to procure Newco to enter into the New Agreement with Great Poka and Cheung Kong. That New Agreement set out the terms on which the Lot was to be redeveloped. It was the appellant’s subsidiary that was to embark on that venture, not the appellant. Although clause 5.02 also provided that, in the event Newco did not enter into the New Agreement, the appellant would do so, this was clearly a provision for the benefit and protection of Great Poka, since that company was granted a power of attorney to execute the New Agreement on behalf of the appellant. (5) The Redevelopment Agreement would be terminated in the event that Newco did not execute the New Agreement (clause 6.01) and, in that case, the appellant would be obliged to refund the Initial Payment (clause 6.02). 43. The provisions of the Redevelopment Agreement referred to demonstrate that the appellant’s intention, as reflected in the Minutes, was carried into effect by way of the joint venture to redevelop the Lot. Prodes was incorporated to fulfil the role of Newco. The appellant duly assigned the Lot to Prodes on 14 November 1994 and, on 24 November 1994, Prodes duly entered into the New Agreement with Great Poka and Cheung Kong. Significantly, the appellant was not a party to the New Agreement and the provisions of clause 5.02 by which the appellant might have become a party to it were never triggered so that the intention must have remained that Prodes would carry out the redevelopment rather than the appellant. The profits of the joint venture envisaged by the New Agreement were to be shared equally between Prodes, on the one hand, and Great Poka and Cheung Kong, on the other. 44. In these circumstances, the submission of Mr Paul Shieh SC,[34] that on a proper reading of the Redevelopment Agreement, it is clear that irrespective of whether Prodes had any intention to trade the appellant itself had the requisite intention to trade, cannot be accepted.[35] It emerged in the course of the hearing that there were two variants of this argument being advanced on behalf of the respondent. The first, more straightforward argument, is that the appellant was embarking on a venture in the nature of a trade by itself participating in the joint venture. The second alternative (which will be addressed in Section D.3 below) is that the appellant was engaged in a trade in the nature of procuring its subsidiary to enter into the joint venture agreement. 45. As regards the respondent’s first line of argument, the whole structure of the contractual arrangements was for Prodes to carry out the property redevelopment and for the appellant to drop out of the picture. That this is what in fact happened supports the inference that this remained the appellant’s intention throughout. Mr Shieh’s reliance on the fact that the Redevelopment Agreement did not provide for the sale of the Lot by the appellant to Cheung Kong or Great Poka, or to Prodes,[36] does not alter the fact that the arrangement was for the appellant to assign the Lot to Prodes for it to carry out the joint venture. Nor does it matter that Prodes was incorporated after the execution of the Redevelopment Agreement,[37] since the critical question is one of the intention of the appellant in relation to the Lot and that intention is to be gleaned not only from the terms of the agreement but also the events subsequent to its execution. 46. There is nothing in these facts to suggest that the appellant’s intention to dispose of the Lot to its subsidiary and to use that subsidiary as the vehicle to carry out the redevelopment joint venture of the Lot ever changed. As Mr Clifford Smith SC[38] submitted, the power of procuring its subsidiary to enter into the New Agreement lay entirely within the hands of the appellant, so that, even after the Redevelopment Agreement was executed, it would only be in the event that the appellant positively changed its intention from one of disposing of the Lot as a capital asset to one of trading the Lot itself that it might, by reason of clause 5.02, become a party to the New Agreement. As we have seen, it never did. 47. It is also material to note that the consideration received by the appellant for its grant of the redevelopment right to Great Poka (i.e. the Initial Payment of HK$165,104,100) and on its disposal of the Lot to Prodes (the sum of HK$314,315,900) amounted to a total of HK$479,420,000 being the amount at which the Lot was valued on 23 April 1994, taking into account the land premium paid (HK$61,420,000). In other words, it was intended that the appellant would receive no more than if it had sold the Lot, having paid the land premium, in the open market to a third party to redevelop. D.3 The respondent’s alternative trading argument 48. In the course of the hearing, Mr Shieh sought to advance an alternative argument as to the nature of the venture on which the appellant was engaged. The argument is found in the following submission in the Case of the Respondent: “The case is not one whereby [the appellant] had realised the capital value of the Lot and then left it completely to the independent judgment or discretion of Prodes to decide whether to redevelop the Lot. The whole deal was preordained by the terms of the Redevelopment Agreement (to which only [the appellant], Great Poka and Cheung Kong were parties) before Prodes had come into the picture. [The appellant] either (i) procured Newco to sign the New Agreement or (ii) took up the role of a joint venture partner with Cheung Kong carrying out the redevelopment/trade itself. Either way, [the appellant] was in a business venture as a trader.”[39] 49. The point at (ii) in the passage quoted is the respondent’s first line of argument, addressed in Section D.2 above. The respondent’s alternative argument, at (i) in the passage quoted, is that the appellant was engaged in a trade or business of procurement, whereby it procured a subsidiary to enter into the property redevelopment joint venture and undertook various positive obligations under the Redevelopment Agreement. The profit arising on the venture was said to be the Initial Payment and Mr Shieh identified its source as the appellant’s ability to grant contractual rights and undertake obligations. 50. This argument is different to the basis on which the Court of Appeal majority, Chung J or the Board minority decided in favour of the respondent and Mr Shieh candidly acknowledged that the alternative point he was advancing in this Court had not been raised in this form either before the Court of First Instance or the Court of Appeal. 51. Leaving aside the question of whether it was open to the respondent to advance a wholly new point on appeal to this Court, the alternative trading argument is, with respect, untenable. In advancing the argument, Mr Shieh invited us to look at the transaction commercially and as a matter of economics. However, doing so, the appellant’s alternative argument lacks reality. It is to be remembered that one is asking the question, “What trading or business venture has the taxpayer embarked upon?” Here, the appellant was a toy manufacturer. It had been holding long-term a capital asset in the form of the Property. The substance of the transaction was that the appellant was disposing of that capital asset. It had enhanced the value of the asset prior to disposal by way of assignment to a subsidiary, who was intended to be a participant in a joint venture with Great Poka and Cheung Kong to redevelop the Lot, taking advantage of its enhanced value. The appellant was not intended to be a participant in that joint venture unless, by actions within its own control, it decided that its subsidiary would not participate. It was no part of the appellant’s business to act as a procurer of joint venture participants for property developers. The rights it granted and the obligations it undertook were designed to enable the appellant to dispose of the Lot to its subsidiary and then to drop out of the joint venture project to redevelop the Lot. None of those rights or obligations supports a conclusion that the appellant was intending to embark on a venture in the nature of a trade in respect of the Lot. D.4 The true and only reasonable conclusion on the facts 52. For these reasons, in my view, the true and only reasonable conclusion on the undisputed evidence and primary facts is that the appellant did not change its intention in relation to the Lot and did not enter into a venture in the nature of a trade in disposing of it. In this regard, I respectfully agree with the conclusion of Lam J, dissenting, in the Court of Appeal.[40] E. Not necessary to address the question of law for which leave granted 53. In light of the conclusion reached in Section D.4 above, I would substitute the contrary conclusion that the appellant did not engage in a trade in disposing of the Lot in place of that reached by the Board of Review and the courts below. It is therefore unnecessary to address the question of law for which leave to appeal was granted. F. Conclusion and disposition 54. For the above reasons, I would therefore allow the appellant’s appeal and annul the assessment. 55. I would direct any submissions as to costs be filed in writing within 14 days of the date of the handing down of this judgment, to be dealt with on the papers. Mr Justice Cheung PJ: 56. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Gummow NPJ: 57. I agree with the judgment of Mr Justice Fok PJ. Mr Clifford Smith SC and Mr Justin Lam, instructed by Pang & Associates, for the Appellant Mr Paul Shieh SC and Mr Mike Lui, instructed by the Department of Justice, for the Respondent [1] CACV 115/2017, [2018] HKCA 301, Judgment dated 1 June 2018 (“CA Judgment”), at [2]. [2] In Board of Review Case No. D18/15, under reference B/R 18/11, Decision dated 20 October 2015 (“BOR Decision”) [3] Mr Miu Liong, Nelson, Barrister-at-Law, and Mr Mark Richard Charlton Sutherland, Barrister-at-Law, FCIArb. [4] The Chairman, Mr Kenneth Kwok Hing Wai, SBS, BBS, SC, JP. [5] BOR Decision at [123]. [6] HCIA 1/2016, before Chung J. [7] Neither party had sought to persuade the judge to uphold this. [8] CACV 115/2017, [2018] HKCA 301, Judgment dated 1 June 2018 (“CA Judgment”). [9] Cheung & McWalters JJA. [10] Concerning the appellant’s right, on the assumption the Lot was part of its trading stock, to have the value of the land disposed of taken into account in calculating its profit. [11] CACV 115/2017, [2018] HKCA 544. [12] FAMV 56 & 57/2018, [2018] HKCFA 55 (Ribeiro, Fok & Cheung PJJ). [13] Hong Kong Court of Final Appeal Ordinance (Cap.484), s.22(1)(b). [14] FAMV 56 & 57/2018, [2018] HKCFA 55, Determination dated 13 November 2018 at [2]. [15] Ibid. at [1]. [16] (2016) 19 HKCFAR 54 at [23], [44]-[47], [73]. [17] Ibid. at [23], [48]-[49], [73]. [18] Ibid. at [50]. [19] (2004) 7 HKCFAR 275. [20] BOR Decision at [123]. [21] Ibid. at [121]-[122]. [22] Ibid. at [127]. [23] CFI Judgment at [52]. [24] Ibid. at [46]. [25] CA Judgment at [5.2] and [6.1] to [6.23]. [26] Ibid. at [6.16]. [27] Ibid. at [6.22]. [28] Church Body of Hong Kong Sheng Kung Hui v Commissioner of Inland Revenue (supra.) at [23], [44], [48]-[49]. [29] Adams v Cape Industries plc [1990] Ch 433 at 536 per Slade LJ. [30] Prest v Petrodel Resources Ltd [2013] 2 AC 415 at [35] per Lord Sumption JSC. [31] As was the case, for example, in Commissioner of Inland Revenue v Tai Hing Cotton Mill (Development) Ltd (2007) 10 HKCFAR 704. [32] (2007) 10 HKCFAR 417 at [134]. [33] CA Judgment at [6.22]. [34] Appearing with Mr Mike Lui for the respondent. [35] Case of the Respondent at [22]. [36] Ibid. at [22.2]-[22.3]. [37] Ibid. at [22.4]. [38] Appearing with Mr Justin Lam for the appellant. [39] Case of the Respondent at [22.4(d)] (emphasis in original). [40] CA Judgment at [30]. 
Mr Justice Fok PJ: 1. This is the judgment of the Court. This appeal arises out of the appellant’s conviction for the murder of Madam Yeung Sau-yu (“Madam Yeung”) and raises the questions of law set out below concerning the trial judge’s directions to the jury on the defence of provocation. A. Introduction A.1 The killing of Madam Yeung and the appellant’s trial and conviction 2. The appellant was charged with murdering Madam Yeung on 12 September 2009 at the premises in which he cohabited with her in Cheung Sha Wan. Madam Yeung’s body was discovered by the police on the evening of 12 September 2009 when they gained access to the premises. She was certified dead at the scene. An autopsy report prepared by a forensic pathologist stated the cause of Madam Yeung’s death to be multiple cut wounds leading to death as a result of profuse haemorrhage and damage to the vital organs. The evidence was that there were at least 213 cut wounds all over her body and also signs of manual strangulation, which may have contributed to her death. The forensic pathologist agreed with the appellant’s trial counsel that the nature of the attack and the nature and number of wounds indicated “a frenzied attack” on Madam Yeung. 3. The appellant had departed Hong Kong for the Mainland after Madam Yeung was killed but on his return to Hong Kong on the evening of 13 September 2009 he was arrested. He admitted killing Madam Yeung but raised the defence of provocation. It was his case that he was provoked into killing Madam Yeung because of her infidelity and her taunting him. His offer to plead guilty to manslaughter on the basis of provocation was not accepted by the prosecution and his trial for murder[1] proceeded before V. Bokhary J (“the Judge”) and a jury in March 2014. 4. The appellant was aged 45 at the time of his trial. He had been married in the Mainland and had a daughter with his former wife, who left him to come to Hong Kong in 2004. He met Madam Yeung, whose husband was in Hong Kong, in 2005 and became intimate with her in 2007. The appellant, his daughter and Madam Yeung came to settle in Hong Kong in April 2009. According to the evidence, which the jury were directed to accept (a matter to which we return in Section A.3 below), shortly thereafter, the relationship between the appellant and Madam Yeung deteriorated. He suspected her of having an affair and of lying about what she was doing and who she was with. Matters came to a head on the evening of 11 September 2009. He had come home to find evidence that Madam Yeung had had sexual relations with another man in their bed. On her return home they argued and, although they briefly had intercourse, the argument continued and escalated: Madam Yeung suggested that the appellant should jump to his death; on being challenged as to her lying as to who she had dinner with, Madam Yeung taunted the appellant as to his looks and sexual prowess compared to the other men she had been with; on confronting her with a used condom he had found discarded at the premises that day, Madam Yeung thrust this into his mouth and told him to eat it. The appellant then slapped Madam Yeung and squeezed her neck for about 10 seconds, after which she continued to taunt him about his former wife’s infidelity and suggested he was not in fact the father of his daughter. On hearing this, the appellant said he became enraged and fetched a chopper from the kitchen and, in a fury, chopped randomly at Madam Yeung. 5. By their unanimous verdict convicting the appellant of murdering Madam Yeung, the jury were satisfied that the prosecution had proved that the defence of provocation did not apply in the present case. The Judge accordingly sentenced the appellant to life imprisonment. Since the principal issues in this appeal concern the Judge’s directions to the jury on the issue of provocation, it will be necessary to deal with that direction in some detail later in this judgment. Before doing so, it is convenient to summarise in general terms, by way of introduction, the relevant law as to the defence of provocation. A.2 The defence of provocation 6. The doctrine of provocation operates as a partial defence to a charge of murder, reducing the offence of murder for which the penalty of life imprisonment is fixed by statute to the lesser offence of manslaughter for which the penalty is in the discretion of the sentencing judge. Originally a creature of the common law, the doctrine is now subject to legislation. In Hong Kong, section 4 of the Homicide Ordinance (Cap.339) (which is in identical terms to section 3 of the Homicide Act 1957[2]) provides: “Where on a charge of murder there is evidence on which the jury can find that the person charged was provoked (whether by things done or by things said or by both together) to lose his self-control, the question whether the provocation was enough to make a reasonable man do as he did shall be left to be determined by the jury; and in determining that question the jury shall take into account everything both done and said according to the effect which, in their opinion, it would have on a reasonable man.” 7. As Lord Diplock observed in DPP v Camplin [1978] AC 705,[3] the statutory provision brought about two important changes in the common law. The first was to abolish previous common law rules on what could constitute provocation and, in particular, the rule that words alone were insufficient to do so. The second was that if there is any evidence that the defendant at the time of the act causing death lost his self-control as a result of some provocation, however slight it appeared to the judge, he was obliged to leave it to the jury to decide, as a matter of opinion and not law, “whether a reasonable man might have reacted to that provocation as the accused did”. 8. Statutorily so defined, the ingredients of the defence of provocation have been the subject of authority from the House of Lords and the Privy Council, which has been followed by this Court, and are not in issue in this appeal. In Ho Hoi Shing v HKSAR (2008) 11 HKCFAR 354, Chan PJ, giving the reasons for the Court’s decision, held (at [31]): (1) The defence of provocation involves the determination of two issues: was the accused provoked to lose his self-control, and was the provocation enough to make a reasonable man do as he did? (2) The first issue is a subjective question involving a finding of fact upon the evidence before the court. (3) The second issue is an objective question requiring the application of the standard of self-control of an ordinary person of the same age and sex as the accused. (4) This is the effect of section 4 of the Homicide Ordinance as explained in DPP v Camplin, Luc Thiet Thuan v The Queen [1996] 2 HKCLR 45 and A-G for Jersey v Holley [2005] 2 AC 580. (5) If, on the first issue, the jury are sure that the accused was not provoked to lose his self-control, that is the end of the defence. (6) However, on the second issue, if they accept that the accused was or might have been provoked, they then have to consider whether a person having the power of self-control to be expected of an ordinary person (of the same age and sex as the accused) would have lost his self-control and reacted in the way the accused did. (a) If they consider that such a person would or might have behaved as the accused did, the defence of provocation is made out. (b) If they are sure that such a person would not have reacted in a similar way, the defence fails. 9. The subjective question as to whether the appellant was provoked to lose his self-control is not in issue in the present appeal. Nor is it in issue that, as the Court held in Ho Hoi Shing (at [33]), there are two elements of the objective question, namely: (1) First, an assessment of the gravity of the provocation, which involves a consideration of everything both done and said as required by s.4 of the Ordinance; and (2) Second, an assessment of how a person with ordinary power of self-control would or might react to provocation of that gravity. 10. In so holding, the Court was approving and following the explanation of this aspect of the defence of Lord Nicholls of Birkenhead, who in giving the majority judgment in A-G for Jersey v Holley, adopted the comments of Lord Hobhouse of Woodborough in his dissenting judgment in R v Smith (Morgan) [2001] 1 AC 146 at p.185F. Because of the particular issues raised in this appeal, it is convenient to set out Lord Hobhouse’s summary in R v Smith (Morgan) of the constituents of the defence of provocation (at p.205C-E), which was in the following terms: “(a) The defendant must have been provoked (whether by things done or by things said or by both together) to lose his self-control and kill or do whatever other act is alleged to render him guilty of murder. (b) This is a factual question upon which all relevant evidence is admissible including any evidence which tends to support the conclusion that the defendant either may have or did not lose his self-control. (c) If the jury conclude that the defendant may have been provoked to lose his self-control and do as he did, the jury should, as an exercise of judgment, but taking into account all the evidence, form a view as to the gravity of the provocation for the defendant in all the circumstances. (d) Finally, the jury should decide whether in their opinion, having regard to the actual provocation ((a) and (b) above) and their view as to its gravity (c above), a person having ordinary powers of self-control would have done what the defendant did.” 11. Constituents (a) and (b) in the above passage relate to the subjective question of loss of self-control, whilst constituents (c) and (d) relate to the objective question. As will become apparent, the focus of this appeal was on those two constituents (or elements) of the objective question. However, it is not suggested that there is any error in the way in which they have been described in Ho Hoi Shing or the authorities it followed and applied. Instead, this appeal is concerned with the manner in which a judge is to direct a jury in respect of them. A.3 The Judge’s directions to the jury on provocation 12. In her summing up to the jury, the Judge initially introduced the defence of provocation “in general terms, leaving until later the evidence and submissions relating to provocation in the present case.” In this introductory part of her summing up, the Judge said: “Very briefly, the matters put forward by the defence as provocation consists essentially of the deceased’s affair with another man, coupled with a deterioration of relationship between the accused and the deceased; the suspicion that she was having sexual relations with another man; his discovery of the used condom and the bed in a mess, with pubic hair on the bed sheet and a man’s hair on the pillow; her flaunting the affair and taunting him with it saying that it was none of his business; her thrusting a used condom in his mouth and telling him to eat it; comparing his appearance and sexual prowess unfavourably with other men by saying that he was no longer exciting to be with anymore; her saying that his daughter had been fathered by another man; and her refusal to say who she was having sexual relations with. That is a rough summary; the details are in his testimony, which I will come to later.” 13. The Judge then proceeded to direct the jury on the issue of provocation, reminding the jury that this only arose if they felt sure that the appellant unlawfully killed Madam Yeung with the requisite intent for murder. On the first issue, the Judge took the unorthodox step of directing the jury as follows: “On the issue of provocation, the first question is whether the accused was provoked into a sudden and temporary loss of self-control by things said and done by the deceased. The facts are for you, but the law is for me, and I have come to the conclusion, as a matter of law, that there is no evidence on which it can be safely said that the accused was not provoked into a sudden and temporary loss of self-control. So I repeat. I have come to the conclusion, as a matter of law, that there is no evidence on which it can safely be said that the accused was not provoked into a sudden and temporary loss of self-control. So I direct you to proceed on the basis that the accused was so provoked into a sudden and temporary loss of self-control. That is a direction of law which you are bound to follow.” This was a misdirection by the Judge, since this first question was a matter for the jury to determine as a matter of fact and not for the Judge as a matter of law, not least because the veracity of the appellant’s evidence was in issue. However, it was an error regarded by the majority of the Court of Appeal to be in favour of the appellant and hence immaterial.[4] 14. As to the second issue, the Judge directed the jury as follows: “It is a question of fact for you to decide as the sole judges of fact and it is whether an ordinary person of the accused’s age and sex would have been provoked to do as he did. By the expression ‘an ordinary person’, the law means a person who has the power of self-control to be expected of an ordinary sober person who is of the accused’s age and sex. Whenever I say an ‘ordinary person’, that is what I will mean. What is to be expected of an ordinary person? The law expects people to exercise control over their emotions. If, for example, a person has an unusually volatile, excitable or violent nature, he cannot rely on that as an excuse. So the ordinary person in this context is a person who is not exceptionally excitable or given to violence, but possess[ed] of such powers of self-control as everyone is entitled to expect that his fellow citizens will exercise in society as it is today. Therefore, when considering this question you must take into account everything that was said and done according to the effect which, in your opinion, it would have on that ordinary person. If you are sure that what was done and said would not have caused an ordinary, sober person of the accused’s age and sex to do as he did, the prosecution will have disproved provocation. Then, providing the prosecution has made you sure of the ingredients of the offence of murder, you will find the accused guilty of murder. If, on the other hand, you think that what was done and said would or might have caused an ordinary, sober person of the accused’s age and sex to do as he did, you will not find the accused guilty of murder, but find him guilty of the lesser offence of manslaughter.” 15. The Judge then summarised the evidence of the witnesses. In respect of the appellant’s evidence, the Judge began by summarising his evidence-in-chief, including the evidence of his background, age, education and first marriage, his daughter born of that marriage, his relationship with Madam Yeung in the Mainland and his coming to Hong Kong with his daughter and Madam Yeung in April 2009. The Judge then summarised the evidence of the deterioration in the relationship between the appellant and Madam Yeung after their arrival in Hong Kong and his suspicions of her infidelity. She then addressed in detail the evidence of the events of 11 September 2009 and the confrontation between the appellant and Madam Yeung at home that evening, including the details of what the appellant said Madam Yeung had said and done to him, leading to his chopping her in a rage. The Judge then summarised the thrust of the prosecution’s cross-examination of the appellant and said this: “Members of the jury, the accused denies telling any lies when interviewed, and in all the circumstances there is no basis on which you can safely conclude that he had lied as to what was said and done between him and the deceased. In respect of what was said and done between the accused and the deceased, I direct you that the only safe course is for you to proceed on the basis of the account given by him in the witness-box, with such support the defence draws from what he did say when interviewed. Having so directed you, I do not need to deal with the re-examination of the accused by his own counsel save to remind you to take it fully into account as you see fit.” This was also a misdirection by the Judge, since it was for the jury to decide whether to accept that the evidence of the appellant, or some parts of it, was or might be true (a matter put in issue by the prosecution). Again, however, it was an error regarded by the majority of the Court of Appeal to be in favour of the appellant and hence immaterial.[5] 16. The Judge then directed the jury as to the evidence relevant to the appellant’s intent in killing Madam Yeung and concluded her instruction to them on the issue of provocation as follows: “I have already directed you to proceed on the basis that he had been provoked into a sudden and temporary loss of self-control; but that on its own would not reduce what would otherwise be murder to the lesser offence of manslaughter. There is still a question of whether an ordinary person of the accused’s age and sex would have done what he did. If you feel sure that an ordinary person of his age and sex, even in the accused’s circumstances and even having undergone all the provocation which the accused told you that he had undergone at the time of the killing and leading up to it over a considerable period, would not have done what the accused did, then the doctrine of provocation would not operate to reduce murder to manslaughter. In that situation, if you feel sure that the accused intended to kill the deceased or at least to cause her really serious injury, you will find him guilty of murder. As I have told you, by the expression ‘an ordinary person’, the law means a person who has the power of self-control to be expected of an ordinary sober person who is of the accused’s age and sex.” A.4 The Court of Appeal 17. The appellant appealed against his conviction to the Court of Appeal.[6] Two grounds of appeal were pursued, one of which related to the defence of provocation.[7] That ground was that the Judge failed to direct the jury to assess the gravity of the provocation when assessing the second, objective, limb of the defence of provocation, namely whether the provocation was enough to make a reasonable man do as he did. 18. The majority of the Court of Appeal (Lunn VP and Macrae JA) concluded that the Judge’s direction, set out in paragraph [16] above, did bring home to the jury that, in the context of the second limb of the defence of provocation, they were required to have regard, as the first element, to the gravity of the provocation to the appellant himself.[8] 19. McWalters JA, however, dissenting, held that the Judge’s direction was inadequate[9] because the Judge did not direct the jury to consider and form a view as to the gravity of the provocation to which the appellant had been subjected (constituent (c) of Lord Hobhouse’s summary of the defence) as a discrete first step before going on to determine the sufficiency of the provocation as regards the loss of self-control (constituent (d) of Lord Hobhouse’s summary).[10] A.5 The certified question of law 20. This difference of opinion in the Court of Appeal formed the basis of the appellant’s application for leave to appeal to this Court. The Court of Appeal, again by a majority, with McWalters JA dissenting, refused to certify the following question as one of great and general importance, namely: “Is it a correct approach in law to combine the two steps (being constituents (c) and (d) in R v Smith (Morgan) [2001] 1 Cr App R 13 [224] per Lord Hobhouse), into a single step?”[11] 21. On the appellant’s further application, the Appeal Committee granted leave to appeal on the following question of law, namely: “Whether, in respect of the second limb of the defence of provocation (namely as to ‘whether the provocation was enough to make a reasonable man do as he did’, Homicide Ordinance, Cap.339, s.4), it is a misdirection of law not to direct the jury to form a discrete view as to the gravity of the provocation on the defendant, having the particular attributes he has, separately from and before determining whether, objectively, having regard to the actual provocation and their view as to its gravity, a person with ordinary powers of self-control would have done what the defendant did?”[12] A.6 A further question of law arising in the course of the appeal 22. In the course of Mr William Tam SC’s submissions on behalf of the respondent, in the light of an answer given by him to a question from the bench, the Court became concerned that a new and additional question of law of importance was raised that ought to be determined in this appeal. Having adjourned briefly to consider the position, the Court exceptionally granted leave to appeal on the following further question, namely: “Whether in directing the jury in relation to the question in s.4 [of the Homicide Ordinance], Cap.339 as to whether the provocation was enough to make a reasonable man do as the defendant did, it may in certain cases be incumbent on the trial judge to direct the jury that the words ‘do as he did’ refer to the formation of an intent to kill or cause really serious bodily harm rather than the precise form of physical reaction? If so, whether the present case is such a case?” B. The contentions on appeal 23. The appellant contends for affirmative answers to the certified question set out in paragraph [21] above (“the gravity question”) and also to the further question set out in paragraph [22] above (“the ‘do as he did’ question”), whilst the respondent contends for negative answers to both. B.1 The appellant’s contentions on the gravity question 24. In relation to the gravity question, it is submitted in the Appellant’s Case[13] that the jury are “to be directed to consider the ‘objective’ limb of the provocation in the two distinct steps identified at sub-paragraphs (c) and (d) in R v Smith (Morgan).” Specifically, it is submitted that the jury are, first, to weigh up how serious the provocation was for the defendant “as a distinct step, and to take into account whether what was said and/or done might have affected the defendant more than other people.” Then, it is “only after the jury has weighed up the seriousness of the provocation on the defendant” that it is to consider, “having regard to the actual provocation and [its] view on how serious that provocation was for the defendant, whether an ordinary person would have acted as the defendant did in those circumstances.” 25. This approach is said to be supported by the Crown Court Bench Book in use in England and Wales prior to the abolition of the defence of provocation and its replacement by a defence of loss of control by the Coroners and Justice Act 2009, sections 54 to 56. It is also supported by the reasoning of McWalters JA in the Court of Appeal (which reasoning the appellant adopts) at paragraphs [87] and [92] of his judgment and in his analysis of why he considered the Judge’s direction to the jury to be inadequate (at paragraphs [98(iii)] to [100] of his judgment). 26. The crux of McWalters JA’s reasoning is to be seen in the following two passages in his judgment where he is addressing the second objective question which the jury must consider in relation to the issue of provocation: “87. … at this stage the issue is gaining an appreciation of the impact that the provocation had on the defendant as a precursor to determining whether it was enough to make an ordinary person do as the defendant did. This is what is described by Lord Hobhouse in his sub-paragraph (c), namely the step to be taken by the jury of forming a view as to gravity of the provocation to the defendant. … 92. … by combining Lord Hobhouse’s sub-paragraphs (c) and (d) into one direction on the objective test the risk is created that the importance of sub-paragraph (c) may not be properly appreciated and the role it plays in providing subjective context in the determination of the self-control issue with which sub-paragraph (d) is concerned may not be properly brought home to the jury. For this reason it is important that this task of the jury is seen as a discrete step that must be taken before the jury goes on to determine the sufficiency of the provocation vis-à-vis the loss of self-control. If they do not take this step separately from the final step then their determination of the issue of self-control may be at risk of being flawed by not taking full account of the context.” (Italics in original; underlining added.) B.2 The contentions on the ‘do as he did’ question 27. In relation to the ‘do as he did’ question, adopting the point identified in the further question of law set out above, it was contended on behalf of the appellant that in cases where the precise form of the physical reaction by the defendant to the provocation is of an extreme or unusually prolonged or violent nature, it is incumbent on a trial judge to direct the jury in certain cases that the words “do as he did” in section 4 of the Homicide Ordinance means the forming by the defendant of the intent to kill or cause really serious bodily harm, rather than the precise form of the physical reaction. It was further contended that it was incumbent on the Judge in this case to make such a direction because the appellant’s physical reaction to the provocation in the form of 213 cut wounds was so extreme and unusually prolonged or violent that there was a real risk that the jury would, without a specific direction to disregard the precise form of the physical reaction, inevitably find on the second, objective, question that no ordinary person would have reacted as the appellant did. 28. It was contended on behalf of the respondent, on the other hand, that section 4 of the Homicide Ordinance was drafted in terms which framed the objective test for the jury by reference to the form of physical reaction, as opposed to the formation of an intent to kill or cause really serious bodily harm, by the use of the express words “do as he did”. It was further contended that this conclusion is supported by English and Privy Council decisions on section 3 of the Homicide Act 1957 (and equivalent provisions) and Hong Kong decisions on section 4 of the Homicide Ordinance. C. The gravity question 29. The ultimate question in relation to the gravity question is whether the reasoning of McWalters JA in the Court of Appeal, set out in paragraph [26] above, is sound. 30. In seeking to support that reasoning, Mr McCoy SC,[14] on behalf of the appellant, submitted that it is consistent with the words used by Lord Hobhouse in describing constituent (c) of the defence of provocation in R v Smith (Morgan), namely that “the jury should, as an exercise of judgment, … form a view as to the gravity of the provocation for the defendant in all the circumstances” (underlining added). This requires, Mr McCoy submitted, constituent (c) to be assessed subjectively and then the question asked, objectively, as constituent (d), whether the hypothetical ordinary person would have lost his self-control in the face of provocation of that particular gravity. C.1 The underlying rationale of the defence 31. The terms of section 4 of the Homicide Ordinance are set out in paragraph [6] in Section A.2 above. The underlying rationale of the defence of provocation on a charge of murder is that the law deems that there may be circumstances in which an impulsive homicidal act done by a person who has been provoked into losing his self-control should be excused the ultimate penalty, now a mandatory sentence to life imprisonment, reserved for the crime of murder. The law does so, not by treating the defendant’s intention in killing to be otherwise than that for murder,[15] but by reducing his culpability for the intentional killing to that of the crime of manslaughter. Thus, the doctrine derives from “the law’s compassion to human infirmity”[16] and is “a concession to human frailty”,[17] the policy of the law being to recognise that, in some instances, things said or done may cause a person to lose their self-control so that the resulting intentional act causing death is to be regarded as one which any person with ordinary powers of self-control would or might have done. If this is so, then the defence applies. 32. But the law’s compassion in this respect is not unlimited or indiscriminate. A person who is naturally quick-tempered, overly sensitive or prone to violence may, as a matter of fact, have been provoked into a loss of self-control. Notwithstanding that, the defence of provocation will not apply if it is determined that the defendant reacted in a way which falls below the standard which a jury considers is the appropriate degree of self-control which everybody is expected to exercise in society as it is today. This is an important part of the assessment because, were it otherwise, the defence of provocation would be more than a mere concession to human frailty and would instead develop into a charter for excessive and deadly reaction by individuals who are overly sensitive or lacking in reasonable self-control. C.2 The subjective question of loss of self-control 33. As explained in Section A.2 above, the question of fact of whether a particular defendant lost his self-control and carried out a homicidal act is a subjective question which depends on the evidence of what was said and done to the defendant and his reaction to that. It is a question for the jury to decide and, at this stage, since it is a subjective inquiry, it will be relevant to consider any personal idiosyncrasies of the particular defendant that may have a bearing on whether in fact he lost his self-control. So, for example, evidence of the defendant’s propensity to react strongly to a particular thing said or done to him may be relevant. This may result in a finding that the defendant did in fact lash out in a homicidal rage having lost his self-control, although what was said or done to him was of a relatively trivial nature, because of some reason particular to him. But this is a consequence of the question being subjective. 34. In the present case, the Judge effectively usurped the role of the jury on this first question by directing them that they were to proceed on the basis that the appellant was in fact provoked into a sudden and temporary loss of self-control. As already indicated, this was a misdirection since that decision was not for the Judge to take but for the jury, in the light of all the relevant admissible evidence concerning what was said and done to the appellant and his reaction to it. It seems likely, as the majority of the Court of Appeal held, that the Judge’s misdirection was the result of her attempt to be helpful to the appellant. Be that as it may, it was not a direction that should have been given. C.3 The two elements of the objective question 35. Once it is determined that a defendant has lost self-control and consequently killed, the next question is whether, having regard to the gravity of the provocation, a person having the power of self-control to be expected of an ordinary person (of the same age and sex as the accused) would or might have lost his self-control and reacted in the way the accused did (in the sense discussed below). As explained in Section A.2 above, there are two elements of this question. That both those elements are to be assessed objectively is supported by (a) the principle that the criminal law is based on a generally accepted standard of conduct applicable to all, (b) the wording and policy of the statutory provision itself, and also (c) the authorities interpreting it. C.3a The underlying principle of the criminal law 36. Thus, the criminal law is generally concerned with setting standards of human behaviour that all persons in society are bound to comply with: see R v Thieu Kham Tran [2010] 3 S.C.R. 350. This principle was recognised in the Report of the Royal Commission on Capital Punishment (1953) and its adoption of “a fundamental principle of the criminal law that it should be based on a generally accepted standard of conduct applicable to all citizens alike”.[18] It is a principle which, therefore, supports the adoption of an objective standard. C.3b The wording of section 4 37. To this end, in section 4 of the Homicide Ordinance, the jury are required to answer the question “whether the provocation was enough to make a reasonable man do as he did” and, in answering that question, the jury must “take into account everything both done and said according to the effect which, in their opinion, it would have on the reasonable man”. The question which the jury must ask themselves is one of sufficiency: was the provocation enough to push the defendant ‘over the edge’ and to cause him to lose the self-control to be expected of an ordinary member of society? In answering that question, the jury must have regard to the effect of the provocative words and actions on “the reasonable man”. 38. As Lord Hoffmann noted in his judgment in R v Smith (Morgan): “… the concept of the ‘reasonable man’ has never been more than a way of explaining the law to a jury; an anthropomorphic image to convey to them, with a suitable degree of vividness, the legal principle that even under provocation, people must conform to an objective standard of behaviour which society is entitled to expect: see Lord Diplock in Camplin [1978] AC 705, 714. In referring to ‘the reasonable man’ section 3 invokes that standard.”[19] It is to be noted here that it has frequently been pointed out that the statutory reference to a “reasonable man” is not ideal since it may be said that a reasonable person does not commit homicide. For this reason, it has been understood to refer to the ordinary person or, in context, the person having ordinary powers of self-control: see Stingel v The Queen (1990) 171 CLR 312 at p.328 and R v Thieu Kham Tran at [30]. C.3c The authorities interpreting section 4 39. In DPP v Camplin, Lord Diplock noted that: “… for the purposes of the law of provocation the ‘reasonable man’ has never been confined to the adult male. It means an ordinary person of either sex, not exceptionally excitable or pugnacious, but possessed of such powers of self-control as everyone is entitled to expect that his fellow citizens will exercise in society as it is today. A crucial factor in the defence of provocation from earliest times has been the relationship between the gravity of provocation and the way in which the accused retaliated, both being judged by the social standards of the day.”[20] (Emphasis added.) 40. In R v Smith (Morgan), Lord Millett explained the approach required of the jury in these terms: “The first question the jury must consider is whether the accused was provoked by something, whether done or said, into losing his self control and reacting as he did. If he was, the next question is whether that something would or might have been sufficient to produce the like reaction in a person similarly placed but possessing the powers of self-control of an ordinary person. This does not require the jury to conjure up a picture of a hypothetical ordinary person or the judge to direct them which characteristics of the accused should be attributed to him and which should be disregarded. The question might perhaps be more easily answered if it were reformulated: would or might the provocation have produced the like reaction from the accused if he had exercised normal powers of self-control.”[21] (Italics in original; underlining added.) 41. It will be noted that Lord Millett referred to “a person similarly placed” rather than the defendant himself. This is clearly a reference to an objective standard, albeit one which, as will be addressed below, incorporates certain features of the particular defendant. The objective standard is necessary, otherwise an unreasonable defendant who over-reacted to an entirely trivial provocation, but which he subjectively assessed to be a gross provocation, might be able to rely on the defence. This is incontrovertibly not the purpose of the defence of provocation. In his speech, Lord Millett cited academic opinion which supports the objective measurement of both the gravity of the provocation and the loss of self-control: “I agree with Professor Ashworth in the article to which I have already referred (at p 312) that, while mitigation of the offences of those who are incapable of exercising ordinary self-control is desirable, the defence of provocation is not an appropriate vehicle. Where an individual who is congenitally incapable of exercising reasonable self-control is provoked by a petty affront, his loss of self-control must be ascribed to his own personality rather than to the provocation he received. In ‘A Rationale of the Law of Homicide’ (1937) 37 Columbia LR 701, 1261, 1281 Wechsler and Michael write: ‘Other things being equal, the greater the provocation [measured objectively], the more ground there is for attributing the intensity of the actor’s passions and his lack of self-control on the homicidal occasion to the extraordinary character of the situation in which he was placed rather than to any extraordinary deficiency in his own character.’ Professor Ashworth observes that the converse also holds true: where the provocation is objectively trivial, the defendant’s loss of self-control should be attributed to his own deficiency rather than the provocation. He concludes that ‘congenitally incapable individuals have an independent claim to mitigation’, and that ‘the defence of provocation is for those who are in a broad sense mentally normal’.”[22] 42. Similarly, in A-G v Holley, Lord Nicholls clearly considered both elements of the evaluative ingredient of the statutory defence to be objective, when he said: “Broken down, this objective ingredient has two elements. The first element calls for an assessment of the gravity of the provocation. The second element calls for application of an external standard of self-control: ‘whether the provocation was enough to make a reasonable man do as he did.’”[23] And he emphasised the need for uniformity in the objective standard required by the “reasonable man” standard thus: “Under the statute the sufficiency of the provocation (‘whether the provocation was enough to make a reasonable man do as [the defendant] did’) is to be judged by one standard, not a standard which varies from defendant to defendant.”[24] 43. Australian courts, when interpreting statutory language which asked whether the provocation was “of such a nature as to be sufficient to deprive an ordinary person of the power of self-control”,[25] have taken the same view. (1) In Stingel v The Queen, for example, the High Court of Australia said this language was “clearly intended to involve an objective threshold test”[26] and noted that: “The central question posed by the objective test – i.e. of such a nature as to be sufficient – obviously cannot be answered without the identification of the content and relevant implications of the wrongful act or insult and an objective assessment of its gravity in the circumstances of the particular case.”[27] (Italics in original; underlining added.) (2) In Buttigieg, the Queensland Court of Appeal referred to matters which would be relevant to affecting the gravity of the particular wrongful act or insult “in the mind of the hypothetical ‘ordinary person’”. [28] It is therefore the gravity to the objective ordinary person that is to be assessed and evaluated by the jury. (3) The majority of the High Court of Australia made the same point in Masciantonio v The Queen emphasising, in the application of the objective test, the effect of like provocation on “the ordinary person”. [29] 44. Similarly, in R v Rongonui [2000] 2 NZLR 385, in his observations as to the appropriate directions to a jury in a case of provocation, Tipping J described the appropriate manner in which the level of gravity of the provocation was to be assessed in these terms: “… It is desirable to add that this means the provocation must have been of sufficient gravity to deprive of their self-control a person with the accused’s characteristic but who is expected to display the power of self-control of an ordinary person. To assist the jury further, it is helpful to say that this requires the jury to assess the gravity of the provocation from the point of view of a person with the accused’s characteristic.”[30] Although the then statutory defence of provocation in New Zealand[31] was in different terms to section 4 of the Homicide Ordinance, it involved an inquiry into the sufficiency of the provocation and this observation of Tipping J supports the conclusion that the gravity of the provocation in a given case is to be assessed objectively. C.4 Assessing the gravity of the provocation 45. Whilst the assessment of the gravity of the provocation is an objective one, it is nevertheless necessary to have regard to aspects of the particular defendant to whom the provocation is directed. This is because of one of the two changes brought about by the introduction of the statutory defence, namely the abolition of the common law rule that words alone were insufficient to constitute provocation. Hence, as Lord Diplock explained in DPP v Camplin: “… now that the law has been changed so as to permit of words being treated as provocation even though unaccompanied by any other acts, the gravity of verbal provocation may well depend upon the particular characteristics or circumstances of the person to whom a taunt or insult is addressed. To taunt a person because of his race, his physical infirmities or some shameful incident in his past may well be considered by the jury to be more offensive to the person addressed, however equable his temperament, if the facts on which the taunt is founded are true than it would be if they were not. It would stultify much of the mitigation of the previous harshness of the common law in ruling out verbal provocation as capable of reducing murder to manslaughter if the jury could not take into consideration all those factors which in their opinion would affect the gravity of taunts or insults when applied to the person whom they are addressed.”[32] And as Lord Simon of Glaisdale said in the same case: “The effect of an insult will often depend entirely on a characteristic of the person to whom the insult is directed.”[33] 46. It is for this reason that, for the purposes of assessing the gravity of the provocation, one must take the reasonable man, who is affected by that provocation, to be “a person similarly placed” to the defendant. As Lord Millett put it in R v Smith (Morgan): “… the jury must take the accused as they find him, warts and all. When considering whether a person of ordinary self-control would have been provoked to react as the accused did, the jury must have regard to what Lord Simon in Camplin called ‘the entire factual situation’. The question for the jury is whether a person of ordinary self-control would have reacted as the accused reacted if he were similarly placed, that is to say, having the history, experiences, background, features and attributes of the accused. This is a question of opinion on which the jury may bring their collective experience and good sense to bear without further evidence: see Camplin [1978] AC 706, 716D, 720F-G, and 727G-H. Accordingly, I respectfully agree with Lord Hoffmann that the question is whether the defendant’s behaviour fell below the standard which could reasonably be expected of him, but only if that is taken to mean of him exercising normal self-control.”[34] (Italics in original.) 47. Thus, the reasonable man who is expected to exercise the ordinary powers of self-control is a person having the history, experiences, background, features and attributes of the defendant relevant to the provocation. This remains an objective test but one which fairly incorporates the features of the defendant that might affect the gravity of the provocation. This was illustrated by Lord Nicholls in A-G v Holley in the following passage of his judgment: “Hence if a homosexual man is taunted for his homosexuality it is for the jury to consider whether a homosexual man having ordinary powers of self-control might, in comparable circumstances, be provoked to lose his self-control and react to the provocation as the defendant did. Authority for this proposition, if needed, is the ‘glue-sniffer’ case of R v Morhall [1996] AC 90. There the deceased nagged the defendant about his addiction to glue-sniffing. The problem before the House of Lords was whether this addiction should have been taken into account at the defendant’s trial as affecting the gravity of the provocation: see p 97D. Lord Goff of Chieveley, with whose speech all members of the House agreed, said it should. The thrust of his reasoning was that, for this purpose, ‘the entire factual situation’ was to be taken into account. This includes matters not falling strictly within the description ‘characteristics’. It also includes matters which are discreditable to the defendant. Lord Goff said, at p 99: ‘suppose that a man who has been in prison for a sexual offence, for example rape, has after his release been taunted by another man with reference to that offence. It is difficult to see why, on ordinary principles, his characteristic or history as an offender of that kind should not be taken into account as going to the gravity of the provocation.’”[35] 48. By attributing to the reasonable man the relevant history, experiences, background, features and attributes of the defendant and by applying contemporary societal standards, the provocation is properly contextualised, so that the jury ask themselves the question of whether the provocation was enough to make a reasonable man do as he did in the context of the provocation which caused the loss of self-control that in fact occurred. As already observed, this inquiry of sufficiency necessarily involves an assessment of the gravity of the provocation – since its gravity and the requisite standard of self-control are, as Lord Millett put it in R v Smith (Morgan), “two sides of the same coin”[36] – but the assessment of the gravity of the provocation nevertheless remains objective. This is helpfully illustrated by the following passage in the judgment of Charron J delivering the judgment of the Supreme Court of Canada in R v Thieu Kham Tran: “For example, in determining the appropriate objective standard, it will be relevant for the trier of fact to know that the alleged provocation occurred in circumstances where the deceased was wrongfully firing the accused from his long-term employment. This context is necessary to set the appropriate standard. But the standard does not vary depending on the accused’s peculiar relationship or particular feelings about his employer or his employment. Personal circumstances may be relevant to determining whether the accused was in fact provoked – the subjective element of the defence – but they do not shift the ordinary person standard to suit the individual accused. In other words, there is an important distinction between contextualizing the objective standard, which is necessary and proper, and individualizing it, which only serves to defeat its purpose.”[37] (Emphasis added.) C.5 Answering the gravity question 49. In view of the inquiry required by the words of section 4 of the Homicide Ordinance as interpreted by the authorities referred to above, it is clear that the subjective assessment of the gravity contended for by the appellant must be rejected. 50. The approach contended for by the appellant would subvert the proper evaluative exercise that the jury are required to undertake when considering a defence of provocation. It would, as Dickson CJ noted in R v Hill [1986] 1 S.C.R. 313, subvert the logic of the objective test because it would lead to the anomalous result that “[a] well-tempered, reasonable person would not be entitled to benefit from the provocation defence … while an ill-tempered or exceptionally excitable person would find his or her culpability mitigated by provocation and would be guilty only of manslaughter”.[38] 51. Were the gravity of the provocation to the particular defendant to be assessed in terms of the degree of provocation to him subjectively, it would be difficult, if not impossible, for a jury to determine the level of permissible response to an objectively trivial insult without lowering the uniform standard of self-control that every member of society is required to exercise. 52. The use of the objective standard to assess the gravity of the provocation and whether it was sufficient to cause any ordinary member of society to lose self-control necessarily sets a benchmark that not everyone will be able to meet. But, as Lord Nicholls put it in A-G v Holley, continuing from the passage cited above: “Of course, assessing the conduct of a glue-sniffing defendant against the standard of a glue-sniffing man having ordinary powers of self-control may mean the defendant is assessed against a standard of self-control he cannot attain. He may be exceptionally excitable or pugnacious. But this is so with every defendant who seeks to rely upon provocation as a defence. The objective standard of self-control is the standard set by the common law and, since 1957, by the statutory reference to a ‘reasonable man’. It is of general application. Inherent in the use of this prescribed standard as a uniform standard applicable to all defendants is the possibility that an individual defendant may be temperamentally unable to achieve this standard.”[39] 53. It therefore follows that the passage in Lord Hobhouse’s speech in R v Smith (Morgan) setting out constituent (c) of the objective question is not to be understood as requiring an assessment of the gravity of the provocation from the defendant’s subjective perspective. So reading the phrase “the gravity of the provocation for the defendant” is to treat those words as if they were embodied in statute and to construe them literally, without reference to the context in which Lord Hobhouse’s statement was made. That context includes Lord Hobhouse’s reference to Lord Diplock’s suggested jury direction in DPP v Camplin which Lord Hobhouse considered to be “loyal to the drafting of section 3 which is concerned with the effect the provocation would have on the reasonable/ordinary man”.[40] It also includes the context of the other authorities referred to in Section C.3c above. Rather than a literal reading as Mr McCoy contends for, the phrase in question should be understood as indicating that Lord Hobhouse intended no more than that there should be an assessment of the gravity of the provocation by reference to the position in which the defendant finds himself. 54. For all these reasons, therefore, we would answer the gravity question in the negative. It is not a misdirection of law to fail to direct the jury to form a discrete view as to the gravity of the provocation on the defendant, having the particular attributes he has, separately from and before determining whether, objectively, having regard to the actual provocation and their view as to its gravity, a person with ordinary powers of self-control would or might have done what the defendant did. C.6 The adequacy of the Judge’s direction on gravity 55. Having answered the gravity question as above, much (if not all) of the force of the appellant’s challenge to the adequacy of the Judge’s directions to the jury concerning the gravity of the provocation falls away. As set out in Section A.3 above, the Judge fairly summarised all the evidence of Madam Yeung’s provocative acts and words and set out the relevant circumstances of the appellant and his co-habitation and relationship with Madam Yeung and his relationship with his daughter by his former wife. The Judge’s summary would properly have put the provocative acts and words in context. That is all that was necessary since, it must be remembered, the jury were not being asked to consider whether the appellant himself was reasonable in reacting as he did but whether, in the face of the provocation, a person with the powers of self-control to be expected of an ordinary member of society today, would or might have done so. 56. The particular passage in the summing up which encapsulated the relevant inquiry of sufficiency for the jury was that set out at [16] above, in particular: “There is still a question of whether an ordinary person of the accused’s age and sex would have done what he did. If you feel sure that an ordinary person of his age and sex, even in the accused’s circumstances and even having undergone all the provocation which the accused told you that he had undergone at the time of the killing and leading up to it over a considerable period, would not have done what the accused did, then the doctrine of provocation would not operate to reduce murder to manslaughter.” (Emphasis added.) 57. The underlined passage above must be understood in its context in the summing up read as a whole. It followed a detailed recapitulation by the Judge of the evidence of the appellant regarding the events on the evening in question and an overly (see [15] above) favourable direction by the Judge that the jury should proceed on the basis of the appellant’s account of events. The repeated emphasis of “even” in the appellant’s circumstances and “even” having undergone the provocation described, would fairly have invited the jury to consider the gravity of the provocation to the ordinary person in the position of the appellant before assessing whether that provocation was sufficient to cause such a person to lose their self-control. 58. With respect, we therefore do not agree with McWalters JA’s description of this as “a general and unfocused direction”.[41] The present case was not one in which it was suggested there were any special features of the appellant that bore on the gravity of the provocation and with which the hypothetical reasonable man should be invested. Instead, the Judge provided detailed directions to the jury regarding the defence case that he was the victim of gross deceit by his co-habitee, a disrespected and cheated upon lover, whose parentage of a beloved child was cast in doubt, for these matters to be weighed in the relevant sufficiency exercise. 59. The direction which the Judge gave the jury in the present case followed the specimen direction on provocation provided to judges by the Judicial Institute.[42] That direction properly identifies the subjective and objective questions which the jury must address in considering a defence of provocation. In relation to the objective question, the two constituent elements of that question are covered by the question of sufficiency outlined in question 5 of that specimen direction. As the specimen direction recognises, there will, of course, be cases where special directions will be called for, for example, where there are features of the defendant which might mean that the provocative conduct would have been more provoking for such a person. But in view of the answer to the gravity question set out in Section C.5 above, it is not necessary to consider further the appropriateness of the specimen direction. C.7 Miscellaneous points on the gravity question 60. It is true that the Crown Court Bench Book, Specimen Directions to the Jury (2008), at [51] in respect of “Provocation (Murder)”[43] did construct a jury direction which split Lord Hobhouse’s constituents (c) and (d) into two separate questions. Mr McCoy placed heavy reliance on this specimen direction in support of the appellant’s case on the gravity question. The relevant portion of that direction reads as follows: “5. But if you conclude that D was or might have been provoked, in the sense which I have explained, you must then go on to weigh up how serious the provocation was for this defendant. Is there anything about this defendant which may have made what was [said and/or done] affect him more than it might have affected other people? (Here, identify any matters which may be relevant to the gravity of the provocation to this defendant.) 6. Finally, having regard to the actual provocation and to your view of how serious that provocation was for this defendant, you must ask yourselves whether a person having the powers of self-control to be expected of an ordinary, sober person, of D’s age and sex …, would have been provoked to lose his self-control and do as this defendant did … . If you are sure that such a person would not have done so, the prosecution will have disproved provocation, and D is guilty of murder … . If, however, you conclude that such a person would or might have reacted and done as D did, your verdict would be ‘Not guilty of murder, but guilty of manslaughter [by reason of provocation]’.” (Emphasis in original; Notes omitted.) 61. However, what other bench books provide by way of specimen directions is beside the point. Specimen directions are merely guides for judges to follow and not authoritative statements of substantive law. Indeed, as the title page of the Hong Kong Specimen Directions notes: “Save where the Court of Final Appeal or the Court of Appeal has so ruled, these Specimen Directions have no legal authority.” More significantly, the Crown Court Bench Book appears to have treated Lord Hobhouse’s description of the constituents literally and without regard to the context in which he gave that description. Insofar as the direction suggests a subjective assessment of the gravity of the provocation to the defendant, this flies in the face of a substantial weight of authority as shown in Section C.3c above. To that extent, it provides, with respect, dubious support for the appellant’s case on this issue. 62. Mr McCoy submitted that the key passage in Lord Hobhouse’s speech in R v Smith (Morgan) that explained the true nature of his constituent (c) of the defence of provocation was that at p.196C-D of the report, namely: “The answer is that the role of the second question is being misunderstood. Its purpose is, as previously stated, to provide a standard of ordinary self-control so as to compare the reaction of the defendant as he was in fact provoked to lose his self-control with the reaction of a person with ordinary powers of self-control to provocation of equal gravity. Its purpose is not to create for the jury some impossible self-contradictory chimera designed ultimately to displace the concept of reasonableness altogether.” 63. This passage does not, however, support the appellant’s argument. On the contrary, it demonstrates that the comparison is between the defendant’s loss of self-control in the face of the provocation with the reaction of the hypothetical person exercising the ordinary powers of self-control to provocation of equal gravity. That latter exercise is a wholly objective exercise and does not admit of a subjective assessment of the gravity of the provocation. As explained above, were this not so, the hyper-sensitive defendant over-reacting to a trivial provocation would provide the relevant yardstick of behaviour and this is not the way the defence operates. 64. Similarly, Mr McCoy’s reliance on the Court of Appeal decisions in HKSAR v Lo Chun Siu[44] and HKSAR v Poon Man Sum[45] does not, in our view, assist the appellant. The gravity question was not in issue in either of those cases and there is no discussion and resolution of the issue now before this Court. In any event, the description given by McWalters J (as he then was), delivering the judgment of the court in Lo Chun Siu, of the first element of the objective question is consistent with that element being objective and not subjective.[46] Nor does Poon Man Sum support the subjective approach contended for by the appellant. The failure in that case was to direct the jury to consider the gravity of the provocation in the context of the relevant evidence which included an earlier threat to commit suicide.[47] D. The ‘do as he did’ question 65. The question is set out in paragraph [22] above. As is apparent from the parties’ contentions on it (at Section B.2 above), the central issue concerns the meaning of the words “do as he did” in section 4 of the Homicide Ordinance. The proper construction of those words is important because, depending on their true construction, there is a risk that the jury may embark on a line of reasoning that would, on that construction, be impermissible. The question framed by the Court raises the issue of whether it may be necessary for a trial judge to direct the jury in a specific manner in order to avoid that risk. D.1 The issue of construction 66. The wording of section 4 of the Homicide Ordinance is set out in paragraph [6] above. The second, objective, question arises for the jury’s consideration when there is evidence on which the jury can find that the person charged with murder was provoked to lose his self-control and that issue is whether “the provocation was enough to make a reasonable man do as he did” (underlining added). What is the proper construction of those words? 67. To answer that question, the Court must construe the language used in the light of its context and purpose, that being the modern approach to statutory construction which has been repeatedly endorsed and applied by this Court: See, e.g., Medical Council of Hong Kong v Chow Siu Shek (2000) 3 HKCFAR 144 at 154B-C; HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at [11]-[14]; HKSAR v Fugro Geotechnical Services Ltd (2014) 17 HKCFAR 755 at [19]. D.1a The latent ambiguity in the words 68. The words “do as he did” in section 4 are inherently ambiguous: see the discussion in Professor Glanville Williams’ Textbook on Criminal Law (2nd Ed., pp.543-544) and Professor Andrew Ashworth’s Principles of Criminal Law (3rd Ed., pp.283-284). There is a range of meanings that they could convey, namely that the provocation was sufficiently grave to cause a person with ordinary powers of self-control: (1) To lose self-control (“the loss of self-control meaning”); (2) To kill the victim (i.e. to form the intent to kill or cause grievous bodily harm and act on that intent) by whatever means (“the killing simpliciter meaning”); (3) To kill the victim using the means that the defendant did, e.g. by stabbing or shooting or strangulation (“the means of killing meaning”); (4) To kill the victim in exactly the manner the defendant did, e.g. by 6 shots of the gun or 213 chops of the knife (“the precise method of killing meaning”). 69. Meaning (1), the loss of self-control meaning, can immediately be discounted as the proper construction of “do as he did” since it is difficult to see how the words convey only the meaning that the defendant must have lost his self-control and no more. If the intention was to restrict the meaning to loss of self-control, one might have expected the draftsman to have adopted wording along the lines of the New Zealand Crimes Act 1961, section 169(2)(a), which raises the question of whether “[the provocation] was sufficient to deprive a person having the power of self-control of an ordinary person … of the power of self-control” (see paragraph [96] below). Moreover, to say that a person has lost his self-control does not give any indication of what he has done in consequence of that loss, whereas the use of the verb “to … do” in section 4 indicates a need to focus on some form of action. 70. This leaves the debate between meanings (2), (3) and (4) above, each of which involves the act of killing but with different degrees of specificity. D.1b The reasonable relationship rule 71. A rule known as the “reasonable relationship rule”, or proportionality rule, was often tied to meaning (4) of “do as he did” (the precise method of killing meaning), since a defendant whose act of killing was regarded as disproportionate was deprived of the defence, suggesting that examining exactly what the defendant did was necessary to operating the rule. 72. In Mancini v DPP [1942] AC 1, Viscount Simon LC stated the common law rule that “the mode of resentment must bear a reasonable relationship to the provocation if the offence is to be reduced to manslaughter.”[48] And in Phillips v The Queen [1969] 2 AC 130, Lord Diplock cited that statement in support of the following proposition: “… it is beyond question that at common law by which the matter was regulated both in Jamaica and in England until the legislation cited above, the relationship between the degree of retaliation and the nature of the provocation was a relevant factor in determining whether the offence proved was manslaughter and not murder.”[49] 73. This rule was described as being “firmly established” as part of English law prior to the enactment of section 3 of the Homicide Act 1957[50] and was applied by the Privy Council on an appeal from Hong Kong in Lee Chun-Chuen v The Queen [1963] AC 220 at p.231, a case decided under the common law doctrine and before the enactment of section 4 of the Homicide Ordinance. The rule was exemplified in the jury direction which Devlin J (as he then was) gave in R v Duffy, and which Lord Goddard in that case ([1949] 1 All ER 932) described as “impeccable”, namely: “in considering whether provocation has or has not been made out, you must consider the retaliation in provocation – that is to say, whether the mode of resentment bears some proper and reasonable relationship to the sort of provocation that has been given. Fists might be answered with fists, but not with a deadly weapon, and that is a factor you have to bear in mind when you are considering the question of provocation.”[51] 74. In this regard, it is a question of some importance to consider whether the reasonable relationship rule, either in whole or in part, has survived the enactment of the defence of provocation. In England, the Court of Appeal in R v Walker [1969] 1 WLR 311 saw force in the submission, which it thought “may well be correct”, that section 3 of the Homicide Act 1957 had “replaced the reasonable relationship rule, and that that rule, if it ever was a rule, has now gone, and has been replaced by the words in the section”.[52] That obiter view was, of course, equivocal but in any event it begs the question as to what the words in the section mean. 75. Meaning (4) of “do as he did” (the precise method of killing meaning) would derive considerable support from a conclusion that the rule does continue to operate, whilst a conclusion that the rule has been superseded and no longer applies would lend support for either meaning (2) (the killing simpliciter meaning) or (3) (the means of killing meaning). D.1c The English and Privy Council authorities on section 3 of the Homicide Act 76. In DPP v Camplin, Lord Diplock framed a “proper direction” to a jury on provocation in these terms: “The judge should state what the question is using the very terms of the section. He should then explain to them that the reasonable man referred to in the question is a person having the power of self-control to be expected of an ordinary person of the sex and age of the accused, but in other respects sharing such of the accused’s characteristics as they think would affect the gravity of the provocation to him; and that the question is not merely whether such a person would in like circumstances be provoked to lose his self-control but also whether he would react to the provocation as the accused did.”[53] (Underlining added.) 77. Lord Diplock had earlier given the judgment of the Privy Council in Phillips v The Queen, which considered the reasonable relationship rule in Mancini v DPP. There, special leave to appeal to the Privy Council was given to resolve inconsistent decisions of the Court of Appeal of Jamaica as to whether it was a misdirection for a judge to direct a jury in a provocation case that the retaliation must bear some proper and reasonable relationship to the provocation received. Lord Diplock was of the opinion that the Jamaican statutory provision (identical to section 3 of the Homicide Act 1957) only changed the common law doctrine of provocation in two respects: first, in abolishing the common law rule that words alone could not amount to provocation and, second, in leaving to the jury exclusively the question of deciding whether or not a reasonable man would have reacted to the provocation in the way in which the defendant did. Thus, as to the latter, Lord Diplock explained that the second question for the jury was, “‘Would a reasonable man have reacted to the same provocation in the same way as the defendant did?’”[54] 78. Lord Diplock then addressed the argument raised by the appellant in Phillips that “once a reasonable man had lost his self-control his actions ceased to be those of a reasonable man and that accordingly he was no longer fully responsible in law for them whatever he did”[55] (underlining added). This argument was, in effect, that there was no longer any role for the reasonable relationship rule. This argument was rejected by Lord Diplock, who held: “This argument is based on the premise that loss of self-control is not a matter of degree but is absolute; there is no intermediate stage between icy detachment and going berserk. This premise, unless the argument is purely semantic, must be based upon human experience and is, in their Lordships’ view, false. The average man reacts to provocation according to its degree with angry words, with a blow of the hand, possibly if the provocation is gross and there is a dangerous weapon to hand, with that weapon. It is not insignificant that the appellant himself described his own instantaneous reaction to the victim’s provocation in spitting on his mother as ‘I spin around quickly was to punch her with my hand.’”[56] His Lordship then described the judge’s direction to the jury (“whether a reasonable man would have reacted to the provocation in the way that the appellant did”), which had “followed closely” the statutory language as “impeccable”.[57] 79. Pausing here, it would appear that Phillips provides some support for the suggestion that the reasonable relationship rule has survived the enactment of the statutory defence of provocation. However, Lord Diplock added the following important passage: “Since the passing of the legislation it may be prudent to avoid the use of the precise words of Viscount Simon in Mancini v Director of Public Prosecutions [1942] A.C.1 ‘the mode of resentment must bear a reasonable relationship to the provocation’ unless they are used in a context which makes it clear to the jury that this is not a rule of law which they are bound to follow, but merely a consideration which may or may not commend itself to them.”[58] 80. It is important to note from this passage that Lord Diplock was categorically stating that the reasonable relationship rule was now not a rule of law which a jury was bound to follow. That statement of the law does clearly differ from the reasonable relationship rule as stated in Mancini. In Phillips, the judge did not direct the jury in terms of the reasonable relationship rule and the Privy Council held that there was no error in the summing up on provocation and dismissed the appeal. 81. This passage was followed by the English Court of Appeal in R v Brown [1972] 2 QB 229, where it was contended that the judge had misdirected the jury on the reasonable relationship rule. The Court of Appeal noted that the point had not required decision in R v Walker but did in the case before it. Having referred to the passages from Phillips quoted above, the Court of Appeal held: “In the view of this court, when considering whether the provocation was enough to make a reasonable man do as the accused did it is relevant for a jury to compare the words or acts or both of these things which are put forward as provocation with the nature of the act committed by the accused. It may be for instance that a jury might find that the accused’s act was so disproportionate to the provocation alleged that no reasonable man would have so acted. We think therefore that a jury should be instructed to consider the relationship of the accused’s acts to the provocation when asking themselves the question ‘Was it enough to make a reasonable man do as he did?’ We feel that Lord Diplock’s warning should be followed and that it would be better not to use the precise words of Viscount Simon unless it is made quite clear that it is not a rule of law which the jury have to follow.”[59] (Underlining added.) 82. The decision in Brown therefore provides support for the continued relevance of the proportionality of the retaliation to the provocation, albeit in a qualified way as being “not a rule of law” and albeit limited to the “nature of the act committed by the accused”. To similar effect is the decision in R v Acott [1997] 1 WLR 306, where Lord Steyn (with whom the other members of the House of Lords agreed) said this: “Moreover, although there is no longer a rule of proportionality as between provocation and retaliation, the concept of proportionality is nevertheless still an important factual element in the objective inquiry. It necessarily requires of the jury an assessment of the seriousness of the provocation.”[60] 83. On the reasoning set out in paragraph [75] above, this supports the argument that the proper construction of the words “do as he did” is not meaning (4) but nevertheless still leaves open the question of whether the act of retaliation is to be expressed in terms of meaning (2) or (3) set out above. D.1d The Hong Kong authorities 84. Phillips was cited to the Court of Appeal in R v Vu Van Thang [1991] 2 HKC 90 but the issue of the relationship between the provocation and the reaction was not raised, since on the facts of that case it was held that it had not been necessary for the judge to leave the defence of provocation to the jury. 85. That issue was, however, raised in the appeal in HKSAR v Liang Yaoqiang [2014] 4 HKC 145. That case concerned the appellant in this appeal and was his first appeal to the Court of Appeal against his conviction for murder before Beeson J and a jury.[61] In that appeal, one of the grounds of appeal concerned the judge’s failure to correct the following submission made in the prosecution’s closing speech to the jury: “… if you decide he was, or may have been, provoked, I submit that no reasonable person would have done as he did. The infliction of 209 or thereabouts chop wounds to the victim’s body is out of all proportion to any possible provocation.”[62] 86. The Court of Appeal’s judgment on this ground of appeal (which was delivered by Stock VP) was strictly obiter since the appeal was allowed on another ground of appeal. However, the Court of Appeal considered the argument that was advanced on behalf of the appellant that: “… once it is accepted by the jury that the provocation offered would or may cause a reasonable person to lose his self-control, no further inquiry is warranted into the nature and degree of the consequent conduct since, by definition as it were, a loss of self-control envisages the likelihood or at least the possibility of an uncontrolled frenzy of extraordinary violence.” [63] 87. The Court of Appeal rejected this argument, citing the jury direction provided by Lord Diplock in DPP v Camplin and the passages from Phillips and Brown quoted above.[64] However, given the obiter nature of the discussion, there is no substantive reasoning in the Court of Appeal’s judgment that bears on the issue of construction now before this Court. D.1e The position in Australia 88. The appellant relied heavily on the approach in the Australian cases dealing with the defence of provocation. Those cases have clearly taken a different approach to that of the English cases considered in Section D.1c above and support a conclusion that the relevant inquiry concerns the sufficiency of the provocation to cause the accused to form an intent to kill or do grievous bodily harm, rather than the precise form of physical reaction. 89. Thus, in Johnson v The Queen (1976) 136 CLR 619, the High Court of Australia considered the relevant New South Wales statute providing a defence of provocation.[65] It held, relevantly, that the accused bore the onus of proving on a balance of probabilities the matters in the proviso to the section, namely: “(a) That such provocation was not intentionally caused by any word or act on the part of the accused; (b) That it was reasonably calculated to deprive an ordinary person of the power of self-control, and did in fact deprive the accused of such power, and, (c) That the act causing death was done suddenly, in the heat of passion caused by such provocation, without intent to take life.” It also held that the requirement that the mode of retaliation should bear a reasonable proportion to the act of provocation was not a separate element which the accused must establish, but was a relevant matter to paragraphs (b) and (c) of the proviso. 90. Importantly, Barwick CJ observed in his judgment: “To take into account the mode and extent of retaliation when determining whether an ordinary man, subjected to the like acts of provocation in all the circumstances in which the accused then stood, would have lost self-control to the point of doing something akin to what the accused has done is one thing. To require that it be established positively and, as a separate issue or element, whether the act of the accused was in fact proportionate to the provocation, is quite another; or to require the Crown as a specific matter to negative that proportion is quite another. This is particularly so if it be assumed that it has already been concluded that the accused had relevantly lost self-control. In considering whether an ordinary man would have lost self-control so as to form an intent to kill and to kill in the manner in which the accused did so, the jury may think the provocation was such that an ordinary man might react in the way in which the accused reacted. After all, it is the induced intent to kill rather than the induced fatal act which is the critical consideration. As I have stressed, the provocation is irrelevant unless there was an intent to kill or to do grievous bodily harm: and the provocation to be operative must have caused that intent. No doubt the question is difficult, namely: would the actual provocation, which has caused the accused so far to have lost self-control as to have formed an intent to kill, have caused an ordinary man to have lost his self-control and to have formed that intent? To express the objective test as whether the provocation would have caused an ordinary man to have so far lost self-control as to have done an act of the kind or degree done by the accused, tends somewhat to obscure the necessity of there being an intent to kill or do grievous bodily harm. Emphasis on the essential causation of the requisite intent may, on the one hand, make provocation more difficult to establish and, on the other hand, lessen perhaps the importance in the mind of a tribunal of the particular way in which the intent to kill or to do grievous bodily harm was effectuated.”[66] 91. In the subsequent case of Masciantonio v The Queen (1994-1995) 183 CLR 58, the High Court of Australia cited Barwick CJ’s judgment with approval in the following terms: “… as Barwick CJ pointed out in Johnson v R in considering whether an ordinary person could have reacted in the way in which the accused did, it is the formation of an intent to kill or do grievous bodily harm which is the important consideration rather than the precise form of physical reaction.”[67] (Underlining added.) 92. In that case, the High Court held that the relevant question to be measured in terms of the ordinary person was whether the accused had been provoked to lose self-control so as to form the intent to kill or do grievous bodily harm rather than the duration or precise physical form of his reaction. That the focus is on the formation of the intent to kill or do grievous bodily harm is shown in the following passage: “The question is not whether an ordinary person, having lost his self-control, would have regained his composure sooner than the accused nor is it whether he would have inflicted a lesser number of wounds. It is whether an ordinary person could have lost self-control to the extent that the accused did. That is to say, the question is whether the provocation, measured in gravity by reference to the personal situation of the accused, could have caused an ordinary person to form an intention to kill or do grievous bodily harm and to act upon that intention, as the accused did, so as to give effect to it. The associated question whether, in the sequence of events, an accused, having lost his self-control, had regained it so that the continued infliction of injury was in fact no longer provoked, is not a question to be answered by reference to the ordinary person. It is to be answered by reference to the conduct of the accused himself and to common experience of human affairs. It is the nature and extent – the kind and degree – of the reaction which could be caused in an ordinary person by the provocation which is significant, rather than the duration of the reaction or the precise physical form which that reaction might take. And in considering that matter, the question whether an ordinary person could form an intention to kill or do grievous bodily harm is of greater significance than the question whether an ordinary person could adopt the means adopted by the accused to carry out the intention.”[68] (Italics in original; underlining added.) 93. These passages in Masciantonio demonstrate that it is not an element of provocation in the statutory provisions there under consideration that the retaliation should be proportionate to the provocative incident. They therefore suggest support for meaning (2) of “do as he did” rather than meanings (3) or (4). To this end, in R v Barrett (2007) 171 A Crim R 315, the Supreme Court of Victoria held that jury directions should not include a suggestion that “the law will only concede the existence of provocation if there is some kind of proportion between the provocation and the murderous act or acts.” Instead, the direction should conclude: “The provocation must be of a kind that might in the same circumstances cause an ordinary person to form an intention to kill or to cause really serious injury. The question is not whether an ordinary person would react to the provocation by using the precise method of response as did the accused but whether, in those circumstances, an ordinary person might form an intention to kill or to cause really serious injury.”[69] 94. However, it is to be noted that, in Masciantonio, in the passage immediately before the reference to Barwick CJ’s dictum in Johnson quoted in paragraph [91] above, the majority stated: “… it is now well established that the question of proportionality is absorbed in the application of the test of the effect of the provocation upon the ordinary person. As Lord Diplock said in Phillips v The Queen Viscount Simon’s remark was ‘an elliptic way of saying that the reaction of the defendant to the provocation must not exceed what would have been the reaction of a reasonable man’.”[70] 95. This proposition, that the question of proportionality is absorbed in the application of the “ordinary person” test, demonstrates that there is still a place for that concept in the inquiry for the jury. The suggested absorption of the question of proportionality into the “ordinary person” test bears some similarity to the proposition of Lord Steyn in R v Acott (quoted in paragraph [82] above), namely that “the concept of proportionality is nevertheless still an important factual element in the objective inquiry.” D.1f The position in New Zealand 96. The question of the relation of the retaliation to the provocation has also been considered in New Zealand, where under the applicable statutory provision[71] it has been held that proportionality is not required as a matter of law. 97. In R v Rongonui [2000] 2 NZLR 385, Elias CJ discussed the notion of proportionality of the response to provocation. She stated these propositions at the outset of her analysis (at [137]): (1) Proportionality is not required as a matter of law and it is wrong for the judge to suggest that there is any requirement of reasonably proportionate retaliation. (2) Lack of proportionality may be evidence which helps the jury determine whether or not the accused committed the homicide under provocation (i.e. the subjective question of whether the defendant in fact lost his self-control). 98. Elias CJ went on to consider (at [138]) the submission that “[i]n the circumstances of the frenzied attack described by the pathologist … a requirement to take proportionality of the response into account was fatal to the defence of provocation because the response could only have been seen as disproportionate.” She considered the Privy Council decision in Phillips, noting the statutory language of section 3 of the Homicide Act 1957. She contrasted (at [140]), the New Zealand statutory defence of provocation where “the objective limb is concerned only with the sufficiency of the provocation to deprive the accused of self-control.” Hence, she cautioned (at [141]), “[s]ome care is necessary in the application of Phillips given the different statutory provisions.” 99. Elias CJ concluded (at [142]) that the direction in Rongonui “would have suggested to the jury that the disproportionality of the actual response was something it must take into account”. The material part of that objectionable direction in Rongonui (set out at [68]) was in the following terms: “When you are considering suggestions of provocation or evidence of the sort that we have been discussing for the last two weeks, it is usual to look at the nature of the threat or the provocation itself and the nature of the response, and it’s perhaps common sense to suggest that there should be some degree of proportionality between the threat and the response. You consider what would the person with the ordinary power of self-control do under the circumstances, would he or she retreat, would he or she do something less violent than what happened in this case. It is entirely a matter for you, it is not a question of law at all. But naturally you will look at all the circumstances both of the build up as to what happened and what happened at the time when you are assessing the suggested provocation.” A direction of this nature gives rise to the risk that the jury may embark on the impermissible line of reasoning discussed in Section D.4 below. 100. In the later case of R v Timoti [2006] 1 NZLR 323, the question of proportionality in provocation was again considered. In relation to the subjective question of whether the accused lost his self-control, the Supreme Court of New Zealand held (at [34]) that the relationship between the level of provocation and the level of response was “apt to be equivocal” and continued (at [35]): “Hence an abstract direction on proportionality is not likely to be helpful to the jury; the more so if it does not distinguish the factual question from the evaluative question. A case-specific direction on the factual question which invites the jury to consider whether the relationship between the degree of provocation and the level of response assists them in deciding whether the accused did lose the power of self-control will be of assistance; and a rehearsal of the competing contentions of the parties in case-specific terms will be of particular assistance to the jury in deciding this first factual issue. If it is decided in favour of the Crown, that is, the Crown has established that the accused did not lose the power of self-control, the evaluative question will not arise.” 101. The court then considered the second, evaluative, question (i.e. the objective question) and posed the question of whether the concept of proportionality is capable of assisting the jury in their evaluation of whether the provocation was sufficient to deprive the statutory hypothetical person of the power of self-control, noting that on this question the actual mode of response of the accused to the provocation was irrelevant. In this regard, the court explained (at [36]-[38]): “[36] The question now being addressed is not whether the accused did lose the power of self-control; rather it is whether the provocation was sufficient to cause the hypothetical person to do so. The accused’s actual response is no more than a distraction on this issue. The question concerns the response of the hypothetical person to the provocation actually found to have been given. Hence a direction which invites a comparison between the level of provocation and the accused’s response to it is inappropriate and apt to cause confusion on this second (evaluative) question. [37] Obviously the question whether the provocation was sufficient to cause the hypothetical person to lose the power of self-control to the point of committing the act of homicide carries within it a value judgment to which proportionality considerations are relevant. But the concept of sufficiency adequately incorporates this without any need for a confusing proportionality gloss. This approach is supported by observations made by the High Court of Australia in Masciantonio v The Queen. In that case, even against a ‘do as he did’ criterion, Brennan, Deane, Dawson and Gaudron JJ, in their joint judgment, aptly said it was well established that the question of proportionality was ‘absorbed’ into the application of the test involved in what we have called the evaluative question. [38] In summary, therefore, questions of proportionality may be of some assistance to a properly directed jury on the factual question but are likely to be more confusing than helpful to them on the evaluative question. On that question a direction couched in terms of proportionality between the provocation and the accused’s response should not be given.” (Underlining added.) 102. It should be noted that, in Timoti, the Supreme Court of New Zealand drew attention (at [43]) to the comments of the Privy Council in Phillips that there are degrees of self-control but suggested that these comments should be read in the light of the materially different wording of section 3 of the Homicide Act 1957: “focusing attention on what the accused did in response to the provocation. There is a material difference between the concept of provocation causing the accused to ‘do as he did’ and provocation causing the accused to lose the power of self-control, as in our Act.” 103. The Supreme Court of New Zealand therefore rejected (at [46]) the need to consider the degree of loss of self-control beyond loss to the point of forming the necessary murderous intent as well as performing an action which caused death: “When s 169(2)(b) speaks of provocation which has deprived the offender of the power of self-control and has thereby induced the act of homicide, it is implicit that the act of homicide encompasses not only the physical action which caused the death but also the necessary murderous intent. It follows that for the purposes of the evaluative question the provocation must have been sufficient to cause the statutory hypothetical person to lose the power of self-control to the point of forming the necessary murderous intent as well as performing an action which caused death. In short, the provocation must have been sufficient to cause in the hypothetical person loss of self-control inducing both a murderous act and murderous intent. What is not relevant to the evaluative question is the degree of the loss of self-control beyond that point. Hence the extent of loss of self-control manifested by the accused is not relevant. Referring back to our discussion of R v Anderson, the degree or method or continuance of the actions which caused the death are not relevant to the inquiry.” 104. Hence, the court concluded that the following portion of the summing up in Timoti (set out in [30]) was a misdirection: “You must consider, as a weighty factor, whether the accused’s acts leading to Mr Ruarau’s death bear any proper or reasonable relationship to the sort of provocation said to have been given by the accused’s mother and Mr Wuatai. The extent of loss of self-control has to be considered, in proportion to the alleged provocation …” Again, a direction of this nature may give rise to the risk that the jury may embark on the impermissible line of reasoning discussed in Section D.4 below. 105. The discussion above shows that the New Zealand courts have rejected any requirement of proportionality between the provocation and the response beyond the need for the provocation to have been sufficient “to cause in the hypothetical person loss of self-control inducing both a murderous act and murderous intent.” This reasoning, albeit relating to materially different statutory wording, would support meaning (2) of “do as he did” and would certainly suggest that meanings (3) and (4) are not the proper construction of those words. 106. Of particular relevance to our statutory wording, however, is the passage in paragraph [37] of the judgment in Timoti (quoted above) noting that the concept of sufficiency (with which section 4 of the Homicide Ordinance is concerned, see paragraph [37] above) “incorporates a value judgment to which proportionality considerations are relevant” and the observation that this is consistent with the idea that, in relation to the “do as he did” criterion, proportionality is absorbed into the application of the objective test of the effect of the provocation on the ordinary person. D.2 The policy underlying the defence 107. The underlying rationale of the defence of provocation has been addressed in Section C.1 above. The subjective question of whether in any given case a defendant has been provoked to lose his self-control and kill in consequence does not present any conceptual difficulties. This is a question which the jury must answer by reference to all the available evidence of what was said and done by the victim to the defendant and what the defendant did in retaliation. At this stage, the nature of the provocation and the precise form and extent of the retaliation is necessarily relevant evidence, although as was pointed out in Timoti, the position may be equivocal. A disproportionately vicious retaliation may be evidence that self-control was in fact lost, although the possibility remains that such retaliation could have been carried out as a cool-headed and deliberately sadistic killing or as an attempt to camouflage the killing as one committed under a loss of self-control. This question must be left to the jury to decide on the evidence. 108. However, in respect of the objective question, the notion that the retaliation must bear a particular relationship to the provocation is one beset with conceptual difficulties. Where the person with ordinary powers of self-control has been provoked to the degree necessary to cause him to lose that self-control, there is an inherent contradiction in then going on to ask whether his actions in that state of loss of self-control bear some proportionate relationship to the provocation offered by the victim. As Elias CJ put it in Rongonui, this may seek to “invoke a rationality already lost.”[72] If a person has been provoked to such an extent that the ordinary person could equally have lost his self-control to the point of forming the intent to kill the provocateur and acting on that intent, it is difficult to see why the extent of the defendant’s reaction should deprive him of the defence. To ask of the hypothetical “reasonable man” in section 4 whether the provocation might have been sufficient to make him lose his self-control to the extent of making a choice as to weapon (gun, knife or chapati pan) or to the extent of stabbing, say, once or twice in carrying out his intention to kill the victim rather than, say, 10 or 20 or even 200 times seems artificial and, more importantly, conceptually extremely difficult for a jury. The precise mode of retaliation may also be fortuitously dependent on the lethal instrument near at hand. 109. As a matter of underlying policy, where the jury have already reached the conclusion that the homicidal act was carried out intentionally by the defendant when he had lost his self-control, the real focus of the evaluative, objective, question is whether a person with the ordinary powers of self-control might have reacted to the retaliation by forming the intent to kill or cause really serious bodily harm to the victim and acting on that intent. To the extent that an extreme homicidal response negates a finding of a subjective loss of self-control, the defence will not apply. But save in that circumstance, there is no compelling reason of logic or policy to require the jury to consider whether the homicidal actions (which have already been established were in fact brought about a loss of self-control) bear a reasonable relationship to the provocation. On the contrary, the difficulties inherent in such a question strongly suggest a conclusion that no such question is necessary. D.3 The proper construction of “do as he did” in section 4 110. The meaning of the words “do as he did” being inherently ambiguous, the context and purpose of section 4 is critical to determining their proper construction. 111. As a matter of context, section 4 is addressing a partial defence in a case of murder. The context is derived from the opening words of the section (“Where on a charge of murder …”). The focus of the defence is then on whether there is evidence on which the jury can find “that the person charged” (i.e. charged with murder) was provoked “to lose his self-control” and “to make a reasonable man do as he did”. The statutory purpose of section 4 is to give effect to the underlying policy of the law to excuse certain impulsive homicidal acts resulting from substantial provocation from the ultimate penalty for murder (see Section C.1 above). The section also altered the common law in the two respects addressed in paragraph [7] above. 112. As the discussion of the English and Privy Council cases in Section D.1c above shows, it is debatable whether the reasonable relationship rule still applies under the statutory defence of provocation. The cases certainly support the proposition that there is no rule of law to that effect. The idea that the mode of resentment must be proportionate in type to the provocation given is, at the very least, at odds with the abolition of the common law rule that words alone could not amount to provocation. Once it is accepted that words themselves could provoke the loss of self-control with an intention to kill, the reasonable relationship rule must (if it survives in any form) at least require some reformulation in order to operate in a case of provocative words. 113. That being the case, it is difficult to see a proper basis for concluding that meaning (4) (the precise method of killing meaning) can be the correct construction of “do as he did”. The conceptual difficulties in seeking to find a relationship between provocation that brings about a loss of self-control triggering a homicidal reaction with the reasonableness of the choice of weapon or mode of carrying out a murderous intent weighs against the conclusion that meaning (3) (the mode of killing meaning) is the proper construction of those words. 114. In the light of the context and purpose of section 4 and the underlying policy of the law to mitigate the penal consequences of a killing carried out in the frenzy of a loss of self-control, meaning (2) (the killing simpliciter meaning) provides the most logical and readily understandable meaning of the words “do as he did”. This requires the jury to consider whether, in the light of the gravity of the provocation on a person in the position of the defendant but with the ordinary powers of self-control to be expected in society today, that person might have formed the intention to kill or cause grievous bodily harm to the victim and have acted on that intention. This is a simpler test in which the magnitude of the response is absorbed into the evaluation of whether the provocation was sufficient to cause a person with ordinary powers of self-control to lose that self-control to the point of forming the intent to kill or cause grievous bodily harm and acting on that intent. It avoids the need for illogical and extremely difficult mental gymnastics on the part of the jury in trying to grade degrees of provocation and degrees of reasonable retaliation to such provocation. The defence of provocation is neither intended to be, nor capable of being, a matter of scientific measurement. Rather, it is intended that juries apply their common sense to the evidence to determine the result in any given case. D.4 The impermissible line of reasoning 115. In view of the proper construction of the words “do as he did” as set out in Section D.3 above, there is a risk that, without a proper direction, the jury may embark on a line of reasoning that would, on that construction, be impermissible. Thus, the line of reasoning represented by the prosecution’s closing speech to the jury in the appellant’s original trial (see paragraph [85] above), which asks it to reason that: (a) notwithstanding that he was in fact provoked to lose his self-control; (b) the scale of the retaliation was out of proportion to the provocation offered; so that (c) the objective question should necessarily be answered against him; is an impermissible line of reasoning. The jury should not be led into thinking that killing by inflicting 213 wounds would deprive the accused of the provocation defence, whereas killing by some lesser number of wounds would not. D.5 Answering the ‘do as he did’ question 116. The ‘do as he did’ question does not ask if there was a misdirection by the Judge in the present case and, as will be evident from Section A.3 above, the Judge did not suggest to the jury that there was any requirement of reasonably proportionate retaliation. Instead, the Judge directed the jury in accordance with the language of section 4 of the Homicide Ordinance. 117. However, what the ‘do as he did’ question does raise is the broader question of whether it is incumbent on a judge to give the jury directions in cases where there is a risk that the jury might pursue the impermissible line of reasoning described in Section D.4 above. The short answer to that first part of the ‘do as he did’ question is that it is incumbent on a Judge to direct a jury in terms that will instruct them not to do so since that would be a line of reasoning inconsistent with the proper construction of section 4. 118. The second part of the ‘do as he did’ question asks whether the present case is such a case. The answer to that question lies in whether the nature and extent of the homicidal act in a given case is such that might lead to the impermissible line of reasoning. It also depends on the course of any cross-examination of the defendant and the content of the closing speeches of counsel. This question must therefore be answered on a case by case basis and, as explained in Section D.6 below, it is apparent that this was such a case. D.6 The adequacy of the Judge’s summing up on “do as he did” 119. The Judge’s summing up on provocation, set out at Section A.3 above, did not invite it to adopt the impermissible line of reasoning. Neither, however, did the Judge’s directions warn the jury not to adopt it. This was, however, a case where there was clearly a risk that the jury might do so. 120. This is amply demonstrated by the response Mr Tam gave to a question from the bench as to how the jury’s verdict in the present case could be explained in light of the Judge’s directions (albeit misdirections) to take it from her as a matter of law that the appellant was in fact provoked into a sudden and temporary loss of self-control and that the jury should proceed on the basis of the appellant’s evidence as to what was said and done between him and Madam Yeung. Mr Tam rightly surmised that, assuming the jury acted faithfully in accordance with those directions of the Judge, a plausible explanation for the verdict of guilty of murder is that the jury thought that no reasonable man would have struck Madam Yeung 213 times with the chopper. 121. The risk was increased by defence counsel’s submissions that tended to suggest the defence would avail the accused only if the jury were prepared to find that another person in the defendant’s position would have behaved exactly as the defendant did. In his closing speech, he submitted: “But the point is … whether a reasonable man could, or might, have done the same thing. That’s all you need. Would someone else in his position, might he have done the same thing? … Is it not possible, or could it not be the case that another person in exactly the same position, shoes, as the defendant, could or would not he have behaved exactly the same as the defendant behaved in this case? Have the prosecution proved to you that no reasonable man would behave – would not behave as the defendant behaved when he did what he did that day? … Of course, we say that cumulative provocation, acts and words were said, provocation did take place and another person, another reasonable man in the defendant’s position could or might have behaved exactly as the defendant has behaved.” 122. Confusingly for the jury, these submissions (on which they received no directions from the Judge) were inconsistent with an earlier submission made by counsel for the defence in the following terms: “This is very important. If you decide you find the deceased’s actions could have caused – that’s all you need, legally that’s all you need – could have caused the defendant to lose his self-control, you need not consider whether or not his actions therefore after was [proportional] or reasonable. What I’m saying, if you conclude that the actions could have caused the defendant to do what he did, the fact that there were 213 blows is neither here nor there. As her Ladyship will tell you, the law states there is no hard and fast rule that the mode of resentment must bear a reasonable relationship to the provocation. The focus that you must have is on the point immediately prior to the defendant’s attack on the deceased, bearing in mind what took place well before then.” 123. In the circumstances, there is clearly a risk in the present case that the impermissible line of reasoning might have been adopted and, for these reasons, the Judge’s direction on the ‘do as he did’ question was inadequate. D.7 The need for particular directions in certain cases and the type of direction that should be given 124. These issues do not arise in every case involving provocation. However, in cases where a real risk of the impermissible line of reasoning identified in Section D.6 above exists, a suitable direction which counteracts that risk should be given to the jury. What exactly is said will depend on the evidence adduced and on what has been advanced by counsel but a direction along the following lines, which for illustrative purposes is adapted to the present case, serves as a guide: “In addressing the issue whether [those words/that conduct] was or may have been such as to cause an ordinary person to do as the defendant did, it is not a question of whether that ordinary person might have reacted in exactly the same way as did the defendant. It is sufficient were you to find that the provocation which was presented would or might cause an ordinary person, placed in the same situation as the defendant, to lose self-control and cause [the victim’s] death with the intention of doing so or at least with an intention to cause really serious bodily harm. If that is what you find, then you must find the defendant not guilty of murder but guilty of manslaughter and the possibility that the ordinary man would not have [stabbed 200 times but only, say, ten times]/[used the chopper but might have used some other implement] is irrelevant.” 125. To this suggestion, we would add that in every case where the defence of provocation falls to be left to the jury, the trial judge should: (1) Before closing speeches, discuss with counsel whether the evidence and/or their submissions will call for such a direction and, if the judge decides that such a direction is to be given, should discuss with counsel the proposed terms of such a direction; and (2) In the event that such a direction is to be given, provide the jury with a written form of the direction to be presented as part of a fuller written direction as to the law of murder and the defence of provocation. Common sense dictates that the task of the jury is much facilitated by written handouts which clearly and succinctly specify what has to be proved before they may convict and, sometimes, by written routes to verdict. We do not suggest that such handouts or routes to verdict are necessary for every criminal trial and we do not suggest that of itself the failure to provide a handout or route to verdict impeaches the safety of a conviction. But, save in cases where the legal issue is simple – and murder cases are seldom simple – we encourage their use. E. Disposition of the appeal 126. For these reasons, we would answer the gravity question as set out in Section C.5 above and the ‘do as he did’ question as set out in Section D.5 above. 127. This is not a case in which it would be appropriate to substitute a conviction of manslaughter, as was submitted on behalf of the appellant. This was the course adopted by the High Court of Australia in Johnson. However, in the present case, where the issue of credibility of the facts on which the defence was based ought to have been left to the jury to determine, the only appropriate order on the quashing of the murder conviction is one of re-trial. F. Conclusion 128. For these reasons, we would allow the appellant’s appeal and quash his conviction for murder and order that he be re-tried. Mr Gerard McCoy SC, Mr Richard Donald and Ms Denise Souza, instructed by Tse Yuen Ting Wong, assigned by the Director of Legal Aid, appeared for the Appellant on 11 January 2017 Mr Gerard McCoy SC and Ms Denise Souza, instructed by Tse Yuen Ting Wong, assigned by the Director of Legal Aid, appeared for the Appellant on 19 January 2017 Mr William Tam SC, DDPP, and Mr Ivan Cheung, PP, of the Department of Justice, for the Respondent [1] In HCCC 457/2013 (this was in fact a re-trial of the appellant, the first trial before Beeson J and a jury in October 2010 having miscarried: see [2014] 4 HKC 145). [2] Now repealed and replaced by the defence of “loss of control”: see Coroners and Justice Act 2009, ss.54-56. [3] At p.716B-D; DPP v Camplin involved a 15 year old boy who had killed by hitting the victim with a chapati pan after the latter had buggered the defendant in spite of his resistance and then laughed at him. [4] CACC 131/2014, Judgment dated 30 June 2015 (“CA Judgment”) per Lunn VP at [18] and Macrae JA at [47]. [5] CA Judgment per Lunn VP at [18] and Macrae JA at [49]-[50]. [6] In CACC 131/2014, before Lunn VP, Macrae and McWalters JJA. [7] The other ground of appeal, unanimously rejected by the Court of Appeal, related to the adequacy of the Judge’s direction as to the alternative verdict of unlawful act manslaughter. [8] CA Judgment at [7], [56] and [68]. [9] Ibid. at [100]-[101]. [10] Ibid. at [92]-[93]. [11] CACC 131/2014, Judgment dated 23 October 2015. [12] FAMC 60/2015, Determination dated 26 May 2016 (Ribeiro, Tang & Fok PJJ). [13] At [16]. [14] Appearing with Mr Richard Donald and Ms Denise Souza. [15] Lee Chun-chuen v The Queen [1963] AC 220 at p.228 (a case decided under the common law before the enactment of section 4 of the Homicide Ordinance). [16] Per Tindal CJ in R v Hayward (1833) 6 C. & P. 157, 159, cited by Lord Diplock in R v Camplin (supra.) at p.713H. [17] R v. Smith (Morgan) (supra.) per Lord Hoffmann at p.159F and per Lord Millett at p.207A; A-G v Holley (supra.) per Lord Nicholls of Birkenhead at [3]. [18] (Cmd 8932), which led to the enactment of the Homicide Act 1957; see the passage cited by Lord Hobhouse in R v Smith (Morgan) at 194C. [19] [2001] 1 AC 146 at 172F-G. [20] [1978] AC 705 at 716H-717B. [21] [2001] 1 AC 146 at 211E-F. [22] Ibid. at 214A-D. [23] [2005] 2 AC 580 at [6]. [24] Ibid. at [22]. [25] An inquiry of sufficiency equivalent to that under section 4 of the Homicide Ordinance: see at [36] above. [26] (1990) 171 CLR 312 at 324. [27] Ibid. at 325. [28] (1993) 69 A Crim R 21 at 28. [29] (1994-1995) 183 CLR 58 at 66-67 and 69. [30] [2000] 2 NZLR 385 at [235]. [31] Crimes Act 1961, s.169(2), see [2000] 2 NZLR 385 at [65]; now repealed, see Crimes (Provocation Repeal) Amendment Act 2009. [32] [1978] AC 705 at 717C-E. [33] Ibid. at 726C. [34] [2001] 1 AC 146 at 209H-210C. [35] [2005] 2 AC 580 at [11]. [36] [2001] 1 AC 146 at 211G. [37] [2010] 3 S.C.R. 350 at [35]. [38] [1986] 1 S.C.R. 313 at 324. [39] [2005] 2 AC 580 at [12]. [40] [2001] 1 AC 146 at 198A (emphasis in original). [41] CA Judgment at [98]. [42] Specimen Directions in Jury Trials – Direction 51, Provocation (Murder); issued September 2013. [43] The Court was informed this was the last version of this particular direction prior to the repeal of the statutory defence of provocation in England and Wales. [44] CACC 90/2013, unrep., Judgment dated 6 June 2014 (Stock VP, Barnes and McWalters JJ). [45] CACC 152/2014, unrep., Reasons for Judgment dated 4 May 2015 (Lunn VP, Macrae and McWalters JJA). [46] See paragraph [175(2)]. [47] See pp.26-27 of the unreported judgment, at paragraphs [57]-[58]. [48] At p.9. [49] At pp.136H-137A. [50] R v Walker [1969] 1 WLR 311 at 316D. [51] At p.933A-B. [52] At p.316E-F. [53] [1978] AC 705 at p.718E-F. [54] At p.137D. [55] At p.137H. [56] At pp.137H-138B. [57] At p.138B-C. [58] At pp.138C-D. [59] At p.234B-C. [60] At pp.312H-313A. [61] In CACC 393/2010. [62] At [61]. [63] At [62]. [64] At [67]-[70]. [65] At p.649, Crimes Act 1900 (NSW) s.23: “(1) Where, on the trial of a person for murder, it appears that the act causing death was induced by the use of grossly insulting language, or gestures, on the part of the deceased, the jury may consider the provocation offered, as in the case of provocation by a blow. (2) Where, on any such trial, it appears that the act or omission causing death does not amount to murder, but does amount to manslaughter, the jury may acquit the accused of murder, and find him guilty of manslaughter, and he shall be liable to punishment accordingly: Provided always that in no case shall the crime be reduced from murder to manslaughter, by reason of provocation, unless the jury find: (a) That such provocation was not intentionally caused by any word or act on the part of the accused; (b) That it was reasonably calculated to deprive an ordinary person of the power of self-control, and did in fact deprive the accused of such power, and (c) That the act causing death was done suddenly, in the heat of passion caused by such provocation, without intent to take life.” [66] At pp.639-640. [67] At p.67. [68] At pp.69-70. [69] At [131]. [70] At p.67. [71] Crimes Act 1961, which materially provides: “169. Provocation – (1) Culpable homicide that would otherwise be murder may be reduced to manslaughter if the person who caused the death did so under provocation. (2) Anything done or said may be provocation if – (a) In the circumstances of the case it was sufficient to deprive a person having the power of self-control of an ordinary person, but otherwise having the characteristics of the offender, of the power of self-control; and (b) It did in fact deprive the offender of the power of self-control and thereby induced him to commit the act of homicide. (3) Whether there is any evidence of provocation is a question of law. (4) Whether, if there is evidence of provocation, the provocation was sufficient as aforesaid, and whether it did in fact deprive the offender of the power of self-control and thereby induced him to commit the act of homicide, are questions of fact. (5) No one shall be held to give provocation to another by lawfully exercising any power conferred by law, or by doing anything which the offender incited him to do in order to provide the offender with an excuse for killing or doing bodily harm to any person. (6) This section shall apply in any case where the provocation was given by the person killed, and also in any case where the offender, under provocation given by one person, by accident or mistake killed another person. (7) The fact that by virtue of this section one party to a homicide has not been or is not liable to be convicted of murder shall not affect the question whether the homicide amounted to murder in the case of any other party to it. [72] At [136]. 
The Court: Introduction 1. On 4 August 2015, the appellant, Haresh Murlidhar Harjani, was convicted in the District Court[1] of conspiring with others to deal with property, knowing or having reasonable grounds to believe that that property represented the proceeds of an indictable offence, contrary to sections 25(1) and (3) of the Organized and Serious Crimes Ordinance, Cap 455 (“OSCO”) and sections 159A and 159C of the Crimes Ordinance, Cap 200. On 7 October 2015, he was sentenced to 3 years and 9 months’ imprisonment. 2. Section 25(1) of OSCO provides that: “Subject to section 25A, a person commits an offence if, knowing or having reasonable grounds to believe that any property in whole or in part directly or indirectly represents any person’s proceeds of an indictable offence, he deals with that property.” 3. In reaching his decision, the Judge relied upon the judgment of this Court in HKSAR v Pang Hung Fai[2](“Pang Hung Fai”), interpreting the effect of it as requiring a substantial objective element when deciding whether a person had “reasonable grounds to believe” that property represented the proceeds of an indictable offence. He rejected the suggestion that if it were or might be the case that a defendant honestly believed that the property was not thus tainted, he was entitled upon that basis alone to be acquitted. Although he found that the appellant did not know that the property represented the proceeds of an offence, he concluded that he had shut his eyes to obvious indicia of illegality and had reasonable grounds to believe that the property was thus tainted. 4. On 12 September 2017, the appellant’s appeal against conviction was dismissed but the Court of Appeal[3] held that the Judge had adopted an erroneous approach to the question of “reasonable grounds to believe”. In purporting to apply the decision of this Court in HKSAR v Yeung Ka Sing Carson[4](“Carson Yeung”), the Court of Appeal concluded that it sufficed to mandate an acquittal if a defendant honestly believed, or may honestly have believed, that the property with which he dealt did not represent the proceeds of an indictable offence, even if that belief were, objectively assessed, unreasonable. However, the court concluded that, notwithstanding the Judge’s erroneous approach to the law, it was apparent from the Reasons for Verdict that he had found that the appellant did not hold that honest belief and, accordingly, dismissed the appeal. 5. The application for leave to appeal to this Court was restricted to substantial and grave injustice limb of section 32 of the Court of Final Appeal Ordinance, Cap 484. However, since it was apparent in the light of the Court of Appeal’s judgment that the judgments in Pang Hung Fai and Carson Yeung merited clarification, and that other issues relevant to the application of section 25 of OSCO and section 159A of the Crimes Ordinance presented themselves for consideration, this Court raised a number of questions for argument and determination, namely: (1) What is the meaning of “having reasonable grounds to believe that any property … represents any person’s proceeds of an indictable offence” (abbreviated to “the property is tainted”) in section 25(1) of OSCO? (2) What is the relevance of the defendant’s actual belief in determining whether the statutory test is satisfied? (3) To what extent is “wilful blindness” relevant in determining whether the statutory test is satisfied? (4) In the light of section 159A of the Crimes Ordinance: (a) given the requirements of subsection (2), can there be an offence of conspiracy to deal with property having reasonable grounds to believe that such property … represents any person’s proceeds of an indictable offence? (b) given the requirements of subsection (1)(a), where defendants have reasonable grounds to believe that property is tainted, will they be guilty of conspiracy if they agree to deal with the property notwithstanding that those grounds may not exist at the time of the dealing? The facts and the charge 6. The particulars of the charge, as ultimately amended, read as follows: “HARJANI Haresh Murlidhar, between the 26th day of April, 2014 and the 21st day of July, 2014, both dates inclusive, in Hong Kong, conspired with CASTELINO Brian Mario, DIALLO Ibrahima and other persons unknown, to deal with property, namely a sum of $539,375 United States currency in account numbered 076-402-63220051 of State Bank of India held in the name of Sino Investment and Trading Limited, knowing or having reasonable grounds to believe that the said property, in whole or in part directly or indirectly represented the proceeds of an indictable offence.” 7. The facts which gave rise to the charge were not complex. A contract was made by exchange of emails for the sale by Dohigh Trading Limited (“Dohigh”) to Dragon Asia Fertilizer Limited (“Dragon Asia”) of a shipment of fertilizer to be shipped from a Chinese port to Bangladesh for US$10,788,000. Dragon Asia was required to make a down payment of 5% of the sum, with the balance to be paid by letter of credit. Those emails were hacked and modified so as to deceive Dragon Asia into paying the required deposit into a bank account of Sino Investment and Trading Limited (“SIAT”) at the Hong Kong branch of State Bank of India (“SBI”) and into nominating SIAT as the beneficiary of the letter of credit. Accordingly, on 9 July 2014, the sum of US$539,375 (representing the 5% deposit less bank charges of US$25) was diverted and paid into the US Dollar bank account of SIAT at the Hong Kong branch of SBI instead of to Dohigh. 8. SIAT was a company incorporated in Hong Kong by the appellant and one Castelino Brian Mario (“Brian”) in 2012. Brian, who remained in India throughout, instructed transfers totalling US$327,175 to be made from that account to SIAT’s Hong Kong Dollar account. The appellant and Brian were the signatories for both bank accounts. On 12 July 2014, the appellant came to Hong Kong from Sri Lanka and, after he withdrew cash totalling HK$236,000 between 15 and 18 July 2014, he was arrested on 21 July 2014 and charged with conspiracy to deal with the US$539,375 knowing or having reasonable grounds to believe that it represented the proceeds of an indictable offence, contrary to sections 25(1) and (3) of the OSCO and sections 159A and 159C of the Crimes Ordinance. 9. Initially, the charge named the appellant and Brian as the two co‑conspirators. The prosecution had intended to call as a witness, one Diallo Ibrahima (“Diallo”) with a view to rebutting the appellant’s evidence that he had been asked to handle matters relating to the contract by someone he knew as “Daniel”. However, when shown defence exhibits consisting of extensive email and skype communications passing between Diallo and the appellant relating to the funds and to use of bank accounts, the prosecution dropped Diallo as their witness. After Diallo was produced and cross-examined by the defence, the prosecution amended the charge to name Diallo as a co-conspirator along with Brian, the appellant and other persons unknown (such persons being those who had manipulated the emails). It was the prosecution’s case (accepted by the Judge) that Diallo and “Daniel” as well as the person who used the email or skype names of “saint cool” or “coolsaintt” were the same person. The Judge also found that Diallo “knew there was fraud going on”.[5] 10. The appellant admitted that Dragon Asia had been deceived by fraudsters and that the deposits and withdrawals had occurred, as instructed by Brian. The appellant’s case was that he was a legitimate businessman and that both he and Brian had dealt with Diallo but believed that the latter was an agent acting bona fide on behalf of the principals in the fertilizer deal with the funds in question deriving from a genuine commercial transaction. 11. The appellant claimed that he was asked to execute the letter of credit and to receive the deposit and the letter of credit proceeds in the SIAT account as an account that had been designated by Dragon Asia, the letter of credit’s applicant. The reward for so doing and for providing inspection services at shipment would be 15% of the contract price, with SIAT getting 12% (US$1,294,560) and Diallo receiving the other 3% (US$323,640). After deducting this 15%, the balance of the monies would be remitted to the supplier. He claimed that the US$539,375 was received by SIAT as part of the aforesaid 15% which it would earn if the deal went through, but which would be returned in case it did not. The fraud came to light on 17 July 2014 when Dohigh informed Dragon Asia that it had not been paid and Dragon Asia revoked the letter of credit. The diverted deposit amount had to be replaced and the fertilizer shipment was postponed from August to September. The appellant claimed that he was deceived by Diallo into unwittingly receiving the proceeds of that fraud. The judgments below and this appeal 12. The debate in the courts below focussed on the mental element of the substantive offence, in particular on the true construction of the words in section 25(1) of OSCO “knowing or having reasonable grounds to believe that any property in whole or in part directly or indirectly represents any person’s proceeds of an indictable offence”. Defence counsel’s submission was that “an honest belief would be a defence to both the substantive charge and the alleged conspiracy”, even if the appellant’s belief was not reasonable due to lack of due diligence on his part.[6] The prosecution’s case was, on the contrary, that “only an honest and reasonable belief can be considered as a valid ground” for exculpation.[7] 13. The Judge held in favour of the prosecution’s construction, basing his approach on his reading of Pang Hung Fai, which was then the leading authority. His Honour held that: “The approved test on ‘having reasonable grounds to believe’ is that the jury should consider such grounds the accused had for believing and that anyone looking at those grounds objectively would believe the property represented proceeds of an indictable offence. An alternative interpretation is whether the accused ought to have known the property represented proceeds of an indictable offence.”[8] 14. The Judge found that Diallo had not told the appellant about the fraud and that it was not possible to hold that he knew that the US$539,375 represented the proceeds of an indictable offence. Nonetheless, he was satisfied that the appellant had reasonable grounds to believe that the property was thus tainted. The appellant was an educated man and an experienced businessman. He had shut his eyes to the obvious. There were ample indicia that the funds were from a tainted source. He said: “SIAT was only incorporated in Hong Kong in 2012. It had no established business or reputation. It was not an L/C specialist. Why would someone pick SIAT for the job? Why would someone entrust an unrelated company (SIAT) to receive first the USD539,375 and later a much greater sum of over [USD]10M through execution of the L/C? And more puzzling still, why would someone give SIAT 12% (USD1,294,560) and coolsaintt 3% (USD323,640) as commission for such a simple task? I am sure it was only because Defendant could offer SIAT’s account as available for coolsaintt’s side to use that account to receive money for conduit. That was the reason (and the only reason) why SIAT was named as beneficiary on the Contract and the L/C. I am sure Defendant was aware of such reason.”[9] “Their communication shows that as early as 26 January 2014, Defendant upon request had offered coolsaintt his SIAT account in SBI. Coolsaintt asked Defendant to confirm its correctness on 26 April 2014. … By then, nothing had been mentioned of the fertiliser deal. The first vague reference of the matter appeared in coolsaintt’s email on 3 July 2014 … when he asked Defendant to check if the money had been wired in.”[10] … “In his emails to coolsaintt, Defendant complained that he was asked to move money to many accounts. He worried that would arouse the bank to investigate and thus endanger his own account.”[11] 15. As for the fact that the appellant had never asked coolsaintt why SIAT had been chosen for the job and the appellant’s explanation that he did not feel any need to, it being, as a matter of commercial practice, beyond the role of an agent to do so, the Judge rejected this explanation saying that: “… I am sure the truth is that Defendant did not want to ask coolsaintt anything which would jeopardize their business. … I am sure Defendant knew he/SIAT was chosen only because of the availability of SIAT’s account to receive the down payment and the L/C money. …”[12] … “… The whole situation here is, however, so questionable that it calls for immediate rigorous proof or due diligence enquiry. Someone for no good reasons wanted to use SIAT’s name on the Contract and the L/C as beneficiary and required SIAT’s account to receive the down payment and the L/C money. A huge and totally disproportionate reward (12% of USD10,778,000) was offered to Defendant. Defendant in such circumstance ought to think or dig deep into the matter. Yet, he did not but simply turned a blind eye to the situation. ... He was unreasonably contented with the scarce information he had of the highly suspicious deal coolsaintt brought him. His belief that the money was legitimate was, therefore, ill-founded.”[13] “… I am sure anyone looking objectively at the grounds Defendant had would believe differently from Defendant, who was in fact turning a blind eye to a highly questionable situation which he must have so appreciated. Anyone looking objectively at such grounds Defendant had would believe the money involved represented in whole or in part, directly or [indirectly] the proceeds of an indictable offence. Defendant ought to have known that but he chose not to. He was only concerned with the profit.”[14] … “… In the present case, I find Defendant’s so-called ‘belief’ was just ill-founded and equivalent to turning a blind eye to obvious suspicion. It, therefore, could hardly be said to be sincere or in good faith.”[15] 16. The Court of Appeal held that the Judge was wrong to require that any genuinely held belief by a defendant must also be reasonably held. Instead, in purporting to apply Carson Yeung, it held that: “The correct approach ... is to use those facts that suggest that no reasonable person would believe in the legitimacy of this transaction as part of his assessment of the credibility of the appellant’s claim that he in fact genuinely held beliefs about the legitimacy of the transaction.”[16] 17. The issue, in the eyes of the Court of Appeal, was: “... whether the judge used reasonableness in the permissible way as a means of assessing the plausibility of the appellant’s evidence en route to determining whether it was or might be true, or whether he used it in the impermissible way of requiring that his asserted beliefs be not only genuinely held but also be reasonably held.”[17] 18. This interpretation of the effect of the judgment in Carson Yeung is also evident from the Court of Appeal’s comment upon the Judge’s encapsulation of the contending submissions of the parties in the court below, contrasting the submissions of Mr Beel, counsel for the appellant at trial with those of counsel for the prosecution. That encapsulation ran as follows: “In court, Counsel says even if Defendant just honestly (but not reasonably) believed the money was from a legitimate source, that would suffice as a defence to the substantive offence of money laundering as well as to conspiracy to launder money. (Prosecutor disagrees. He says it is only a valid consideration for either offence if Defendant was honest and reasonable in holding his belief of the source of the money as legitimate).”[18] 19. As to these divergent contentions, the Court of Appeal said: “This would suggest that Mr Beel was urging upon the judge a correct statement of the law whilst the prosecutor was inviting the judge to apply the law in an incorrect manner.”[19] 20. After analysing the Judge’s findings, the Court of Appeal dismissed the appeal, holding that: “... notwithstanding that the judge erred by not applying the correct law to his assessment of how the appellant’s claim as to the beliefs he held impacted upon proof that he had the mens rea of reasonable grounds to believe, the appeal must be dismissed. This is because, by a perfectly proper route, the judge reached a concluded view that the appellant’s claim as to his beliefs was not truthful. Having properly rejected that claim the judge correctly applied the Pang Hung Fai law in finding that the mens rea element of the offence had been proved beyond reasonable doubt.”[20] 21. The appellant lodged an application for leave to appeal to this Court on the substantial and grave injustice basis, contending that the Court of Appeal did not have any valid alternative basis for upholding the conviction.[21] The Appeal Committee[22] granted leave on this basis and additionally, certified and granted leave on a question of law, held to be one of great and general importance, as to: “Whether wilful blindness is a sufficient basis for sustaining liability under section 25(1) of the Organized and Serious Crimes Ordinance, Cap 455.” 22. On 23 November 2018, shortly before the appeal came on for hearing on 27 November 2018,[23] the Court wrote to the parties, noting that this was a conspiracy case and that “a question appears to have been raised as to possible differences in the elements of the offence required to be proved in a case charged as a conspiracy as compared with a case charged as a substantive money laundering offence”. The Court indicated that it might wish to be addressed “as to whether the aforesaid question concerning possible differences in the elements of the offence arises as a matter of law in the present case, touching on the issue of wilful blindness and on mens rea generally, especially in the light of section 159A(2) [of the Crimes Ordinance, Cap 200]”. 23. The parties were not ready to deal with these questions at the hearing, so the appeal was adjourned to a date to be fixed, with the Court specifying by letter dated 27 November 2018, the expanded matters on which it wished to be assisted, namely, the questions set out at paragraph [5] above. Question 1: What is the meaning of “having reasonable grounds to believe that any property … represents any person’s proceeds of an indictable offence” (abbreviated to “the property is tainted”) in section 25(1) of OSCO? Question 2: What is the relevance of the defendant’s actual belief in determining whether the statutory test is satisfied? 24. The Court has addressed these questions in Pang Hung Fai and Carson Yeung but the decision of the Court of Appeal in this case shows that clarification is necessary as to the effect of these decisions. 25. In Pang Hung Fai [24], in a judgment with which all members of this Court agreed, Spigelman NPJ said that on most occasions the test propounded by the Appeal Committee in Seng Yuet Fong v HKSAR[25] would suffice: “To convict, the jury had to find that the accused had grounds for believing; and there was the additional requirement that the grounds must be reasonable: That is, that anyone looking at those grounds objectively would so believe.” 26. In Carson Yeung[26], this Court endorsed that proposition. We remain of the view that the Seng Yuet Fong test correctly represents the law. In the interests of clarity, however, we would reformulate the test as follows: (i) What facts or circumstances, including those personal to the defendant, were known to the defendant that may have affected his belief as to whether the property was the proceeds of crime (“tainted”)? (ii) Would any reasonable person who shared the defendant’s knowledge be bound to believe that the property was tainted? (iii) If the answer to question (ii) is “yes” the defendant is guilty. If it is “no” the defendant is not guilty. 27. Thus the first issue that the judge or jury (“the court”) must address is what matters the defendant knew of that might have affected his belief as to whether the property was clean or tainted. This question is subjective only in as much as it requires the tribunal to make findings as to the knowledge of the defendant at the time of the relevant transaction. Where the defendant gives evidence of facts and matters that affected his belief about the nature of the property, the court has to decide whether he is, or may be, telling the truth about the existence of these facts and matters. 28. The second issue is whether any reasonable person who shared the defendant’s knowledge would have been bound to believe that the property was tainted. This question is objective. Where the court finds that the defendant was, or may have been, telling the truth about the existence of facts and matters that he claims affected his belief, the court must take those facts and matters into account when answering the question, would any reasonable person with knowledge of those facts and matters have believed that the property was tainted? If the answer to the question is “yes” the defendant is guilty. If it is “no” the defendant is not guilty. 29. Applying these principles in practice will normally be relatively straightforward where the defendant does not give or adduce evidence. The court has first to find what relevant facts or circumstances were known to the defendant and then decide whether those facts or circumstances would have led any reasonable person to believe that the property in question was tainted. If the answer is “yes” the defendant will be convicted. When the judge comes to sentence he or she will be likely to do so on the basis that the defendant must also have believed that the property was the proceeds of crime. 30. Difficulty can arise in practice where the defendant gives evidence that he did not believe that the property was tainted. Although the test in law is objective – “would any reasonable person believe the property was tainted?” – in applying that test the court must give due consideration to the evidence given by the defendant as to what he believed and why. The court has to consider two interrelated questions: (i) is the defendant telling the truth when he says that he did not believe that the property was tainted and (ii) could a reasonable person in the position of the defendant have failed to believe that the property was tainted? 31. Normally the court will give the same answer to each question. If the court concludes that no reasonable person in the position of the defendant could have failed to believe that the property was tainted the court is likely to reject the defendant’s assertion that he did not have this belief. Applying the statutory test the defendant will be convicted. 32. Conversely, where the court accepts that the defendant did not believe that the property was tainted, this is likely to be in circumstances where the court has concluded that a reasonable person in the position of the defendant would not necessarily have believed that the property was tainted. Applying the statutory test the defendant will be acquitted. 33. A rare case may arise where the court concludes that any reasonable person in the position of the defendant would have believed that the property was tainted but nonetheless accepts the evidence of the defendant when he says that he did not have this belief. This is only likely to arise in circumstances where it is apparent that the defendant lacks the reasoning abilities of the normal person. In such circumstances, applying the statutory test, the defendant should be convicted but the fact that he did not himself believe that the property was tainted may well be a mitigating factor when he is sentenced. 34. In the present case, the Judge held[27] that the appellant did not know that the funds that he handled represented the proceeds of an indictable offence and, accordingly, proceeded to consider whether he had “reasonable grounds to believe” that the funds were tainted. He had no difficulty in finding that this test was satisfied. He held that the appellant “ought to have known” that the funds were tainted but deliberately turned a blind eye to their provenance[28]. Nowhere did the Judge make an express finding as to the belief of the appellant as to the nature of the funds that were to be paid into his account. On the Judge’s view of the law he did not need to do so. It sufficed to convict the appellant that he had “reasonable grounds to believe” that the funds were tainted. 35. The Court of Appeal considered that the Judge had fallen into error in setting out his understanding of the meaning of “reasonable grounds to believe”. The court further held that the Judge would have erred in law had he convicted the appellant notwithstanding having made a finding that the appellant did not believe that the funds were tainted. It found, however, on true analysis of his judgment, he had not accepted the appellant’s evidence that he “held a genuine belief in the legitimacy of this transaction”[29]. On this basis the Court of Appeal upheld the judgment. 36. As we shall explain, it was the Court of Appeal and not the Judge that erred in explaining the test of “reasonable grounds to believe”. The confusion that is demonstrated by the difference between the reasoning of the Judge and that of the Court of Appeal in this case has its origin in some passages in the judgment of Spigelman NPJ in Pang Hung Fai. 37. In that case the Defendant ‘Pang’ had permitted a friend ‘Kwok’ to use his Hong Kong bank account to receive and then to remit to Cambodia very substantial funds. These funds were the proceeds of a fraud perpetrated by Kwok. Pang was charged with breach of section 25 of OSCO. The case against him was not that he knew that the sums were tainted but that he had reasonable grounds to believe that they were tainted. 38. The prosecution case, not untypically, was founded largely on the adverse conclusions that would reasonably be drawn from the nature of the transactions themselves. Pang gave evidence that he did not suspect that the funds were tainted because “on entirely reasonable grounds, he had trusted Kwok implicitly”[30]. He gave detailed evidence of why he trusted Kwok. This included the fact that they had been friends for over 30 years, he had always found him to be scrupulous, and had never known him to do anything dishonest or dishonourable nor to have his integrity or honesty questioned. 39. The relevant issue was whether Pang’s alleged trust in Kwok’s honesty could be taken into account when considering whether he had reasonable grounds to conclude that the funds were tainted. As to this, Spigelman NPJ held: “83. I can see no reason why, when applying the words of the statutory formulation, what McWalters J called: ‘the personal beliefs, perceptions and prejudices’ of the accused should be entirely omitted from consideration. That does not mean that any such ‘perception or evaluation’, to use the words of the ground of appeal, is entitled to weight, let alone determinative weight. 84. The error, I believe, is in the reasoning process by which the statutory word ‘grounds’ has been replaced by the word ‘facts’. As I have said above, ‘facts’ is a narrower concept than ‘grounds’. It may be that a ‘belief, perception or prejudice’ is not a ‘fact’. However, such matters fit quite readily within the concept of a ‘ground’, which a particular person can be said to have ‘had’. 85. When assessing the whole of the evidence, the judge or jury can give such weight to an accused’s belief, perception or prejudice as s/he believes is warranted. No doubt, in many cases, that decision maker will entirely discount such evidence of the accused. Nevertheless, they are ‘grounds’ which stand or fall by the test of reasonableness. … 88. For present purposes, Lord Sumption JSC put it precisely in Hayes v Willoughby [2013] 1 WLR 935, [14]: ‘Reasonableness is an external objective standard applied to the outcome of a person’s thoughts or intentions’. 89. In my opinion, contrary to the conclusion of the Court of Appeal, this approach does not convert an objective test into a subjective test. …” 40. We would make the following comments in relation to this passage. The critical question in Pang Hung Fai was whether any reasonable person with Pang’s knowledge and experience, would have concluded that the funds were tainted. Pang gave evidence that, prior to the transaction, he believed that Kwok was honest and that the facts of the transaction did not cause him to change that belief. Spigelman NPJ held that the fact that Pang had believed Kwok to be honest was a relevant “ground” to be taken into consideration when deciding whether, nonetheless, any reasonable person would have concluded that the funds were tainted. 41. As Spigelman NPJ made plain, however, it did not follow that any weight, let alone decisive weight, should be attached to Pang’s belief that Kwok was honest. The question remained as to whether, despite having such belief prior to the transaction, any reasonable person would have concluded that the funds were tainted. That was an objective question. 42. How much weight should be given to Pang’s belief that Kwok was honest depended critically on the grounds for holding that belief. Had Pang only met Kwok a week before the transaction the fact that he considered Kwok to be honest would have carried little or no weight. What was significant was the long relationship between them before the transaction, which provided grounds to believe that Kwok was honest. Spigelman NPJ criticized the trial judge for failing to take proper account of these matters[31]. Both Pang’s trust in Kwok and the reasons for that trust should together have been weighed in the balance when deciding whether any reasonable person, having the grounds that Pang had, would have concluded that the funds were tainted. 43. The grounds that Pang had for believing that Kwok was honest were equally grounds that he had for believing that the funds were not tainted. Where, as in Pang Hung Fai, a defendant gives evidence that he held a relevant belief or perception at the time of the impugned agreement the appropriate course for the trial judge or the jury is to have regard to all the grounds that the defendant had for holding such belief or perception when considering the fundamental question – “would anyone, looking objectively at the grounds that the defendant had, conclude that the property was tainted?” As Spigelman NPJ emphasized the test is an objective one. 44. We turn to consider the decision of this Court in Carson Yeung. That case also involved charges of contravening section 25(1) of OSCO by using bank accounts to deal with very large sums of money. A number of issues fell to be considered by this Court. The relevant issue required the Court to reconsider the meaning of “having reasonable grounds to believe”in the light of the decision in Pang Hung Fai. In particular, the Court had to consider to what extent the Court had to make findings about the defendant’s state of mind when deciding whether he had “reasonable grounds to believe”. 45. At paragraphs 110 to 114 the Court gave three examples of cases where it was necessary to consider the defendant’s state of mind. In each case the reason why it was necessary to do so was to decide whether there were circumstances known to the defendant which would or could have affected his or her belief as to whether the relevant property was tainted. What was material was not whether the defendant subjectively believed that the property was not tainted but whether the defendant was aware of facts and circumstances that might reasonably lead to the conclusion that the property was not tainted. 46. The first case considered[32] was HKSAR v Yan Suiling[33]. In that case the defendant’s explanation in respect of the impugned transactions was that they were pursuant to transactions designed to circumvent the Mainland’s exchange control regulations. Had this been true it would have supported her case that she had no grounds for believing that the relevant funds were the proceeds of an indictable offence. The Court did not believe her explanation. 47. The second case considered[34] was HKSAR v Li Kwok Cheung George[35]. In that case it was the defendants’ case that the property dealt with consisted of clean funds provided by a financial services company, albeit that they were intended to be used for the purposes of fraud. If this was true, as the Court found, there were no reasonable grounds to believe that the funds in question were the proceeds of an indictable offence. Once again the Court was concerned with the relevant background facts known to the defendants, not whether they subjectively believed that the funds were not tainted. 48. The third case considered was Pang Hung Fai itself. This Court cited the following passage from the judgment of Spigelman NPJ[36]: “106. … Although earlier in his judgement, the trial judge sets out the close personal and business relationship between the two men, extending as it did over decades, none of these factors are contained in the list of facts available for consideration by the ‘commonsense, right-thinking member of the community’ at [101]-[107]. Accordingly, even the matters referred to in [102], [103] and [104] are out of context. More importantly, none of these other aspects of the relationship are listed as facts which the ‘reasonable person’ took into account. This error is probably a result of asking the HKSAR v Shing Siu Ming questions in the wrong order. 107. In my opinion the appellant’s contention that the list of grounds considered by the trial judge was too narrow should be upheld.” 49. This citation exemplifies the point that we have made at paragraph [42] above. It is the facts and circumstances that a defendant asserts led him to a particular belief or perception that are significant rather than the subjective belief or perception itself. 50. In Carson Yeung this Court went on to say this[37]: “If … the evidence provided by the defendant as to what he perceived and believed is accepted as true or as evidence which may be true; and if true would be inconsistent with him having reasonable grounds to believe that the property in question represents the proceeds of crime, an acquittal is called for since an essential mens rea element cannot be established against the defendant.” 51. The phrase that we have emphasized is of critical importance. A belief or perception held by the defendant will only be inconsistent with his having reasonable grounds to believe that property is tainted if that belief or perception is itself founded on reasonable grounds. That is why the important question is not merely what beliefs or perceptions the defendant may have had but the grounds advanced by the defendant for holding the alleged beliefs or perceptions. 52. In Carson Yeung this Court continued as follows[38]: “No difficulty arises if the defendant’s evidence provides a plausible explanation for his dealing with the property. But what of the situation where the defendant’s asserted perceptions or beliefs, even though believed, would strike others as excessively naïve or gullible or foolish in the light of objective facts known to the defendant which would have led others readily to believe that reasonable grounds existed to believe that the property was tainted? The facts of HKSAR v Pang Hung Fai furnish a possible example. Many people might well have considered it suspicious for Kwok to ask Pang to receive payments into Pang’s bank account and then to remit the funds to Cambodia shortly afterwards for no apparent reason. Kwok, as Pang must have known, undoubtedly had his own bank accounts and, given his substantial business interests in Cambodia, undoubtedly had ample means of transferring funds between Hong Kong and that country. However, given the evidence of his long and particularly close friendship with Kwok described above, if there had been a retrial, Pang might have been able plausibly to claim that he had asked no questions because he trusted Kwok implicitly, believing him to be an honourable and substantial businessman and that he had no reason to suspect that the funds were the proceeds of crime. The principle which requires the focus to remain on whether the defendant had the requisite reasonable grounds to believe dictates that even in such cases, the offence is not made out.” 53. The Court of Appeal in the present case referred to the passages of Carson Yeung set out above[39]. It went on to advance the following propositions. “40. As a result of the Court of Final Appeal decision in Yeung Ka Sing Carson the law is now settled on the mens rea element of the section 25(1) OSCO offence. 41. When a defendant gives evidence and testifies as to his perceptions and beliefs in respect of the property with which he dealt, or as here conspired to deal, the court is called upon to determine two issues, namely: (i) does it accept that the defendant’s evidence is true or may be true; and (ii) if it does so accept, would that evidence be inconsistent with the defendant having reasonable grounds to believe that the property in question represents the proceeds of crime? 42. If both questions are answered in the affirmative then the defendant will be entitled to an acquittal. The second question will almost inevitably be answered in the affirmative as the whole purpose of the defendant testifying will be to advance a defence which, if accepted, will secure his acquittal. 43. Thus the primary focus of the court will be on the first question. The judgments of the Court of Final Appeal, both in Pang Hung Fai and in Yeung Ka Sing Carson, make it clear that it is for the court to assess the claim made by a defendant as to his perceptions and beliefs and to decide what weight it should give to that claim. 44. In Pang Hung Fai, Spigelman NPJ, in giving a judgment with which the other members of the court agreed, emphasized that having regard to a defendant’s perceptions and beliefs did not mean that those perceptions or beliefs were ‘entitled to weight let alone determinative weight.’ He went on to say at paragraph 85 of his judgment: ‘85. When assessing the whole of the evidence, the judge or jury can give such weight to an accused’s belief, perception or prejudice as s/he believes is warranted. No doubt, in many cases, that decision maker will entirely discount such evidence of the accused. Nevertheless, they are ‘grounds’ which stand or fall by the test of reasonableness.’ 45. Thus, as with the testimony of any witness, it falls to the judge to determine whether what the defendant said is true or, where the witness is a defendant, might be true. In performing this task the judge, again as with any witness, has regard to the inherent plausibility of the defendant’s testimony. 46. The more reasonable a defendant’s testimony the more plausible it will be and the more plausible it is the more the court will be inclined to accept that it is or may be true. In this way the reasonableness of a defendant’s claim as to his perceptions or beliefs is relevant to the court’s determination of whether the defendant’s evidence as to his perceptions and beliefs is true and that he did genuinely hold the perceptions or beliefs that he claims he held. 47. However, what the court cannot do is require that the defendant’s perceptions or belief be reasonably held in addition to being genuinely held. This was made clear by the Court of Final Appeal in Yeung Ka Sing Carson at paragraph 119 of its judgment …”. 54. Thus the Court of Appeal has interpreted Pang Hung Fai and Carson Yeung as imposing a subjective test of belief, so that if the court concludes that the defendant believed, or may have believed, that the property was not tainted he will be entitled to be acquitted, even if that belief was not reasonable. On this basis the Court of Appeal held (i) that Mr Beel for the appellant was correct to contend that an honest belief that the funds were clean would suffice to establish a defence and (ii) that prosecuting counsel was wrong to submit that only an honest and reasonable belief that the funds were clean could provide a defence. 55. Whilst we can see how paragraph 119 of Carson Yeung (at paragraph [52] above)might give rise to the interpretation which the Court of Appeal applied, the interpretation of Pang Hung Fai and Carson Yeung is not in fact correct and does not reflect what was intended. Its starting point in paragraph 42 is that the defendant’s perception or belief is likely to be determinative of the question of whether or not there were reasonable grounds to believe that the property was tainted. This ignores Spigelman NPJ’s statements, quoted by the Court of Appeal, that the defendant’s perceptions would not necessarily carry any weight, let alone determinative weight. It is contrary to the observation of Spigelman NPJ that the test is objective, and it conflicts with the second limb of the test in Seng Yuet Fong, approved by this Court in Pang Hung Fai and Carson Yeung (at paragraphs [25]-[26] above). The Court of Appeal has failed to appreciate the words that we have emphasized in paragraph 119 of Carson Yeung that on the facts of Pang Hung Fai, “given the evidence of his long and particularly close friendship with Kwok …, Pang might have been able plausibly to claim that he had asked no questions because he trusted Kwok implicitly, … and that he had no reason to suspect that the funds were the proceeds of crime” (emphasis in bold added). These are the words of the application of an objective, not a subjective test. 56. What this Court was emphasizing at paragraph 119 of Carson Yeung was the need to apply the test of “reasonable grounds to believe”from the viewpoint of the defendant, having regard to all the facts and circumstances known to him, and not from the viewpoint of an objective bystander considering simply the adverse inferences to be drawn from the details of the transaction itself. This Court was certainly not intending to indicate that the defendant was entitled to be acquitted if he believed, or may have believed, that the property was not tainted, where he did not have reasonable grounds for such belief. 57. Once the court has determined, having due regard to the defendant’s evidence if he has testified, what facts and matters affected, or may have affected, his belief as to whether or not the relevant property was tainted, including any facts or matters that may have led him to form personal beliefs, perceptions or prejudices, the court must then ask the objective question of whether, any reasonable person, affected by all those facts and matters, would have been bound to conclude that the property was tainted. 58. In a case such as Pang Hung Fai the court has to weigh the matters of which the defendant had personal knowledge, which would incline a reasonable person to believe that the transaction was clean, against the particulars of the impugned transaction itself, which would incline a reasonable man to believe that the transaction was tainted. If on balancing all these matters any reasonable person would be bound to come to the conclusion that the transaction was tainted, the defendant will properly be convicted. If, on balance, a reasonable man might conclude that the transaction was clean, an acquittal must follow. 59. The interpretation that this Court has given to “reasonable grounds to believe” accords with the interpretation that the Supreme Court of the United Kingdom has recently given in R v Lane[40] to the similar phrase “reasonable cause to suspect”. 60. Section 17 of the United Kingdom Terrorism Act 2000 provides: “A person commits an offence if– (a) he enters into or becomes concerned in an arrangement as a result of which money or other property is made available or is to be made available to another, and (b) he knows or has reasonable cause to suspect that it will or may be used for the purposes of terrorism.” 61. The Supreme Court identified the question raised by the appeal as follows[41]: “The question which arises on this appeal concerns the correct meaning of the expression ‘has reasonable cause to suspect’ in s 17(b). Does it mean that the accused must actually suspect, and for reasonable cause, that the money may be used for the purposes of terrorism? Or is it sufficient that on the information known to him there exists, assessed objectively, reasonable cause to suspect that that may be the use to which it is put?” 62. The Supreme Court, after considering the language of the statute and its statutory context, concluded that the latter alternative was the correct one. In so doing it distinguished, and circumscribed, the effect of the decision of the Supreme Court on mens rea in R v Saik[42] (“Saik”). 63. The Supreme Court observed that the relevant provision did not create an offence of strict liability,[43] adding the following comment, which is relevant in the context of the present case: “The requirement that there exist objectively assessed cause for suspicion focuses attention on what information the accused had. As the Crown agreed before this court, that requirement is satisfied when, on the information available to the accused, a reasonable person would (not might or could) suspect that the money might be used for terrorism. The state of mind of such a person is, whilst clearly less culpable than that of a person who knows that the money may be used for that purpose, not accurately described as in no way blameworthy.” 64. In the present case the Judge rightly preferred the submissions of law made by counsel for the prosecution to those made by counsel for the defence. He applied the right test to his findings of fact and properly convicted the appellant or would properly have done so had the appellant been charged with the substantive offence of contravening section 25(1) of OSCO rather than with conspiracy. The effect of charging conspiracy is considered below. Question 3: To what extent is “wilful blindness” relevant in determining whether the statutory test is satisfied? 65. The legal doctrine of “wilful blindness” has been used by courts in cases where actual knowledge of a fact or of particular circumstances is an essential ingredient of the offence. Under the doctrine a defendant is treated as having such knowledge if, having been put on enquiry, he has deliberately refrained from making enquiries that would have resulted in his acquiring the knowledge in question. The doctrine is controversial inasmuch as it has attracted academic and judicial criticism. 66. A useful summary of the doctrine whereby, on given facts, the defendant’s wilful blindness can properly be equated with actual knowledge is set out in Simester and Sullivan’s Criminal Law – Theory and Doctrine.[44] The authors put this as follows: “… This doctrine applies where the defendant intentionally chooses not to inquire whether something is true because he has no real doubt what the answer is going to be. Its effect is to attribute knowledge of the circumstance to the defendant. In other words, where the wilful blindness doctrine applies, the law will treat the defendant as having actual knowledge …, and not merely the reckless knowledge that he otherwise would have. The conditions under which the doctrine applies are not capable of being stated precisely. Broadly speaking, if there is an obvious way of finding something out and the defendant deliberately shuts his eyes to a risk by failing to find out, he will not be permitted to exculpate himself by claiming that he did not know the truth. Wilful blindness covers the case where D ‘deliberately refrained from making inquiries the results of which he might not care to have’. … On the other hand, wilful blindness cannot be invoked just because D should have inquired into the facts, or even if he suspected the truth; otherwise knowledge would effectively be indistinguishable from recklessness. However, the line between recklessness and wilful blindness is a fine one. It appears that the defendant is wilfully blind in two situations. The first is if he shuts his eyes and fails to enquire because he is virtually certain what the answer will be. … Alternatively, the wilful blindness doctrine will also apply if the means of knowledge are easily to hand, and D realises the likely truth of a matter but refrains from enquiry in order not to know. … It should be emphasised that wilful blindness is a doctrine of substantive rather than evidential law. It imputes knowledge to the defendant for legal purposes where there is not, in fact, such knowledge. Even if the doctrine does not apply, however, there remains the further possibility that a jury may infer actual knowledge on the part of the defendant (i.e. that in fact he accepted the truth and had no serious doubt), from evidence that D recognised the likely circumstances and made no further enquiry.”[45] 67. Given that the conditions under which the doctrine applies are not capable of precise formulation, counsel have helpfully placed before us a number of citations from various jurisdictions. They generally support the summary cited in the preceding paragraph but two may be extracted which are succinct and particularly useful. In The Queen v Crabbe[46], the High Court of Australia approved the following statement by Professor Glanville Williams: “A court can properly find wilful blindness only where it can almost be said that the defendant actually knew. He suspected the fact; he realised its probability; but he refrained from obtaining the final confirmation because he wanted in the event to be able to deny knowledge. This, and this alone, is wilful blindness. It requires in effect a finding that the defendant intended to cheat the administration of justice.”[47] 68. In Pereira v DPP, the High Court of Australia sounded some warnings, namely, that the concept does not detract from the requirement to prove actual knowledge; that it is the knowledge of the accused person that must be proved and not that of a hypothetical person in the position of the accused; and that where knowledge is to be inferred, it must be the only rational inference available: “Even where, as with the present charges, actual knowledge is either a specified element of the offence charged or a necessary element of the guilty mind required for the offence, it may be established as a matter of inference from the circumstances surrounding the commission of the alleged offence. However, three matters should be noted. First, in such cases the question remains one of actual knowledge … . It is never the case that something less than knowledge may be treated as satisfying a requirement of actual knowledge. Secondly, the question is that of the knowledge of the accused and not that which might be postulated of a hypothetical person in the position of the accused, although, of course, that may not be an irrelevant consideration. Finally, where knowledge is inferred from the circumstances surrounding the commission of the alleged offence, knowledge must be the only rational inference available. All that having been said, the fact remains that a combination of suspicious circumstances and failure to make inquiry may sustain an inference of knowledge of the actual or likely existence of the relevant matter. In a case where a jury is invited to draw such an inference, a failure to make inquiry may sometimes, as a matter of lawyer’s shorthand, be referred to as wilful blindness. Where that expression is used, care should be taken to ensure that a jury is not distracted by it from a consideration of the matter in issue as a matter of fact to be proved beyond reasonable doubt.”[48] 69. Despite the fact that the Reasons for Verdict are replete with findings that the appellant frequently and deliberately turned a blind eye to highly suspicious circumstances, the Judge, having found that Diallo had not told the appellant about the fraud, concluded that the appellant “cannot be held to know the USD539,375 represented proceeds of an indictable offence.”[49] But he held that the appellant had reasonable grounds to believe that the funds represented the proceeds of an indictable offence. Since, on one view, it might be said that it was open to the Judge to infer actual knowledge from his findings of wilful blindness, and since the doctrine has not been the subject of analysis by this Court in the context of section 25(1) of OSCO, the Court invited the parties to address the issue. 70. There is no reason, in principle, why the doctrine may not be applied to the knowledge limb of section 25(1) of OSCO, in other words as a basis from which to infer actual knowledge that the property represents the proceeds of an indictable offence, provided: (a) it is proved that the property was in fact the proceeds of an indictable offence, since one cannot know something to be so unless it is in fact so; (b) it is remembered that it is actual knowledge that must be proved; (c) it is remembered that it is not a question of whether the defendant should have made inquiries or whether the reasonable man would have made inquiries, although the latter issue may be relevant, though not conclusive, in deciding whether the defendant deliberately refrained from making inquiries; and (d) knowledge is the only reasonable inference to be drawn from the fact of wilful blindness and the other evidence in the case. 71. Where, however, an offence is charged under section 25(1) of the OSCO it will normally not be necessary or helpful to apply the controversial doctrine of wilful blindness in order to establish knowledge. This is because of the statutory alternative basis for conviction of “reasonable grounds to believe”. Where a person has been “wilfully blind” to the existence of a fact or of particular circumstances, he will almost always have had reasonable grounds to believe that the fact or circumstances in question existed. Thus it is better to concentrate on the statutory test of whether the defendant had “reasonable grounds to believe” rather than to introduce the doctrine of “wilful blindness”. 72. In coming to the conclusion that the appellant had reasonable grounds to believe that the funds represented the proceeds of an indictable offence, the Judge itemised various indicia of illegality which must have been obvious to the appellant and his Reasons for Verdict are peppered with references to turning a blind eye to obviously suspect circumstances. He referred, for example, to a situation which was “so questionable that it calls for immediate rigorous proof or due diligence enquiry,”[50] such that the “[d]efendant in such circumstance ought to think or dig deep into the matter. Yet, he did not but simply turned a blind eye to the situation. … He was unreasonably contented with the scarce information he had of the highly suspicious deal … . I am sure anyone looking objectively at the grounds Defendant had would believe differently from Defendant, who was in fact turning a blind eye to a highly questionable situation which he must have so appreciated. Anyone looking objectively at such grounds Defendant had would believe the money involved represented in whole or in part, directly or [indirectly] the proceeds of an indictable offence. Defendant ought to have known that but he chose not to. … In the present case, I find Defendant’s so-called ‘belief’ was just ill-founded and equivalent to turning a blind eye to obvious suspicion.”[51] 73. Wilful blindness also featured prominently in the Judge’s Reasons for Sentence: “24. … Defendant’s criminality lies in his letting others to use SIAT’s Account A to receive the money coming from a highly suspicious offer which he accepted blindly. However, Defendant was not, as his counsel puts it, a duped victim himself. The communication records between Defendant and coolsaintt reveal that Defendant was always willing and ready to offer bank accounts under his control in different places for coolsaintt’s clients to deposit money. I reject Defendant’s explanation that it was for confirming house business. I find that Defendant offered different accounts for money-conduit dealing. The present case demonstrates Defendant blindly accepted whatever the agent coolsaintt told him. Defendant chose not to exercise any scrutiny or judgment which he was capable of rendering if he wanted to. 25. … Defendant is not as culpable as the fraudsters or as coolsaintt, who knew the fraud. Defendant is culpable as he cared not about the source of the funds.”[52] 74. While, as demonstrated above, the judge made frequent reference to the appellant turning a blind eye, he was doing so merely to underline his conclusion that the appellant had “reasonable grounds to believe” that the relevant funds were tainted. He was not applying the legal doctrine of “wilful blindness” to establish knowledge. This is clear from the fact that he made an express finding that the appellant did not know that the property was tainted. Question 4(a): Given the requirements of section 159A(2) of the Crimes Ordinance, can there be an offence of conspiracy to deal with property having reasonable grounds to believe that such property … represents any person’s proceeds of an indictable offence? 75. Section 159A of the Crimes Ordinance provides that: “(1) Subject to the following provisions of this Part, if a person agrees with any other person or persons that a course of conduct shall be pursued which, if the agreement is carried out in accordance with their intentions, either – (a) will necessarily amount to or involve the commission of any offence or offences by one or more of the parties to the agreement; or (b) would do so but for the existence of facts which render the commission of the offence or any of the offences impossible, he is guilty of conspiracy to commit the offence or offences in question. (2) Where liability for any offence may be incurred without knowledge on the part of the person committing it of any particular fact or circumstance necessary for the commission of the offence, a person shall nevertheless not be guilty of conspiracy to commit that offence by virtue of subsection (1) unless he and at least one other party to the agreement intend or know that that fact or circumstance shall or will exist at the time when the conduct constituting the offence is to take place.” 76. By reference to the terms of subsection (2), the question, more specifically framed, is whether criminal liability under section 25(1) of OSCO may be incurred “without knowledge on the part of the person committing it of any particular fact or circumstance necessary for the commission of the offence”. The question is generated by the phrase in section 25(1) “having reasonable grounds to believe”, since it renders a person liable to conviction when he does not know that the property with which he deals represents the proceeds of an indictable offence, but only has reasonable grounds to believe that it does. 77. As the Court recently explained in HKSAR v Lai Kam Fat,[53] the genesis of subsection (2) is the decision of the Court of Criminal Appeal in Churchill v Walton[54], prior to which “[c]onspirators charged with agreeing to commit a strict liability offence could be convicted on the mere intention to agree.”[55] The appellant in that case had been charged with conspiring to contravene a provision of the Customs and Excise Act 1952 which prohibited the use of oils as fuel for vehicles if, in anticipation of domestic use, there had been a rebate of duty unless the amount of the rebate had been paid to the Commissioners of Customs and Excise. It was an offence of strict liability. The trial judge directed the jury that all that was necessary to warrant a conviction for conspiracy to use the oil in road vehicles where the amount of the rebate had not been paid to the Commissioners, was proof of the agreement to use oil which had not in fact borne the requisite tax and was not going to do so, adding “[i]t does not involve any proof of dishonest intent or of knowledge that such use would cheat the revenue … .”[56] The House of Lords held this to be a misdirection, since defendants could only be guilty of conspiracy “[i]f what they agreed to do was, on the facts known to them, an unlawful act,” and that “[i]f, on the facts known to them, what they agreed to do was lawful, they are not rendered artificially guilty by the existence of other facts, not known to them, giving a different and criminal quality to the act agreed upon”.[57] 78. In the subsequent report by the Law Commission (England and Wales) that principle was endorsed: “What the prosecution ought to have to prove is that the defendant agreed with another person that a course of conduct should be pursued which would result, if completed, in the commission of a criminal offence, and further that they both knew any facts which they would need to know to make them aware that the agreed course of conduct would result in the commission of the offence.”[58] 79. The rationale for this approach was explained by Lord Nicholls in Saik: “The rationale underlying this approach is that conspiracy imposes criminal liability on the basis of a person’s intention. This is a different harm from the commission of the substantive offence. So it is right that the intention which is being criminalised in the offence of conspiracy should itself be blameworthy. This should be so, irrespective of the provisions of the substantive offence in that regard.”[59] 80. Following the Law Commission’s report, the Criminal Law Act 1977 was enacted. Section 159A(1) of the Crimes Ordinance is in the same terms as section 1 of that Act. 81. Lord Nicholls’ analysis or explanation of section 1(2) of the Act is, for present purposes, particularly instructive: “7. Under this subsection conspiracy involves a third mental element: intention or knowledge that a fact or circumstance necessary for the commission of the substantive offence will exist. Take the offence of handling stolen goods. One of its ingredients is that the goods must have been stolen. That is a fact necessary for the commission of the offence. Section 1(2) requires that the conspirator must intend or know that this fact will exist when the conduct constituting the offence takes place. 8. It follows from this requirement of intention or knowledge that proof of the mental element needed for the commission of a substantive offence will not always suffice on a charge of conspiracy to commit that offence. In respect of a material fact or circumstance conspiracy has its own mental element. In conspiracy this mental element is set as high as ‘intend or know’. This subsumes any lesser mental element, such as suspicion, required by the substantive offence in respect of a material fact or [circumstance]. In this respect the mental element of conspiracy is distinct from and supersedes the mental element in the substantive offence. When this is so, the lesser mental element in the substantive offence becomes otiose on a charge of conspiracy. It is an immaterial averment. To include it in the particulars of the offence of conspiracy is potentially confusing and should be avoided. 9. The phrase ‘fact or circumstance necessary for the commission of the offence’ is opaque. Difficulties have sometimes arisen in its application. The key seems to lie in the distinction apparent in the subsection between ‘intend or know’ on the one hand and any particular ‘fact or circumstance necessary for the commission of the offence’ on the other hand. The latter is directed at an element of the actus reus of the offence. A mental element of the offence is not itself a ‘fact or circumstance’ for the purposes of the subsection. 10. This contrast can be illustrated by the offence of entering into an arrangement whereby the retention by another person (A) of A’s proceeds of crime is facilitated, knowing or suspecting A has been engaged in crime: section 93A of the Criminal Justice Act 1988, now repealed. The requirement that the defendant must know or suspect A’s criminal history is an element of the offence, but it is a mental element. The need for the defendant to have this state of mind is not a fact or circumstance within section 1(2). Another ingredient of the offence is that the property involved must be the proceeds of crime. That is a fact necessary for the commission of the offence and section 1(2) applies to that fact.” [60] 82. It is not clear from the appellant’s submissions whether it is sought to be argued that the fact of the criminal provenance of the property (that is, that the relevant property in whole or in part directly or indirectly represents any person’s proceeds of an indictable offence) is a “fact or circumstance” of which an accused must have knowledge for the purposes of section 159A(2). If this were required, then the answer to the Question posed under this section would have to be “No”, since a person who only has reasonable grounds to believe under the reasonable grounds limb of section 25(1) of OSCO would hardly ever (if not never) be able to have the requisite knowledge or intention. 83. However, the law in the United Kingdom relating to money laundering offences differs in a material aspect from the effect of section 25(1) of OSCO. The relevant offences created by the Criminal Justice Act 1988 and the Drug Trafficking Act 1994 require proof that the property dealt with in fact represents the proceeds of criminal conduct or of drug trafficking, as the case may be,[61] whereas it is not an ingredient of a section 25(1) OSCO offence that the property must be the proceeds of crime. The criminal provenance of the property is not a “particular fact or circumstance necessary for the commission of the offence”: HKSAR v Wong Ping Shui & another[62]; Oei Hengky Wiryo v HKSAR (No.2)[63]; HKSAR v Li Kwok Cheung George[64]; HKSAR v Yeung Ka Sing Carson[65]. Further, as held in Saik, the “fact or circumstance necessary for the commission of the offence” is “directed at an element of the actus reus of the offence” and “[a] mental element of the offence is not itself a ‘fact or circumstance’ for the purposes of the subsection.”[66] In the context of section 25(1) of OSCO, the actus reus is the act of dealing with the property whereas “[t]he defendant’s mens rea is ... established if he is shown to know or to have reasonable grounds to believe that ‘any property’ in whole or in part represents ‘any person’s proceeds of an indictable offence’.”[67] 84. Accordingly, the law in Hong Kong does not require it to be proved under the reasonable grounds limb of section 25(1) of OSCO that the relevant property dealt with was tainted. It is not, in the language of section 159A(2), a “fact or circumstance necessary for the commission” of that offence. This being the case, section 159A(2) simply does not apply to the reasonable grounds limb of section 25(1) of OSCO insofar as this fact or circumstance is concerned. We would add that this was the conclusion reached by the Court of Appeal in HKSAR v Lung Ming Chu.[68] 85. The answer to this Question is “Yes”. Question 4(b): Given the requirements of subsection (1)(a) of section 159A, Crimes Ordinance, where defendants have reasonable grounds to believe that property is tainted, will they be guilty of conspiracy if they agree to deal with the property notwithstanding that those grounds may not exist at the time of the dealing? 86. As seen above, there are two limbs to the offence set out in section 25(1) of OSCO: the “knowing” limb and the “reasonable grounds” limb. We are in this appeal only concerned with the “reasonable grounds” part of section 25(1). Notwithstanding that section 159A of the Crimes Ordinance sets out a composite scheme dealing with the requisites to establish the offence of a statutory conspiracy, section 159A(2) has no application as regards the “reasonable grounds” limb of section 25(1), leaving only section 159A(1) to be applied. It is in these circumstances that we have to consider Question 4(b). There are two facets to be considered: first, is it possible as a matter of law to have a conspiracy to commit the offence contained in the “reasonable grounds” limb of section 25(1) of OSCO in the first place; secondly, if this is possible, in what circumstances can the statutory conspiracy be committed? Before these facets can be considered and Question 4(b) determined, it is first necessary to break down the requirements of section 159A(1). 87. At common law, prior to the passing of the statutory provision, the offence of conspiracy essentially consisted of the making of an agreement between two or more persons to do an unlawful act.[69] Statute has now made the definition more precise. Section 159A(1)[70] can be broken down into the following components that have to be proved in order to establish the offence: (1) An agreement between two or more persons. (2) The future course of conduct agreed under that agreement to be pursued. (3) The intentions of the parties under the agreement. (4) Finally, it must be shown that if the agreement is carried out in accordance with the parties’ intentions, this will necessarily amount to or involve the commission of an offence by one or more of the parties to the agreement or would do so but for the existence of facts which render the commission of the relevant offence impossible. 88. The following observations can be made in relation to these components of the offence. 89. Paragraph [87(1)] causes little difficulty. The question here is simply: have the parties made an agreement? How the agreement is proved in a conspiracy is another matter and this aspect will be dealt with later.[71] 90. The reference in paragraph [87(2)] to a course of conduct means the terms agreed between the parties as to what is to be done in the future under the agreement reached. The question is: what was agreed to be done under the agreement? 91. Then comes paragraph [87(3)]: the identification of the intentions of the parties under the agreement. The course of conduct agreed between the parties under the agreement will obviously be a part of the parties’ intentions. However, the term “intentions” goes beyond just considering the physical acts agreed to be done under the agreement. They must relate also to the consequences the parties intend by their agreed acts. For instance, to take a common example: A and B make an agreement that the following week they will put poison in V’s drink to give to him. The agreed course of conduct is the physical act of putting poison in V’s drink and giving it to him; the intended consequence is that V will take that drink when it is given to him and is poisoned to death. That the intended consequences are important to be considered is perhaps an obvious statement to make. This is also the view of learned academic writers: see Smith, Hogan and Ormerod’s Criminal Law.[72] Using the poison in the drink example, the authors are of the view that if consequences are not considered but instead only the physical acts comprising the course of conduct, this would be “absurd”. The passage continues, “To avoid the absurdity, ‘course of conduct’ must be read to include the intended consequences – in this case, the death of V”. Smith, Hogan and Ormerod take the view that the intended consequences should be regarded as part of the “course of conduct” referred to in section 159A. Intended consequences may perhaps not easily fit into the words “course of conduct” but they would certainly be part of the parties’ intentions. It is unnecessary to resolve this semantic debate as long as intended consequences are properly to be taken into account. The intentions of the parties are however also relevant in considering the element of the offence set out in paragraph [87(4)] above. 92. Before dealing with the component referred to in paragraph [87(4)] above, we should just make an additional point regarding the consideration of the parties’ agreed course of conduct and their intentions. It is also necessary in the consideration of these components to look at the facts and circumstances that the parties had in mind in the performance of their agreement. This is again a matter of common sense – one needs to examine the agreement in context. Thus, in our murder example, one needs to examine what facts and circumstances the parties had in mind when their agreed course of conduct came to be pursued and carried out in accordance with their intentions. These may comprise, for instance, the fact that A or B will arrange to meet V, be in a position to put the poison in his drink and make him drink it. A consideration of the facts and circumstances that the parties had in mind are important in the consideration of whether an offence would necessarily be committed, the aspect referred to in paragraph [87(4)] above. As Smith, Hogan and Ormerod state:[73] “To consider D’s liability for conspiracy, we therefore have to imagine that the course of conduct he intended was pursued and to ask, would it, when completed, have necessarily amounted to or involved the commission of any offence.” 93. The essence of a conspiracy being the agreement to commit an unlawful act, the establishment of the unlawful act is the final crucial step. The course of conduct under the agreement, if[74] this agreement is performed in accordance with the parties’ intentions thereunder, must[75] amount to or involve the commission of a crime by one or more parties to the agreement (or would do so but for the existence of facts which render the commission of the offence impossible). A number of points can be made in relation to this aspect: (1) The particular offence needs to be identified. All offences triable in Hong Kong are covered: section 159A(3). In the present case, the relevant offence is the “reasonable grounds” limb of section 25(1) of OSCO. (2) In analysing whether or not an offence will “necessarily” be committed, the constituent elements of the particular offence will need to be satisfied. Even without the requirements of section 159A(2), which as we have held has no application in the present case, there will be a need to satisfy both the actus reus and mens rea elements of the offence. The actus reus part requires no further elaboration. On mens rea, apart from the intention to carry out the agreement reached between the parties, the parties’ state of mind must also satisfy the mental ingredients of the substantive offence. It was put in the following way by Lord Nicholls of Birkenhead in Saik:[76] “Thus under this subsection[77] the mental element of the offence, apart from the mental element involved in making an agreement, comprises the intention to pursue a course of conduct which will necessarily involve commission of the crime in question by one or more of the conspirators. The conspirators must intend to do the act prohibited by the substantive offence. The conspirators’ state of mind must also satisfy the mental ingredients of the substantive offence.” Accordingly, it must be shown that at the time the conspiratorial agreement is made, if the course of conduct agreed on were to be performed in accordance with their intentions, which on this analysis would necessarily be in the future a crime would be committed and this would include the necessary mens rea of that crime. 94. The example used above[78] can be analysed with reference to the requisite components of section 159A(1): (1) The agreement is made between A and B. (2) The agreed course of conduct is to put poison in V’s drink next week. (3) Their intentions are that he will drink the poison and will die. (4) If the agreed course of conduct is pursued and the agreement is carried out in accordance with the said intentions, the offence of murder will necessarily be committed. The requisite actus reus for the offence of murder consists of the act of killing V by putting poison in his drink, which it is intended V will take. The requisite mens rea consists of the intention to kill V next week, this intention being in existence at the time the agreement was made. 95. It is important to emphasize that for the purpose of establishing a conspiracy, the relevant time in the analysis is the time the agreement is made. If the components that must be proved exist at the time the agreement is made, the statutory offence is established. The offence lies in the making of the agreement and the intention of the parties to carry out that agreement in the future; the offence is established notwithstanding that there may be changes made thereafter or if there are any changes in intention or if the agreement is simply not performed.[79] Moreover, even if the agreement is impossible to perform because circumstances change, an offence is committed: see section 159A(1)(b). 96. Analysed as above, one can accept that there can be a conspiracy to commit virtually any type of offence. The question for us is however whether it is possible to charge a conspiracy to commit the offence under the “reasonable grounds” limb of section 25(1) of OSCO. The appellant says it is not possible. 97. At first blush, it is difficult to see why it is not so possible. In the context in which that provision operates, being the dealing of property, an agreement to deal with that property (by whatever means) would fulfill the first two components as required by section 159A(1). As far as the intentions of the parties are concerned, these include the intention to deal with the relevant property at the agreed time. So much can be accepted. 98. But is there a problem with the need to show that an offence is “necessarily” committed? The argument is put in the following way by the appellant in the present case:[80] “105. But, in the context of the second limb of s 25 of OSCO, where there is an agreement to deal with property at some time in the future such dealing will not necessarily involve the commission of the substantive offence. That is because the question of whether the substantive offence is committed will depend on the state of mind of the person dealing with the property at the time of the dealing. The state of mind of a person at the time of his entering the agreement will not necessarily be his state of mind at the time of the dealing. 106. It would be artificial and unrealistic to contend that at the time the agreement was entered into the parties knew and intended that at the time of the dealing, the dealer would have reasonable grounds for believing that the property was tainted.” 99. This submission focuses on the required state of mind for the substantive offence. As stated earlier,[81] it must be shown that at the time the conspiratorial agreement is made, if the agreement was performed as the parties intended it to be performed, that an offence would necessarily be committed and this would include the requisite mental state for the offence. In our murder conspiracy example, the requisite mental state would be the intention to kill V the following week and this was the intention of the parties to the agreement at the time it was made. Where the “reasonable grounds” limb of section 25(1) of OSCO is involved, can it ever be proved that at the time the agreement is made, the parties would have the requisite mental state for the commission of the offence when their agreement was performed? 100. Conceptually, the answer is “Yes”. For instance, if it can be proved that at the time the agreement is made, the parties had at that time reasonable grounds to believe that the property they planned to deal with was and would remain tainted and they agreed to deal with such property in the future, then an offence – the “reasonable grounds” limb of section 25(1) – would necessarily be committed. Both the actus reus of the offence (dealing with the property) and the mens rea (having reasonable grounds to believe that the property is tainted) would be established. Put another way, in this scenario, the analysis can be summarised as follows: “Knowing what I know about the character of the property I will be dealing with (and this includes the reasonable grounds to believe it is and will remain tainted), I intend to deal with that property next week.” This will be a part of the relevant facts and circumstances envisaged by the parties when the agreement is carried out in accordance with their intentions. At the time an agreement is made, the parties must have an idea of the nature or character of the property they intend to deal in. 101. The following may help further illustrate this. A and B make an agreement that next week, when monies become available from a dubious source, they will be deposited by B into A’s accounts and, after they are deposited, A will then withdraw the monies on B’s instructions. Both A and B have at the stage the agreement is made reasonable grounds to believe that these monies are and will remain tainted, and with these grounds, they intend to pursue the said course of conduct and carry out their intentions to so deal with the monies. The constituent components of section 159A(1) are satisfied: (1) An agreement is made between A and B. (2) The course of conduct agreed to be pursued between the parties consists of dealing with the monies that will materialise the following week by B depositing them into A’s accounts and then A withdrawing the monies on B’s instructions. (3) The intentions of the parties as at the date their agreement is made are to deal with the monies in accordance with the course of conduct agreed to be pursued having reasonable grounds to believe that they are and will remain tainted monies. That is to say, they intend to deal with the monies next week in the way they have agreed having reasonable grounds to believe that they are and will remain tainted monies. (4) If the agreed course of conduct is pursued and the agreement between A and B is carried out in accordance with the parties’ said intentions, this will necessarily amount to or involve the commission of a crime (namely the “reasonable grounds” limb of section 25(1) of OSCO). The actus reus of the offence (dealing with the monies as aforesaid) and the mens rea (having reasonable grounds to believe that the property is tainted) will be established. 102. The appellant’s argument[82] proceeds on the basis that the state of mind of the person dealing with the property at the time that person enters into the conspiratorial agreement may be different to the relevant state of mind at the time of the dealing. This is acutely so when one is dealing with the “reasonable grounds” limb of section 25(1) of OSCO. The facts available to A at the time of the agreement to suggest there exist reasonable grounds to believe that the property is tainted may change by the time the property is actually dealt with. How can it then be said that a crime will “necessarily” be committed? Or, as the appellant also contends, how can it be said that the parties knew and intended that at the time of dealing, the dealer would then have reasonable grounds to believe that the property was tainted? 103. Superficially attractive though the appellant’s arguments appear to be, they are not with respect sound and cannot be accepted: (1) The focus must be on the parties’ intentions at the time the agreement is made. If the evidence shows that the intention was to deal with the relevant property at the appointed time in the future, knowing the character and nature of that property, a conspiracy is established: see paragraphs [100] and [101] above. For the avoidance of doubt, although one must focus on the parties’ intentions as at the date of the agreement, it is necessary to demonstrate that the parties had the intention to deal with the relevant property in the future, that property having the nature and character of property which there are reasonable grounds to believe is and will remain tainted property. (2) The fallacy of the appellant’s approach is that it does not follow what is required to be considered under section 159A(1). It is no answer to refer to the fact that circumstances may change after the time the agreement is made and that therefore the state of mind of the conspirators or any of them may not be the same as at the date of the agreement. It is crucial to look at the matter at the date the agreement is made and not take into account subsequent events: see paragraph [95] above. (3) The reference in section 159A(2) to intention and knowledge of facts and circumstances is a reference to the actus reus, not the mens rea, of an offence.[83] The more important, indeed the critical, inquiry is to identify the intentions of the parties under the agreement they have made and then proceed to assess whether an offence would necessarily be committed. If they intended to deal with property that had the character and nature of property which they had reasonable grounds to believe was and would remain tainted, an offence is committed. Thus, if a defendant who has reasonable grounds to believe property is tainted agrees unconditionally with another to deal with that property in the future, pursuing a course of conduct which, if carried out accordingly with this intention, this will necessarily amount to the commission of an offence under the reasonable grounds limb of section 25(1) of OSCO. 104. From the above, it can now be seen that a conspiracy to commit the offence under the “reasonable grounds” limb of section 25(1) of OSCO is as a matter of law possible, depending on the circumstances. Naturally, whether or not a conspiracy can be established depends entirely on the evidence. Here it is worthwhile reminding ourselves how criminal conspiracies are often sought to be proved. In many ways, there is an air of artificiality about the offence of conspiracy in that while the analysis must be conducted in relation to the agreement reached by the alleged conspirators, the terms thereof and the parties’ intentions thereunder are proved only rarely by adducing direct evidence of the making of the agreement. In a civil case, agreements and their terms are proved by direct evidence of the making of the agreement, whether it is made orally or in writing. By contrast, in a criminal conspiracy, which is regarded as a continuing offence,[84] the evidence of each of the requisite components in section 159A(1) is often inferred from the overt acts of the parties, usually over a period of time.[85] As was put by Lord Diplock in DPP v Knuller (Publishing, Printing and Promotions) Ltd[86]: “In most cases of conspiracy, the prior agreement is itself only inferred from the acts which have been done in pursuance of it.”[87] From such overt acts are also to be drawn inferences going to the state of mind of the parties.[88] 105. On the facts of the present case, the appellant made available his bank account or accounts as a conduit for funds from and to third parties and in the period leading up to his receipt and transfer of those funds at all material times, he had reasonable grounds to believe that the funds were tainted. This was a part of the character of the property with which he was to deal. It is in that factual context that he agreed to make his bank account available. He now argues that between the time he agreed to make his bank account available for apparently tainted funds and the time the funds arrived, the grounds for belief may change. And yet if one asks the question what was his intention at the time of the agreement, the answer must be that he intended to make his account available as a conduit knowing the character of these funds. If he had been asked at the date of the agreement, “will you make your account knowing the character of the funds?”, he would answer “of course I will”. Why is that not an agreement which, if carried out in accordance with that intention, would necessarily amount to or involve the commission of an offence? 106. In the present case, the inferences to be drawn against the appellant are strong. The trial judge found that Diallo knew of the fraud,[89] although he was not satisfied that the appellant did. The evidence was of a frequent exchange of emails concerning the receipt into a SIAT account of US$539,375 with Diallo asking the appellant to “get the money out the next day if it came through”;[90] of the appellant expressing anxiety that Diallo’s request to move money to various accounts would cause the bank to investigate;[91] of an absence of any discussion between the appellant and Diallo as to who was to buy what;[92] of reasons given by the appellant for the receipt of such a large sum which made no commercial sense; of no questions asked as to why SIAT was chosen to receive the funds;[93] and of a situation which from beginning to end reeked of illegality. In the context of the evidence adduced in the present case, it would be entirely unrealistic to suppose that either Diallo or the appellant contemplated a set of circumstances prevailing at the date of dealing other than one which continued to reek of that illegality. There could be no realistic doubt that those were the circumstances in which they intended to deal. Accordingly, there is no basis, on the facts of the present case, to conclude other than that the course of conduct agreed by the appellant and his co-conspirators would, if executed in accordance with their intentions, necessarily involve the commission of an offence contrary to section 25(1) of OSCO. 107. A final point should be made. On the facts of the present case, the knowledge on the part of the alleged conspirators to the agreement – Diallo and the appellant – differ in this respect: in relation to the monies they intended to deal with, while the appellant had reasonable grounds to believe that the property was tainted, Diallo had actual knowledge that the monies were tainted. Is this difference in the mental state of the conspirators relevant as far as establishing the offence of conspiracy is concerned? We do not believe it is. In this scenario, both the appellant and Diallo still intended to deal with property that was illicit for the purposes of section 25(1) of OSCO. In the appellant’s case, he intended to deal with property which he had reasonable grounds to believe was tainted. In Diallo’s case, he intended to deal with tainted property. This situation is similar to the example given by Lord Hope of Craighead in Saik;[94] he saw no difficulty in concluding that the equivalent of section 159A(1) was satisfied. Obviously, if there was a significant mismatch between the intentions of the parties to the agreement, a conspiracy may not be established at all. For example, in a money laundering situation, if A intended only to deal in clean monies whereas B intended to deal in tainted monies, there is no coincidence of intentions between them. A conspiracy would not then be proved. However, that is not the present case. 108. The answer to Question 4(b) is therefore “Yes”, subject to the discussion in this section above. The substantial and grave injustice ground 109. The Judge held that “[t]he ‘having reasonable grounds to believe’ test is anyone looking at the grounds Defendant had would believe the property Defendant dealt with represented proceeds of an indictable offence. … I am sure anyone looking objectively at the grounds Defendant had would believe differently from Defendant, who was in fact turning a blind eye to a highly questionable situation which he must have so appreciated.”[95] He added, later, that: “In the present case, I find Defendant’s so-called ‘belief’ was just ill-founded and equivalent to turning a blind eye to obvious suspicion. It, therefore, could hardly be said to be sincere or in good faith.”[96] Accordingly, he expressed himself satisfied that the appellant had conspired to deal with the funds having reasonable grounds to believe they represented the proceeds of an indictable offence.[97] 110. It will be recalled that the Court of Appeal held that the Judge was wrong to require that any genuinely held belief by a defendant must also be reasonably held but upheld the conviction on the basis that it was evident from the Reasons for Verdict that the Judge had disbelieved the appellant’s claim as to his beliefs.[98] 111. The contention that there has been substantial and grave injustice is that the Court of Appeal did not have any valid alternative basis for upholding the conviction by which, if we understand the submissions correctly, is meant that there was no valid basis upon which to conclude that the appellant did not believe the transaction in question to be a genuine business transaction and that, in any event, one cannot properly discern from the Reasons for Verdict that the Judge had made that finding. 112. It is not necessary for present purposes to particularise such of the evidence and those parts of the Reasons for Verdict upon which the appellant relies in support of this ground. It seems to us plainly obvious that the Judge did not believe the appellant’s exculpatory assertions and that there was ample evidence to justify the conclusion he reached as to the suspect facts and circumstances of which the appellant was well aware and that they constituted reasonable grounds to believe the funds to be the proceeds of an indictable offence. One might add that there was, as it happens, no need for the Court of Appeal to reach for an alternative basis upon which the conviction might be justified. That is because, as we have earlier explained, the court fell into error in its approach to the application of section 25(1) of OSCO. The Judge’s approach accorded more accurately with the tests we have propounded. Conclusion 113. Since none of the issues raised in this appeal is resolved in the appellant’s favour, the appeal must be dismissed. Mr Peter Duncan SC, Mr Trevor Beel and Ms Priscilla Lau, instructed by Jal N. Karbhari & Co., assigned by the Director of Legal Aid, for the Appellant Mr Jonathan Caplan QC, Counsel on fiat, and Mr Anthony Chau, ADPP (Ag.), of the Department of Justice, for the Respondent [1] His Honour Judge J Lam, DCCC 1047/2014, Reasons for Verdict dated 4 August 2015 (“Reasons for Verdict”). [2] (2014) 17 HKCFAR 778. [3] Macrae JA (as he then was), McWalters and Pang JJA, [2017] 5 HKLRD 326, CACC 352/2015, Judgment dated 12 September 2017 (“Court of Appeal”). [4] (2016) 19 HKCFAR 279. [5] Reasons for Verdict para 123. [6] Reasons for Verdict para 220. [7] Ibid. para 221 (emphasis in original). [8] Ibid. para 223. [9] Ibid. para 182. [10] Ibid. para 183. [11] Ibid. para 186. [12] Ibid. para 197. [13] Ibid. para 202. [14] Ibid. para 203. [15] Ibid. para 228. [16] Court of Appeal para 57. [17] Ibid. para 48. [18] Reasons for Verdict para115(b), Court of Appeal para 49. [19] Court of Appeal para 50. [20] Ibid. para 65. [21] Amended Application for Leave to Appeal dated 19 April 2018. [22] Ribeiro and Tang PJJ, Stock NPJ, [2018] HKCFA 26, FAMC 60/2017, Determination dated 16 May 2018. [23] Ma CJ, Ribeiro and Fok PJJ, Stock and Lord Phillips NPJJ, FACC 17/2018. [24] (2014) 17 HKCFAR 778 paras 52 and 56. [25] [1999] 2 HKC 833 at p.836E-F. [26] (2016) 19 HKCFAR 279 para 103. [27] Reasons for Verdict para 191. [28] Ibid. paras 203, 211, 215 and 223. [29] Court of Appeal para 64. [30] Pang Hung Fai para 24. [31] Ibid. para 106. [32] Carson Yeung para 112. [33] (2012) 15 HKCFAR 146. [34] Carson Yeung para 113. [35] (2014) 17 HKCFAR 319. [36] Pang Hung Fai paras 106-107. [37] Carson Yeung para 118 (emphasis in bold added). [38] Ibid. para 119 (emphasis in bold added). [39] Court of Appeal paras 28 and 29. [40] [2018] UKSC 36. [41] Ibid. para 4. [42] [2006] UKHL 18, [2007] 1 AC 18. [43] R v Lane [2018] UKSC 36 para 24 (underlining in original). [44] 6th ed, Bloomsbury (2016) at pp.157-159. [45] Ibid. at pp.158-159 (emphasis in original). [46] (1985) 156 CLR 464 at pp.470-471. [47] Criminal Law: The General Part 2nd ed, Stevens & Sons (1961) at p.159. [48] (1988) 82 ALR 217 at pp.219-220. [49] Reasons for Verdict para 191 (emphasis in original). [50] Reasons for Verdict para 202. [51] Ibid. paras 202, 203 and 228. [52] His Honour Judge J Lam, DCCC 1047/2014, Reasons for Sentence dated 7 October 2015 paras 24 and 25 (emphasis added). [53] [2019] HKCFA 36, FACC 1/2019 at paras 32-41. [54] [1967] 2 AC 224. [55] Making Sense of Mens Rea in Statutory Conspiracies, Professor David Ormerod (2006) 59 CLP 185 at p.190. [56] Churchill v Walton [1967] 2 AC 224 at p.231G. [57] Ibid. at p.237 (emphasis added). [58] ‘Criminal Law: Report on Conspiracy and Criminal Law Reform’ – The Law Commission (1976) Law Com. No 76 para 1.39. [59] Saik para 13. [60] Ibid. paras 7-10. [61] R v Montila [2004] 1 WLR 3141. [62] (2001) 4 HKCFAR 29. [63] (2007) 10 HKCFAR 98. [64] (2014) 17 HKCFAR 319. [65] (2016) 19 HKCFAR 279. [66] [2007] 1 AC 18 para 9. [67] Carson Yeung para 42. [68] [2009] 3 HKC 137. [69] See Churchill v Walton [1967] 2 AC 224 at p.232E. [70] It contains the same wording as the equivalent English legislation, being s.1(1) of the Criminal Law Act 1977 (as amended by the Criminal Attempts Act 1981). [71] See para [104] below. [72] 15th ed (2018) at p.446 under the heading “Consequences as part of the course of conduct”. [73] Ibid. at p.445 (para 11.3.3.2). [74] The word “if” requires the analysis to be made by reference to the time the agreement is made but looking to the future to the date of performance. [75] The word “necessarily” is critical here. [76] Saik para 4. [77] The English equivalent of section 159A(1). [78] In para [91]. [79] See Saik paras 3, 44, 75, 93 and 112. [80] In paras 105 and 106 of the Appellant’s Submission on the Issues raised by the Court (emphasis in original). [81] In para [93(2)] above. [82] See para [98] above. [83] See Saik at para 9: “A mental element of the offence is not itself a ‘fact or circumstance’ for the purposes of the subsection”. [84] See Smith, Hogan and Ormerod at p.438 (para 11.3.3.1) referring to DPP v Doot [1973] AC 807 at pp.822H‑823A. [85] Such as in the present case, where the indictment pleads the conspiracy to have taken place between 26 April 2014 and 21 July 2014. [86] [1973] AC 435 at p.477D-E. [87] There are numerous other references in the authorities to similar effect, for example, Churchill v Walton at p.232F; DPP v Doot at p.818A; Smith, Hogan and Ormerod at pp.444-445 (“Proof of the agreement”); Glanville Williams: Textbook of Criminal Law (4th ed.) para 19-006. [88] See Saik para 63. [89] Reasons for Verdict para 123. [90] Ibid. para 139. [91] Ibid. para 172. [92] Ibid. para 180. [93] Ibid. para 197. [94] At paras. 76 and 77. [95] Reasons for Verdict para 203. [96] Ibid. para 228. [97] Ibid. para 229. [98] See paras [16]-[20] above. 
Mr Justice Ribeiro PJ: 1. I agree with the judgements of Mr Justice Tang and Mr Justice Fok PJJ and with the additional observations of Mr Justice Chan NPJ. Mr Justice Tang PJ: Introduction 2. Section 14 of the Inland Revenue Ordinance, Cap 112, provides that profits tax shall be chargeable — on every person carrying a trade in Hong Kong in respect of his assessable profits. Trade is defined in s 2 as including “every trade and manufacture, and every adventure and concern in the nature of trade”. 3. The respondents in this appeal by the Commissioner of Inland Revenue (“the Commissioner”) are the Church Body of the Hong Kong Sheng Kung Hui and the Hong Kong Sheng Kung Hui Foundation. They are respectively the incorporation of the Anglican Church in Hong Kong (“the Church Body”) and the incorporation of the Anglican Bishop of Hong Kong (“the Foundation”). I will refer to them collectively as HKSKH. 4. The Church Body and the Foundation had since the 1930s been the respective owner of a large estate in Tai Po (“the Old Lots”)[1] which comprised agricultural land and restricted building land[2] on which was built the well-known St Christopher’s Home, an orphanage which was established in 1935. In time, with the urbanization of the New Territories and the ease of travel, the Old Lots became highly desirable for residential development. It was said to be an agreed fact that since the 1970s the taxpayers had planned to develop the Old Lots[3] but I believe it is more accurate to say that HKSKH began exploring the possibility of developing the Old Lots in the 1970s. The earlier plans all involved a measure of institutional use. However, since at least September 1989, the plans only involved a residential development.[4] The Commissioner accepted that at the time of the acquisition of the Old Lots, the intention was to hold them indefinitely,[5] and that the Old Lots were capital assets. 5. Before the Old Lots could be used for a substantial residential development two hurdles had to be overcome. First, permission was needed under s 16 of the Town Planning Ordinance Cap 131, without which large scale development of the Old Lots could not take place. Moreover, any permission granted would control the intensity or type of the permitted development.[6] Secondly, the lease restrictions have to be relaxed by the government as landlord[7] which normally requires the payment of a premium, said to be calculated on the difference in value between the Old Lots with their original lease restrictions and the New Lot with the new and relaxed restrictions. The procedure is commonly known as a surrender and regrant.[8] In December 1990, the taxpayers employed a firm of architects to apply to the District Lands Office, Tai Po (“DLO/TP”) for a surrender and regrant. The basic terms were communicated to the architect in August 1991 and the premium was assessed in October 1992 at $838,260,000. The draft special conditions for the new grant were also supplied for comment. In May 1993, the premium was reduced to $704,240,000. The Old Lots were surrendered to Government on 17 November 1993 in return for the New Lot.[9] On 2 July 1993, a number of property developers were invited to tender offers to either purchase the New Lot[10] (“Option A”) or to enter into a joint venture agreement to develop the New Lot (“Option B”). On 23 July 1993, Cheung Kong (Holdings) Limited (“Cheung Kong”)[11] submitted their tender on both options. And on 12 August 1993, the taxpayers accepted Option B, the joint venture offer.[12] And on 3 December 1993, HKSKH entered into a joint venture agreement with Cheung Kong. Pursuant to the joint venture agreement,[13] HKSKH eventually became entitled to 129 units and 94 car parking spaces (the units) in the development.[14] Some of these properties have been sold and the proceeds divided between the Church Body and the Foundation in the agreed proportion. The Church Body and the Foundation were assessed for profits tax for the years of assessment 1998/99 to 2004/05 inclusive. The profits tax payable by the Church Body was assessed at $75,881,426, and for the Foundation, $108,912,965.[15] 6. On appeal to the Board of Review the taxpayers contended that there was no change of intention at all, alternatively, that the change of intention only occurred in 1993, when it accepted Cheung Kong’s tender on 12 August or 3 December when it entered into the joint venture agreement.[16] The date of any change of intention is important because the amount of profits tax payable would vary according to the value at the time of change of intention.[17] However, liability to pay profits tax could only arise upon a sale[18] in the course of trade and the earliest date for a sale was 12 August 1993 if, as seems likely, HKSKH had committed itself to a sale and a joint venture. Since HKSKH had accepted Option B, which entailed a joint venture agreement, it might be thought that it had entered into a venture in the nature of trade.[19] The Commissioner contended on the other hand that HKSKH had changed their intention and embarked on trade or business in:[20] (a) February 1984 at the earliest; (b) January 1987; (c) December 1987; or (d) September 1989 at the latest. 7. The Board of Review held that the Church Body and the Foundation had changed their intention by September 1989 at the latest, alternatively, December 1990.[21] Before the Board, the parties agreed that as at 28 September 1989, the value of the Old Lots was $192.5 million.[22] We were told that the amount of tax at stake in this appeal is around $185 million.[23] 8. After HKSKH’s appeal was dismissed by Reyes J on 27 January 2010, HKSKH appealed to the Court of Appeal. In the Court of Appeal, Mr Denis Chang SC appearing for HKSKH relied on a line of cases[24] from which he submitted one could deduce what he called the “enhancement for realisation principle”. The Court of Appeal[25] held that the true and only conclusion was that there was no change of intention from capital holding to trading/business by September 1989 or in December 1990 and remitted the matter to the Board to consider whether the change of intention occurred in August 1993 or December 1993 or alternatively some other date or dates (other than September 1989 or December 1990). Cheung JA who delivered the only reasoned judgment (with which Yuen JA and Au J agreed) said the Board erred in holding: “… that there was a change of intention in 1989 or 1990 when, on the facts found by the Board, all that the taxpayers had done was to have engaged in the process of realizing the Old Lots.”[26] The Certified Questions 9. On the Commissioner’s application for leave to appeal to this court, the appeal committee[27] granted leave to appeal on the following questions: (1) Does any “enhancement for realisation principle” arise from the authorities cited in paragraph 9 of the Judgment of the Court of Appeal, and if so, what is its scope? (See: §§9, 10 and 12.6 of the Judgment). (2) In determining whether a taxpayer has changed his intention regarding an asset from holding it for investment to holding it for trading, is the Board of Review required to refer to and apply the “enhancement for realisation principle” (as understood by the Court of Appeal or otherwise), and if the Board fails to do so, does this justify the appellate court’s interference with the Board’s finding of fact? (See: §§10, 12.2 to 12.19 of the Judgment). (3) Does a finding of fact on change of intention based solely on “enhancement activities” necessarily amount to an error of law made by the Board of Review? (See: §10.7 of the Judgment). The Evidence 10. The Church Body became the owner of Lot No 429 in DD 34 in the 1930s and the Foundation, Lot No 432 RP in DD 34 in the 1930s. Lot No 1302 RP in DD 36 was donated to the Foundation in 1957. The site area of the Old Lots was 182,798.469 sq m. St Christopher’s home, the orphanage, was built on the Old Lots. In addition, the Foundation also owned Taxlord Lot T-77 in DD 34 which was adjacent to the Old Lots. The Old Lots were surrendered in return for the New Lot on 17 November 1993. Prior to the surrender, HKSKH invited tenders from developers to purchase outright (Option A) or to enter into a joint venture with the developer to develop the New Lot (Option B). On 12 August 1993, HKSKH accepted Cheung Kong’s tender on Option B and entered into a joint venture agreement dated 3 December 1993.[28] Upon the completion of the development the units were assigned to HKSKH, the tax assessments, the subject of this appeal, arose out of the subsequent sale of some of the units. 11. According to the agreed facts, HKSKH had planned to develop the Old Lots since the 1970s. The details are set out in para 15 of the Board’s Decision (“the Decision”). As early as July 1978, the development of the Old Lots involved a high class private residential development and a Diocesan Retirement village, a special school and additional facilities to the Home. By January 1981 under plan/proposal M-1, the residential development comprised 588 units and a clubhouse on an area of 1,205,790 sq ft, and the institutional use included “existing blocks, staff quarters, children living units and special school for church members, care and attention home, retreat home, youth camp, etc.” totalling 223,413 sq ft. 12. Then in January 1986 (plan/proposal SK-A), the institutional use included the Home and a retreat centre with an area of 17,000 sq m and the residential use comprised 19 blocks of 10-12 storey towers, 11 blocks of 8-10 storey towers, 38 houses, (totalling 876 units) and a club home (sic), supermarkets, food centre, nursery and kindergartens. 13. In June 1986, there were minor changes. In December 1987 (plan/proposal SK-F), the institutional use had shrunk to 5,000 sq m for the Home,[29] and the residential use increased to 131,533 sq m with a total of 1,014 units. 14. By September 1989 (plan/proposal SK-H), there was no longer any institutional use and the residential use was reduced to 109,679.08 sq m, with 20 blocks of multi-storey towers and 20 houses (totalling 838 units) and supermarkets, laundry, coffee shop, food centre and shopping mall. 15. In May 1990 (plan/proposal SK-J), the residential use was reduced to 60,000 sq m for 2-storey houses and multi-storey (maximum 575 units) and a clubhouse. The Board’s Decision 16. The Board traced the evolution of the plans regarding institutional use and discussed internal documents of HKSKH and concluded at para 67 that: “It is clear from the appellants’ own documents that, as from September 1989 at the latest, the development of the Old Lots and the re-provisioning of the Home, or the facilities provided by the Home, became separate projects.” 17. It is not clear which of the plans/proposals was given a s 16 approval, and on the basis of which, the eventual surrender and regrant was granted but the premium for the surrender and re-grant was agreed in May 1993. The Deerhill Bay Development, comprising 22 houses, 5 low-rise and 5 high-rise buildings with 381 units, was eventually completed and the occupation permit issued in August 1998.[30] Pursuant to the joint venture agreement as amended, HKSKH were assigned the units, and the sale of some of the units has given rise to the dispute over HKSKH’s liability to pay profits tax. 18. Since it was common ground that the Old Lots were acquired and held by HKSKH not for the purpose of trading but as it were as an investment or a capital asset, they would remain “an investment unless the owner changes his intention to that of trading. If findings of this kind are to be made, precision is required. There must be evidence which establishes that change of intention. An investment does not turn into trading stock because it is sold.”[31] The Commissioner has rightly accepted that the enhanced value obtained from the mere realisation of an investment or a capital asset does not become assessable to tax,[32] and the issue is whether the owner sold merely as owner or as trader.[33] The authorities cited on behalf of the Commissioner in this context included a dictum from Gibbs CJ in FCT v Whitfords at 368 that “If the taxpayer does no more than realise an asset, the profits are not taxable. It does not matter that the taxpayer goes about the realisation in an enterprising way, so to secure the best price”.[34] In Taylor v Good,[35] Russell LJ[36] after examining a long line of authorities,[37] said he would not regard an owner of land as being engaged in trade if “not being himself a developer, (he) merely takes steps to enhance the value of the property in the eyes of a developer who might wish to buy for development”.[38] Whether an owner sells as owner or as trader is a question of fact which “depends on the interaction between the various factors that are present in any given case”.[39] But as Browne-Wilkinson VC went on to say there are factors (commonly called badges of trade) which provide “common sense guidance to the conclusion which is appropriate”.[40] 19. The Board concluded that the change of intention to trade took place by September 1989 at the latest because “as from September 1989 at the latest, the development of the Old Lots and the re-provisioning of the Home, or the facilities provided by the Homes, became separate projects”.[41] And that HKSKH continued to market the Old Lots “in an organised and coherent way with a view to maximising the income from the development. They sought and subsequently obtained a new grant by surrendering the Old Lots, thereby substituting the Old Lots by the New Lot”.[42] The Board held in the alternative that the intention was changed in December 1990 when HKSKH’s architect applied for a surrender and regrant.[43] 20. It is implicit in the Decision that HKSKH had sold as trader, presumably when they accepted Option B on 12 August 1993 or when HKSKH entered in the joint venture agreement dated 3 December 1993 but the Board did not appear to realise that the issue they had to decide was when HKSKH first intended to sell not merely as owner but as a trader and what was the evidence which established that change of intention. 21. The Old Lots were agricultural land with restricted building rights, given their size, their situation at a scenic spot off the Tai Po Road, they had great potential for residential development and were highly valuable.[44] It is common sense that the achievable price might vary according to whether planning permission had been obtained and if so the extent of the development permitted, and the amount of premium payable to government for changes to the lease terms which would enable the development to be complete. Neither planning permission nor relaxation of the lease restrictions could be taken for granted, and it is plain common sense for HKSKH as owner to seek planning permission and the lifting of lease restrictions[45] before marketing them. 22. It is in this context that the so-called enhancement for realisation principle requires consideration and I turn to the first question. The first question 23. It is unhelpful to ask whether there is an “enhancement for realisation principle” and I would not use that expression. I believe the expression “enhancement for realisation” describes a state of affairs which may provide a guide to the ultimate decision whether in the realisation of the relevant asset the owner was engaged in “trade — (or any) adventure (or) concern in the nature of trade”.[46] It is not controversial that an owner of a capital asset who sells it “as is” would not be liable for profits tax. Nor is it likely that, the owner of a home, who has done it up before putting it onto the market, or one who obtained planning permission for an extension before marketing the property, would be faced with the enquiry whether he was selling merely as owner or was he engaged in trade. In such simple cases, a finding that the owner was “trading” would be inconceivable. But, I agree with Mr Fung SC for the Commissioner, that the answer depends “on the facts of the particular case. It is essentially a question of fact and degree”.[47] But guidance is available on how the question should be answered in the particular case. These are the so-called badges of trade. Useful guidance is provided by McHugh NPJ in Lee Yee Shing v CIR, [48] namely: “Whether the taxpayer… (7) Has expended time, money or effort in selling the asset or commodity that goes beyond what might be expected of a non-trader seeking to sell an asset of that class?” (emphasis added)(the 7th badge of trade) 24. Mr Fung, has rightly accepted that if an owner of a capital asset, in selling it has done no more than what a non-trader owner might have done in similar circumstances, it would be difficult to infer that the owner intended to trade. It follows that if the facts of this case showed that HKSKH had done no more, the Decision based on such inference could not be supported. Mr Fung’s concession is well supported by the authorities referred to in para 18 above. I will add a few observations. First, I will examine Taylor more closely. There, the taxpayer purchased a house at an auction for £5,100. It was his case that he purchased the house as a possible family home, and not for the purpose of trade but when it was found to be unsuitable, he sold it about 4 years later after obtaining planning permission to erect in its place 90 houses, for £54,500. He was assessed to income tax on the profits by the Commissioners on the basis that there was intent to trade from the date of the purchase. However, the Crown conceded on appeal that the house was bought as an investment thus any intention to trade had to post-date the purchase. Megarry J remitted the case back to the Commissioners to determine when the intention to trade was first established. The Court of Appeal set aside the remittal and importantly, Russell LJ said at 560G: “that activities such as those in the present case, designed only to enhance the value of the land in the market, are (not) to be taken as pointing to, still less as establishing, an adventure in the nature of trade.” 25. Russell LJ also noted that in Hudson’s Bay Co Ltd v Stevens,[49] the court upheld the finding that there was no trade in buying and selling and that an owner who sold land purchased as an investment is no different in substance from aperson who has inherited land, if he dealt with it merely as owner even if he might have expended money in getting the property up for sale.[50] 26. Also, in Whitfords, Gibbs CJ said he “should make it clear that [he] regard it as established that profit yielded by the mere realisation of a capital asset” is not taxable[51] and emphasised the importance of the words “mere” or “merely” which are often used in such context. Although whether the taxpayer sold merely as owner is essentially a question of fact and degree, it is necessary to identify evidence which establishes the change of intention.[52] McHugh NPJ’s 7th badge of trade provides guidance for the identification of such evidence. Conduct of the taxpayer going beyond what a non-trader owner might have done in similar circumstances is such evidence but if there is no such evidence, it is difficult to imagine a case where a finding of trading or intention to trade could be supported. In the present case, in my view HKSKH’s conduct in 1989 and 1990 had not gone beyond what might be expected of a non-trader owner in similar circumstances. 27. The Board of Review concluded that there was a change of intention from capital holding to trading/business and that took place by September 1989. The significant event relied on by the Board for this conclusion appeared to be the fact that by September 1989 “the development of the Old Lots and the re-provisioning of the home, or the facilities provided by the homes, become separate projects”.[53] With respect, the fact that no part of the Old Lots was required for HKSKH’s use might explain why HKSKH decided to sell all of the Old Lots and Rowlatt J’s succinct statement “merely realizing is not trading”[54] shows that a mere sale of a capital asset is not trading. I do not understand why the decision to sell should be thought to support the conclusion that this was the date by which an intention to trade must have commenced. The same could be said of the Board’s alternative finding. If, as I think, a non trading owner might apply for a surrender and regrant, the fact that HKSKH made a similar application cannot support a finding of intention to trade. 28. The Board went on to say at para 70 that:[55] “(1) Mr Li Fook Hing was appointed a co-chairman of the Tai Po Kau Joint Development Committee in May 1989, after the re-provisioning of the Home had been separated from the development of the Old Lots and the retirement village project had been frozen for a long time. (2) It is clear from the evidence of Mr Li Fook Hing that he approached the matter on commercial principles, with the laudable object of raising as much income as possible for HKSKH and its charitable activities. (3) The appellants continued to retain the services of professional advisers including architects and lawyers to work on the development of the Old Lots. (4) They actively marketed the disposal of the Old Lots by approaching leading developers in Hong Kong for offers and tenders. (5) They sought and subsequently obtained town planning permission. (6) The appellants have performed activities in relation to the Old Lots in an organised and coherent way with a view to maximising the income from the development. (7) They sought and subsequently obtained a new grant by surrendering the Old Lots, thereby substituting the Old Lots by the New Lot. (8) They have chosen to carry on a separate adventure or enterprise of a lucrative commercial and trade character, different and distinct from their charitable work.” 29. I do not believe the conduct of HKSKH described in the first 7 sentences went beyond what a non-trader owner might have done in similar circumstances and they do not support a finding of change of intention. The Board rightly applauded Mr Li Fook Hing’s object of raising as much income as possible for HKSKH and its charitable activities but failed to consider whether what Mr Li and his committee had done went beyond what a non-trader owner might have done in similar circumstances. The eighth sentence does not follow from the earlier sentences and no useful purpose will be served by further consideration. 30. The Board also considered the badges of trade. “Badges of trade” is a convenient expression to describe the various factors which might help a tribunal to decide whether there was trading or intention to trade.[56] As Browne-Wilkinson VC said “the most they can do is provide common sense guidance to the conclusion which is appropriate”.[57] The badges of trade are not to be applied mechanically since the relevance and importance of each badge of trade may vary according to the circumstances of the particular case. The Board has quoted[58] McHugh NPJ’s 9 badges of trade in full, namely, “whether the taxpayer: (1) has frequently engaged in similar transactions? (2) has held the asset or commodity for a lengthy period? (3) has acquired an asset or commodity that is normally the subject of trading rather than investment? (4) has bought large quantities or numbers of the commodity or asset? (5) has sold the commodity or asset for reasons that would not exist if the taxpayer had an intention to resell at the time of acquisition? (6) has sought to add re-sale value to the asset by additions or repair? (7) has expended time, money or effort in selling the asset or commodity that goes beyond what might be expected of a non-trader seeking to sell an asset of that class? (8) has conceded an actual intention to resell at a profit when the asset or commodity was acquired? (9) has purchased the asset or commodity for personal use or pleasure or for income?” 31. It is obvious that some of these badges of trade have little relevance, for example, the second badge, given that the Old Lots were capital assets, it was irrelevant how long they had been held. Moreover, it is obvious that since the issue is whether HKSKH had sold merely as owner or as a trader, the 7th badge was of critical importance. Inexplicably, the Board went on to consider each of the badges of trade with the exception of the 7th badge and concluded that “there was a change of intention from capital holding to trade/business”.[59] 32. This is what the Board said at para 71 : “71. We turn now to the ‘badges of trade’ listed by McHugh NPJ and quoted by us in paragraph 58 (c) above. This is not a mechanical exercise of counting the number of scores. What we are required to do, in the words of McHugh NPJ, is to ‘make a value judgement after examining all the circumstances involved in the activities claimed to be a trade’. In considering the ‘badges of trade’, we must not lose sight of the fact that some of the factors are more relevant to the question of intention at the time of acquisition. In the cases before us, it is common ground that at the respective times of acquisition, the appellants’ intention was to hold the Old Lots indefinitely. The issue here is whether there was a change of intention: (a) Whether the appellants have frequently engaged in similar transactions – no. (b) Whether the appellants have held land for a lengthy period – yes for the Old Lots but no for the New Lot. (c) Whether the appellants have acquired an asset that is normally the subject of trading rather than investment – land can be the subject of trading or investment. It is normal to seek surrender and re-grant in trading cases. (d) Whether the appellants have bought or acquired large quantities of land – there is no evidence on whether the appellants hold other land. (e) Whether the appellants have sold the asset (or parts thereof) for reasons that would not exist if they had any intention to resell at the time of acquisition – no for the New Lot, see paragraph 70 above. (f) Whether the appellants have sought to add re-sale value to the asset by additions or repair – yes, see paragraph 70 above. (g) Whether the appellants have conceded an actual intention to resell at a profit when the asset was acquired – no. (h) Whether the appellants have acquired the asset for personal use or pleasure or for income – for ‘personal’ use in the provision of charitable activities in respect of the Old Lots, for re-sale in respect of the New Lot.” 33. Given the obvious importance of the 7th badge of trade, its omission robbed the Board’s conclusion of any validity. The Board’s treatment of the other trading badges also suffered from the omission. The distinction drawn between the Old Lots and the New Lot in (b), (c), (e) and (h), as well as the references to para 70 in (e) and (f), is, I believe the product of the Board’s mistaken belief that the decision of HKSKH to sell the Old Lots after institutional use of the Old Lots was abandoned in 1989 was itself evidence of an intention to trade. Also, the comparison of the surrender and regrant with “additions or repairs” is inapt. There was no attempt to evaluate any of the badges of trade. I am unable to gather from para 71 which of the badges of trade, the board regarded as support for their conclusion. I will say no more. 34. Contrary to Mr Fung’s submission, there is no support for the Board’s conclusion and it does not help to say that the Board had taken everything into consideration. It is obvious that the Board failed to realise that the issue they had to decide was whether HKSKH intended to sell merely as owner or as trader. 35. On the facts of the present case, with respect, I agree with the Court of Appeal that the Board’s conclusion is one which no reasonable Board of Review properly instructed in the law could find.[60] 36. I can deal with the other two certified questions briefly. Second question 37. I have already said I would not use the expression “enhancement for realisation principle”. However, the Board failed to consider whether HKSKH had intended to sell merely as owner or as a trader, and for that reason, its conclusion was vitiated and properly set aside. Third question 38. Since there was no evidence to support a finding of change of intention in September 1989 or December 1990, the Court of Appeal was right to have set aside the Decision and made the order for remittal. Disposition 39. Since writing the above I have had the advantage of reading the judgment of Mr Justice Fok PJ and the observations of Mr Justice Chan NPJ, for the above reasons and those given by Mr Justice Fok PJ and Mr Justice Chan NPJ, I would dismiss the Commissioner’s appeal and make an order nisi that the respondents are to have the costs of the appeal, such costs to be taxed unless agreed. Mr Justice Fok PJ: A. Introduction 40. I agree with the judgment of Mr Justice Tang PJ and the additional observations of Mr Justice Chan NPJ. 41. This appeal concerns a charge to profits tax, the correctness of which depended on the validity of a finding of fact by the Board of Review. Although the Court of First Instance upheld the finding of fact on an appeal by way of Case Stated, the Court of Appeal overturned that finding, applying what was described as the “enhancement for realisation principle” (addressed in more detail below). The questions that arise in this appeal are, first, whether such a principle exists and, secondly, regardless of the answer to that question, whether the Court of Appeal was correct to overturn the Board of Review’s finding of fact. 42. As will be seen from the summary of the facts more fully set out in the judgment of Tang PJ, which I gratefully adopt, the essential facts were these. The taxpayers owned land which, when originally acquired, was intended to be held indefinitely. Many years later, the taxpayers planned to redevelop the land and took steps to obtain planning permission and to apply for a land exchange with a view to redeveloping it and generating as much income as possible. The taxpayers accepted a tender from a property developer and then entered into a joint venture agreement to develop an extensive residential complex. The taxpayers sold the residential units and car parking spaces allocated to them under the joint venture and made substantial profits. These profits were assessed to profits tax by the Commissioner whose determinations confirming the relevant assessments were the subject of the proceedings in the courts below. B. The material issue of fact 43. Profits tax is chargeable only on profits arising in or derived from the carrying on by a taxpayer of “a trade, profession or business” in Hong Kong and profits arising from the sale of capital assets are excluded from such charge: Inland Revenue Ordinance (Cap.112), section 14(1). 44. It clearly follows from this statutory charging provision that a landowner may sell his land at an enhanced price above his acquisition cost but not be subject to tax on the profits thereby generated unless in doing so he is embarking on a trade or business of selling land. So the material issue of fact in the present case was whether the taxpayers were carrying on a trade or business when they made the profits sought to be taxed, or whether those profits arose from the sale of a capital asset. 45. The question of whether an activity amounts to the carrying on of a trade or business is a question of fact and degree to be answered by the relevant fact-finding body on a consideration of all the circumstances: see Lee Yee Shing v Commissioner of Inland Revenue[61] and Real Estate Investments (NT) Ltd v Commissioner of Inland Revenue[62]. 46. An intention to trade is essential. As Lord Wilberforce said in Simmons v IRC:[63] “Trading requires an intention to trade: normally the question to be asked is whether this intention existed at the time of the acquisition of the asset. Was it acquired with the intention of disposing of it at a profit, or was it acquired as a permanent investment? Often it is necessary to ask further questions: a permanent investment may be sold in order to acquire another investment thought to be more satisfactory; that does not involve an operation of trade, whether the first investment is sold at a profit or at a loss. Intentions may be changed. What was first an investment may be put into the trading stock – and, I suppose, vice versa. If findings of this kind are to be made precision is required, since a shift of an asset from one category to another will involve changes in the company’s accounts, and, possibly, a liability to tax: see Sharkey v. Wernher [1956] A.C. 58. What I think is not possible is for an asset to be both trading stock and permanent investment at the same time, nor to possess an indeterminate status – neither trading stock nor permanent asset. It must be one or other, even though, and this seems to me legitimate and intelligible, the company, in whatever character it acquires the asset, may reserve an intention to change its character. To do so would, in fact, amount to little more than making explicit what is necessarily implicit in all commercial operations, namely that situations are open to review.”[64] 47. As this passage shows: (1) the relevant time to consider intention is when the relevant asset is sold; (2) the intention then may be different to the intention when the asset was originally acquired; but (3) if a change of intention is to be relied upon as the basis for a finding of an intention to trade, precision in the fact finding process is required. C. Disposal of land may or may not be in the nature of trade 48. It is well-settled that an owner of land may dispose of his land at a higher price than that for which he acquired it and not be liable for profits tax on the gain, since his gain is “a mere enhancement of value”[65] which may simply be the result of market forces. Moreover, he may expend money improving the property in advance of such disposal without being held to have embarked on an adventure in the nature of trade. So, a landowner may lay out roads and sewers on his land or sub-divide it into smaller lots prior to sale, or re-invest the sale proceeds from part of the land to further improve the remaining parts of the land for further sales, without being found to have been carrying on a trade or business.[66] 49. Equally, however, a landowner may act in relation to the sale of his land in such a way that he will be found to have disposed of it in the course of a trade or business even if he did not himself buy the land but instead inherited it or has held the land for a long time for his own use.[67] This may be so even if the disposal is a ‘one-off’ transaction.[68] D. Determining whether there is an intention to trade 50. As indicated above, in determining whether an activity amounts to trading, the fact-finding tribunal must consider all the circumstances involved in the activity. It will then have to make a “value judgment”[69] as to whether this constitutes trading and whether the requisite intention to trade can be inferred. Regardless of what is claimed to be the intention subjectively, the question falls to be determined objectively having regard to all the surrounding circumstances.[70] 51. For this purpose, various factors have been identified as constituting “badges of trade”, the presence or absence of which may assist in the ultimate determination of whether there is an intention to trade or the carrying on of a trade. In Lee Yee Shing, McHugh NPJ identified the following “badges” at [60], namely: “… whether the taxpayer: (1) has frequently engaged in similar transactions? (2) has held the asset or commodity for a lengthy period? (3) has acquired an asset or commodity that is normally the subject of trading rather than investment? (4) has bought large quantities or numbers of the commodity or asset? (5) has sold the commodity or asset for reasons that would not exist if the taxpayer had an intention to resell at the time of acquisition? (6) has sought to add re-sale value to the asset by additions or repair? (7) has expended time, money or effort in selling the asset or commodity that goes beyond what might be expected of a non-trader seeking to sell an asset of that class? (8) has conceded an actual intention to resell at a profit when the asset or commodity was acquired? (9) has purchased the asset or commodity for personal use or pleasure or for income?” These are very similar to the “badges of trade” listed by Sir Nicolas Browne-Wilkinson V-C (as he then was) in Marson (Inspector of Taxes) v Morton,[71] which McHugh NPJ also set out in his judgment in Lee Yee Shing at [62]. 52. It is important to note that Sir Nicolas Browne-Wilkinson V-C stated that it was clear the question of whether or not there was an adventure in the nature of trade depended on (a) “all the facts and circumstances of each particular case” and (b) “the interaction between the various factors that are present in any given case”.[72] He was also at pains to emphasise that “the factors … are in no sense a comprehensive list of all relevant matters, nor is any one of them … decisive in all cases”.[73] As Lord Bridge has observed, “the law has never succeeded in establishing precise rules which can be applied to all situations to distinguish between trading stock and capital assets.”[74] Indeed, it is perhaps unfortunate that the various factors are referred to as “badges of trade”, since that phrase tends to suggest that the mere presence of one or more of those badges may mean that an activity is in the nature of a trade. This is not the intent of the list of factors, the purpose of which is to identify the facts and matters to which a fact-finding tribunal will look holistically in order to determine if the inference of an intention to trade is or is not to be drawn. E. Is there an “enhancement for realisation” principle? 53. The Court of Appeal held that the Board of Review had erred in finding an intention to trade on the part of the taxpayers by September 1989 at the latest or, alternatively, by December 1990. In reaching that conclusion, the Court of Appeal, purportedly following a line of cases culminating in Taylor v Good (Inspector of Taxes),[75] applied what was described as the “enhancement for realisation” principle.[76] At [10.7] in his judgment, Cheung JA (with whom Yuen JA and Au J agreed) explained the principle in these terms: “In my view the ambit of the Taylor line of authorities is that the activities relating to the enhancement of the value of the property for the purpose of sale would not necessarily point towards a change of intention to one of trading (rendering the transaction an adventure in the nature of a trade) when the property was initially held for investment and later disposed of. If a finding of change of intention is solely based on such enhancement activities then this amounts to an error of law in the context of a tax appeal.” It will be observed that the two sentences quoted above are somewhat contradictory in that the first sentence is equivocal (“not necessarily”) but the second is unequivocal (“If … solely based … then this amounts to an error of law”). The decision of the Court of Appeal proceeds, though, on the basis that the “enhancement for realisation” principle is that stated, unequivocally, in the second sentence. 54. In my view, there is no “enhancement for realisation” principle as held by the Court of Appeal (and as the taxpayers contended for in this appeal). As McHugh NPJ said in Lee Yee Shing, “No principle of law defines trade.”[77] Whether time, effort or money expended on an asset to enhance its sale price is or is not such as to justify a finding of intention to trade must be a matter of fact and degree and depend on the extent of such expenditure. This is expressly acknowledged by McHugh NPJ’s 7th badge of trade set out above. It follows, therefore, that enhancement for realisation, if going beyond what might be expected of a non-trader preparing to sell a long term capital asset, may be sufficient to support a finding of fact that the landowner has formed the intention to sell the land in the course of a trade or business. That finding of fact, based as it must be on the drawing of an inference that the landowner has formed the intention to trade, will depend on whether the fact-finding tribunal is satisfied that, in the colourful words of Asquith LJ (as he then was) the scheme to sell “moved out of the zone of contemplation – out of the sphere of the tentative, the provisional and the exploratory – into the valley of decision.”[78] However, in reaching that decision, the fact-finding tribunal will look to all the facts and circumstances of the case and their interaction before reaching an ultimate conclusion on the issue. 55. Not only is the “enhancement for realisation” principle contrary to the holistic approach to take into account all the facts and circumstances of the particular case, it is inconsistent with judicial statements which suggest that the act of expending time, effort or money on an asset to enhance its sale price may on its own be sufficient to support a finding of an intent to trade. Ultimately, it will depend on whether, on the particular facts, an inference of trading can properly be drawn. (1) In Pilkington v Randall, Salmon LJ (as he then was) said: “I do not read the decision of this Court in Hudson’s Bay Co. v Stevens, or the decision of Rowlatt, J., in Rand v Alberni Land Co., Ltd. as laying down a proposition of law to the effect that, whenever a property owner develops his land by making roads and laying sewers and selling plots, he can never be carrying on a trade.”[79] (2) And in Federal Commissioner of Taxation v Whitfords Beach Pty Ltd[80], in concluding that the profits in question were taxable under the second limb of section 26(a) of the Income Tax Assessment Act 1936 (Cth), if not under section 25(1) (which was the conclusion of the majority of the High Court), Mason J (as he then was) said: “In this respect I do not agree with the proposition which appears to be founded on remarks in some of the judgments that sale of land which has been subdivided is necessarily no more than the realization of an asset merely because it is an enterprising way of realizing the asset to the best advantage. That may be so in the case where an area of land is merely divided into several allotments. But it is not so in a case such as the present where the planned subdivision takes place on a massive scale, involving the laying out and construction of roads, the provision of parklands, services and other improvements. All this amounts to development and improvement of the land to such a marked degree that it is impossible to say that it is mere realization of an asset.”[81] 56. I regard Taylor v Good (Inspector of Taxes), which it should be noted was decided before the House of Lords’ decision in Simmons v IRC, as a decision on its own facts rather than as laying down the principle of law applied by the Court of Appeal in the present case. The cases cited by Russell LJ[82] (who gave the only reasoned judgment of the Court of Appeal), are not authority for the purported principle. Instead, the issue before the English Court of Appeal was whether the activities of the taxpayer “in this case”[83] after the purchase of the property (which was accepted by the Crown was not part of an adventure in the nature of trade) could be regarded as constituting such an adventure. The Court of Appeal held that those activities[84] were not of such a quality or degree as properly to be regarded as constituting an adventure in the nature of trade. The case was therefore clearly a decision on its own facts. 57. The Court of Appeal cited (at [9.8] and [10.7]) Board of Review Case No. D65/87[85] as an example of a case in which Taylor v Good (Inspector of Taxes) was applied. However, this was a decision on its own facts, involving findings by the Board of Review, in relation to particular properties acquired as capital assets, that the taxpayer did not form any intention to trade in respect of them. It does not, therefore, lend support to the existence of the “enhancement for realisation” principle. F. Was the Court of Appeal nevertheless right to overturn the Board of Review’s finding of fact? 58. To interfere with the Board of Review’s finding that the taxpayers formed the intention to trade the land by September 1989 at the latest or, alternatively, by December 1990, the Court of Appeal had to find that, on the facts found by the Board of Review, the true and only reasonable conclusion was that there was no such intention by then.[86] 59. On the other hand, as the Commissioner contended in this case, if the primary facts as found were capable of supporting two alternative inferences, it was not open to the appellate tribunal to substitute its preferred inference for that legitimately drawn by the fact-finding tribunal.[87] That was the view of Reyes J in the Court of First Instance leading him to uphold the Board of Review’s decision and dismiss the appeal.[88] 60. Notwithstanding its error in applying the so-called “enhancement for realisation” principle, was the Court of Appeal right to overturn the Board of Review’s finding of fact that, by September 1989 at the latest or, alternatively, by December 1990, the taxpayers had formed the intention to trade in the land? In my judgment, it was. 61. It was not in dispute that the taxpayers acquired the land as a long term capital asset. It was the Commissioner’s contention that the taxpayers had changed their intention in relation to the land so that, in disposing of it, they were carrying on a trade. This involved the proposition that, as a matter of fact, there was a change of intention on the part of the taxpayers. The question is whether this change of intention could properly be inferred from the primary facts found by the Board of Review. In this respect, it must be borne in mind that it is a requirement of drawing an inference that: (1) the inference must be grounded on clear findings of primary fact; and (2) the inference must be a logical consequence of those facts.[89] 62. A preliminary point to note is that the Board of Review’s finding that the taxpayers had formed the intention to trade by September 1989 at the latest or, alternatively, by December 1990 is not formulated with particular precision, a requirement stressed by Lord Wilberforce (see above). By definition, the alternative finding is inconsistent with the finding of a change of intention by the earlier date “at the latest”. Moreover, a period of 15 months in the life-span of a trade or business is not insignificant: values material to a tax assessment may fluctuate substantially and substantial activities may take place within a window of time of that magnitude. If such a finding were to be taken further, greater precision, therefore, would be required. 63. The primary facts which led the Board of Review to conclude that the taxpayers had the intention to trade by one or other of September 1989 or December 1990 were those set out in paragraph 70 of the Board of Review’s decision (set out by Tang PJ in paragraph 28 above). However, the agreed facts before the Board of Review included the fact that: on 2 July 1993, the taxpayers invited various property developers to submit tender offers either to purchase the New Lot or to enter into a joint venture agreement for development of the New Lot; on 23 July 1993, Cheung Kong (Holdings) Limited submitted two tender offers which included Option A, being a sale and purchase offer; on 12 August 1993, the taxpayers accepted Option B, which was Cheung Kong’s joint venture offer; on 3 December 1993, the taxpayers entered into a joint venture agreement with Cheung Kong for the development of the New Lot into the private residential development later known as “Deerhill Bay”; on 18 March 1998, a supplemental agreement allocated 129 residential units and 194 car parking spaces to the taxpayers; the occupation permit for the development was issued in August 1998 and between that date and 2006, the taxpayers sold their residential units and car parking spaces, thereby generating the profits subject to the disputed profits tax assessments. 64. In those circumstances, had the taxpayers accepted Option A, an outright sale of the land, it is difficult to see that they would be trading in the land and, in my judgment, it could not be said that a firm intention to commit to one method of disposal rather than the other had been formed before such time they had determined to accept Option B, involving the entry into a joint venture agreement with Cheung Kong to participate in the property redevelopment. Moreover, I do not regard the activities of the taxpayers identified in paragraph 70 of the Board of Review’s decision as having gone beyond what a non-trading property owner might do by way of improving his property with a view to its disposal at the best possible price. Consequently, the inference that the Board of Review drew as to a change of intention by September 1989 at the latest or, alternatively, by December 1990 is not logical and cannot therefore be supported. On the contrary, the primary facts found by the Board of Review do not show that what the taxpayers had done, whether by September 1989 or December 1990, went beyond what might be expected of a non-trader preparing to sell a long term capital asset. 65. Furthermore, the Board of Review’s curious omission, when considering the “badges of trade” (in paragraph 71 of its decision[90]), to consider the 7th badge of trade listed by McHugh NPJ (see above) materially undermines its conclusion as to the proper inference to draw regarding the taxpayers’ intention. 66. I therefore consider that the Court of Appeal was right, notwithstanding its error in applying the so-called “enhancement for realisation” principle, in concluding that the true and only reasonable conclusion was that there was no change of intention from capital holding to trading or carrying on a business by September 1989 or December 1990 and in allowing the appeal from the Court of First Instance to the extent it did. It also follows, since there is every reason to think that, although there was no change of intention by September 1989 or December 1990, there may have been such a change subsequent to those dates, that the Court of Appeal was right to remit the matter to the Board of Review to ascertain when that change of intention occurred. G. Conclusion 67. For these reasons, I would dismiss this appeal with costs. Mr Justice Chan NPJ: 68. I agree with the judgment of Mr Justice Tang PJ and the judgment of Mr Justice Fok PJ. I would like to add the following observations. 69. When considering whether a person is liable to pay profits tax, the starting point must be the statutory provision, s.14(1) of the Inland Revenue Ordinance, Cap 112. It provides, among other things, that tax is payable on profits arising from a trade, but excludes profits arising from the sale of capital assets. “Trade” is defined in s.2 as including every adventure and concern in the nature of trade. Most disputes involve the determination of whether what was done in a particular case amounts to an adventure in the nature of trade or merely a realization of capital assets. 70. It is not disputed that a single one off transaction can be an adventure in the nature of trade (see e.g. Marson v Morton, [1986] 1 WLR 1343, 1347H). However, as Lord Wilberforce said in Simmons v IRC [1980] 1 WLR 1196, 1199A, “Trading requires an intention to trade”. Evidence of intention is necessary to establish whether a transaction in question is an adventure in the nature of trade and not merely a realization of capital assets. The intention at the time of acquisition of an asset which was later sold at a profit is usually “a very strong pointer” (Marson v Morton, 1348H; see also All Best Wishes Ltd v IRC (1992) 3 HKTC 750, 771), but that intention may be changed (Simmons v IRC, 1199, 1202). 71. Whether there has been a change of intention is a question of fact. The answer depends on all the facts and circumstances of each case and the interaction between the various factors present. There are “features or badges” which are relevant in the determination of this question, but no list of relevant factors can be exhaustive and no single factor can be decisive (see Marson v Morton, 1348B). To arrive at a proper assessment on the facts, the correct approach is “to stand back, having looked at those matters and look at the whole picture” and ask the question: whether there was an adventure in the nature of trade (Marson v Morton, p1349C). 72. I do not think Mr Denis Chang SC is right in submitting that there is a principle of law which he calls “enhancement for realization principle” that must be applied to cases involving enhancement activities and that failure to apply such principle will result in a conclusion that the transaction in question was an adventure in the nature of trade to be set aside. The authorities cited by Mr Chang do not support his propositions. I believe the true effect of the authorities is as follows. 73. Where property was originally acquired as capital investment, the owner can take steps to improve his property or, as it is sometimes put, engage in activities to enhance the value of the asset so as to obtain the best or maximum price when he disposes of it. This in itself would not convert the disposal into an adventure in the nature of trade unless such steps or activities go beyond what would be regarded as the mere realization of capital. Where, on the facts of the case, the steps and activities taken have gone beyond the mere realization of capital, they would be regarded as evidence pointing towards the conclusion that there was a change of intention on the part of that person to embark on an adventure in the nature of trade. (See Williams J in Scottish Australian Mining Ltd v Federal Commissioner of Taxation [1950] 81 CLR 188, 195; Lord Donovan in McClelland v Tax Commissioners [1971] 1 WLR 191, 198; Russell LJ in Taylor v Good [1973] 49 TC 277, 296; Gibbs CJ in Federal Commissioner of Taxation v Whitfords Beach Pty Ltd (1982) 150 CLR 355, 367; Nourse LJ in Kirkham v Williams [1991] 1 WLR 863, 868H.) Whether the steps or activities taken in a particular case can be so regarded must depend on all the circumstances of that case including the subject matter of the asset and the nature and extent of the steps and activities taken. Enhancement activities are only one of the factors, albeit an important factor, to be considered in ascertaining whether the owner has the intention to carry on an adventure in the nature of trade. 74. It is common ground that the property concerned in this case was intended as an investment at the time it was acquired. The respondents had since the 1970s made several plans and taken various steps in its proposed development. But on the facts of this case, contrary to Reyes J’s view, I do not think it was open to the Board of Review to conclude that there was a change of intention on the part of the respondents in September 1989 or at the latest December 1990 to develop the property concerned as a trade. The Board had, wrongly in my view, regarded the fact that by September 1989, the development of the property concerned became a separate project as the critical turning point in the events. This can be seen from paragraphs 68 and 70 of its Decision. In these two paragraphs, the Board seemed to suggest that not only was there no explanation on the part of the respondents as to why they chose to proceed with the development of the property after that date, but the respondents had also proceeded with such development actively in an “organized and coherent” manner by appointing professional advisors and Mr Li Fook Hing as chairman of a development committee, approaching developers and applying for town planning permission and for a new grant. The Board had obviously overlooked the fact that the property concerned was an agricultural and restricted building land which would not yield a very attractive price if it were to be sold as such. The steps and activities taken by the respondents were necessary for finding out the potential of the property concerned and ascertaining the maximum value which it could fetch. With respect, I also have difficulty in understanding paragraph 71 of the Board’s Decision. If the Board was there purporting to apply the badges of trade as proposed by McHugh NPJ in Lee Yee Shing v Commissioner of Inland Revenue(2008) 11 HKCFAR 6 to the facts of this case, the simple answers given in that paragraph are, without elaboration, hardly helpful at all. In my view the true and only reasonable conclusion is that paragraphs 72 to 74 of the Board’s Decision cannot be supported on the basis of the respondents' activities up to September 1989 or December 1990. Mr Justice Gummow NPJ: 75. I agree with the judgment of Mr Justice Fok PJ, and thus with the judgment of Mr Justice Tang PJ, and the additional observations of Mr Justice Chan NPJ. Mr Eugene Fung SC and Mr Wilson Leung, instructed by Department of Justice, for the respondent /appellant Mr Denis Chang SC and Mr Newman Lam, instructed by PC Woo & Co, for the 1st and 2ndappellants / 1st and 2ndrespondents [1] Lot 429 in DD 34 was owned by the Church Body and Lot No 432 RP in DD 34 and Lot No 1302 RP in DD 36 owned by the Foundation. Lot No 1302 RP was donated to the Foundation in 1957, BR para 49(c). The total area of the Old Lots was 182,798.469 sq m. See Copies of application for town planning permission dated 28 December 1987, LWC-5 (s 16 application). [2] Board of Review (“BR”) para 16. [3] BR para 15, where the different plans at different times were set out. They will be discussed more fully below. [4] With attendant commercial uses. [5] BR para 50. [6] BR para 66(5) mentioned a s 16 town planning application in relation to proposal SK-H (see para 14 below) and an application in December 1987 in connection with proposal SK-F (see para 13 below). Presumably there were other applications in relation to other proposals. [7] At Government’s discretion. [8] BR para 16. As is commonly known the mechanism whereby lease restrictions are removed are usually achieved by means of a surrender of Old Lots for a new grant, a New Lot. Since many old agricultural lots were small and had irregular shapes and inadequate access, government often used the mechanism of a surrender and regrant to reshape the land and to provide for roads or access. Sometimes, the New Lot will be smaller in area than the surrendered lots. When government agrees to relax lease restrictions they would indicate the amount of premium payable, which although subject to contract, would normally remain the same provided they were accepted within a stated period. [9] BR para 16(e). [10] To be granted. [11] Or a subsidiary formed for the purpose. For convenience I will refer to it as Cheung Kong. [12] BR para 20. [13] As amended in 1998. [14] The Deerhill Bay. [15] BR paras 9(2), 27, 29 and 30. It is not clear on what figures the Commissioner’s assessment was based. [16] Court of Appeal (“CA”) para 5. [17] For example, the parties agreed that the value of the Old Lots was $222.48 million as at 1 May 1990, $1.11 billion (exclusive of premium) as at 12 August 1993 and $2.3 billion (premium paid) as at 3 December 1993. The lower the value at the time of change of intention, the higher the profit and the tax payable. [18] Though for the purpose of determining the amount of profits tax payable, the date when the intention to trade was found was important, because the value on that date will be basis for the calculation of profits, if any. [19] If, however, it had accepted Option A, it may be that they had done no more than selling a capital asset and there was no trading at all. As Cheung JA suggested in the CA at 12.14, it was only when Option B was accepted that the development moved out of the zone of contemplation into the valley of decision employing the language of Asquith LJ in Cunliffe v Goodman [1950] 2 KB 237 at 254. [20] BR para 61. [21] BR paras 73-74. [22] BR para 75. [23] Appellant’s written case para 24. [24] Including Taylor v Good (Inspector of Taxes)[1973] STC 383 and Hudson’s Bay Co Ltd v Stevens (1909) 5 TC 424. [25] Cheung, Yuen JJA and Au J. [26] CA para 12.19. [27] Ribeiro, Tang and Fok PJJ. [28] We do not have copies of the tender documents nor the joint venture agreement. Presumably, the premium was paid pursuant to the tender and the provisions of the joint venture agreement. [29] The Board said the proposed area for the Home fell outside the Old Lots and would be situated at the Taxlord Lot No.T-77. See BR 66(5), footnote 7. [30] CA para 2.13. [31] Per Mr Andrew Li QC as he then was in D65/87 IRBRD Vol 3 66, based on the observations of Lord Wilberforce and Lord Salmon in Simmons v IRC [1980] 9 1 WLR 1196 at 1199B and 1203H respectively. [32] Para 108, the Commissioner’s written case, citing in support Californian Copper Syndicate v Harris (1904) 5 TC 159 at 165-166; Whitfords Beach at 367-368, 372, 394-395; D65/87 at 80; Kirkham at 868H. [33] See for example, Farwell LJ in Hudson’s Bay at 438, and Mason J in FCT v Whitfords Beach Pty Ltd at 371. [34] In McClelland v Commissioner of Taxation [1971] 1 WLR 191, Lord Donovan, delivering the majority judgment of the Privy council on appeal from Australia, made a similar statement at 197E. [35] [1974] 1 WLR 556. [36] Whose judgment was agreed to by Stamp and Orr LJJ. [37] Including Hudson’s Bay Co. Ltd. v Stevens (1909) 5 TC 424, CA. [38] At 560D. [39] Browne-Wilkinson VC in Marson v Morton [1986] 1 WLR 1343 at 1348B. [40] 1348D. [41] BR para 67. [42] BR para 70. [43] BR para 74. [44] The Old Lots had an agreed value of $192.5 million as at 28 September 1989. BR para 75. The valuation must have reflected in part its development value. As at 12 August 1993, the value was $1.11 billion exclusive of the premium. We do not know how much of the increase is attributable to a rise in the market and how much to the fact that the requisite permissions had been obtained. Presumably the valuation in 1989 took into account their redevelopment potential discounted by the uncertainty in obtaining the necessary permissions. [45] By means of a surrender and regrant and an indication of the premium payable which is likely to affect the price at which the land could be sold. [46] Definition of “trade” s 2 Inland Revenue Ordinance Cap 112. [47] Pilkington v Randall (1966) 42 TC 662 at 674. [48] (2008) 11 HKCFAR 6 at 28 para 60, with the express agreement of Ribeiro PJ and Sir Noel Power NPJ. Bokhary and Chan PJJ who delivered a separate judgment said at para 40 that they “found (McHugh NPJ’s) helpful generally”. [49] (1909) 5 TC 424, CA. [50] At 559H. [51] Whitfords was concerned with the interpretation of s 25(1) and s 26(a) of the Income Tax Assessment Act 1936 with which we are not concerned, but Gibbs CJ’s words are apposite to determining whether an owner was selling as owner or as trader. [52] See per Mr Andrew LI QC cited in para 18 above. [53] BR para 67. [54] Alabama Coal, Iron, Land and Colonization Co. Ltd. v Mylam (1926) 11 TC 232, approved Commissioner of Taxes v British Australian Wool Realization Association, Ltd [1931] AC 224 at 252. [55] I have numbered the sentences for ease of reference. [56] Lee Yee Shing v CIR (2008) 11 HKCFAR 6 per McHugh NPJ at para 60. [57] Marson v Morton [1986] 1 WLR 1343 at 1348D. [58] At para 58(c). [59] BR para 72. [60] CA para 12.1. [61] (2008) 11 HKCFAR 6 at [38] and [56]. [62] (2008) 11 HKCFAR 433 at [40] and [55]. [63] [1980] 1 WLR 1196 at 1199A-D. [64] This passage was cited with approval by this Court in Lee Yee Shing v Commissioner of Inland Revenue (supra) at [57] and in Real Estate Investments (NT) Ltd v Commissioner of Inland Revenue (supra) at [39]. [65] Californian Copper Syndicate (Limited and Reduced) v Harris (1904) 5 TC 159 per the Lord Justice Clerk (Lord Kingsburgh) at 166. [66] Hudson’s Bay Co Ltd v Stevens (1909) 5 TC 424 at 437-438; and Rand v Alberni Land Co Ltd (1920) 7 TC 629 at 638-639. [67] Alabama Coal Co Ltd v Mylam (1926) 11 TC 232; All Best Wishes Ltd v Commissioner of Inland Revenue (1992) 3 HKTC 750. [68] Marson (Inspector of Taxes) v Morton [1986] 1 WLR 1343 at 1347H; and see, by way of example, Hong Kong Oxygen & Acetylene Co Ltd v Commissioner of Inland Revenue [2001] 1 HKLRD 489. [69] Lee Yee Shing per McHugh NPJ at [56]. [70] All best Wishes Ltd v Commissioner of Inland Revenue (supra) at 771. [71] [1986] 1 WLR 1343 at 1348D-1349C. [72] Ibid. at 1348B. [73] Ibid. at 1348C; he emphasised this again at 1349C. [74] Waylee Investment Ltd v Commissioner of Inland Revenue [1990] STC 780 at 784g. [75] [1973] STC 383, [1974] 1 WLR 556. [76] CA Judgment at [9.9]. [77] (2008) 11 HKCFAR 6 at [56]. [78] Cunliffe v Goodman [1950] 2 KB 237 at 254; see also Crawford Realty Limited v Commissioner of Inland Revenue (1991) 3 HKTC 674 at 693, where Barnett J drew a distinction between “enhancement” of an asset (which would not be trading) and such extensive enhancement as to constitute “substitution” (which would be trading). [79] (1966) 42 TC 662 at 673; to the same effect, in the same case at first instance, Cross J (as he then was) said (ibid. at 669), “I do not think that one can lay down hard and fast rules, such as that the construction of roads and sewers and the installation of services can never be enough to make the case one of embarking upon trade.” [80] (1982) 150 CLR 355. [81] Ibid. at 385. [82] [1974] 1 WLR 556 at 559G-560D, namely: Hudson’s Bay Co Ltd v Stevens, Rand v Alberni Land Co Ltd, Alabama Coal Co Ltd v Mylam and Pilkington v Randall. [83] Ibid. at 561C. [84] Ibid. at 558D-E, namely: a first application for planning consent to use the land for residential purposes; preparation of plans for the lay-out of houses and a successful application for planning permission for such lay-out; and the procuring of co-operation from a neighbouring landowner to facilitate suitable road access. [85] IRBRD Vol.3 66. [86] Edwards v Bairstow [1956] AC 14 at 36; Kwong Mile Services Ltd v Commissioner of Inland Revenue (2004) 7 HKCFAR 275 at [31] and [36]. [87] Richfield International Land and Investment Co Ltd v Inland Revenue Commissioner [1989] STC 820 at 824h. [88] HCIA 2/2009, unrep., Judgment dated 27.1.2010 at [38]. [89] Nina Kung v Wong Din Shin (2005) 8 HKCFAR 387 at [185]; whilst in a criminal case there is an additional third requirement that the inference must be irresistible(Winnie Lo v HKSAR (2012) 15 HKCFAR 16 at [115]), the first two requirements remain essential in respect of any case. [90] Set out by Tang PJ in paragraph 32 above. 
Press Summary (English) Press Summary (Chinese) FACV No. 8 of 2012 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION FINAL APPEAL NO. 8 OF 2012 (CIVIL) (ON APPEAL FROM CACV NO. 192 OF 2010) Between: Chief Justice Ma : 1. I agree with the reasons given by Mr Justice Gleeson NPJ for dismissing the appeal. Mr Justice Ribeiro PJ : 2. I agree with the reasons given by Mr Justice Gleeson NPJ for dismissing the appeal. Mr Justice Litton NPJ : 3. The application for judicial review in this case focusses on two “decisions” of the Council, as expressed in two letters from the Institute’s solicitors: (i) one dated 12 December 2009 (comprising six pages of close print) “the first decision” and (ii) the other dated 8 January 2010 (comprising two pages of close print) “the second decision”: A curious way of impeaching the decisions of a public body by way of judicial review. If a decision of such a body is capable of being reviewed by the courts, it would normally be capable of being expressed in a few words. And, being thus expressed, the judge dealing with the application for leave under Ord 53 r 3 of the Rules of the High Court, Cap 4, would readily see which of the recognized grounds formulated by Lord Diplock in CCSU v Minister for the Civil Service [1985] AC 374 at 410 might cover the case. 4. When Mr Gerard McCoy SC, counsel for the appellants, was asked at the hearing which of the three grounds articulated in CCSU v Minister for the Civil Service he invoked in support of his case, he said the first and the third: that is to say “illegality”[1] and “procedural impropriety”[2]. And then later, in the course of his submissions to this Court, Mr McCoy said he relied also on the second ground: “irrationality”[3]. When an application for judicial review is cast over such a wide field, it calls into question the very foundation of the application. 5. The impugned decision-maker in this case is the Council: The body which, in its discretion, might initiate disciplinary proceedings against certified public accountants under s 34(1) of the Professional Accountants Ordinance, Cap 50. The “first decision” 6. The first impugned decision as set out in the application for leave under Ord 53 r 3 is the solicitors’ letter of 12 December 2009. This is summarized in the application for leave dated 15 January 2010 as a decision of the Institute (the respondent to the proceedings) “rejecting all the Applicants’ complaints including inter alia the constitution of the Investigation Committee”. But the Council had no function under the statute to resolve complaints. Its function was to consider the Investigation Committee’s report and to decide whether to proceed further by initiating the disciplinary process as laid down in the statute: see para 5 above. The appellants’ complaints related solely to the investigation process. Those complaints might at most affect the weight of the Investigation Committee’s report and its conclusion that a prima facie case had been made out against the appellants: But, at the end of the day, what the Council had to decide was simply whether it should refer the case to the Disciplinary Panels, thereby initiating the disciplinary process, having regard to the seriousness of the matters disclosed in the report as a whole and all relevant matters. 7. The appellants had been told (by letter dated 8 January 2010) that the Council would next meet on 19 January 2010 and that if they had further submissions to make they should do so by 12 January. 8. The appellants pre-empted the Council’s meeting and launched their application for judicial review on 15 January. It is therefore plain that there was no ground whatever to impugn the first decision. The “second decision” 9. As regards the second decision (set out in the lawyers’ letter of 8 January 2010), it is summarized thus in the application for judicial review: “The Decision of the Respondent … refusing to consider re-constitution of a new Investigation Committee with different membership so that the statutory task of determining whether a prima facie case for disciplinary proceedings exists be conducted afresh by the new Investigation Committee”. No statutory duty falling upon the Council has been identified. And yet the relief sought in the application for judicial review was an order of mandamus: “requiring the Respondent to re-constitute an Investigation Committee … etc” Conclusion 10. The application for judicial review was hopelessly flawed from the outset. It ought never to have been made. 11. I agree with Mr Justice Gleeson NPJ’s judgment. Mr Justice Bokhary NPJ : 12. This appeal was ably argued but had to be dismissed. As to why, I agree with the reasons which Mr Justice Gleeson NPJ gives. Mr Justice Gleeson NPJ : 13. This is an appeal from a decision of the Court of Appeal (Hon Tang VP, Chu JA and Lam J), which dismissed an appeal to that court from Reyes J. Reyes J dismissed an application for judicial review brought in respect of an enquiry into the professional conduct of a firm of accountants. At the conclusion of argument in this Court, the Court made orders that the appeal be dismissed and that the appellants pay the costs of the respondents, and said that it would publish its reasons at a later date. What follows is a statement of my reasons for joining in those orders. 14. The second and third appellants, Mr Cheng and Mr Lai, are partners in Messrs HLB Hodgson Impey Cheng (“HLB”), a firm of Certified Public Accountants, which is the first appellant. The respondent, The Hong Kong Institute of Certified Public Accountants (“the Institute”) is a body corporate with a membership comprising all registered certified public accountants. As at February 2010, there were 29,933 members of the Institute, and 1,507 registered firms and corporate practices. 15. The Institute was established by the Professional Accountants Ordinance, Cap 50 (“the PAO”), as the profession’s principal regulator. The affairs of the Institute are governed by a Council. The members of the Council are 14 certified public accountants elected at the Annual General Meeting of the Institute and four lay persons appointed by the Chief Executive of the Hong Kong Special Administrative Region. The Institute maintains a permanent staff including a Compliance Department which assists in investigations and disciplinary proceedings resulting from complaints about the conduct of members of the Institute or registered firms or corporate practices. 16. This matter arose out of a reference to the Institute by the Securities and Futures Commission (“SFC”) in September 2006 of certain concerns over the standard of HLB’s work as auditors of Tiffit Securities (Hong Kong) Ltd (“Tiffit”), a corporation licensed to carry on the business of dealing in securities. In July 2006, the SFC discovered a material shortfall in the securities held by Tiffit on behalf of its clients, and suspected misappropriation. On 24 July 2006, administrators were appointed, and, in May 2007, Tiffit was wound up. HLB had issued clean audit reports for Tiffit for the years ended 31 March 2004 and 31 March 2005, and had completed the 2006 audit. The decisions the subject of the judicial review application were made by the Council of the Institute in the course of dealing with the reference. The PAO 17. The starting point for a consideration of a challenge to the lawfulness of an exercise of statutory power is the statute. It is therefore necessary to begin with an examination of the scheme of the PAO in relation to complaints and disciplinary proceedings. 18. Part V of the PAO deals with Disciplinary Proceedings. It comprises ss 33 to 41B. 19. Section 34(1) lists a series of complaints that may be made about a certified public accountant, and provides that such complaints shall be made to the Registrar, appointed under s 21, who shall submit the complaint to the Council which may, in its discretion, refer the complaint to the Disciplinary Panels. If the Council decides not to refer the complaint to the Disciplinary Panels, the complainant may request it to make the reference and the Council is obliged to do so unless it is of opinion that no prima facie case has been made out or that the complaint is frivolous or vexatious. One of the grounds of complaint is failure to observe a professional standard. 20. Section 33 establishes the Disciplinary Panels, one (Panel A) consisting of lay persons appointed by the Chief Executive and one (Panel B) consisting of certified public accountants appointed by the Council. Where a complaint is referred to the Disciplinary Panels, the Council is required to constitute a Disciplinary Committee consisting of three persons from Panel A and two persons from Panel B. 21. Sections 36 and 37 deal with the proceedings of a Disciplinary Committee. Hearings, subject to certain exceptions, are in public. A Committee may take evidence on oath and compel the attendance of witnesses. The complainant and the person whose conduct is the subject of the proceedings may be legally represented. The case against that person is presented either by the complainant or the complainant’s lawyer, or the Registrar or the Registrar’s lawyer, or a member of an Investigation Committee or the member’s lawyer, or some other person appointed by the Investigation Committee to represent it. 22. The powers of a Disciplinary Committee, if satisfied that a complaint referred to it is proved, are set out in s 35. They include a power to order that the name of an accountant be removed from the register, either permanently or for a period, or to impose a fine or administer a reprimand. An appeal lies to the Court of Appeal which may confirm, vary or reverse the order appealed against. 23. In brief, a complaint under s 34 is made to the Registrar who is obliged to submit it to the Council. The Council assesses complaints and decides whether they should go further. If the Council decides a complaint should go further then the complaint is dealt with after a full hearing on its merits by a Disciplinary Committee which is empowered to make orders against the accountant the subject of the complaint. If such an order is made, the accountant has a right of appeal to the Court of Appeal. 24. Part VA of the PAO deals with investigations made on the initiative of the Council, which, after certain procedures have been followed, may result in the conduct of an accountant being referred to the Disciplinary Panels and dealt with under s 34, as if the matter were a complaint made to the Registrar by an Investigation Committee. Part VA comprises ss 42A to 42H. 25. Section 42C(2) provides that where the Council reasonably suspects or believes that an accountant, or firm, has acted in a manner described in certain of the provisions of s 34, the Council may, in its discretion, constitute an Investigation Committee and direct the Committee, having considered the matter, to inform the Council as to whether in its opinion, were a complaint made against him or it, the accountant or firm would have a case to answer. Section 42C(1) provides that where an Investigation Committee so informs the Council, the Council may in its discretion refer the matter to the Disciplinary Panels and constitute a Disciplinary Committee pursuant to s 33 and the Disciplinary Committee shall deal with it as if it were a complaint under s 34, as mentioned above. 26. An Investigation Committee is constituted as follows. Section 42B establishes two Investigation Panels, one (Panel A) comprising lay persons appointed by the Chief Executive and one (Panel B) comprising accountants appointed by the Council. An Investigation Committee is made up of three persons from Panel A and two persons from Panel B. 27. An Investigation Committee is given certain investigative powers (s 42D), which include a power to compel production of documents and otherwise obtain information, including information from an accountant or firm. 28. An Investigation Committee may delegate all or any of the powers given to it by s 42D. 29. An Investigation Committee does not conduct hearings. Its task is to consider the matter and inform the Council whether in its opinion, were a complaint made against an accountant or firm, there would be a case to answer (s 42C(2)). If it so informs the Council, the Council may in its discretion refer the matter to the Disciplinary Panels and constitute a Disciplinary Committee (s 42C(1)). 30. It is not open to doubt that, on the true construction of the PAO, an Investigation Committee, in exercising its statutory function of considering the matter, reaching an opinion about the conduct of an accountant or firm, and informing the Council of its opinion, is obliged to act fairly and to be, and appear to be, independent and impartial. At the same time, having regard to the nature of the case which the appellants seek to make, it is necessary to keep in mind the following features of the statutory scheme which bear upon the practical content of those obligations. First, neither the Investigation Committee nor the Council is required or empowered to make a decision on the merits of a case, or potential case, against an accountant or a firm, or to make orders against an accountant or firm. That is the role of the Disciplinary Committee. Secondly, the Investigation Committee is to form an opinion about whether there is a case to answer, and the Council, having been informed of that opinion, is to decide whether in its discretion it should refer the matter to the Disciplinary Panels and constitute a Disciplinary Committee. An opinion adverse to an accountant or a firm formed by the Investigation Committee does not necessarily result in such a reference by the Council, which has its own statutory discretion to exercise. The Council may not necessarily agree with the opinion of the Investigation Committee. An unfavourable opinion of an Investigation Committee is neither a necessary nor a sufficient condition for a referral. The range of discretionary considerations that might lead the Council to decide not to refer a matter to the Disciplinary Panels is bounded by the purposes of the Ordinance, but it is clear that under the statute it does not automatically make a reference on being informed of an adverse opinion. Thirdly, the nature and scope of the investigative work undertaken by an Investigation Committee, and the method of investigation adopted, is left largely in its own discretion. It has a capacity to delegate certain of its investigative powers. There is nothing to prevent it from seeking administrative assistance. In any case of substance, it is very likely to require such assistance, especially in view of the fact that a majority of its members are not themselves accountants. 31. Section 51 of the PAO empowers the Council to make rules regulating the conduct of inquiries by the Disciplinary Committee and for other matters relating to such inquiries. Rule 5(b) of the Disciplinary Committee Proceedings Rules provides: “5. The Council, before deciding whether or not to refer a complaint to the Disciplinary Panels: … (b) shall invite the Respondent to submit to the Council in writing any explanation of his conduct or of any other matter alleged in the complaint which he may have to offer.” 32. In that regulatory context, what was said by Lord Denning MR in R v Race Relations Board, ex parte Selvarajan [1975] 1 WLR 1686 is in point: “In recent years we have had to consider the procedure of many bodies who are required to make an investigation and form an opinion. …In all these cases it has been held that the investigating body is under a duty to act fairly: but that which fairness requires depends upon the nature of the investigation and the consequences which it may have on persons affected by it. The fundamental rule is that, if a person may be subjected to pains or penalties, or be exposed to prosecution or proceedings, or deprived of remedies or redress, or in some such way adversely afflicted by the investigation and report, then he should be told the case made against him and be afforded a fair opportunity of answering it. The investigating body is, however, the master of its own procedure. It need not hold a hearing. It can do everything in writing. It need not allow lawyers. It need not put every detail of the case against a man. Suffice it if the broad grounds are given. It need not name its informants. It can give the substance only. Moreover, it need not do everything itself. It can employ secretaries and assistants to do all the preliminary work and leave much to them. But, in the end, the investigating body itself must come to its own decision and make its own report.” The investigation and report 33. The facts summarized below were established by the evidence of Nip Ting Ming, Peter (“Mr Nip”), the Institute’s Deputy Director, Compliance, and K P Morrison, a member of the Investigation Committee. That evidence was accepted by the primary judge. There was a challenge to the relevance of part of Mr Morrison’s evidence. It will be necessary to return to that. 34. The conduct of HLB in respect of the audit was first raised with the Registrar of the Institute, and the President of the Council, by the SFC. The subject of concern was said by the SFC to be the steps taken by HLB for the purpose of verifying the reported stock holdings and account balances of Tiffit’s clients. More than 400 such clients were thought by the SFC to have been victims of misappropriation. The SFC provided the Council with a written statement of its own findings and observations. The Compliance Department of the Institute examined the SFC’s report and observations and recommended to the Council that an Investigation Committee be constituted. The Council so resolved. On 2 April 2007, the Investigation Committee was constituted. Its Chairman was a partner of a major law firm. Its two other lay (non-accountant) members were businessmen. There were also two accountant members, one of whom was Mr Morrison. Each member was required to declare that he had no conflict of interests. 35. The Council appointed Ms Lee, of Elite International Consulting Limited, to act as a consultant to the Investigation Committee. The investigation proceeded from July 2007 until August 2009. Over that period, there were a number of meetings of the Investigation Committee and there was extensive correspondence between the Committee, HLB, HLB’s lawyers and the Council. Meetings of the Committee were attended both by Ms Lee and by officers of the Compliance Department. The Committee sought and obtained from HLB and its lawyers a substantial amount of information about the audits in question. 36. In July 2008, Ms Lee furnished the Investigation Committee with a Discussion Draft headed “Preliminary Findings”. It began by saying that its purpose was to set out for the Committee’s consideration preliminary findings of the review of HLB’s audit working papers and their responses to inquiries made of them. It was a document of some 62 pages. Its conclusions were expressed in such terms as: “The Committee may wish to consider …”. The document identified a number of Statements of Auditing Standards (“SAS”) against which it said the conduct of HLB was to be measured. It went into detail about what were said to be deficiencies revealed by the investigation, and by HLB’s responses to questions about their audit. 37. On 8 July 2008, at a meeting of the Investigation Committee, the Committee directed that the Compliance Department draft for its consideration a report to the Council based on the proposed findings suggested by Ms Lee. 38. On 28 November 2008, the Investigation Committee met to consider a draft report prepared by the Compliance Department. Also present at the meeting were three officers of the Compliance Department. The draft report was of some 30 pages. It referred to the same SAS standards as had been referred to in Ms Lee’s Discussion Draft. The Investigation Committee directed that a number of amendments be made to the draft, and that the draft as amended should be sent to HLB for comment. Before that was done, however, certain further information was requested from HLB about some administrative details. 39. On 28 April 2009, the draft report was sent to HLB by the Chairman of the Investigation Committee, inviting comments by way of written submission. 40. After various extensions of time granted at the request of HLB’s lawyers, the Investigation Committee, on 30 July 2009, decided that it would not give HLB further time, and resolved to make a report to the Council in terms of the draft sent to HLB. A copy of the report was sent to HLB. 41. The report was dated 27 August 2009. It consisted of 38 pages to which were attached a series of annexures extending over some 300 pages. Its conclusion was that the audit procedures evidenced in HLB’s working papers did not adequately support the opinions in the compliance reports it furnished in respect of Tiffit for the years ended 31 March 2004 and 31 March 2005 and that, were a complaint to be made against the appellants under s 34(1)(a)(vi) of the PAO, for failing or neglecting to observe, maintain or otherwise apply a professional standard, they would have a case to answer. 42. On 2 September 2009, the appellants sent a letter, marked “Strictly Private and Confidential”, addressed to the Council of the Institute. The letter was accompanied by what was described as a report to the Council, of some 40 pages, which included allegations of procedural unfairness and apparent bias on the part of the Investigation Committee and certain of its members, a justification of the audit procedures undertaken in respect of Tiffit, and an independent expert report by a securities industry practitioner engaged by HLB to comment on the audit. The HLB report also made the point that the wrong accounting standards had been applied by the Investigation Committee. It said that the relevant standard was the Standard on Assurance Engagements (“SAE”), not the SAS. 43. On 8 September 2009, Mr Hilliard of the Compliance Department wrote to HLB: “In view of the volume of the submissions and the issues raised therein, it was decided that the submissions be referred to the Investigation Committee for its consideration, and that the presentation to the Council of the Investigation Committee’s report dated 27th August 2009 be deferred to allow adequate time for the Committee to properly consider the submissions.” 44. On 12 October 2009, the Investigation Committee met to consider the appellants’ submissions. In the meantime, the Compliance Department had drafted a revised report on the basis that the appellants were correct to say that the relevant standards were the SAE, not the SAS, but that the conduct that was earlier seen as a breach of certain SAS standards would equally be a breach of certain SAE standards. The meeting of the Investigation Committee was also attended by three officers of the Compliance Department. The Investigation Committee reviewed and discussed the revisions to the report of 27 August 2009 proposed by the Compliance Department. The Investigation Committee directed the Compliance Department to prepare a revised report. On 16 October 2009, the Compliance Department sent a draft revised report to the Chairman of the Investigation Committee. On 19 October 2009, the Chairman made comments, proposing certain alterations to the wording of the draft. This resulted in an amended draft report dated 20 October 2009. 45. On 20 October 2009, the Chairman of the Investigation Committee wrote to the appellants: “The Investigation Committee has considered the submissions and has further considered the findings set out in the Report of the Investigation Committee dated 27 August 2009 … [T]he Investigation Committee agrees that the professional standards applicable … are [the SAE]. Accordingly, the Investigation Committee has amended certain sections of the Report. A copy of the amended Report dated 20th October 2009 … and a copy of the amended complaint letter are enclosed.” 46. On 6 November 2009, a new set of lawyers acting for the appellants wrote to the Registrar, copying their letter to each member of the Council. The letter complained that: (a) The Registrar should not have disclosed to the Investigation Committee the appellants’ submissions of 2 September 2009. (b) Two members of the Investigation Committee were associated with the SFC and therefore biased. The letter demanded that the Council “disallow” the Investigation Committee’s (draft) report of 20 October 2009 and refrain from referring the matter to Disciplinary Panels. 47. On 9 November 2009, the appellants’ lawyers lodged disciplinary complaints under s 34 of the PAO against the Registrar, the Director of Compliance, and certain members of the Investigation Committee. 48. On 16 November 2009, the Investigation Committee signed a report in the form of the draft report of 20 October 2009. 49. On 12 December 2009, the lawyers for the Institute sent a detailed response to the appellants’ complaints, in which it was made clear that the Council had not yet dealt with the matter but that it proposed to do so, and inviting the appellants to make any submissions they wished to make “on the substance of the report dated 16 November 2009”. 50. On 5 January 2010, the lawyers for the appellants wrote to the lawyers for the Institute repeating their complaints of bias and making a further complaint (which was later to become the focus of the present proceedings) of “an appearance that the Investigation Committee was simply rubber stamping a draft report which was wholly or substantially prepared by the staff”. The letter also addressed the substance of the allegations of deficiencies in HLB’s audit. The letter concluded by urging the Council: (a) to dissolve the Investigation Committee and decline to accept its Reports; (b) to find that the appellants have no case to answer; and (c) to decline to refer the matter to Disciplinary Panels. 51. On 8 January 2010, the lawyers for the Institute wrote to the lawyers for the appellants saying the Council would next meet on 19 January 2010 and that if the appellants had any further submissions to make they should make them by 12 January 2010. The letter said that the Institute did not intend to give any undertaking that it would not refer the matter to Disciplinary Panels. 52. The appellants thereupon made an application for judicial review and obtained leave from Reyes J on 18 January 2010. The Council of the Institute has never made a decision as to the exercise of its discretionary power under s 42C of the PAO. The Proceedings for Judicial Review 53. The respondent to the application for judicial review is the Institute. The decisions in respect of which relief is sought are identified as follows: “(1) The Decision of the Respondent contained in its letter dated 12 December 2009 (‘the 1st Decision’) rejecting all the applicants’ complaints including inter alia the constitution of the Investigation Committee. (2) The Decision of the Respondent in its letter dated 8 January 2010 (‘the 2nd Decision’) refusing to consider re-constitution of a new Investigation Committee with different membership so that the statutory task of determining whether a prima facie case for disciplinary proceedings exists be conducted afresh by a new Investigation Committee.” 54. The application was amended in certain presently immaterial respects. 55. The substantive relief sought is: (1) An order of certiorari to bring up and quash the 1st and/or 2nd Decisions; and (2) An order of mandamus requiring the respondent to re-constitute an Investigation Committee with different membership so that the statutory task of determining whether a prima facie case exists be conducted afresh by the new Investigation Committee without reference to the earlier reports produced by the existing Investigation Committee. 56. The allegations of actual or apparent bias on the part of some members of the Investigation Committee were not pursued before Reyes J. The focal point of the appellants’ case became the allegation of lack of independence and impartiality on the part of the Investigation Committee said to have been manifested in an appearance of having “rubber stamped” the work of others. 57. The following comments may be made: (a) The decisions the subject of the application for judicial review are two decisions of the Council, each reflected in a letter from the Institute’s lawyers. (b) Neither is a decision under s 42C(1) of the PAO. The Council had not reached the point of considering whether to exercise its power to refer the matter to Disciplinary Panels when these proceedings intervened. In December 2009 and January 2010, the Council had received, from the Investigation Committee and from the appellants, a substantial body of information and commentary on HLB’s audit work in respect of Tiffit. The appellants have not set out to demonstrate that, on that material, it would not have been reasonably open to the Council to decide that a referral was justified. (c) It is difficult to relate either of the impugned decisions to any exercise of statutory power by the Council. The first impugned decision is said to be a decision rejecting the appellants’ complaints about the Investigation Committee. The Council does not have a statutory function of resolving complaints about an Investigation Committee, although in a given case such complaints could be relevant to a decision whether to proceed under s 42C(1). The second impugned decision is in substance a decision not to dissolve the Investigation Committee, re-constitute it and start the investigation afresh. It may be accepted that such a course would have been open to the Council, but it is another thing to say that it had a duty to follow that course before dealing with the question of whether to exercise its discretionary power under s42C(1). (d) It is also difficult to relate the challenges to the decisions to the recognized grounds of judicial review of administrative action. In responding as it did to the demands of the appellants to “disallow” the reports of the Investigation Committee and re-constitute the Committee, the Council was acting within the bounds of its statutory authority. It did nothing illegal. It did not itself act in any way that was unfair. It did not act on irrelevant considerations, or fail to take relevant considerations into account. Its decisions were not affected by fraud. They were not unreasonable. What the Council decided, in substance, was that at a future meeting it would consider whether to exercise its discretionary power to refer the matter to Disciplinary Panels. The decision of Reyes J 58. Reyes J summarized the arguments for the appellants as follows. (It will be noted that, as before the Court of Appeal, and as in this Court, they are substantially framed as arguments about the way in which the Investigation Committee acted, not as arguments going to acts or omissions of the maker of the impugned decisions, the Council). The Investigation Committee, it was argued, considered reports prepared by Ms Lee and the Compliance Department before reaching its own conclusion; it did not exercise its own independent judgment; it “rubber-stamped” the work of Ms Lee and the Compliance Department. In particular, the appellants pointed to the error about the applicable standard as evidence of an uncritical acceptance by the Investigation Committee of the work and opinions of Ms Lee and the Compliance Department. 59. Reyes J rejected to appellants’ arguments on the facts. He pointed to what he described as the practicalities of the operations of an Investigation Committee. It is a part-time body. A majority of its members are not accountants. It is almost inevitable that they will require assistance to evaluate voluminous audit working papers. He also pointed to its function. It is not a decision-making tribunal. There is no formal process it is required to follow. It does not conduct hearings. It forms an opinion and informs the Council of its opinion. The Council is not obliged to accept that opinion. He rejected the suggestion that there was anything untoward or inappropriate in the Investigation Committee’s reliance on extensive assistance from outside. As to the original application of the wrong standard, Reyes J said he did not think a fair-minded observer would conclude from that, or any other of the circumstances, that the Investigation Committee took any of the drafts at face value and blindly copied them without making an independent assessment of the contents. 60. Furthermore, Reyes J referred to the evidence of Mr Morrison and Mr Nip. Both of those witnesses gave detailed evidence of the chronology, of the meetings of the Investigation Committee, and of the communications between the Committee and others, including HLB. In introducing his account of the work done by the Committee, Mr Morrison said: “I attended all meetings of the Investigation Committee in connection with the investigation. Contrary to the assertions made by the Applicants, the reports which were issued by the Investigation Committee on 27th August 2009 and 16th November 2009 reflected the independent judgment of the Investigation Committee, reached after careful consideration of the evidence by the Committee.” He did not rest upon that assertion, but went on, over several paragraphs of evidence, to explain what the Committee did by way of considering reports, asking further questions, identifying issues for further enquiry, and discussing the merits of the matter. It may be added that one of the draft reports before the Committee bears handwritten notes and comments that appear inconsistent with an uncritical acceptance of the draft. When this Court asked whose handwriting it was, the Court was informed that it was that of the Chairman of the Committee. Reyes J said that the Institute’s evidence could not be ignored as self-serving. Weight should be given to it. 61. Reyes J, accepting that the Investigation Committee was required to be independent and impartial, said the real question was what was the “requisite degree” that must be shown. He said: “There is a whole spectrum of independence and impartiality. The more formal the nature of the decision-maker and the more significant the consequences of its decision, the more rigorous the degree of independence and impartiality that such entity must show.” He concluded that there was no basis for the criticisms made by the appellants. 62. In addition, although saying it was strictly unnecessary to his decision, Reyes J expressed the view that the judicial review proceedings were premature. The matter, he said, had a way to go before any binding decision was reached. The work of the Investigation Committee was part of a more extensive process that gave adequate protection to the rights of the appellants. The Court of Appeal 63. In the Court of Appeal, Mr Justice Tang VP, with whom Chu JA and Lam J agreed, examined, and substantially agreed with, the decision of Reyes J on the argument as to the appearance of independence and impartiality of the Investigation Committee. He did not find it necessary to go into the question of prematurity. 64. The Vice-President said that, in the Court of Appeal, the lynchpin of the appellants’ submission was the decision of the Court of Final Appeal in Medical Council of Hong Kong v Helen Chan (2010) 13 HKCFAR 248. That case concerned the procedures of the Medical Council of Hong Kong, which exercised disciplinary powers in respect of doctors. The Council was a deliberative body that decided issues of alleged professional misconduct. It used the services of a legal adviser, who was present at the Council’s deliberations, and produced a draft of the Council’s decision. Mr Justice Bokhary PJ, with whom the other members of the court agreed, found that the practice was not unlawful, noting that it was regulated by the Regulations under which the Council operated. However, he cautioned against participation in the deliberative process of a tribunal by a non-member. He went on to examine a number of authorities dealing with the circumstances in which inappropriate involvement of a non-member will compromise the independence and impartiality of a tribunal, and made certain recommendations for the assistance of bodies such as the Medical Council. 65. The Vice-President referred to the passage in the judgment of Reyes J set out at para 61 above, with which he agreed, and said that, on the evidence which he set out and discussed, it was clear that the Investigation Committee had not acted as a rubber stamp and that he “did not believe the conduct of the Investigation Committee could be described as unfair, lacking in independence or impartiality in any way”. The appeal to this Court 66. In this Court, the appellants advanced three criticisms of the reasoning of the Court of Appeal and the primary judge: (1) It was said that the passage in para 61 above from the judgment of Reyes J, expressly approved by the Court of Appeal, wrongly sanctioned and applied a variable or flexible standard of independence and impartiality, as though the Investigation Committee was not required to be as independent and impartial as, say, a tribunal that decides rights or liabilities. (2) That part of the evidence of Mr Morrison which asserted that, in fact, the Investigation Committee gave independent consideration to the work of Ms Lee and the Compliance Department was said to be irrelevant; the question was one of appearances, and reliance on the whole of Mr Morrison’s evidence revealed a misconception as to the true nature of the enquiry. (3) The Court of Appeal was said to have failed to apply Medical Council of Hong Kong v Helen Chan. 67. The PAO creates a number of procedures by which the conduct of an accountant may come before a Disciplinary Committee, which will then conduct a hearing, with the possibility of an appeal to the Court of Appeal. For present purposes, it is sufficient to consider two. One is initiated by a complaint to the Council under s 34. The Council has a discretion to refer the complaint to the Disciplinary Panels. The other follows a suspicion or belief by the Council itself of certain conduct of a kind that attracts sanctions. In that event, the Council may institute an investigative process through the mechanism of an Investigation Committee. The result of the information obtained by the investigation may or may not be a referral to Disciplinary Panels and the constitution of a Disciplinary Committee. The constitution and role of an Investigation Committee reflects a statutory purpose that it be, and appear to be, independent and impartial. At the same time, the nature of the Investigation Committee and of its work suggests that, at least in substantial cases, it will need assistance, and the assistance of people such as Ms Lee and the officers of the Compliance Department is clearly within legislative contemplation. 68. The concept of independence and impartiality does not vary in its meaning as between different kinds of administrative body, but its practical content varies according to circumstances which include the functions assigned to the body, and its place in a wider statutory scheme. Conduct that may compromise the appearance of independence and impartiality of one person or body required to form an opinion for the purposes of one statutory scheme may not have that effect in the case of another. In Clark v Kelly [2004] 1 AC 681 (PC), a case concerning the role of clerks to justices in Scotland, Lord Hoffmann said, at 691: “Although the position of the clerk attracts requirements of independence under article 6(1), it does not follow that those requirements are identical with those of every other person performing judicial duties. Regard must be had to the nature of the judicial process in which he is engaged, the part he plays in it, the part played by others (including an appellate court) and the institutional and practical safeguards which exist to preserve and to demonstrate the preservation of his independence.” 69. To take a simple example, even when not sequestered, jurors are routinely admonished not to discuss the case with anyone else. There are a number of reasons for this; one is to ensure that their independence and impartiality will not be compromised. No one would suggest, and the appellants did not suggest, that the same requirement of isolation applied to the Investigation Committee. Such a requirement was neither necessary nor appropriate for the work of the Committee. 70. The function of investigation, even where it is required by statute to be performed independently and impartially, almost always involves contact between the investigator and third parties. The circumstances of the case will determine what is a fair and efficient procedure. In this respect, the PAO, understandably, is not prescriptive. The investigative process will be tailored to the individual case. One investigation may call for extensive outside input and assistance. Another may call for practically none. The work of the Investigation Committee in this case lasted more than two years and involved extensive paperwork. Another investigation might be concluded in weeks or even days. In this case, the Investigation Committee made a formal and lengthy report to the Council. That was appropriate in the particular circumstances, but it was not mandated by the PAO. In this case, the Investigation Committee required, or at least received, extensive administrative support. In other cases, that might not be necessary. In the end, what was essential was that the Investigation Committee made up its own mind, and formed its own opinion, on the question it was directed by the Council to consider, and then informed the Council of its opinion. 71. No attempt was made to demonstrate, by reference to the substance of the report of 16 November 2009, that a conclusion that there was a case for the appellants to answer was not reasonably open to the Investigation Committee, or that the material contained in the report could not justify a decision by the Council to refer the matter to the Disciplinary Panels. 72. The complaint that the primary judge and the Court of Appeal wrongly accepted and applied a variable standard of independence and impartiality is puzzling in view of the nature of the factual case that the appellants set out to establish. The central allegation made by the appellants was that the Investigation rubber-stamped, or appeared to have rubber-stamped, the work of outsiders. That allegation was rejected, and correctly rejected, on the evidence. Suppose, however, it had been accepted. What scope would there have been for the application of some supposedly variable standard? 73. The primary judge and the Court of Appeal were addressing an issue of fact. In the circumstances of the present case, did the conduct of the Investigation Committee and those upon whom it relied for assistance create a reasonable apprehension that the Investigation Committee was not independent and impartial and, in particular, that it did not “come to its own decision and make its own report”? In answering that question of fact in the negative, the primary judge and the Court of Appeal rightly recognized that the kind of conduct that may warrant such an apprehension in the case of one kind of statutory body or tribunal may not do so in the case of another. That is all that was meant by the passage quoted in para 61 above. 74. The reliance upon the evidence of Mr Morrison was said to reflect a misunderstanding of the appellants’ case; that case was based on appearances, and Mr Morrison’s statement that he made up his own mind was said to be irrelevant. 75. The reasons for judgment reflect no such misunderstanding. Mr Morrison’s statement in his evidence that he made up his own mind was prefatory to an account of the Investigation Committee’s work and procedure. On any view, that account was relevant and admissible. Where a court, undertaking judicial review, has to deal with an allegation that a decision-making process was such as to create in the mind of a reasonable person an appearance of bias or lack of independence and impartiality, then evidence which gives a full account of the objective facts and circumstances of the process will often be necessary. How otherwise could the likelihood and reasonableness of the supposed apprehension be determined? In Australia, in the application of the reasonable apprehension test in cases of alleged lack of impartiality, it is clearly established that the question is to be asked in terms of a reasonable apprehension on the part of a fair-minded observer with knowledge of all the material facts as they are ultimately found to have been, not merely of such facts as might have appeared at the time to one or other of the parties (Webb v The Queen (1994) 181 CLR 41 per Deane J at 73. See also Johnson v Johnson (2000) 201 CLR 488). 76. In pursuing an allegation that what occurred in the decision-making process of the Investigation Committee created an appearance that the Committee merely rubber-stamped the work of others, the appellants were not entitled to be selective when it came to evidence of what occurred. The court was entitled, and obliged, to consider all the objective facts concerning that process, regardless of whether they were known to the appellants at the time the allegation was first made. The Investigation Committee did not carry out its work in public, like a court. In fact, it was bound by obligations of secrecy (PAO, s 42G). Inevitably, the appellants would be aware of some aspects of the way in which it worked, but not others. In some respects, the extent of their knowledge might have been a matter of chance. However, the nature of the challenge to the work of the Investigation Committee made by the appellants required examination of all the objective facts referred to in the evidence of Mr Morrison and Mr Nip. 77. In relation to the matter of the error as to the relevant professional standards and its later correction, the lawyers for the appellants, in their letter to the Council of 5 January 2010, and again in their application for judicial review, referred to a statement that the Committee “accepted that the professional standards referred to … in [the] report of 27th August were incorrect and amended its report”. The lawyers then said: “This admission is extremely disturbing because what underlies it is an appearance that the Investigation Committee was simply rubber stamping a draft report which was wholly or substantially prepared by the staff of the secretariat or the Institute without even cross checking the most basic ingredients of the complaints which the Investigation Committee was supposed to be investigating, namely, the relevant ‘professional standards’ which were applicable to the subject matter of the investigation. It is hard to believe that the members as experienced as those sitting on the Investigation Committee would all have taken the care and have exercised their independent assessment and judgment and yet had all failed to detect the basic errors which are now admitted. …” 78. In the allegation framed in that manner, there is an overlap between an accusation about what in fact occurred, and one about what appeared to have occurred. It was this passage that prompted Mr Morrison’s statement in his affidavit that he had been informed that the appellants alleged that there had been a rubber-stamping, and his denial of that allegation, made, as was said earlier, by way of introduction to an account of the objective facts. Having regard to the way the allegation was put, evidence from Mr Morrison that the Committee in fact formed its own judgment was admissible. It is true that it was, in a sense, self-serving, but that is not a ground of objection to evidence. Many witnesses give evidence that is self-serving. That goes to its weight, not its admissibility. In a case where a ground of complaint is apprehended bias, it is not relevant for a decision-maker to say: “I did not feel biased”. “No attempt need be made to inquire into the actual thought processes of the judge” (Ebner v Official Trustee in Bankruptcy (2001) 205 CLR 337 at 345). In this case, however, the allegation was made in terms that invited the response that was given, and the response was relevant. 79. The complaint of failure to pay due regard to the decision in Medical Council of Hong Kong v Helen Chan has not been made out. The Court of Appeal described it as the “lynchpin” of the appellants’ argument. That, however, was a case about the procedures of a body that made decisions dealing finally with the merits of charges of professional misconduct. In Clark v Kelly [2004] 1 AC 681 at 692, Lord Hoffmann said: “[T]here are tribunals in which the person whose independence is called into question is not in name or substance a presiding judge in the paradigm sense and in which it may be necessary to look at the decision-making process as a whole, including the right of appeal.” 80. The kind of outside involvement that might be sufficient to call into question the independence and impartiality of a judge, or juror, or disciplinary tribunal, may be quite different from that which would have the same effect in relation to a body whose role is purely investigatory. Conclusion 81. The challenges to the decision of the Court of Appeal have not been made out. 82. Furthermore, the reservations expressed by Reyes J as to what he called the premature nature of the application for judicial review were justified. Looked at from a slightly different point of view, the problem for the appellants was that they were forced into an artificial characterization of the decisions the subject of their application, and of the grounds upon which they were seeking judicial review. They made an energetic, but unsuccessful, attack on the process of the Investigation Committee, but sought judicial review of “decisions” of the Council which amounted in substance to no more than an expression of an intention to exercise its statutory functions under s 42C(1) of the PAO. The statutory condition for that exercise (the receipt of information that, in the opinion of the Investigation Committee, there was a prima facie case against the appellants) was satisfied. There was no statutory obligation on the Council to disregard that information and constitute another Investigation Committee. What, if anything, the Council would have made of the criticisms of the Investigation Committee when it came to make a decision under s 42C(1) is unknown. The matter never reached that stage. 83. The appeal fails. Mr Gerard McCoy SC and Mr Johnny Ma, instructed by Smyth & Co, for the appellants Mr Paul Shieh SC and Mr Bernard Man, instructed by Reed Smith Richards Butler, for the respondent [1] By “illegality” Lord Diplock meant that the decision-maker had misunderstood the law which regulated its power: see p 410-F; sometimes referred to as ultra vires or want of jurisdiction. [2] By “procedural impropriety” Lord Diplock meant that the decision-maker had failed to observe procedural rules laid down in the statute which conferred jurisdiction on the decision-maker. [3] “Irrationality”: A decision so outrageous that no sensible person who had applied his mind to the right question could been made: “Wednesbury unreasonableness.” 
Chief Justice Ma: 1. At the hearing of these appeals, after submissions on the first question of law for which leave to appeal was granted, the Court made the following orders: (1) The appeals are allowed to the extent that the appellants had the right to appeal to the Court of Appeal without first obtaining leave under s.14AA(1) of the High Court Ordinance, Cap.4, against the orders of Chow J dated 18 November 2015 (in HCAL 32 of 2015) and 3 November 2016 (in HCAL 172 of 2015) refusing to grant leave to apply for judicial review. The reasons for judgment will be handed down on a date to be notified. (2) The other questions for which leave to appeal to the Court of Final Appeal as sought in FAMV 415 of 2019 and FAMV 3 of 2020 are to be dealt with on a date to be fixed. 2. The reasons for allowing the appeals are set out in the judgment of Mr Justice Fok PJ. I agree with those reasons and the direction regarding costs at [56] below. Mr Justice Ribeiro PJ: 3. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: A. Introduction 4. The primary issue in these conjoined appeals is one of jurisdiction of the Court of Appeal to entertain an appeal from the Court of First Instance and, in particular, whether a decision to refuse an extension of time to apply for leave to apply for judicial review is a decision to which section 14AA(1) of the High Court Ordinance (Cap.4) (“HCO”) (set out at [10] below) applies. In this judgment, I shall refer to this issue as “the Leave Issue”. 5. If leave is required, a subsequent question arises as to whether the finality provision in section 14AB of the HCO (set out at [12] below), insofar as it relates to judicial review proceedings, is inconsistent with the Court of Final Appeal’s power of final adjudication in Article 82 of the Basic Law.[1] This issue will be referred to as “the Constitutionality Issue”. 6. The circumstances out of which these appeals arise are as follows. In FACV 1/2020: (1) The appellant in FACV 1/2020 (“the appellant H”) applied to the respondent, the Director of Immigration, for a dependant visa. The respondent made a decision to refuse the application on 8 October 2013 and confirmed the decision on 11 August 2014. On 21 August 2015, the appellant H filed a Notice of Application for Leave to Apply for Judicial Review[2] (in HCAL 172/2015) seeking to challenge the respondent’s decision and its confirmation. Being considerably more than three months after the date of the relevant decision sought to be judicially reviewed, in his Form 86 notice, the appellant H sought an extension of time in which to bring the application. (2) Chow J directed that there be what he termed “a rolled up hearing” of (i) the application for an extension of time to commence judicial review proceedings, (ii) the application for leave to apply for judicial review, and (iii) in the event that leave was granted, the substantive judicial review application.[3] (3) By his judgment dated 3 November 2016, Chow J refused to grant the extension of time sought and dismissed the application for leave to apply for judicial review.[4] (4) The appellant H filed a Notice of Appeal against Chow J’s judgment (in CACV 5/2017). However, applying the judgment of the Court of Appeal in MI & IYW v Permanent Secretary for Security[5] (which I shall address further below), the appellant H was directed by the Registrar of Civil Appeals that, if he was minded to pursue the appeal, he should apply for leave to appeal pursuant to section 14AA of the HCO and seek an extension of time for that purpose. The appellant H duly did so by summons dated 22 August 2018. (5) On 29 October 2018, the Court of Appeal refused to grant leave to appeal to it from Chow J’s judgment and struck out the appeal.[6] (6) The appellant H then applied to the Court of Appeal for leave to appeal to the Court of Final Appeal. By its judgment dated 22 November 2019, the Court of Appeal followed its decision in MI and held that, since it had refused leave to appeal to it pursuant to section 14AA of the HCO, section 14AB of the HCO precluded any further appeal to the Court of Final Appeal and dismissed the application for leave to appeal to this Court.[7] (7) The appellant H then applied to this Court for leave to appeal against the Court of Appeal’s judgment striking out his appeal against the judgment of Chow J.[8] 7. In FACV 2/2020: (1) The appellant in FACV 2/2020 (“the appellant AH”) also applied to the respondent for a dependant visa. The respondent made a decision to refuse that application on 14 August 2012 and confirmed the decision on 1 August 2013. On 3 March 2015, the appellant AH filed a Notice of Application for Leave to Apply for Judicial Review seeking to challenge the respondent’s decision and its confirmation (in HCAL 32/2015). Being considerably more than three months after the date of the relevant decision sought to be judicially reviewed, in his Form 86 notice, the appellant AH sought an extension of time in which to bring the application. (2) Chow J heard (i) an application by the appellant AH for an extension of time to commence judicial review proceedings described as “the Extension Application”, and (ii) an application for leave to apply for judicial review “in the event that the Extension Application is granted”.[9] (3) By his judgment dated 18 November 2015, Chow J dismissed the application for the extension of time sought and also dismissed the application for leave to apply for judicial review.[10] (4) The appellant AH applied for an extension of time to seek leave to appeal to the Court of Appeal against Chow J’s judgment. On 23 August 2019, the Court of Appeal dismissed the application,[11] holding: “For these reasons, we do not see any reasonable prospect of success in the applicant’s arguments based on construction of the [No Record Criterion of the Dependant Visa Policy]. Even taking account of these new arguments, we cannot see any ground for disturbing Chow J’s refusal to grant extension of time for an application for judicial review to be brought.”[12] (5) On 1 November 2019, the appellant AH applied by Notice of Motion to the Court of Appeal for leave to appeal to this Court from the CA Judgment in AH (i.e. its judgment dated 23 August 2019). By letter dated 5 December 2019, the Court of Appeal[13] dismissed the Notice of Motion, directing (inter alia) that: “For the reasons set out in H v Director of Immigration [2019] HKCA 1302, the decision of this Court on 23 August 2019 is final and the Notice of Motion of 1 November 2019 is incompetent.” The case referred to (H v Director of Immigration) is the “CA Leave Judgment in H” (see FN7 above) and so this direction was made applying that judgment, namely on the basis that the Court of Appeal had refused leave to appeal to it pursuant to s.14AA of the HCO so that s.14AB of the HCO precluded any further appeal to the Court of Final Appeal. (6) The appellant AH then applied to this Court for leave to appeal against the Court of Appeal’s judgment dated 23 August 2019.[14] 8. The Appeal Committee granted the appellants leave to appeal to this Court on the Leave Issue and the Constitutionality Issue in the following terms: “1. On the basis that this determination is made without prejudice to the consideration by the Court of Final Appeal of the certified questions, we give leave to appeal on the following questions:- (1) Where the Court of First Instance does not grant an extension of time to a party to apply for leave to apply for judicial review, particularly when the hearing of the application to extend takes place at the same time as the hearing for leave to apply for judicial review, is leave to appeal to the Court of Appeal required under s 14AA(1) of the High Court Ordinance Cap 4, or is leave not required by reason of s 14AA(2) or otherwise? (2) If leave to appeal is required, is s 14AB of the High Court Ordinance, insofar as it relates to judicial review proceedings, inconsistent with Article 82 of the Basic Law? 2. Consideration by the Appeal Committee of the other questions sought to be certified by the applicants is to be adjourned until after the determination by the Court of Final Appeal of the two certified questions.”[15] B. The requirement of leave to appeal 9. The default position in relation to civil matters in this jurisdiction is that there is a right of appeal to the Court of Appeal from the Court of First Instance. The HCO provides, by s.14(1): “Subject to subsection (3) and section 14AA, an appeal shall lie as of right to the Court of Appeal from every judgment or order of the Court of First Instance in any civil cause or matter.” 10. That right is qualified by s.14(3) and also s.14AA. The qualifications under s.14(3) concern specific proceedings and orders with which these appeals are not concerned. Under s.14AA, leave to appeal is required for interlocutory appeals – see s.14AA(1), which provides: “Except as provided by rules of court, no appeal lies to the Court of Appeal from an interlocutory judgment or order of the Court of First Instance in any civil cause or matter unless leave to appeal has been granted by the Court of First Instance or the Court of Appeal.” 11. The rationale for the leave requirement is the avoidance of the major delay and expense likely incurred as a result of satellite litigation on interlocutory matters often of only marginal significance to the outcome of the litigation: see Civil Justice Reform – Final Report, at [643]. The requirement of leave to appeal in interlocutory matters therefore relieves the Court of Appeal from having to hear interlocutory appeals unless meritorious or otherwise deserving of being heard. To this end, s.14AA(4) provides: “Leave to appeal for the purpose of subsection (1) shall not be granted unless the court hearing the application for leave is satisfied that – (a) the appeal has a reasonable prospect of success; or (b) there is some other reason in the interests of justice why the appeal should be heard.” 12. In order to ensure that there is finality in a decision as to whether or not leave to appeal should be granted to the Court of Appeal, s.14AB provides: “No appeal lies from a decision of the Court of Appeal as to whether or not leave to appeal to it should be granted.” 13. The rules of court which govern civil appeals to the Court of Appeal are contained in Order 59 of the Rules of the High Court (Cap.4A) (“RHC”). (1) O.59 r.2B regulates applications for leave to appeal in interlocutory matters. So far as practicable, such applications are to be made to the judge who made the order sought to be appealed from. If the first instance judge refuses leave, an application for leave to appeal may be made to the Court of Appeal. (2) O.59 r.2A and r.2C regulate applications to the Court of Appeal for leave to appeal. 14. Recognising that the requirement of leave to appeal to the Court of Appeal in interlocutory matters constitutes a restriction on the right of appeal, O.59 r.21 (derived from s.14AA(2)[16]) qualifies the restriction by providing that a party does not require leave in order to appeal to the Court of Appeal against certain types of order, although interlocutory rather than final. O.59 r.21(1) thus prefaces a list of such orders with the introductory words: “Judgments and orders to which section 14AA(1) of the Ordinance (leave to appeal required for interlocutory appeals) does not apply and accordingly an appeal lies as of right from them are the following –…”. 15. Amongst those exceptions are: “(a) a judgment or order determining in a summary way the substantive rights of a party to an action; … (g) an order under Order 53, rule 3 refusing to grant leave to apply for judicial review;”. 16. So far as O.59 r.21(1)(a) is concerned, O.59 r.21(2) sets out, but expressly without affecting the generality of O.59 r.21(1)(a), a list of judgments and orders determining in a summary way the substantive rights of a party. C. The process for commencing judicial review proceedings 17. The procedure for judicial review applications is governed by provisions of the HCO and the RHC. In the HCO, s.21K(1) provides that: “An application to the Court of First Instance for one or more of the following forms of relief – (a) an order of mandamus, prohibition or certiorari; (b) an injunction under section 21J restraining a person not entitled to do so from acting in an office to which that section applies, shall be made in accordance with rules of court by a procedure to be known as an application for judicial review.” The rules of court referred to in s.21K(1) are those contained in RHC O.53. 18. A threshold requirement that leave be obtained for an application for judicial review is contained in s.21K(3), which provides: “No application for judicial review shall be made unless the leave of the Court of First Instance has been obtained in accordance with rules of court; and the court shall not grant leave to make such an application unless it considers that the applicant has a sufficient interest in the matter to which the application relates.” The primary statutory leave requirement is reflected in the RHC, O.53 r.3(1) providing: “No application for judicial review shall be made unless the leave of the Court has been obtained in accordance with this rule.” 19. The procedure for obtaining leave to apply for judicial review is set out in other provisions of O.53 r.3: (1) The application must be made ex parte in the prescribed form (Form 86 in Appendix A), accompanied by an affidavit verifying the facts relied on: O.53 r.3(2); (2) The leave application can be determined without a hearing unless an oral hearing is requested in the notice of application: O.53 r.3(3); (3) Where leave is refused or granted on terms, the applicant may appeal against the judge’s order within 14 days: O.53 r.3(4);[17] (4) The court shall not grant leave unless it considers the applicant has a sufficient interest in the matter: O.53 r.3(7);[18] and (5) The requirement for leave to apply for judicial review has been described as “an important filter” and leave will only be granted if the court is persuaded that the arguability test is satisfied by the applicant demonstrating that the case is one which enjoys realistic prospects of success: Po Fun Chan v Winnie Cheung.[19] By this filter, the legislative framework seeks to strike a balance between access to the courts for individuals who may have been affected by public law decisions and the public interest that authorities have certainty over the validity of their decisions. 20. Although no express time limit for commencing judicial review proceedings is stipulated in s.21K, O.53 r.4 (headed “Delay in applying for relief”) provides: “(1) An application for leave to apply for judicial review shall be made promptly and in any event within three months from the date when grounds for the application first arose unless the Court considers that there is good reason for extending the period within which the application shall be made. (2) Where the relief sought is an order of certiorari in respect of any judgment, order, conviction or other proceeding, the date when grounds for the application first arose shall be taken to be the date of that judgment, order, conviction or proceeding. (3) The preceding paragraphs are without prejudice to any statutory provision which has the effect of limiting the time within which an application for judicial review may be made.”[20] 21. The terms of O.53 r.4(1) are mandatory (“shall be made”) and the obligation on an applicant for judicial review is to apply “promptly and in any event within three months”. The key time obligation is to act “promptly”. So, depending on the context of a particular administrative decision sought to be challenged, it is possible that an applicant who commences proceedings even within three months may be guilty of undue delay for not having acted promptly.[21] But, in any event, there is, subject to extension for “good reason”, a quantified default time limit of three months from the relevant date from which time starts to run. 22. Undue delay in applying is also referred to in s.21K(6), which sets out the consequences of such delay, providing: “(6) Where the Court of First Instance considers that there has been undue delay in making an application for judicial review, the Court may refuse to grant – (a) leave for the making of the application; or (b) any relief sought on the application, if it considers that the granting of the relief sought would be likely to cause substantial hardship to, or substantially prejudice the rights of, any person or would be detrimental to good administration. (7) Subsection (6) is without prejudice to any enactment or rule of court which has the effect of limiting the time within which an application for judicial review may be made.” The reference to “application for judicial review” in both s.21K(6) and (7) is to be read as also referring, where appropriate, to an application for leave to apply for judicial review: see Reg. v Dairy Tribunal, Ex parte Caswell.[22] 23. The HCO and RHC provisions governing public law proceedings in Hong Kong, including the leave requirement, are derived from the provisions introduced in 1977 in England and Wales by the Supreme Court Act 1981 and O.53 of the Rules of the Supreme Court (“RSC”).[23] In that jurisdiction, prior to 1933, an applicant for judicial review by means of the prerogative writs (of certiorari, prohibition or mandamus) would apply for an order (or rule) nisi and it would then be for the respondent to show cause why the order should not be made absolute.[24] In 1933, that two-stage nisi and absolute procedure was replaced by a requirement that an applicant first obtain the leave of the High Court[25] and that leave requirement remained part of the unified procedure of application for judicial review that was introduced in 1977 upon the recommendation of the Law Commission Report (see FN23). There were time limits for applications for writs of certiorari and mandamus, and latterly for applications for judicial review, under rules which were broadly similar to those found in the SCA 1981 and RSC (in England and Wales) and in HCO s.21K and RHC O.59 (in Hong Kong). 24. I will return later in this judgment to address what happens when there has been undue delay in the making of an application for leave to apply for judicial review and the court considers either that there is, or there is not, good reason for extending the period. D. The Leave Issue D.1 The appellants’ contentions 25. The underlying basis of the Court of Appeal’s decisions in both appeals before us is that the appellants each required leave to appeal against Chow J’s respective judgments. In so deciding, the Court of Appeal followed (as it was bound to do) its earlier decision in MI (as to which see Section D.2 below). 26. The appellants contend that MI was wrongly decided and that, contrary to its conclusions in that case, the Court of Appeal should instead have concluded that the decision of a first instance judge to refuse an extension of time in which to apply for leave to apply for judicial review: (1) determined in a summary way the substantive rights of a party to the action within the meaning of O.59 r.21(1)(a); (2) was a refusal of leave to apply for judicial review within the meaning of O.59 r.21(1)(g); and/or (3) was final, not interlocutory, for the purposes of s.14AA of the HCO. On any of the above three bases, the appellants contend, leave to appeal to the Court of Appeal was not required and their respective appeals, being brought as of right, fell to be determined by that court accordingly. It is convenient to deal first with the argument raised in sub-paragraph (2) above. D2. The Court of Appeal decision in MI 27. The Court of Appeal decision in MI also concerned two separate applications for leave to apply for judicial review, which were made out of time. The first instance judge refused to grant an extension of time and also dismissed each of the applications for leave to apply for judicial review. The applicants both appealed to the Court of Appeal and a common preliminary issue was whether, when a judge refuses to grant an extension of time for an application for judicial review to be brought, an appeal lies as of right or can only be brought with leave obtained under O.59 r.2B. That issue, of course, is the same issue that is before this Court in the present appeals. 28. In MI, the Court of Appeal held[26] that: (1) The decision on an application for extension of time to apply for leave to apply for judicial review is “a discrete application” from the application for leave under O.53 r.3; (2) For an application made within time and without delay, the application for leave falls to be considered by reference to the applicant’s interest in the matter and the reasonable arguability test; (3) For an application made within the three-month time limit but not made promptly, an extension of time is not required but the court can refuse to grant leave on discretionary considerations based on hardship, prejudice or detriment to good administration;[27] and (4) For an application made outside the three-month time limit, an applicant must obtain an extension of time under O.53 r.4(1) before the application for leave can be entertained. 29. On this basis, the Court of Appeal in MI went on to hold that s.14AA of the HCO applies to a refusal to extend time to apply for leave to apply for judicial review.[28] It held that none of the exceptions in O.59 r.21 apply to a decision on an extension of time and that the decision should therefore be regarded as interlocutory in nature.[29] It also held that the refusal of an extension of time under O.53 r.4(1) did not determine the substantive right of an applicant, notwithstanding that it had the collateral effect of preventing an applicant from proceeding further with an application for judicial review.[30] 30. As noted at [28(1)] above, in MI, the Court of Appeal held that a decision on an application for an extension of time is a discrete application, separate from the application for leave to apply for judicial review. That holding was based[31] on the Court of Appeal’s construction of O.53 r.4(1) and its understanding of Lord Goff’s speech[32] in Reg. v Dairy Tribunal, Ex parte Caswell and Lord Slynn’s speech[33] in Reg. v Criminal Injuries Compensation Board, Ex parte A.[34] 31. The crux of the Court of Appeal’s reasoning in MI is set out at [28] of its judgment: “From these authorities, it is indisputable that the Court cannot proceed with a consideration of a leave application made outside the three-month period without first addressing if an extension of time should be granted for good reason. In other words, the application for extension of time is not simply an aspect in the overall multi-faceted assessment of a leave application. It is a prior step though as a matter of convenience the Court will entertain such application together with the application for leave when they are made at the same time. In substance, these remain to be two discrete steps. The Court cannot simply give leave (say on the basis that there would not be hardship, prejudice or detriment under Section 21K) without asking if there is any good reason to extend time under Rule 4(1). Thus, in Reg v Criminal Injuries Compensation Board, Ex p A, supra, at p.340E, Lord Slynn held that when leave was granted, the judge should be taken as having granted an extension of time as well because ‘otherwise he would have had to rule that the application was out of time and to have refused leave’. It follows, as His Lordship held at p.341E, that the question as to if there is good reason for extension of time cannot be re-opened at the substantive application.” 32. There is no material distinction between the relevant procedural circumstances of the intended judicial review applications that led to the Court of Appeal’s decision in MI and those giving rise to the present appeals. The respondent invited the Court to affirm MI and to construe the operation of s.14AA and O.59 r.21(1) in the context of judicial review in the same way the Court of Appeal did there. To decline to do so, it was submitted, would undermine the purposes underlying the judicial review regime of minimising uncertainty in administrative decisions and ensuring the fair distribution of finite judicial resources.[35] The question of whether MI was correctly decided is therefore squarely raised in these appeals. 33. The respondent’s position is that MI was correctly decided by the Court of Appeal contending that, as a matter of statutory interpretation, the provisions of O.53 rr.3 and 4 and s.21K are to be read purposively as imposing a distinction between (i) applications for leave to apply for judicial review which are brought promptly and in any event within three months, for which there is a right of appeal against a refusal of leave by reason of the operation of O.59 r.21(1)(g), and (ii) applications for leave to apply for judicial review which are brought outside that time limit and thus involving undue delay, for which there is a leave to appeal filter under s.14AA of the HCO where an extension of time has not been granted. D.3 Is a request for an extension of time “a discrete application”? 34. With great respect to the Court of Appeal, its decision in MI that an application for an extension of time in which to make an application for leave to apply for judicial review is discrete and separate from the decision to grant or refuse leave to apply is wrong, and involves a misunderstanding of the House of Lords’ decisions in Ex parte Caswell and Ex parte A, for the following reasons. 35. An application for judicial review can only be made when leave to apply has been granted: O.53 r.3(1) and s.21K(3) of the HCO. The application must be made “promptly” and in any event “within three months from the date when grounds for the application first arose”: O.53 r.4(1). But the court retains a discretion to extend the period within which the application may be made where it considers there is “good reason” to do so: ibid. This discretion is consistent with the principle that, “[w]hen the repository of a power fails to comply with the conditions which govern its exercise, it is the function and duty of the court to quash the purported exercise of the power unless there are substantial grounds warranting the refusal of relief, but the grounds on which relief might be refused when the court finds an excess of power are ‘very narrow’”.[36] 36. The consequences of dilatoriness in the commencement of judicial review proceedings is not addressed in O.53 r.4 but is partly addressed in s.21K(6) and (7) of the HCO (see [22] above). The coexistence of the two provisions (O.53 r.4 and s.21K) has been described as “curious”[37] and that of O.53 r.4(3) and s.21K(7) even to have produced a “circulus inextricabilis”.[38] However, as reflected in a previous judgment of this Court,[39] the relationship between these two provisions has been fully considered by the English Court of Appeal in R v Stratford-on-Avon District Council, Ex parte Jackson[40]and by the House of Lords in Ex parte Caswell. 37. Properly construed, O.53 r.4(1) and s.21K(6) and (7) can be reconciled[41] and together provide as follows: (1) Whenever there is a failure to act either “promptly”, or within three months, there is “undue delay” for the purposes of s.21K(6). The time limit for judicial review is not expressed as a single fixed period of three months. Depending on the context, a failure to challenge a particular administrative decision in a shorter period than three months might constitute undue delay for not having been made “promptly”.[42] But, in any event, an application made outside the three-month period will entail undue delay.[43] (2) An applicant commencing judicial review proceedings who has been guilty of undue delay will require an extension of time in which to apply for leave to apply for judicial review. The request for such an extension should be (and, as a matter of practice, is normally) made in the applicant’s Form 86 notice and the issue of whether to grant the request for an extension of time will always, as a matter of practice, be determined together with the leave application.[44] To justify an extension, the applicant must show “good reason” for extending the period. The principles relevant to a determination of whether good reason has been shown are discussed in the judgment of Godfrey Lam J in Re Thomas Lai[45] and the Court of Appeal in AW v Director of Immigration & Anor.[46] Among the various factors likely to be significant are the merits of the substantive application. (3) If the court considers that there is good reason for extending the period, it does not follow that leave to apply will automatically be given. That argument was rejected in R v Stratford-on-Avon District Council, Ex parte Jackson.[47] Instead, the court retains a discretion to refuse to grant leave for the making of the application for judicial review: Ex Parte Jackson at p.1325F-G; Ex Parte Caswell at p.747B-C. The existence of good reason for any undue delay is, therefore, only one facet of the question whether leave to apply should be granted and the grant of leave will also depend on the applicant having a sufficient interest in the matter (see [19(4)] above) and showing that the application satisfies the arguability test (see [19(5)] above). Leave may also be refused on the basis that the granting of the relief sought would be likely to cause hardship or prejudice (as specified in s.21K(6)) or be detrimental to good administration, although this conclusion might be less likely to be reached at the leave stage as the court may not have the necessary material available.[48] (4) Where the court considers that there is no good reason for extending the period, it will refuse leave to apply for judicial review. That this is the consequence of a refusal to extend the period for making the application is clear from the speech of Lord Goff in Ex parte Caswell at p.747B-C, where he said: “It follows that, when an application for leave to apply is not made promptly and in any event within three months, the court may refuse leave on the ground of delay unless it considers that there is good reason for extending the period; …” (Emphasis added). This conclusion is reinforced by reference to Lord Slynn’s speech in Ex parte A at p.340E where he said: “It seems to me, however, that [the judge’s] intention in giving leave must have been to extend the period, otherwise he would have had to rule that the application was out of time and to have refused leave, …” (Emphasis added). It is not the case, contrary to MI, that the court refuses the extension of time as a discrete decision and then makes no decision on the question of leave to apply for judicial review, treating this as a separate matter entirely. If that were so, neither Lord Goff nor Lord Slynn would have expressed themselves in the terms they did in the passages quoted. The respondent’s contention that Lord Goff was “adopting a shorthand”[49] in explaining the consequence of a refusal to extend time is unrealistic, as is the submission that Lord Slynn’s dictum is to be understood in context as treating a decision to extend time for applying as discrete from the grant of leave to apply.[50] Judicial review textbooks also refer consistently to a court refusing leave (or permission) to apply for judicial review when not satisfied that there is good reason for undue delay in making the application.[51] (5) Even if leave to apply is granted, the issue of undue delay on the part of the applicant may be raised again at the substantive hearing of the judicial review application.[52] At that stage, the court may have further evidence, in particular from the respondent, as to the consequences of the grant of any particular relief sought. However, this is not a re-opening of the decision whether to grant leave, since leave has already been given.[53] It is instead a separate consideration of whether, on the substantive hearing, relief should be refused on the basis that the court thinks that relief would be likely to cause substantial hardship or prejudice to the rights of another or be detrimental to good administration (per s.21K(6)). 38. This construction of O.53 r.4(1) and s.21K(6) and (7) is supported by the decision of the Privy Council in Maharaj v National Energy Corporation of Trinidad and Tobago,[54] to which the appellants referred. The appeal concerned the interpretation of the Judicial Review Act 2000 of Trinidad and Tobago but in the course of his judgment for the Board, Lord Lloyd-Jones JSC made the following points of relevance to these appeals and consistent with the analysis set out in the preceding paragraph: (1) The provisions of RSC O.53 and SCA 1981 s.31(6) governing delay should not be applied in a technical manner.[55] (2) The classic exposition of the approach to delay in applications for judicial review in England and Wales remains that in the speech of Lord Goff in Ex parte Caswell and this supports the proposition that, even if there is good reason for extending time, the court may still refuse leave on grounds of prejudice or detriment.[56] (3) Ex parte Caswell provides no justification for insulating the threshold issues of delay and extension of time, on the one hand, from those of prejudice and detriment to good administration, on the other. On the contrary, when issues such as prejudice and detriment to good administration are raised by a respondent in the context of delay, the court may adjourn the question of leave to an inter partes hearing or order a “rolled up” hearing (as to which, see Section F below) at which leave is considered, followed immediately by the substantive application, if leave is granted. In either event, full consideration is given to the issues of extension of time, prejudice and detriment on the basis of the evidence then filed by the parties. Even if leave is granted without full consideration of the issues of prejudice and detriment resulting from delay, these may still be relied upon as a bar to relief at the substantive hearing.[57] 39. Contrary to the Court of Appeal’s conclusion in MI, therefore, whilst a judge cannot proceed with an out of time application for leave to apply for judicial review without considering whether there is good reason to extend time for the making of the application, his decision on that particular issue is not a discrete decision separate to his decision whether or not to grant leave to apply for judicial review. If he thinks there is no good reason for the undue delay, he will refuse leave to apply for judicial review. If he thinks there is good reason for the undue delay, he may grant leave or he may refuse leave to apply for judicial review. The combined effect of the relevant statutory provisions and the authorities which analyse them is that the aspect of delay is intrinsically bound up with the question whether leave to apply for judicial review should be granted. Consistent with this, historically, delay was always regarded as but one factor in the court’s exercise of discretion whether to grant or discharge an order or rule nisi (before 1933) [58] or to grant or refuse leave to apply for one of the prerogative writs or orders (from 1933 to 1977) [59] and any extension of time was not treated as an independent decision to be taken separately. Furthermore, since the merits of the substantive application is one of the factors relevant to the question of whether there is good reason for extending time to apply (see [37(2)] above), it would be highly artificial to treat a decision on an extension of time application as separate and discrete from the decision to grant or refuse leave to apply. 40. It is noteworthy that the respondent has not been able to identify a single decision in which a judge has refused a request for an extension of time to make an application for judicial review without also refusing leave to make the application. The case of R on the application of Melton and others v The School Organisation Committee and Oxford County Council[60] to which the respondent referred is not such a case. Whilst Jackson J (as he then was) said the court came to two conclusions, first, not to extend time by one day and, secondly, that the claimants would have been in breach of the obligation to proceed promptly even if they had begun their proceedings one day earlier, this does not indicate that the consequence of such refusal was not, in substance, a refusal of leave (or permission) to apply for judicial review. The two conclusions reflect the fact that the time limit for judicial review is not a fixed period of time and that applications made within three months may nevertheless involve undue delay. Nor are the case management options discussed by the Court of Appeal in BI v Director of Immigration[61] support for the contention that an extension of time is discrete from the decision on leave, since the Court of Appeal seem to have regarded the ex parte grant of an extension of time as being coupled with a decision to grant leave to apply for judicial review (see [137]). 41. Since the refusal of an extension of time leads inexorably to a decision to refuse leave to apply for judicial review, it is a decision that falls within the terms of O.59 r.21(1)(g) and is therefore a decision to which s.14AA(1) of the HCO does not apply and from which an appeal lies as of right. The conclusion that the refusal of an extension of time in the present cases led to a refusal of leave to make the application for judicial review is confirmed by reference to the formal orders made by Chow J in H’s case (HCAL 172/2015) and AH’s case (HCAL 32/2015). This has always been the case in practice and, as noted above, the respondent has not identified any case in which the request for an extension of time has been dealt with discretely without a decision on the issue of whether to grant leave to make the application as well. 42. This conclusion, that the refusal of an extension of time to make the application for leave to apply for judicial review is part and parcel of a refusal to grant leave to apply and so appealable as of right by reason of O.59 r.21(1)(g), is a realistic characterisation of the decision-making process and accords with both commonsense and practice. As the appellants rightly contend, a court’s decision that leave to apply should be refused because the applicant fails to satisfy the requirement in O.53 r.3(7) of a sufficient interest in the matter is appealable as of right (by reason of O.59 r.21(1)(g)), so it would be inconsistent and anomalous to treat refusal of leave because of undue delay and the refusal of an extension of time under O.53 r.4(1) as not also being subject to an appeal as of right.[62] D.4 The appellants’ other arguments on the Leave Issue 43. In view of the conclusion stated above, it is not strictly necessary to address the two alternative arguments advanced by the appellants to challenge the Court of Appeal’s decision in MI. These were respectively that the refusal of an extension of time to make an application for judicial review: (i) constitutes “a judgment or order determining in a summary way the substantive rights of a party to an action” within O.59 r.21(1)(a); and (ii) is a final judgment and not interlocutory for the purposes of s.14AA of the HCO. 44. The two arguments each proceed on a different footing. Argument (i) proceeds on the basis that the refusal of an extension of time is an interlocutory decision that would prima facie require leave to appeal against it, save that it is treated as an exception under O.59 r.21(1) as a judgment or order to which s.14AA does not apply. Argument (ii), however, proceeds on the basis that the refusal of an extension of time is a final decision and not interlocutory, so therefore falls outside the leave requirement in s.14AA(1) in any event. D.4a Is the refusal of an extension of time a final judgment or order? 45. Whether a judgment or order is final or interlocutory has been described as “a notoriously difficult question”.[63] The usual test applied to determine if a judgment or order is final or interlocutory is the “application approach”, which asks whether the judgment or order, whatever the outcome of the application on which it was made, is finally determinative of the entire cause or matter: Shell Hong Kong Ltd v Yeung Wai Man Kiu Yip Co Ltd & Another.[64] However, a judgment or order will also be regarded as final if it is finally determinative of a crucial or substantive issue in the cause of the matter. A broad common-sense approach is adopted and a court will ask if the issue determined is a “substantive part of the final trial” or a “crucial issue” in the case or a point “that goes to the root of the case” or a “dominant feature of the case”. If so, then even if it does not finally dispose of the whole action, the judgment or order on the issue should be regarded as a final judgment. [65] 46. Looked at in isolation, it is difficult to see a decision on a request for an extension of time to apply for leave to make an application for judicial review as satisfying the test laid down in Shell Hong Kong Ltd v Yeung Wai Man Kiu Yip Co Ltd & Another for a final judgment or order. On its own, it is merely a preliminary step to the decision whether or not to grant leave to make the application. It is neither a dominant feature of the case, nor a point going to the root of the case. Nor can it really be said to be a substantive part of the final trial or even a crucial issue in the case. 47. But it is unnecessary to answer the question posed above because, as already indicated, as a matter of substance and practice, whenever a court refuses an extension of time for the making of an application for judicial review, it also consequently refuses leave to make the application. The relevant decision on which to focus is therefore the refusal under O.53 r.3 of the grant of leave to apply for judicial review. And that decision is one which is expressly excepted from the requirement to obtain leave to appeal to the Court of Appeal under s.14AA of the HCO. D.4b Is the refusal of an extension of time a judgment or order determining in a summary way the substantive rights of a party to an action? 48. Whilst there might be some scope for arguing that a refusal of a request for an extension of time to apply for leave to make a judicial review application, albeit interlocutory, is a judgment or order which determines summarily the substantive rights of the intended applicant for judicial review, it is also unnecessary to answer the question posed above for the same reasons set out in the preceding paragraph. A refusal of a request for an extension of time is not an order made in isolation but is always made at the same time as an order refusing to grant leave to apply for judicial review. That refusal of leave being a judgment or order to which the leave to appeal requirement in s.14AA of the HCO does not apply, by reason of O.59 r.21(1)(g), leave to appeal will lie as of right in any event. Being covered by that express exception, it is redundant to ask whether a refusal of a request for an extension of time to make an application for judicial review is appealable as of right also because it falls within the more general exception in O.59 r.21(1)(a). E. The Constitutionality Issue 49. Since the Court concluded at the hearing that the appellants did not require leave to appeal pursuant to s.14AA of the HCO in order to appeal against Chow J’s judgments refusing to grant leave to apply for judicial review, the question of the compatibility of s.14AB of the HCO with Article 82 of the Basic Law does not arise for consideration. The Court’s general practice is that, where a constitutional issue is raised but does not need to be determined for the purpose of determining the appeal before it, the Court will usually not address that constitutional issue: see, e.g., Fateh Muhammad v Commissioner of Registration & Another (2001) 4 HKCFAR 278 at p.287C-E; and GA v Director of Immigration (2014) 17 HKCFAR 60 at [42]. There is no reason to depart from that practice in these appeals. F. A “rolled up” hearing 50. It is convenient to take this opportunity to say something about the expression, a “rolled up” hearing, since it is not always used consistently in this jurisdiction. 51. In the CFI Judgment in H, Chow J noted that he directed a “rolled up” hearing of: (i) the application for an extension of time; (ii) the application for leave to apply for judicial review; and (iii) the substantive judicial review if leave were granted.[66] There are other examples of these three matters being directed to be dealt with together in a hearing described as “rolled up”.[67] But in another case, involving an application for (i) an extension of time and (ii) for leave to apply for judicial review, the application was described as being dealt with on a “rolled up” basis.[68] 52. The more orthodox use of the term “rolled up hearing” is where the court deals with stage (ii), the application for leave to apply for judicial review, and also stage (iii), the substantive judicial review application if leave is granted, together at the same hearing.[69] It is this type of two-stage process,[70] dealing with leave and the substantive hearing (assuming leave is granted), that is properly termed a “rolled up” hearing. Whilst it is certainly not a term of art, it might be preferable if its use were, in future, limited to such a two-stage process only and not used to refer to a hearing including an application for an extension of time to apply for leave to make the judicial review application, so as to avoid the erroneous impression that the hearing of an application for an extension of time is a discrete decision separate to the decision whether to grant or refuse leave to apply. An application which involves undue delay will always require a request for an extension of time as part of the application for leave to apply for judicial review. A hearing of such an application for leave, including the request for an extension of time, is a single composite hearing of the necessary issues to be determined before the court can be satisfied it should grant leave to make the judicial review application or refuse such leave. G. Disposition 53. For the above reasons, I would answer the first question of law for which leave to appeal was granted (set out at [8] above) as follows: Leave to appeal is not required because the decision not to grant an extension of time to apply for leave to apply for judicial review leads to, and is part and parcel of, a decision to refuse to grant leave to apply and so is a decision to which s.14AA of the HCO does not apply by reason of RHC O.59 r.21(1)(g). It is not necessary to address the second question of law for which leave to appeal was granted. 54. To the extent explained in Section D.3 above, the decision in MI is overruled and should not be followed. 55. I would allow the appeals to the extent that the appellants had a right of appeal to the Court of Appeal, without leave, against Chow J’s refusal to grant them respectively leave to apply for judicial review of the Director of Immigration’s respective decisions to refuse to grant them a dependant visa. The further course of the appellants’ appeals will depend on the Appeal Committee’s consideration of the other questions of law for which leave to appeal to this Court is sought by the appellants. 56. Failing agreement between the parties, I would direct the filing of written submissions as to costs which, unless directed otherwise, the Court will deal with on the papers and without a further hearing. The appellants should file their submissions within 14 days of the handing down of judgment in these appeals. The respondent’s submissions should be filed within 14 days thereafter and any reply submissions from the appellants within 7 days of the filing of the respondent’s submissions. Mr Justice Cheung PJ: 57. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Gleeson NPJ: 58. I agree with the judgment of Mr Justice Fok PJ. FACV 1/2020 Mr Philip Dykes SC and Mr Timothy Parker, instructed by Daly & Associates, assigned by the Director of Legal Aid, for the Appellant (“H”) Mr Abraham Chan SC, Mr Jonathan Chang SC and Mr Harrison Miao, instructed by the Department of Justice, for the Respondent FACV 2/2020 Mr Philip Dykes SC and Mr Geoffrey Yeung, instructed by Daly & Associates, assigned by the Director of Legal Aid, for the Appellant (“AH”) Mr Abraham Chan SC, Mr Jonathan Chang SC and Mr Harrison Miao, instructed by the Department of Justice, for the Respondent [1] “The power of final adjudication of the Hong Kong Special Administrative Region shall be vested in the Court of Final Appeal of the Region, which may as required invite judges from other common law jurisdictions to sit on the Court of Final Appeal.” [2] Also referred to as a Form 86 notice. [3] HCAL 172/2015, Judgment dated 3 November 2016 (“CFI Judgment in H”) at [3]. [4] Ibid. at [64]. [5] [2018] HKCA 419, [2018] 4 HKC 440 (“MI”). [6] CACV 5/2017, [2018] HKCA 750, (Lam VP, Kwan & Poon JJA) Judgment dated 29 October 2018 (“CA Judgment in H”). [7] CACV 5/2017, [2019] HKCA 1302, (Poon Ag CJHC, Lam & Kwan VPP) Judgment dated 22 November 2019 (“CA Leave Judgment in H”). [8] FAMV 415/2019. [9] HCAL 32/2015, Judgment dated 18 November 2015 (“CFI Judgment in AH”) at [1]. [10] Ibid. at [57]. [11] HCMP 196/2016, [2019] HKCA 952, (Poon Ag CJHC & Lam VP) Judgment dated 23 August 2019 (“CA Judgment in AH”). [12] Ibid. at [25]. [13] (Lam VP, Barma & Au JJA). [14] FAMV 3/2020. [15] FAMV 415/2019 & 3/2020, [2020] HKCFA 5, (Ma CJ, Fok & Cheung PJJ) Determination dated 10 March 2020. [16] This provides: “Rules of court may specify a judgment or order of any prescribed description to which subsection (1) does not apply and accordingly an appeal lies as of right from the judgment or order.” [17] Such an appeal is as of right and leave to appeal to the Court of Appeal is not required. [18] Reflecting the requirement of “a sufficient interest” in s.21K(3) (supra). [19] (2007) 10 HKCFAR 676 at [14]-[17]. [20] There is no statutory provision which applies any other time limit for an application for judicial review of the subject matter of the intended proceedings in these appeals, so the relevant time limit here is that provided for in O.53 r.4(1). [21] It should be noted that, in England and Wales, doubt was expressed as to whether the promptness test was sufficiently certain to meet the requirements of EU law and the ECHR: R v Hammersmith and Fulham LBC Ex parte Burkett [2002] UKHL 23, [2002] 1 WLR 1593 at [53] and [59]. The ECJ has held that the promptness test breaches EU law since it is subject to the discretion of the court and gives rise to uncertainty: Uniplex (UK) Ltd v NHS Business Services Authority (C-406/08) [2010] 2 CMLR 47 at [41]-[42]. But the promptness test continues to apply in domestic law in England and Wales: R (on the application of Berky) v Newport City Council [2012] EWCA Civ 378 at [34]-[35] and [53]; Springer v University Hospitals of Leicester NHS Trust [2018] EWCA Civ 436 at [52]. [22] [1990] 2 AC 738 per Lord Goff of Chieveley at p.746G-H. HCO s.21K is materially the same as s.31 of the Supreme Court Act 1981 (now called the Senior Courts Act 1981) (“SCA 1981”) and s.21K(6) and (7) reproduce s.31(6) and (7) of the SCA 1981. [23] Pursuant to recommendations in the Law Commission’s Report on Remedies in Administrative Law, Law Com. No. 73, Cmnd. 6407 (March 1976): see Hong Kong Civil Procedure 2020, Vol.1, at [53/14/1]. [24] De Smith’s Judicial Review (8th Ed.), Lord Woolf et al. (eds.) (London: Sweet & Maxwell, 2018) at [15-080]. [25] Ibid. at [15-081]; Applications for judicial review: the requirement of leave, A.P. Le Sueur and Maurice Sunkin [1992] PL 102. [26] [2018] 4 HKC 440 at [29]. [27] The s.21K(6) considerations. [28] [2018] 4 HKC 440 at [45]. [29] Ibid. at [44]. [30] Ibid. at [39]. [31] Ibid. at [17]-[29]. [32] [1990] 2 AC 738 at pp.746-747. [33] [1999] 2 AC 330 at p.341. [34] The Court of Appeal were correct in referring to the English decisions, since that jurisprudence is based on materially identical statutory provisions to RHC O.53 r.4 and HCO s.21K(6) and (7), namely O.53 r.4 of the Rules of the Supreme Court and s.31(6) and (7) of the Supreme Court Act 1981: see Reg. v Dairy Tribunal, Ex parte Caswell [1990] 2 AC 738 at pp.744H-745E for the relevant English provisions and [20] and [22] above for the Hong Kong provisions. [35] Consolidated Case for the Respondent at Section C. [36] Shiu Wing Steel Ltd v Director of Environmental Protection & Airport Authority (No.2) (2006) 9 HKCFAR 478 at [77]. [37] Ex parte A [1999] 2 AC 330 at p.340A. [38] I.e. a vicious circle: Ex parte Caswell [1989] 1 WLR 1089 per Lloyd LJ at p.1094F. [39] Shiu Wing Steel Ltd v Director of Environmental Protection & Airport Authority (No.2) (2006) 9 HKCFAR 478 at [78]. [40] [1985] 1 WLR 1319 per Ackner LJ (as he then was) at p.1325. [41] Ex parte Caswell [1990] 2 AC 738 at p.747F. [42] See [20] and FN21 above. [43] Ex parte Jackson [1985] 1 WLR 1319 at p.1325E-F. [44] Which may be at an oral hearing or on the papers alone (see [19(2)] above). [45] [2014] 6 HKC 1 at [43]-[45]. [46] [2016] 2 HKC 393, CACV 63/2015, 3 November 2015, at [23]-[28]. [47] [1985] 1 WLR 1319 per Ackner LJ (as he then was) at p.1325, with whose conclusion Lord Goff agreed in Ex parte Caswell at pp.746E-747B. [48] Ex parte Caswell [1990] 2 AC 738 at p.747C-E. [49] Consolidated Case for the Respondent at [73(3)]. [50] Ibid. at [75]. [51] Patterson and Karim on Judicial Review (3rd Ed.), Sam Karim QC (ed.) (London: LexisNexis, 2019) at [1.15]; Administrative Law (8th Ed.), Professor Paul Craig (London: Sweet & Maxwell, 2016) at [27-48]; Judicial Remedies in Public Law (5th Ed.), Sir Clive Lewis (London: Sweet & Maxwell, 2015) at [9-042] and [12-009]; Judicial Review Handbook (6th Ed.) Michael Fordham QC (Oxford: Hart Publishing, 2012) at [26.1.1] and [26.1.7]; and De Smith’s Judicial Review (8th Ed.), Lord Woolf et al. (eds.) (London: Sweet & Maxwell, 2018) at [16-058]. [52] Ex parte Jackson [1985] 1 WLR 1319 at p.1325G; Ex parte Caswell [1990] 2 AC 738 at p.747G-H. [53] Ex parte A [1999] 2 AC 330 at p.341D-E. [54] [2019] UKPC 5; [2019] 1 WLR 983. [55] Ibid. at [29], applying R v Comr for Local Administration, Ex p Croydon London Borough Council [1989] 1 All ER 1033 per Woolf LJ (as he then was) at p.1046. [56] Ibid. at [26]. [57] Ibid. at [41]. [58] See, e.g., R v The Stainforth and Keadby Canal Company (1813) 1 M. & S. 32, 105 ER 12; R v The Trustees of the Rochdale and Halifax Turnpike Road (1848) 12 Q.B. 448, 116 ER 935; The Queen v Sheward (1880) 9 QBD 741; and R v London County Council, Ex parte Swan and Edgar (1927) Limited (1929) 141 The Law Times 590 (30 November 1929). [59] See, e.g., R v Ashford, Kent, Justices, Ex parte Richley [1955] 1 WLR 562; R v Dorset Quarter Sessions Appeals Committee, Ex parte Weymouth Corporation [1960] 2 QB 230; R v Inner London Crown Court, Ex parte Greenwich London Borough Council [1976] QB 540. [60] [2001] EWHC Admin 245; [2001] A.C.D. 77. [61] [2016] 2 HKLRD 520 at [136]. [62] Consolidated Printed Case of the Appellants at [64]. [63] Hip Hing Timber Co Ltd v Tang Man Kit & Another (2004) 7 HKCFAR 212 per Lord Millett NPJ for the Court at [38]. [64] (2003) 6 HKCFAR 222 at [26]. [65] Ibid. at [27]-[31]; applied in HKSAR v Mak Wan Ling (No.1) (2019) 22 HKCFAR 51 at [12] (concerning when a judgment is a final judgment within s.31 of the Hong Kong Court of Final Appeal Ordinance (Cap.484)). [66] CFI Judgment in H at [3]. [67] See, e.g., Win More Shipping Limited v Director of Marine, HCAL 1520/2018, [2019] HKCFI 1137 (unrep., 2 May 2019) per Chow J at [30]; and Nisa Azizan v Director of Immigration, HCAL 203/2015, [2016] HKEC 891 (unrep., 15 April 2016) per Chow J at [4]. [68] AMZ v Torture Claims Appeal Board, HCAL 607/2017, [2018] HKCFI 700 (unrep., 3 April 2018) per DHCJ Marlene Ng (as she then was) at [7]. [69] See: Hong Kong Civil Court Practice, W.S. Clarke (Hong Kong: LexisNexis) (updated in June 2020 in Issue 92) at [3732.1] referring to the decision of Hartmann J (as he then was) in Lo Siu Lan & Another v Hong Kong Housing Authority, HCAL 154/2004 (unrep., 15 December 2004) as an early example in Hong Kong of this procedure. See also: Chee Fei Ming v Director of Food and Environmental Hygiene (No.2) [2016] 3 HKLRD 412 at [6(a)]; BI v Director of Immigration [2016] 2 HKLRD 520 at [135]; and Leung Kwok Hung v President of the Legislative Council (No.2) (2014) 17 HKCFAR 841 at [19(8)]. [70] Described as a “common phenomenon”: see Judicial Review Handbook (6th Ed.) Michael Fordham QC (Oxford: Hart Publishing, 2012) at [21.2.3]. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Fok PJ and the orders contained in the judgment. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: 3. The fish farming industry in Hong Kong is regulated and protected by the Marine Fish Culture Ordinance.[1] This appeal is concerned with one aspect of the regulation of fish farms (or as they are more properly referred to under the Ordinance, fish culture zones), concerning the validity of a licence to engage in fish culture after its expiry date where there is an appeal concerning its cancellation. Specifically, when a licence has been cancelled and an appeal against that cancellation is instituted, does the validity of the licence automatically continue until the determination of the appeal even if that date is after the expiry date of the licence? Although, on its face, of narrow ambit, this appeal involves a question of statutory construction of relevance to other licensing regimes and hence is of some general importance. A. The background facts 4. The respondent to this appeal was the holder of a licence (“the Licence”) to culture marine fish within Tiu Cham Wan, which is one of the areas of the waters of Hong Kong designated under the Ordinance as a fish culture zone. The licence was issued by the Director of Agriculture, Fisheries and Conservation (“the Director”) pursuant to section 8 of the Ordinance[2] and was valid for one year from 17 December 2014 until 16 December 2015. It was subject to endorsements as to the size and number of structures permitted on the raft in the fish culture zone (“the Licence Restrictions”) and to printed conditions on its reverse. 5. For reasons unconnected with this appeal, a decision was taken by the Director to cancel the Licence. It seems that pisciculture may have taken a back seat to culinary pursuits: there was evidence before the Magistrate of a departure from the original use of the respondent’s raft at the fish farm in that the structures exceeded the Licence Restrictions in dimension and equipment was present not related to fish culture, such as kitchenware, large dining tables and large televisions.[3] On 24 November 2015, the Agriculture, Fisheries and Conservation Department (“AFCD”) gave written notice to the respondent of the cancellation of the Licence with immediate effect pursuant to section 9 of the Ordinance (see below). The respondent had 28 days within which to appeal.[4] The respondent wished to challenge that decision and, on 18 December 2015, lodged an appeal to the Administrative Appeals Board (“the Board”) against the cancellation of the Licence pursuant to section 16(1)(c) of the Ordinance (see below). 6. In the meantime, the date for application to renew the Licence had passed. This is because, under regulation 4(2) of the Marine Fish Culture Regulations,[5] an application to renew a licence must, unless the Director permits otherwise, be made not later than one month before the date of its expiry. This was also one of the conditions printed on the reverse of the Licence. The Licence was due to expire on 16 December 2015, so the latest date on which the respondent could apply for renewal (unless otherwise permitted by the Director) was 16 November 2015. No such permission was ever given by the Director but the respondent sought the renewal of the Licence on 6 January 2016. On 11 January 2016, the Director replied to the renewal request stating that the AFCD could not process it because the Licence had already expired on 16 December 2015 and the application to renew had not been made on time. In consequence, on 28 January 2016, the respondent lodged a further appeal to the Board, this appeal being against the refusal to renew the Licence. 7. Subsequent to the proceedings giving rise to this appeal (see Section B below), by its decision handed down on 8 August 2017, the Board dismissed the respondent’s appeals against both the cancellation of the Licence and the refusal to renew the Licence.[6] 8. In the period between 17 December 2015 and 27 January 2016, the respondent continued to cause or permit a raft to remain in the Tiu Cham Wan fish culture zone and it is that act which gave rise to these proceedings. B. The proceedings below 9. The respondent was charged with causing or permitting a raft to remain in the Tiu Cham Wan fish culture zone between 17 December 2015 and 27 January 2016 otherwise than under and in accordance with a licence or permit granted by the Director, contrary to sections 13(2) [7] and 21(3)[8] of the Ordinance. The defence at the trial before Deputy Special Magistrate Andrew Hung was that, by virtue of the provisions of sections 8 and 16 of the Ordinance, the Licence should be deemed to have been in force during the period of the charge by reason of the respondent’s appeal lodged (on 18 December 2015) against the decision to cancel the Licence[9] and also by reason of its appeal lodged (on 28 January 2016) against the decision to refuse renewal of the Licence.[10] 10. On 29 September 2016, after trial, the Magistrate convicted the respondent of the offence and, on 12 October 2016, imposed a fine of HK$2,500. He rejected the defence advanced by the respondent on the basis of sections 8 and 16 of the Ordinance.[11] 11. The respondent appealed to the Court of First Instance.[12] The intermediate appeal was heard by Deputy High Court Judge Stanley Chan. On 24 November 2017, the Deputy Judge allowed the appeal and quashed the respondent’s conviction. The Deputy Judge’s reasoning will be examined below in the context of a consideration of the respondent’s arguments on this appeal. C. Relevant provisions of the Ordinance 12. The proper construction of sections 9 and 16 of the Ordinance lies at the heart of this appeal, so it is convenient to set them out at length. Section 9, dealing with the cancellation of licences, provides as follows: “(1) The Director may cancel a licence – (a) on any ground specified in section 8(6) which would have entitled him to refuse to grant or renew a licence; (b) on the ground that the licensee has – (i) contravened any of the provisions of this Ordinance or any regulations made thereunder; (ii) contravened his licence; (iii) failed or is unable or incompetent to carry out the purposes of his licence; or (iv) failed to provide adequate management or supervision of any raft or impoundment in respect of which the licence is valid. (2) Where the Director cancels a licence under subsection (1) he shall, where practicable, send to the person who was the holder of the licence a notice of the cancellation and state in the notice the reasons for the cancellation.” 13. Section 16, dealing with the right of appeal, provides as follows: “(1) Any person aggrieved by a decision made in respect of him by the Director to – (a) refuse to grant or renew a licence under section 8(6); (b) refuse to approve the transfer of a licence under section 8A(3)(b); (c) cancel a licence under section 9(1); (d) refuse to grant a permit under section 14(1); or (e) cancel or refuse to renew a permit under section 14(2), may appeal to the Administrative Appeals Board against that decision. (2) Where an appeal is made under subsection (1) against a decision of the Director to – (a) cancel a licence or permit, the decision shall not become effective pending; (b) refuse to renew a licence or permit, the licence or permit (if expired) shall be deemed to continue in force according to its terms and conditions until; or (c) refuse to approve the transfer of a licence (including a licence to which section 8A(6) applies), the licence shall continue in force pending, the determination of the appeal by the Administrative Appeals Board. (3) Where – (a) an appeal has been made under this section; and (b) the period of validity of the licence or permit to which the appeal relates would have expired but for subsection (2)(b) or (c), then the licensee or permittee, as the case may be, shall be liable for the fee prescribed – (c) for the period from the day the licence or permit would have expired to the day of the determination of the Administrative Appeals Board or withdrawal of the appeal, whichever first occurs; and (d) on a pro rata basis, irrespective of the outcome of the appeal.” 14. It is also convenient, in the context of section 9(1), to set out the provisions of section 8(6), specifying specific grounds on which the Director might refuse to grant or renew a licence, namely: “(6) The Director may refuse to grant or renew a licence if it appears to him – (a) that, having regard to the size or location of a fish culture zone, the grant or renewal of a licence would cause overcrowding of the fish culture zone or would otherwise not be in the best interests of fish culture; (b) that any raft or impoundment used or to be used by the applicant for the purpose of fish culture does not comply with any of the provisions of this Ordinance or any regulations made thereunder.” D. The certified question of law for this Court 15. By notice of motion dated 1 December 2017, the prosecution (hereinafter referred to as “the appellant”) sought a certificate for leave to apply to the Court of Final Appeal against the Judgment of the Deputy Judge allowing the respondent’s magistracy appeal. By Judgment dated 7 February 2018, the Deputy Judge refused that application and dismissed the notice of motion.[13] The Deputy Judge also made a costs order against the appellant in respect of that application. 16. On 15 May 2018, on the appellant’s application, the Appeal Committee[14] granted leave to appeal to this Court in respect of the following question of law, namely: “What is the effect, if any, on an appeal made under section 16(1)(c) of the Marine Fish Culture Ordinance, Cap 353 against a decision to cancel a licence under section 9(1) on the period of validity of a licence if as at the time the appeal was made, the date of expiry of the licence has passed and no application to renew the licence has been made?” E. The Deputy Judge’s reasoning in allowing the magistracy appeal 17. The Deputy Judge’s reasoning in allowing the magistracy appeal is supported by the respondent in this appeal and so it is convenient to set out that reasoning in order to analyse the Deputy Judge’s construction of sections 9 and 16 of the Ordinance. 18. The Deputy Judge regarded it as important that the Director sought to cancel the Licence before it had expired. Had the Licence been cancelled after its expiry without the respondent having applied for a renewal, the outcome would (in his view) have been different.[15] Instead, in this case, the Director cancelled the Licence on 24 November 2015 with immediate effect when it was still valid and so the Deputy Judge considered: “… that the key to the case naturally freezes at the point when action was taken to cancel licence on 24 November rather than whether or not subsequent renewal application was made by the [respondent] or the [Director’s] decision to refuse renewal.”[16] 19. The Deputy Judge took the view that, since the respondent had appealed against the cancellation of the Licence, considerations relating to its renewal were no longer material. He held: “The reason is that under section 16(2)(a), the decision to cancel a licence shall not become effective pending the determination of the appeal by the Administrative Appeals Board. Upon my enquiry, the Administrative Appeals Board has yet to give a determination until this appeal hearing, nor will the [prosecution] be able to tell the exact date of determination. Since the [respondent] had sought to appeal against the cancellation of licence, I consider that the matters relating to renewal would naturally be set aside. Nevertheless, according to section 16(3), the licensee is still liable for the licence fee prescribed.”[17] 20. The Deputy Judge then concluded: “I find that since the decision to cancel licence had become ineffective by reason of the appeal made by the [respondent] to the Administrative Appeals Board, it is not tenable for the [Director], by ‘technical’ means, to insist the [respondent] had permitted the subject raft to unlawfully remain or exist in the zone when the [respondent] was unable to renew the licence within the prescribed time frame.”[18] 21. Accordingly, the Deputy Judge allowed the respondent’s appeal, quashed the conviction and set aside the fine imposed by the magistrate. F. The respondent’s contentions on this appeal 22. The above reasoning of the Deputy Judge is supported by the respondent in this appeal. Essentially, the respondent’s argument proceeds on the basis that, the Licence having been cancelled by the Director with immediate effect, before the respondent lodged an appeal against such cancellation, there was no expiry date for the Licence as it had been terminated.[19] 23. The respondent then contends that, when it lodged its appeal against the cancellation, the effect of section 16(2)(a) was that “the Director’s decision to cancel the Licence is to [sic] ‘put on hold’ until the outcome of the appeal by the Board”.[20] The respondent’s argument continues: “Once the Director’s decision to cancel the Licence is ‘put on hold’, it makes the Licence still valid for the time being (as if it was not cancelled by the Director) because the Board needs to consider and rule upon the contentions to be raised by the Director and the [respondent] respectively at the appeal.”[21] 24. The respondent contends that there was no duty on it to apply to renew the Licence after it had lodged an appeal against the cancellation. The argument seems to be that, by reason of the operation of section 16(2)(a) of the Ordinance, contrasted with section 16(2)(b), the original terms of the Licence (including its expiry date) ceased to apply, so that there was no ongoing licence which the respondent could apply to renew.[22] Instead, it would appear to be the respondent’s contention that a new (presumably ad hoc) licence (devoid of terms and requiring no payment of fees) would come into existence pending the determination of the appeal to the Board against the cancellation. The respondent submits that this conclusion is supported by the provisions of section 16(3) of the Ordinance concerning payment of the fee for a licence pending an appeal to the Board.[23] G. The proper construction of section 16 of the Ordinance 25. Section 16 of the Ordinance (set out above) gives persons aggrieved by certain decisions of the Director relating to the regulation of marine fish culture in Hong Kong a right of appeal to the Board. Pursuant to section 16(1)(c), a person aggrieved by a decision made in respect of him by the Director to cancel a licence under section 9(1) may appeal to the Board against that decision. The other decisions for which section 16(1) provides an avenue of appeal are where the intended appeal is against: a refusal to grant or renew a licence under section 8(6); a refusal to approve the transfer of a licence under section 8A(3)(b); or, a refusal to grant a permit or to cancel or refuse to renew a permit under sections 14(1) and (2) respectively. 26. It is important to note that different language is used to describe the consequences flowing from the lodging of an appeal to the Board depending on the nature of the decision or action of the Director appealed against. These consequences are addressed in section 16(2) but it is also relevant to take into consideration the provisions of section 16(3) in this regard. 27. In the case of an appeal against a decision of the Director to cancel a licence or permit under section 9(1), “the decision shall not become effective pending … the determination of the appeal by the Administrative Appeals Board” (section 16(2)(a), emphasis added). On its face, this relates back to the cancellation decision and simply provides that the cancellation does not take effect until an appeal against that cancellation is determined but it does not say what effect the passage of time has on the licence in the meantime. In other words, in such a situation, a licence continues as if the decision to cancel it had not been made. It continues as if nothing had happened so that the original period of validity of the licence remains unaffected. 28. By contrast, the language of sections 16(2)(b) and 16(2)(c) is different in that, instead of stating that the relevant decision of the Director shall “not become effective”, those sub-sections provide that the relevant licence or permit will “continue in force”. Thus, under section 16(2)(b), if the Director refuses to renew a licence or permit (under one of the specified statutory provisions) and the aggrieved person appeals that decision, then “the licence or permit (if expired) shall be deemed to continue in force according to its terms and conditions until … the determination of the appeal by the Administrative Appeals Board” (emphasis added). Similarly, under section 16(2)(c), if the Director refuses to approve the transfer of a licence and that refusal is appealed, then “the licence shall continue in force pending, the determination of the appeal by the Administrative Appeals Board” (emphasis added). Accordingly, the original period of validity of the licence is extended, unlike the position under section 16(2)(a). 29. This difference of language is reflected in section 16(3) which deals with the liability for the fee for a licence or permit. Where an appeal is lodged and “the period of validity of the licence or permit to which the appeal relates would have expired but for subsection (2)(b) or (c)”, then the licensee or permittee is liable for a fee calculated in accordance with sections 16(3)(c) and (d). 30. These provisions are important because they acknowledge that the effect of an appeal in the case of a refusal to renew a licence or permit (section 16(2)(b)) or a refusal to approve the transfer of a licence (section 16(2)(c)) is to override the effect of the expiry date of the licence or permit as the case may be. This is borne out by: (1) the provision that the licence or permit “continue in force” (in both sections 16(2)(b) and (c)); (2) the words “if expired” in brackets (in section 16(2)(b)); and (3) the provision that the licensee or permittee be liable for a pro rata fee during the period from the original expiry date until the determination of the appeal, notwithstanding the fact that the licence or permit would have expired but for the appeal under those sub-sections (sections 16(3)(a) and (b)). 31. As a matter of statutory language, it is clear that the provisions in sections 16(3)(a) and (b) relate only to an appeal under section 16(2)(b) or (c) and do not relate to an appeal under section 16(2)(a). There is therefore no express statutory provision for a licence or permit that is cancelled to “continue in force” until the decision of the Board on the appeal with the fee consequences provided for in section 16(3). Instead, the contrasting statutory language shows clearly that the cancellation decision “shall not become effective” until the decision of the Board on the appeal. On its face, this contrasting statutory language suggests that a cancelled licence will in the meantime operate, subject to its terms including its expiry date, so that a licence holder can continue to operate his fish farm until the expiry of the licence as if it had not been cancelled. 32. This ability to continue to operate the fish farm does not, however, absolve the licensee from the need to apply to renew his licence if he wishes to conduct that operation beyond the expiry date of his licence. Such renewal application is governed by the requirements of the Regulations and, if aggrieved by a refusal to renew, may give rise to an appeal having the effects set out in section 16(2)(b) or (c), including the consequential fee obligation set out in section 16(3). 33. The absence of a corresponding fee obligation in section 16(3) in relation to an appeal against a cancellation of a licence under section 16(2)(a) is telling. This statutory language clearly indicates that such an appeal does not have the effect of extending the licence beyond its expiry date. 34. As is now well-established, statutory language must be construed having regard to its context and purpose and this was common ground between the parties. In relation to the Ordinance, its purpose is clear: it is designed to put in place an effective licensing regime for the regulation and protection of marine fish culture in Hong Kong waters. This statutory purpose, however, does not definitively support or undermine the Deputy Judge’s analysis and the respondent’s contentions set out above. However, that analysis and those contentions would lead to some surprising results which strongly indicate that this was not the legislative intention of the relevant provisions of the Ordinance. 35. Thus, given the fact that the appeal process before the Board typically takes many months (and in the present case the two appeals lodged by the respondent were only determined after nearly 20 months), the extension during that period of a cancelled licence without statutory power to collect any additional fee would be contrary to common sense. It would also give rise to an obvious risk of abuse. The effect of an appeal against a cancellation decision being to relate back to the date of that decision and to neutralise the effect of such a cancellation, a non-compliant licence holder could commit a breach of his licence before its expiry and then by appealing against the cancellation decision, take advantage of section 16(2)(a) (on the Deputy Judge’s construction) to secure an extension of the original licence pending the determination of the appeal. Moreover, on the construction of section 16(3) set out in paragraph [31] above, he would be in a more advantageous position to that of a licence holder appealing against the Director’s refusal to renew his licence who would have to pay a fee for the period during which section 16(2)(b) deemed his licence to continue. There is no obvious purpose to such an odd outcome and it is therefore highly unlikely to have been intended. 36. Statutory interpretation is also contextual by reference to other provisions in the same legislation and in legislation of a similar nature. Reference has already been made above to various contextual points demonstrating the difference in the statutory language between section 16(2)(a), on the one hand, and sections 16(2)(b) and (c) on the other. That difference in language is more than just a matter of form, as shown by the provision for fees in section 16(3). 37. By way of analogous example, reference may be made to the provisions of section 27 of the Road Traffic Ordinance[24] relating to the grant of passenger service licences. Such licences for public light buses, private buses and school private light buses set out the date of commencement and expiry.[25] Section 31 of the RTO empowers the Commissioner of Transport in certain circumstances to cancel a passenger service licence. The cancellation comes into effect 21 days from the date of notification to the licensee.[26] The licensee is entitled to apply for a review of the Commissioner of Transport’s decision by a Transport Tribunal and, if he does, the cancellation shall not take effect pending the review.[27] A finding in this appeal that the lodging of an appeal against cancellation extends the life of the licence until determination of the appeal, even if that date is after the expiry date of the licence, might arguably mean that an application to review cancellation of a passenger service licence would extend the life of such a licence. Since the reason for cancellation may involve a serious breach of the licence, that is unlikely to have been the intention of the legislation and this illustrates, by analogy in a different statutory context, the absurdity inherent in the respondent’s case in this appeal. 38. Both parties also referred to other ordinances to make contextual points. The respondent pointed to language in various ordinances to support the contention that when the legislature wishes to make a licence invalid notwithstanding a pending appeal against cancellation, the legislation expressly so provides: see section 30(2) of the Adoption Ordinance[28] and section 16A(2) of the Pesticides Ordinance[29]. It is doubtful, however, whether these ordinances can be regarded as legislation of a similar nature to the Ordinance under consideration in the present appeal. In any event, these statutory provisions merely show a legislative policy, in the context of certain licensed or authorised activity (in the cited instances, for the welfare of children and in the interests of public health), which favours maintaining suspension of the licence or authorisation despite the lodging of an appeal. Whilst it is clear that policy was not adopted in the Ordinance, it does not follow from that fact that an appeal under section 16(2)(a) against cancellation of a licence to culture marine fish prolongs the licence beyond its expiry date. 39. For its part, the appellant relied on provisions in the Fisheries Protection Ordinance[30] and the Protection of Endangered Species of Animals and Plants Ordinance[31] to show that, in that legislation covering similar subject matters to that of the Ordinance, a distinction is drawn between the effect of a pending appeal against a cancellation of a licence and that against a renewal of a licence and, where an extension of the validity of a licence until the disposal of an appeal is intended, the legislation states this expressly.[32] 40. It is doubtful that the wider context of other ordinances cited by the parties otherwise adds much to the arguments in support of or against this appeal. However, there is some force in the contention that express language is typically used where, because of a pending appeal, an extension to the validity of a licence is provided for notwithstanding the cancellation of the licence. H. Answering the certified question and determining the appeal 41. In my view, the answer to the certified question (set out at paragraph [16] above) flows from the proper construction of section 16 of the Ordinance (addressed above) and is as follows: (1) Where an appeal is timeously lodged against a decision to cancel a licence under the Ordinance (i.e. within 28 days of receipt of notice of the decision) and within the validity of the licence, the cancellation does not become effective pending the determination of the appeal notwithstanding that the cancellation may on its face state that it takes immediate effect. Nevertheless, the expiry date and other terms of the licence remain applicable and, if no application to renew the licence is made, the licence will expire on the expiry date when that date arrives. (2) If a cancellation is made during the currency of a licence but an appeal is only lodged against the cancellation decision after the expiry date of the licence, the licence will have run its original course and expired before the lodging of the appeal. In those circumstances, the lodging of the appeal (within the 28-day time limit) will have the effect of nullifying the cancellation. However, as indicated in sub-paragraph (1) above, the licence will have continued to run and will have expired on its expiry date. If the licensee continued to operate a fish farm after the cancellation, there would be no offence until the expiry date of the licence. Thereafter, however, the continued operation of a fish farm would be an offence since there would no longer be a valid licence in place. (3) Although for the reasons explained in Section I below this is not a matter before the Court in this case, it is necessary to mention in this context the contrasting consequences of an appeal under section 16(2)(b) or (c). Where, for example, an application for renewal is refused under section 8(6) and that refusal is appealed, the effect of section 16(2)(b) is that the licence will be deemed to continue in force according to its terms and conditions until the determination of the appeal. This applies even if the licence has expired as at the date of the institution of the appeal. In that event, the appeal will relate back to the expiry date of the licence and the licence is deemed to continue in force thereafter. The continuation of the licence is on its original terms so a breach of the licence conditions thereafter will render it liable to enforcement action by the Director (including cancellation, which may in turn give rise to the consequences of section 16(2)(a)). 42. With respect to him, the Deputy Judge’s conclusion that the position was frozen by the Director’s decision to cancel the respondent’s licence so that the subsequent appeal against that decision had the effect of maintaining the validity of the licence until the determination of the appeal and regardless of the absence of any application to renew the licence was wrong and based on a misconstruction of section 16 of the Ordinance. The Deputy Judge placed an erroneous construction on section 16(2)(a) by holding that, once an appeal against a cancellation decision is lodged, the other provisions of the Ordinance relating to the renewal of the licence were “set aside”. 43. On the contrary, it was open to the respondent to apply to renew its licence in accordance with the Regulations. Had it done so, and had renewal been refused under section 8(6) and that refusal timeously appealed, then the provisions of section 16(2)(b) might, subject to arguments based on the effect of a supervening cancellation decision under section 9(1), have led to the licence continuing in force pending the determination of the appeal. However, those facts are not this case (see Section I below). 44. The facts of this case are clear. The respondent’s licence was cancelled. No application to renew the licence was made within time. No appeal against the cancellation was lodged until after the expiry date of the licence. The provisions of section 16(2)(a) operated to suspend the coming into effect of the cancellation decision but did not affect the original period of validity of the licence (or indeed any of its other terms). Accordingly, since not renewed, the licence expired on its expiry date (i.e. 16 December 2015). Thereafter, without a valid licence, the respondent admittedly caused or permitted its raft to remain in the Tiu Cham Wan fish culture zone for the period between 17 December 2015 and 27 January 2016 and so it was properly convicted by the Magistrate of the offence charged. I. A point not argued in this appeal 45. As indicated in paragraph [43] above, the facts of this case are not that the respondent made a timely application to renew its licence which was then refused pursuant to section 8(6) of the Ordinance and which refusal the respondent then appealed pursuant to section 16(1)(a) of the Ordinance (thereby engaging section 16(2)(b)). In this regard, it is to be noted that the reason for refusal of renewal was not one of the grounds listed in section 8(6) of the Ordinance and thus not a decision within section 16(1)(a) of the Ordinance giving rise to a right to appeal. Instead, the refusal was on the basis of the inability to process the application for renewal because the original licence had expired and was no longer valid.[33] So there is some question as to whether an appeal under section 16(1)(a) was open to the respondent in this case. If no such appeal was available, the provisions of section 16(2)(b) would not have any application here in any event. That was certainly the view of the Magistrate.[34] 46. Although there was an appeal against the Director’s refusal to renew the licence, that appeal was lodged on 28 January 2016, after both the expiry date of the licence and the cancellation notice and the lodging of an appeal against that decision. The analysis of the Deputy Judge below proceeded on the basis that the subsequent application to renew the licence and the Director’s refusal of that application were irrelevant[35] and he allowed the appeal on the basis of his analysis of the effect of the cancellation notice and the appeal against that cancellation. The certified question for which leave to appeal was granted (see paragraph [16] above) is accordingly limited to the point decided by the Deputy Judge, namely the effect of an appeal against cancellation and thus the effect of section 16(2)(a) of the Ordinance, and the submissions in the parties’ respective printed cases and at the hearing before this Court were directed to this point alone. 47. The question of whether an out of time application for renewal of a licence which was then refused by the Director under section 8(6) and which refusal was then appealed would, because of the provisions of section 16(2)(b) of the Ordinance, result in the continuation in force of the licence is not before us. I would expressly leave that question open for another occasion on which it might arise. However, it should be noted that there are at least three permutations of the facts that might give rise to a consideration of that question: (1) First, where an application for renewal was made at least one month before expiry of the licence and, on refusal, was subject to an appeal lodged within the validity of the licence. (2) Second, where an application for renewal was not made until less than one month before expiry of the licence and, on refusal, was subject to an appeal lodged outside the validity of the licence. (3) Third, where an application for renewal was not made until after expiry of the licence and, on refusal, was subject to an appeal necessarily also lodged outside the validity of the licence. 48. The effect of section 16(2)(b) of the Ordinance might well be different in each of the above three scenarios and, as I have said, consideration of those questions should await some other case in which they are properly before the Court. Nothing in this judgment concerning the effect of section 16(2)(a) of the Ordinance should be taken to determine the answer to those questions as to the effect of section 16(2)(b) of the Ordinance (or indeed section 16(2)(c)) or analogous statutory appeal provisions in the context of similar licensing regimes. 49. A further point that does not arise is where an application for renewal of a licence is refused but not under section 8(6) (for example, as in this case where the Director refused the renewal because the application was out of time). In that event, it is an open question whether an appeal lies under section 16(1) and, if not, whether the only recourse to a dissatisfied licence holder would be an application for judicial review. If so, there would then be questions as to the availability of interim relief and the consequences in terms of fees during any interim period of operation pending the judicial review. These potentially difficult questions are not before the Court and should be left for determination in another case in which they may be material. J. Disposition 50. For the above reasons, I would allow the appeal against the Deputy Judge’s judgment, reinstate the respondent’s conviction in KTS 7639/2016 and the fine imposed by the Magistrate. I would also set aside the Deputy Judge’s order for costs against the appellant on the application for certification of a question of law. Mr Justice Stock NPJ: 51. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Spigelman NPJ: 52. I agree with the judgment of Mr Justice Fok PJ. Ms Vinci Lam SADPP and Ms Hermina Ng SPP, of the Department of Justice, for the Appellant Mr Henry LW Fung, instructed by Francis Kong & Co., for the Respondent [1] (Cap.353), hereinafter referred to as “the Ordinance”. [2] Section 8(1) provides: “Subject to this Ordinance, the Director may grant to any person, a licence, and renew such licence, to engage in fish culture within a fish culture zone”; section 8(7) provides: “Where the Director refuses to grant or renew a licence under subsection (6) he shall send to the applicant a notice of the refusal and state in the notice the reasons for the refusal.” [3] KTS 7639/2016, Reasons for Verdict, 4 November 2016 (“RV”) at [42]. [4] Section 9(b) of the Administrative Appeals Board Ordinance (Cap.442). [5] (Cap.353A), hereinafter referred to as “the Regulations”. Section 4(2) provides: “Application for renewal of a licence or permit shall, except where the Director otherwise permits in any particular case, be made to the Director not later than 1 month before the date of the expiry thereof.” [6] AAB 62/2015 & 6/2016, Reasons for Determination dated 8 August 2017. [7] Section 13(2) provides: “Subject to subsection (3)(b) [relating to cases of emergency or stress of weather], any person who causes or permits any raft or impoundment to enter or remain or to be constructed in a fish culture zone otherwise than under and in accordance with a licence or permit granted to him commits an offence.” [8] Section 21(3) provides: “Any person who commits an offence under section 12, 13 or 20 is liable to a fine at level 4 and to imprisonment for 6 months.” [9] RV at [49] and [54] to [58]. [10] RV at [49] and [50] to [53]. [11] See the references at FN9 and FN10 above. [12] In HCMA 596/2016, Judgment dated 24 November 2017 (“CFI Judgment”). [13] [2018] HKCFI 242. [14] [2018] HKCFA 25 (Ribeiro PJ, Tang PJ & Stock NPJ). [15] CFI Judgment at [28]. [16] CFI Judgment at [29]. [17] CFI Judgment at [30]. [18] CFI Judgment at [31]. [19] Respondent’s Printed Case at [37]. [20] Ibid. at [40] (emphasis in original). [21] Ibid. at [41] (emphasis in original). [22] Ibid. at [42]. [23] Ibid. at [43]. [24] (Cap.374) (“RTO”). [25] Road Traffic (Public Service Vehicles) Regulations (Cap.374D), section 7. [26] RTO, section 32(1). [27] RTO, sections 33(1) and 32(2). [28] (Cap.290), which provides: “A decision referred to in subsection (1) shall have immediate effect, or have effect from a date specified in the decision (if applicable), notwithstanding any appeal against the decision.” [29] (Cap.133), which provides: “Subject to section 13(4), a decision referred to in subsection (1) has immediate effect, or has effect from a date specified in the decision (if applicable), despite any appeal against the decision.” [30] (Cap.171). [31] (Cap.586). [32] See: sections 33(a), 33(d), 34(a) and 34(b) of (Cap.171); and sections 46(2) and 46(4) of (Cap.586). [33] Letter from AFCD to the respondent dated 11 January 2016 (Exhibit P6 at trial). [34] RV at [51] to [53]. [35] CFI Judgment at [29]. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Gummow NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Gummow NPJ. Mr Justice Tang PJ: 3. I agree with the judgment of Mr Justice Gummow NPJ. Mr Justice Fok PJ: 4. I agree with the judgment of Mr Justice Gummow NPJ. Mr Justice Gummow NPJ: 5. The respondents succeeded at trial before Deputy High Court Judge Saunders in their action as plaintiffs against the appellants for passing-off and also for infringement of registered trade marks contrary to s 18(3) of the Trade Marks Ordinance Cap 559 (“the TMO”)[1]. Subject to modification of the terms of the injunction restraining trade mark infringement, an appeal was dismissed by the Court of Appeal (LamV-P, Barma, McWalters JJA)[2]. 6. Pursuant to a grant of final leave to appeal in this Court, the appellants, the defendants at trial, seek dismissal of the action against them or, failing that, a retrial. The grant was limited to 6 questions, the first five of which concern matters of trade mark law, and the sixth passing-off. For the reasons which follow the appeal should be dismissed. 7. With respect to passing-off, the defendants are, by order made at trial, enjoined from passing-off by acts including the use of “TWG” or “TWG Tea” or any signs which are confusing similar in connection with any tea products and accessories, and in connection with any tea salon, café or restaurant services. The Court of Appeal refused a belated application to limit those terms, which the defendants made after their appeal had been dismissed. A consequence is that the restraint extends to any signs which are confusingly similar to “TWG” or “TWG Tea”. The scope of the restraint serves to mark out the central area of dispute in this litigation. 8. The TMO came into operation on 4 April 2003. It replaced the Trade Marks Ordinance, Cap 43, which had been introduced in 1954 (“the 1954 Ordinance”). Section 10(3) of the TMO provides that “nothing” in that legislation “affects the law relating to passing-off”. Given this relationship between the common law and the statute, it will be convenient in these reasons to deal first with passing-off before turning to the issues the appellants raise respecting refinements in the construction and operation of s 18(3) of the TMO. The findings at trial 9. The plaintiff group of companies (“the Tsit Wing Group” or “the Group” or “Tsit Wing”) is the successor to a Hong Kong business which commenced operations in 1932, initially as a wholesaler in the supply of tea and coffee products, and has continued through three generations of the same family. In recent times the Tsit Wing Group has diversified its operations in various respects. These include the provision of food services, what the trial judge identified as “fast-moving consumer goods” in supermarkets and between 1994 and 2009 the operation of cafés at prime commercial sites under the name “TW Coffee Concept”. However, as the defendants have emphasised, the Group’s activities have not yet included tea salons. In 2011, of the total gross sale figure for the Group of $393m, $114m represented sale of coffee and $111m sales of tea. The trial judge noted that all of these sales of tea and coffee carried a logo containing “TWG” and this appeared on documents such as invoices. The TWG logo was adopted in 2006, and was devised from the initial letters of the expression Tsit Wing Group. 10. The trial judge found that by 2011 the use of “TWG” was “more than sufficient to establish goodwill among those classes of the population in Hong Kong who have had cause to purchase Tsit Wing’s goods and to deal with them”. These included those whose language or first language was Chinese rather than English. His Lordship added that he was “satisfied that the first element of the tort of passing-off has been established” and that the plaintiffs had “established a goodwill and reputation attached to the goods and services which it supplies in the mind of the purchasing public by association with the identifying TWG acronym”, such that it is “recognised by the public as distinctive specifically of Tsit Wing’s goods and services”. These findings were undisturbed by the Court of Appeal. 11. The first defendant was incorporated in Singapore in 2001 and in 2008 adopted “TWG” to identify “The Wellness Group”. Between July 2008 and December 2011 the first defendant established four tea shops in Singapore, had other retail outlets in London, Tokyo, New York, and Abu Dhabi, and supplied its tea to Singapore Airlines and “upmarket” hotels. 12. The year in which the findings at trial respecting the business of the Tsit Wing Group spoke, 2011, is of importance. On 8 December in that year the second defendant (which is a related company of the first defendant and was incorporated in Hong Kong as a vehicle for operations there) opened a “Tea Salon and Boutique” in Hong Kong Central at Podium Level One, IFC Mall. Food and beverage products, especially tea, were sold and consumed there. 13. The photographic evidence shows that persons walking by at the Podium Level one would observe prominently displayed on the exterior of the premises a logo or cartouche being one of the signs a representation of which appears as Annexure 1 to these reasons (“the Cartouche signs”). A feature of the Cartouche signs are the letters “TWG”. On entering the premises they would see the Cartouche signs appearing on containers of tea stacked behind the counter, and on the wall of the area containing tables and chairs for patrons taking tea there. Matter appearing on the elaborate packaging of tea available at these premises included the signs in Annexure 2 (“the Balloon signs”). The matter on the packaging also encouraged further purchases on line, at the site “TWG Tea.com”. To place such an order would require typing of this identification and thus specific attention to “TWG Tea”. The trial judge found that while the defendants wished to market their goods “in an apparently luxurious manner” they could not control the use put by purchasers of their product and were happy to supply to whoever wished to buy their tea. Passing-Off 14. Although the Tsit Wing Group did not presently operate cafés, the trial judge found that there was, if this were necessary for the tort of passing-off, a common field of activity in the retailing of tea. Further, five witnesses gave evidence that when they saw the shop on the Podium Level One they believed that it had been opened by Tsit Wing. Nothing in their cross-examination led the trial judge to doubt that the belief they described was genuine. His Lordship added that the use of “TWG” “renders confusion and deception inevitable”. These findings were not disturbed by the Court of Appeal. The finding of inevitable deception makes it unnecessary to consider here whether, since injunctive relief to restrain passing-off is designed to protect the goodwill of the plaintiffs, and for this purpose confusion may be enough to attract custom to the defendants, confusion not reaching deception may be sufficient for the grant of an injunction. As will appear, likelihood of confusion also plays a significant part in the operation of s 18(3) of the TMO. 15. With respect to the apprehended injury to the goodwill of the Tsit Wing Group in their business, the trial judge held this to be “direct or tangible”. The defendants were likely to expand sales of tea using their sign featuring “TWG” into supermarkets, thereby diverting sales from the plaintiffs. As to the Group, the trial judge also found that while the Group “does not presently operate cafés they will be able to do so in the future and have done so in the past.” As will appear, these matters were sufficient to support the relief for passing-off. 16. However, in addition or in the alternative, the apprehended injury was held to be “intangible”. This “intangible” injury was described by the trial judge as “dilution” by an unfavourable association likely to be drawn by the public between “TWG” and the business of the defendants. Although the trial judge expressly declined to make a finding of fraud by the defendants in opening in Hong Kong with full knowledge of the business there of the Group, he was of the view that the defendants had not operated their business “at the highest moral level” in coming to Hong Kong with apparent disregard of the rights of the Tsit Wing Group. However, it may be noted here that whether such an unfavourable association would be drawn by the public in Hong Kong would depend upon the state of public awareness of these matters. That, in any event, must weaken the finding on this alternative ground of “dilution” to merit further examination. 17. But that is not the gravamen of the complaint by the defendants as appellants in this Court. The defendants attack any reliance upon “dilution” as sufficient injury to goodwill in a passing-off case. In their written submissions they submit that “dilution” is a term which has come to be used in United States law and in academic writing to describe actionable damage to the goodwill of a plaintiff which arises even in the absence of any confusion by customers or consumers between the goods or services of the plaintiff and those of the defendant. The defendants submit that, properly understood, statements respecting “dilution” in certain recent English authorities at first instance[3] and intermediate appellate level[4] do not support such an expansion of the tort of passing-off into one of “unfair competition” and, more to the point for this Court, that such an expansion of the common law should not be made in Hong Kong. Dilution 18. Consideration of these submissions respecting “dilution” requires attention in what follows in these reasons to matters of basic principle. That will show that the defendants correctly submit that what might be called the United States approach to “dilution” does not represent the law of passing-off in Hong Kong. 19. However, that does not result in success for the defendants on their appeal respecting passing-off. This is because the plaintiffs respond by the forensic tactic of confession and avoidance. They submit that (a) the present case is not one where there was a finding of liability on the ground of dilution without confusion and deception and (b) it is well established that the passing-off action protects goodwill against its threatened erosion by the activity of the defendant in cognate fields into which the plaintiff may wish to enter, where that activity causes or is likely to cause deception of those familiar with the mark or other indicia of the plaintiff; and (c) on the findings at trial the present is just such a case. 20. For the reasons developed below, the submission of the plaintiffs should be accepted, and the appeal respecting passing-off disposed of accordingly. However, as intimated earlier, the aspect of “dilution” merits further discussion as a matter of legal principle. Question 6, referred to above at para 6 asks whether sufficient damage for passing-off is sustained by “mere potential dilution of a trade mark”. 21. A proposed development by reference to “dilution” of the tort of passing-off towards a generalised tort of “unfair competition” elicits the warning by Dixon J in Victoria Park Racing and Recreation Grounds Co Ltd v Taylor[5]that in “British jurisdictions” courts of equity have not: “thrown the protection of an injunction around all the intangible elements of value, that is, value in exchange, which may flow from the exercise by an individual of his powers or resources whether in the organization of a business or undertaking or the use of ingenuity, knowledge, skill or labour. This is sufficiently evidenced by the history of the law of copyright and by the fact that the exclusive right to invention, trade marks, designs, trade name and reputation are dealt with in English law as special heads of protected interests and not under a wide generalization.” 22. In Star Industrial Co v Yap Kwee Kor[6], an appeal from Singapore, Lord Diplock emphasised that it had been settled by the speech of Lord Parker in A G Spalding & Bros v A W Gamage Ltd[7]that passing-off provides a remedy not based in fraud as understood at common law or in “property” in an unregistered trade mark, but upon protection from “invasion” of that property of the plaintiff, being the goodwill likely to be injured by the misrepresentation made by the defendant. 23. In GE Trade Mark[8] and Erven Warnink B V v J Townend & Sons (Hull) Ltd[9] (“the Advocaat Case”) Lord Diplock emphasised the accommodation and adjustment here of three interests. First, there is the interest of the plaintiff in protecting the commercial advantage or “goodwill” flowing from its recognition and reputation with customers and prospective customers, secondly there is the interest of the defendant in attracting custom by what to it appears an effective means, and thirdly, the interest of consumers and potential consumers in selecting goods and services without the practise upon them of misrepresentation as to the provenance of the defendant’s goods or services. 24. It follows that to adopt the threat of “dilution” as sufficient for the plaintiff to obtain a remedy to restrain the defendant, in the absence of a misrepresentation to customers, would disturb the accommodation between the three interests by removing the third element from the equation. 25. However, accepting any such removal to be fatal, there remain for passing-off as properly understood questions of what amounts to an actionable “invasion” of “goodwill” by the making of a “misrepresentation”. 26. It is a truism that “goodwill” is a term perhaps insusceptible of a satisfactory and comprehensive definition for all legal purposes. The matter is complicated by the consideration that “goodwill” is also an accounting and business term and accounting and business concepts have been influential in the valuation of goodwill for revenue and other legal purposes. However, at a general level, it has been said that, while goodwill includes whatever adds value to a business, different businesses derive their value from different considerations[10]. The means by which an extensive reputation and goodwill can be turned to account in contemporary circumstances may include licensing and franchising, a matter discussed by Sir Richard Scott V-C in DawnayDay & Co Ltd v Cantor Fitzgerald International[11]. In the present context the concern is with the reputation of the plaintiffs which they may turn to account in cognate fields of business such as the operation of tea salons and other retail outlets; this reputation may be abused by another trader, whose present or threatened activities in those fields are apt to mislead customers. 27. As to the nature of the “misrepresentation”, guidance also is given by Lord Diplock in the Advocaat Case. The formulation of principle in Spalding had proceeded from the formulation which fixed upon misrepresentations that the goods of the defendant were those emanating from the plaintiff and extended that formulation to include a misrepresentation that one class of the plaintiff’s goods (the old line of “Orb” footballs sold for 3 shillings) was another class of the plaintiff’s goods (the improved “Orb” football sold for 10/6). In Advocaat, following upon the Spanish Champagne Cases[12] , Lord Diplock approved the extension of passing-off doctrine so as to include misrepresentations as to the particular character or composition of products marketed by persons including the plaintiff. But, as both Peter Gibson LJ and Sir Thomas Bingham MR later emphasised in the “Elderflower Champagne” Case[13], in this line of cases the issue of deception or likely deception of consumers remains an element of the action. 28. In that regard, Deane J[14] referred to the decision in the Advocaat Case as an instance of: “the adaptation of the traditional doctrine of passing off to meet new circumstances involving the deceptive or confusing use of names, descriptive terms or other indicia to persuade purchasers or customers to believe that goods or services have an association, quality or endorsement which belongs or would belong to goods or services of, or associated with, another or others … .” 29. Four points are to be made here. The first is that these remarks provide no support for an expansion of the tort to encompass “dilution” without deception or confusion of customers and consequent damage to goodwill. The second point is that as long ago as 1929, in Turner v General Motors (Aust)[15] Dixon J remarked: “[B]usiness reputation, in the view of the Courts of Equity, is a right of property which should be protected from misappropriation and that protection is not confined to cases where loss simply consists in the diversion of trade from one to another” As an example, Dixon J referred to the decision of the Court of Appeal in Lloyd’s v Lloyd’s (Southampton) Ltd[16], where the defendant carried on the business of ship brokers, not maritime insurers; in his judgment in Turner[17], Isaacs J referred to “The Times” bicycle case, Walter v Ashton[18]. The third point is that the findings at trial in the present case bring it well within this group of passing-off cases. 30. The fourth point is that the reasoning in the continued line of cases in this field does not depend upon a fraudulent design by the defendant or upon the goods or services of the defendant being of inferior or sub-standard quality. This is confirmed by decisions such as that of Whitford J when granting an interlocutory injunction in LRC International Limited v Lilla Edets Sales Company Ltd[19]. His Lordship found that there would be undoubted confusion were the defendants to market toilet tissues under the name “Marigold”, given the admitted reputation of the plaintiff in “Marigold” for household gloves and plastic pants for babies. It was “almost impossible to imagine that confusion would not in fact result”, given that the fields of business of the parties “are likely to overlap”, and the injunction should be granted even on the basis that the defendant had adopted “Marigold” for its product “in all innocence”. 31. Enough has been said in these reasons to show that in this area of discourse the term “dilution” has been used in several senses. However, it is inapt when used to describe the apprehended injury to the goodwill of the plaintiff by activities of the defendant in related or cognate fields to those in which the plaintiff is engaged and in respect of which its reputation extends, with the consequent scope for misrepresentation to consumers in those fields. Here there is not the “intellectual difficulty” to which Millett LJ referred in Harrods Ltd v Harrodian School Ltd[20], of recognising as sufficient for passing-off damage which does not depend upon confusion of customers. 32. It is convenient now to return to that aspect of the appeal concerning the sufficiency of “dilution” without deception or confusion of customers or the apprehension of that result from threatened conduct of the defendant. 33. The term “dilution” appears in the United States to have been used in a different sense to that just discussed. That has been to describe the gradual whittling away or dispersion of the identity and hold upon the public mind of the trade mark by its use upon non-competitive goods, which has led to unfair competition by other traders, albeit without the likelihood of consumers being confused or deceived[21]. A United States scholar recently described “dilution” as “one of the great mysteries of trademark law” which “judges have had trouble understanding… and scholars have had difficulty in justifying…”; the theory supposes that if, say, TIFFANY is used on soap, motor cars and other diverse goods, while this may not confuse the provenance of those goods with the jewellery company, it becomes more difficult for consumers quickly to recall the original product[22]. 34. It is important to note that these ideas have found expression not in development across the various United States common law jurisdictions but in statute, in particular in federal legislation, the Trademark Dilution Act 1995 and the Trademark Dilution Revision Act 2006. In the United Kingdom, s 10(3) of the Trade Marks Act 1994 (UK) (“the 1994 UK Act”) protects registered marks which are “well known” in the United Kingdom against use by the defendant of an identical or similar mark which is “without due cause” and “takes unfair advantage of, or is detrimental to the distinctive character or repute” of the registered mark. In Hong Kong, similar provision is made in s 18(4) of the TMO. There also is provision in the TMO (s 60) for registration of defensive trade marks which have become “exceptionally well known in Hong Kong” such that use in relation to other goods or services would be likely to detract from their distinctive character. In the present litigation, the plaintiffs have no defensive registration and for infringement there is no reliance by the plaintiffs upon s 18(4); they rely upon the more conventional provision in s 18(3), to which further reference will be made. 35. Should some such notion be adopted by the common law with respect to unregistered marks? As early as 1996 fears were expressed that English courts were inclining towards acceptance of mere harm to the trade value in the plaintiff’s common law mark as sufficient to found a passing-off action, thereby focussing on dilution alone of the commercial interests of the plaintiff, and discarding of the interest of consumers in not being confused or deceived[23]. As indicated above at para [17], in the United Kingdom that continued apprehension, for which the appellants referred to the discussion by Professor Wadlow[24], may be no more than that. 36. Whatever the state of the case law in England, there is no occasion in Hong Kong to introduce such a concept of “dilution” into the common law tort of passing-off. It is true that in the Advocaat Case[25] Lord Diplock supported the development of principles of passing-off in that case on the footing that the common law “ought to proceed upon a parallel rather than a diverging course” from what “over a period of years … a steady trend in legislation” in a particular field reflects as the demand of the public interest. 37. But development, in Advocaat itself, is one thing and distorting by removing a fundamental element of the tort would be another. Such a step would throw into question the saving provision in s 10(3) of the TMO that nothing in that legislation “affects the law relating to passing-off”. Further, the complex inter-action between exclusive or monopoly rights conferred by the various species of intellectual property and statutory competition law concerning control of market power[26] is a matter of legislative concern rather than for the judicial branch. Trade mark Infringement 38. There remain the issues on the appeal which concern trademark infringement, and the particular infringement provision made by s 18(3) of the TMO. 39. Section 3(1) of the TMO provides that: “a ‘trade mark’ (商標) means any sign which is capable of distinguishing the goods or services of one undertaking from those of other undertakings and which is capable of being represented graphically.” It has been noted earlier in these reasons that passing-off protects goodwill not the mark or device per se as a species of property. However, trade marks are, by virtue of registration under the TMO, “personal property” (s 27(1)). This property may be assigned or otherwise transmitted without goodwill of a business (s 27(2)). Registration of their trade marks gives the plaintiffs exclusive rights which are infringed by use in Hong Kong of the trade marks without consent (s 14(1)). Nevertheless, the “personal property” in a registered mark in a sense is subordinate to the common law; a trade mark shall not be registered if its use is liable to be prevented at the suit of another by virtue of the law of passing-off (s 12(5)). Further, s 21 makes special provision to qualify the operation of the infringement with respect to “honest practices” by defendants in comparative advertising. There is no reliance by the defendants upon s 21. 40. Subsections (1), (2), (3) of s 18 disclose three steps or grades in infringement. There are variations between the requisite degree of resemblance between the registered mark and the sign used by the defendant and between the goods or services for which the mark is registered and those in relation to which the sign is used. First, there is the plain case of identity between the mark, the sign, and the goods or services (s 18(1)). Secondly, while the sign is “identical” to the trade mark it is used by the defendant only on goods or services which are “similar” to those of the registration; there is infringement provided such use “is likely to cause confusion on the part of the public” (s 18(2)). Neither of these provisions is in play in the present litigation. 41. Thirdly, and this is the provision upon which the plaintiffs rely, the trade mark and the sign are “similar”, the goods or services of the defendant are “identical or similar” to those for which the mark is registered, and the “use” by the defendant “is likely to cause confusion on the part of the public” (s 18(3)). 42. A defendant “uses” the sign, in particular, and among other activities, by applying it to goods or their packaging, by offering or exposing goods for sale or putting them “on the market” “under the sign”, and by stocking goods “under the sign” for any of those purposes, and by offering or supplying services under the sign, and by using the sign on business papers or in advertising (s 18(5)). A sign is taken to be “applied” to goods if it is “in any manner marked on or incorporated with [the goods]” (s 2(2)). 43. As will appear, there was debate on the present appeal particularly respecting the nature and extent of the requirements in s 18(3) of the similarity of the registered mark and the offending sign, and the likelihood of confusion on the part of the public. At this stage, it is convenient to refer to what appeared to be common ground respecting likelihood of confusion. “Is likely to cause confusion” 44. The phrase “is likely to cause confusion” appeared in s 11 of the Trade Marks Act 1938 (UK) (“the 1938 UK Act”) and is now found in the TMO both with respect to refusal of registration (s 12) and infringement (s 18). With respect to s 11 of the 1938 UK Act, a registration provision, the analysis by Romer J in In Jellinek’s Application[27] proved influential both in the United Kingdom and in other jurisdictions[28]. His Lordship’s analysis supports the proposition with respect to s 18 that while a mere possibility of confusion is not enough, it is sufficient if the result of use by the defendant of the sign in question will be that a number of ordinary persons will entertain a reasonable doubt and be caused to wonder whether it might not be the case that the goods or services in respect of which the defendant’s sign is used have the same provenance as those in respect of which the trade mark is used. 45. In considering this likelihood of confusion on the part of the public by use of a defendant’s sign, s 7(2) of the TMO requires the Court to take into account “all factors relevant in the circumstances”. Uncontroversially, these will include the character of purchasers of the defendant’s goods or services, and the methods by which the goods or services are marketed. 46. The appellants take no issue with the well-known passage from De Cordova v Vick Chemical Coy[29] in which Lord Radcliffe, on behalf of the Privy Council, said: “The likelihood of confusion or deception in such cases is not disproved by placing the two marks side by side and demonstrating how small is the chance of error in any customer who places his order for goods with both the marks clearly before him, for orders are not placed, or are often not placed, under such conditions. It is more useful to observe that in most persons the eye is not an accurate recorder of visual detail, and that marks are remembered rather by general impressions or by some significant detail than by any photographic recollection of the whole” 47. The appellants did, however, emphasise that so far as relevant to s 18(3) of the TMO, these remarks are directed to the issue of confusion of the public, not to similarity of the mark and the sign, as the appellants submitted was the approach of the trial judge. However, there appeared to be no dispute on this appeal respecting the statement by Kitchin LJ in Specsavers International Healthcare Ltd v ASDA Stores[30]. This was accepted by the Court of Appeal in the present case as applicable to s 18 of the TMO. In Specsavers Kitchin LJ adopted the following as a useful and accurate summary of the approach to be taken in assessing the requirement of likelihood of confusion: “(a) the likelihood of confusion must be appreciated globally, taking account of all relevant factors; (b) the matter must be judged through the eyes of the average consumer of the goods or services in question, who is deemed to be reasonably well informed and reasonably circumspect and observant, but who rarely has the chance to make direct comparisons between marks and must instead rely upon the imperfect picture of them he has kept in his mind, and whose attention varies according to the category of goods or services in question; (c) the average consumer normally perceives a mark as a whole and does not proceed to analyse its various details; (d) the visual, aural and conceptual similarities of the marks must normally be assessed by reference to the overall impressions created by the marks bearing in mind their distinctive and dominant components, but it is only when all other components of a complex mark are negligible that it is permissible to make the comparison solely on the basis of the dominant elements; (e) nevertheless, the overall impression conveyed to the public by a composite trade mark may, in certain circumstances, be dominated by one or more of its components; (f) and beyond the usual case, where the overall impression created by a mark depends heavily on the dominant features of the mark, it is quite possible that in a particular case an element corresponding to an earlier trade mark may retain an independent distinctive role in a composite mark, without necessarily constituting a dominant element of that mark; (g) a lesser degree of similarity between the goods or services may be offset by a greater degree of similarity between the marks, and vice versa; (h) there is a greater likelihood of confusion where the earlier mark has a highly distinctive character, either per se or because of the use that has been made of it; (i) mere association, in the strict sense that the later mark brings the earlier mark to mind, is not sufficient; (j) the reputation of a mark does not give grounds for presuming a likelihood of confusion simply because of a likelihood of association in the strict sense; and (k) if the association between the marks causes the public to wrongly believe that the respective goods [or services] come from the same or economically-linked undertakings, there is a likelihood of confusion.” (Emphasis added) 48. It is convenient now to turn to the relief granted at trial and in the Court of Appeal and then to the five trade mark issues on which leave to appeal has been given to this Court. The Registrations and Injunctions 49. The second plaintiff, a member of the TWG Group, is registered as owner of trade marks 300635463 and 300655470 in respect of goods including coffee, tea and sugar (“the Registered Marks”). The validity of the registrations is not in issue. Each registration is of two marks. Section 51 of the TMO authorises the making by the Registrar respecting the registration of “a series of trade marks”, being “a number of trade marks which resemble each other as to their material particulars and differ only as to matters of a non-distinctive character not substantially affecting the identity of the trade mark” (s 51(3)). 50. Representations of the Registered Marks appear respectively as Annexure 3 and 4 to these reasons. Each mark contains overlapping ovals adjacent to the letters “TWG”. In mark “A” of each Registered Mark there are claimed as elements yellow (pantone 109C), brown (pantone 469C), orange (pantone 151C) and white. There is no claim to colour in respect of mark “B” in either of the Registered Marks. 51. The defendants and each of them are presently restrained from infringing any of the Registered Marks. Without prejudice to the generality of that order, they also are restrained from using in the course of trade, in relation to goods or services identical or similar to the category of goods in respect of which the Registered Marks are registered, any of the signs “TWG”, “TWG Tea”, the Cartouche signs and the Balloon signs. 52. The Cartouche signs prominently display “TWG TEA”. They also show “1837”. That date was selected, on the defendants’ evidence, to “celebrate” in 2008 the year in which the East India Company lost its tea trading monopoly and the Chamber of Commerce was established in Singapore. However, the trial judge held that the real intention is choosing the date was to give to customers the false impression that the defendants’ business had existed since 1837. 53. The defendants seek to have the injunctions discharged on the basis that the Court of Appeal, as had the trial judge, erred in law in its construction and application of s 18(3) of the TMO. By order made by this Court on 20 May 2015 leave to appeal was granted in respect of trade mark issues limited to five questions. The Five Trade Mark Questions 54. These are sufficiently indicated by the following: Question 1 - What is the correct construction of s 18(3)? Question 2 - The significance of “TWG” as a feature of the Registered Marks and defendants’ signs - is it a “dominant” feature? Question 3 - The significance of “TWG” compared with other features including figurative features of the Registered Marks and defendants’ signs - do words “generally speak louder than devices”? Question 4 - The role of the claim to colours in the Registered Marks – is a mark registered in black and white registered for all colours? Question 5 - The role of colour in the series registrations of the Registered Marks – does it affect the distinctive character of the series? 55. In Counsel’s oral submissions these questions were somewhat recast. Argument was developed by submitting that in considering s 18(3) the trial judge and the Court of Appeal fell into error of law in these respects: (1) by failing to follow the requisite “step by step” approach in considering the application of the components of s 18(3), namely “similarity” followed by “likelihood of confusion”. (2) by misunderstanding the significance for s 18(3) of passages in cases dealing with earlier legislation in which it had been said to be enough for infringement that the defendant uses in connection with its goods the “essential” or “dominating” feature of the registered mark, and by that misunderstanding (a) discounting the significance in assessing the Registered Marks of colour and of the features other than the letters TWG, and (b) discounting the visual elements in the defendants’ Cartouche signs and Balloon signs. (3) by placing too much emphasis upon aural rather than visual comparison between the Registered Marks on the one hand and the Cartouche signs and the Balloon signs on the other. International arrangements 56. Before turning specifically to the text of s 18(3) and the issues of construction which arise, attention first should be given to any relevant international arrangements. Article 16(1) of the 1994 Agreement on Trade-Related Aspects of Intellectual Property Rights (“the TRIPS Agreement”) states: “The owner of a registered trademark shall have the exclusive right to prevent all third parties not having the owner’s consent from using in the course of trade identical or similar signs for goods or services which are identical or similar to those in respect of which the trademark is registered where such use would result in a likelihood of confusion. In case of the use of an identical sign for identical goods or services, a likelihood of confusion shall be presumed. The rights described above shall not prejudice any existing prior rights, nor shall they affect the possibility of Members making rights available on the basis of use.” (emphasis added) 57. Hong Kong acceded to TRIPS with effect 1 January 1995, as did the European Union and the United Kingdom. If there are several reasonably possible interpretations of a statutory provision such as s 18(3), the Court should favour that which is consistent with the international obligation found in Art 16(1) of TRIPS[31]. Section 27(1) of the 1954 Ordinance as then in force was amended by the Intellectual Property (World Trade Organisation Amendments) Ordinance No 11 of 1996 (“the 1996 Ordinance”), evidently to reflect commitment of Hong Kong under art 16(1) of TRIPS. As amended, s 27(1) spoke of infringing use by a person who “uses in the course of trade a trade mark identical with or nearly resembling [the trade mark], in relation to any goods in respect it is registered, or in relation to goods of the same description, where such use would result in the likelihood of confusion”; the phrase “nearly resembling” meant “a resemblance so used as to be likely to deceive or cause confusion” (s 2(4)) (emphasis added). 58. Article 5(1) of the Directive of the Council of the European Communities of 21 December 1988 to “approximate the trade mark laws of Member States” (“the EU Directive”) stated: “The registered trade mark shall confer on the proprietor exclusive rights therein. The proprietor shall be entitled to prevent all third parties not having his consent from using in the course of trade: (a) any sign which is identical with the trade mark in relation to goods or services which are identical with those for which the trade mark is registered; (b) any sign where, because of its identity with, or similarity to, the trade mark and the identity or similarity of the goods or services covered by the trade mark and the sign, there exists a likelihood of confusion on the part of the public which includes the likelihood of associa­tion between the sign and the trade mark.” (emphasis added) The appellants point out that the European Court of Justice (“the ECJ”) requires the EU Directive to be interpreted in conformity with TRIPS[32] and obliges national laws to be interpreted by national courts as far as possible in conformity with the EU Directive as it is interpreted by the ECJ. 59. The preamble to the 1994 UK Act states that it is made to implement the EU Directive. Section 10(2) of the 1994 UK Act came into force on 31 October 1994, that is to say, shortly before the United Kingdom acceded to TRIPS. It states: “(2) A person infringes a registered trade mark if he uses in the course of trade a sign where because— (a) the sign is identical with the trade mark and is used in relation to goods or services similar to those for which the trade mark is registered, or (b) the sign is similar to the trade mark and is used in relation to goods or services identical with or similar to those for which the trade mark is registered, there exists a likelihood of confusion on the part of the public, which includes the likelihood of association with the trade mark.” (emphasis added) Section 18(3) of the TMO 60. Against this background, the text of s 18(3) of the TMO is as follows: “18(3) A person infringes a registered mark if – (a) he uses in the course of trade or business a sign which is similar to the trade mark in relation to goods or services which are identical or similar to those for which it is registered; and (b) the use of the sign in relation to those goods or services is likely to cause confusion on the part of the public” (emphasis added) 61. The identification in s 18(3) of the defendant by the singular masculine pronoun is to be read as importing the plural and the feminine and neuter genders: Interpretation and General Clauses Ordinance, Cap. 1 s 7. 62. Several points arise from consideration of Article 16(1) of TRIPS and its apparent influence upon Art 5(1) of the EU Directive, s 10(2) of the 1994 UK Act, and, in Hong Kong, the 1996 Ordinance. 63. First, there is in Art 16(1) of the TRIPS and its derivatives the expansion of the scope of trade mark protection beyond the use of a mark upon identical goods or services to use of “similar” signs for “similar” goods or services. Secondly, this expansion of trade mark protection has the qualification that the use by the defendant would be likely to result in confusion. In Re Ping An Securities Ltd[33] the Court noted that this may bring trade mark law closer to the requirement in passing-off to consider the likelihood of false representation or deception in all the circumstances, but, as in the present case, there was no occasion to pursue the matter further. 64. Unlike Art 5(1) of the EU Directive, s 10(2) of the 1994 UK Act, and the 1996 Ordinance, s 18 of the TMO, while teasing out into sub ss (1), (2) and (3) the somewhat compressed language of Art 16(1) of TRIPS, does not in plain terms causally link para (a) and (b) of s 18(3). It uses the coordinate “and” rather than a word such as “because” or “result” which indicates unequivocally that it is the similarity required by para (a) which renders the use likely to cause the confusion specified in para (b). 65. However, the coordinate “and” may also be employed in a cumulative and causal sense. Given the text of Art 16(1) of TRIPS and, in particular, its evident influence upon the 1996 Ordinance, it should be accepted that it is this interpretation which should be given to the present s 18(3) of the TMO. The Appellants’ Submissions Considered 66. To accept the construction of s 18(3) given above in para [65] is not to deny that the appellants are correct is submitting that the steps in para (a) and (b) are distinct and are not to be elided. However, the issue on this appeal is whether, having regard to the way in which the appellants presented their case, the appellants are correct in the further submission that the trial judge and the Court of Appeal erred in failing to consider the issue of similarity between the Cartouche signs and the Balloon signs on the one hand and the Registered Marks on the other, before, having answered that affirmatively, turning to consider the likelihood of confusion. 67. In considering alleged failures by the trial judge and the Court of Appeal in this respect, regard must be had to the issue presented by para 32(a) of the Re-Amended Defence. This averred not that the Cartouche signs and the Balloon signs were not “similar” but that they were not “confusingly similar” to the Registered Marks. The respondents in their written submissions in this appeal, demonstrate by reference to the submissions made at trial and to the Court of Appeal, that it was never contended by the appellants that the absence of similarity was such that para (a) of s 18(3) was not satisfied and there was thus no occasion to consider para (b); rather their case was that the level of similarity, coupled with the context of use, was that there was no likelihood of confusion. This would be sufficient to dispose of the alleged appeal point. 68. However, given the general importance of the case, it is appropriate to consider a submission of law which the appellants make. This is that in considering “similarity” the task is not to locate the respective “dominant” or “essential” features of the registered trade mark and the signs of the defendant, disregarding the significance of the other matters. That, as will appear, may be accepted, but, as will further appear, there was no error in this respect by the approaches to the evidence by the trial judge and the Court of Appeal. 69. The appellants point to the passages in which the trial judge and the Court of Appeal set out para (5) of the statement of principles respecting s 10(2) of the 1994 UK Act by Pumfrey J in Decon Laboratories Ltd v Fred Baker Scientific Ltd[34]. This reads: “As [Sir Wilfred] Greene [MR] said in Saville Perfumery Ltd v June Perfect Ltd (1941) 58 RPC 147, the statutory protection ‘is absolute in the sense that once a mark is shown to offend, the user of it cannot escape by showing that by something outside the actual mark itself he has distinguished his goods from those of the registered proprietor’. If the word ‘mark’ in this phrase is substituted with the word ‘sign’ then this is a statement of the modern law.” 70. However, as the appellants submit, “the statutory protection” of which the Master of the Rolls spoke in Saville Perfumery was that under s 4(1) of the 1938 UK Act. This asked whether the mark used by the defendant “so nearly resembled” the registered mark “as to be likely to cause confusion”. Nevertheless, s 4(1) was interpreted by his Lordship[35] as posing an issue “to be considered by reference not only to the whole mark, but also to its distinguishing or essential features, if any”, and he added that “[in] deciding whether or not a feature is of this class, not only ocular examination, but the evidence of what happens in practice in the particular trade is admissible.” It may be noted that after Saville Perfumery, in De Cordova v Vick Chemical Company [36], Lord Radcliffe said that “a mark is infringed by another trader if even without using the whole of it upon or in connection with his goods, he uses one or more of its essential features” and that the identification of such an essential feature “depends partly on the Court’s own judgment and partly on the burden of evidence that is placed before it.” 71. Whatever the significance to be attached today to Saville Perfumery and Vick, the appellants no doubt are correct that s 18(3) of the TMO provides a statutory text fuller and more nuanced than that of s 4(1) of the 1938 UK Act. However, when applying s 18(3) it cannot be erroneous in assessing “similarity” to consider if there be any striking features of the mark or sign which appear “essential” or “dominant”, but doing so without disregarding the entirety of the mark or sign or stripping it of its context, including evidence of what happens in the particular trade. It is in this sense that Kitchin LJ used the term “dominant” and “dominated” in propositions (d), (e) and (f) of the passage in Specsavers[37] which has been set out in para [47] of these reasons and which the appellants did not dispute. His Lordship was primarily dealing with likelihood of confusion, but it is apparent from the text of propositions (d), (e) and (f) that he also addressed the issue of similarity. 72. The adoption by Pumfrey J in Decon Laboratories[38] of the term “absolute” from Saville Perfumery should be understood in the sense explained above in paras 70 and 71 when construing s 18(3) of the TMO. However, the Court of Appeal gave primary weight to the matters listed in propositions (d), (e) and (f) of Specsavers and taking them as a whole, saw no objection to the identification by the primary judge of the dominant features in the marks and signs so long as he did not disregard their other features in them. 73. This was the manner in which the trial judge had proceeded. He said that the questions of similarity between a registered mark and the allegedly offending sign were closely related to the question of likelihood of confusion, and that, while counsel had dealt with the two issues separately, substantially the same authorities had been cited in respect of the two issues [para 87]. With specific reference to para (a) of s 18(3), the trial judge dealt with “The assessment of similarities” [paras 103-106] and under the heading “Discussion of similarity” [paras 107-114] concluded in para 114 that TWG being the “dominant feature” of the Registered Marks and the Cartouche signs they were “similar” within the meaning of s 18(3)(a). 74. With respect to the Balloon signs, the Court of Appeal added [para 99], using the phrase singular “balloon mark” to identify the Balloon signs: “On the evidence before us, the balloon mark was used by the Defendants in the sale of tea sets which were packed in boxes bearing the cartouche mark. Though the letters “TWG” in the balloon mark is not visually dominant, as submitted by Mr Platts-Mills [counsel for the plaintiffs], this is the only element in the mark which the public is likely to identify with the source of the products. This is particularly so when the actual use of the mark is considered. There is simply no suggestion that the balloon mark has been used on its own and the sale outlet, as we have seen from the photographs, is predominantly identified as a TWG tea boutique. Thus, an average customer is taught by the actual use by the Defendants to place great significance on the “TWG” aspect of the balloon mark.” 75. The trial judge went on [paras 115-132] to “The assessment of the likelihood of confusion”. He said that he was required to view “the matter globally” but was satisfied that “the essence” of the Registered Marks was “TWG” and that its use as the “dominant feature” by the defendants would inevitably lead to confusion, that the different “get up” of the Registered Trade Marks and of the defendants’ signs were not sufficient to prevent confusion, and that differing marketing techniques of the parties, which the defendants had emphasised, were insufficient to deny likelihood of confusion. The trial judge then [para 133] expressed his conclusion in terms indicating that both paras of s 18(3) were satisfied. 76. The appellants complain that in assessing the Registered Marks the trial judge and the Court of Appeal discounted the presence of colours in the Registered Marks when assessing TWG as the essential or dominating feature. However, the respondents properly rely upon what was said in para 49 of the reasons of the Court of Appeal: “[The] coloured versions of the Plaintiffs’ marks were registered together with corresponding monochrome versions as [a] series. In the definition for “series of trade marks” under s51(3) of the TMO, registration of a series is permitted when a number of trade marks resemble each other as to their material particulars and differ only as to matters of a non-distinctive character not substantially affecting the identity of the trade mark. In other words, the colour scheme adopted in the coloured version (which is the only difference between the two versions) is regarded as a matter of non-distinctive character which does not substantially affect the identity of the trade mark.” 77. The upshot is that submissions (1) and (2) of the appellants’ submissions which have been set out above in para 55 must be rejected. 78. There remains submission (3), that too much emphasis was placed upon aural comparison between the Registered Marks and the Cartouche signs and the Balloon signs, at the expense of visual comparison. It may well be true that purchasers of the plaintiffs’ products in the less elaborate surroundings to the defendants’ tea salon of a supermarket are likely to buy goods without making an oral request to a sales assistant and to rely upon visual inspection of what is presented on the supermarket shelves. Further, to place an order on line at the site “TWG Tea.com” requires exercise of the visual not aural senses. Emphasis in some of the older cases to aural comparison may reflect retailing methods of a past age. Nevertheless, with respect to the Balloon signs the Court of Appeal clearly was focused upon visual responses in the passage set out above in para 76. Further, the Court of Appeal concluded that [para 90]: “[Having] regard to the emphasis placed by the Defendants on the letters “TWG” in the overall décor and packaging of the salon and products, even on a visual level, there is considerable similarity with the Plaintiffs’ marks when they are compared as a whole.” (emphasis added) Conclusions 79. The appeal should be dismissed. There should be an order nisi that the respondents are to have the costs of the appeal, to be taxed unless agreed. Should any party wish to have a different order for costs, written submissions should be served on the other party and lodged with the Court within 14 days of the handing down of this Judgment, with liberty on the other party to serve and lodge written submissions within 14 days thereafter. In the absence of such submissions, the order nisi will stand absolute at the expiry of the time limited for these submissions. Chief Justice Ma: 80. For the above reasons, the appeal is dismissed. The Court also makes an order nisi as to costs as set out in para 79 above. (Joseph Fok)				Permanent Judge (William Gummow)				Non-Permanent Judge Mr Martin Howe QC and Mr Douglas Clark, instructed by Hogan Lovells, for the defendants (appellants) Mr Mark Platts-Mills QC, Ms Winnie Tam SC and Mr Philips BF Wong, instructed by Deacons, for the plaintiffs (respondents) Annexure 1 Annexure 2 Annexure 3 Annexure 4 [1] [2013] HKEC 1146. [2] Tsit Wing (Hong Kong) Co Ltd v TWG Tea Co Pte Ltd [2015] 1 HKLRD 414. [3] Irvine v Talksport Ltd [2002] 1 WLR 2355 at 2366-2368 [34]–[38]; Och-Ziff Management Europe Ltd & Anor v Och Capital LLP [2011] FSR 289 at 334-335, [159] – [160]. [4] TaittingerSA v Allbev Ltd [1993] FSR 641 at 674; Harrods Ltd v Harrodian School Ltd [1996] RPC 697 at 724; Dawnay Day & Co Ltd v Cantor Fitzgerald International [2000] RPC 669 at 705-706. [5] (1937) 58 CLR 479 at 509. This passage was adopted by the whole Court in Moorgate Tobacco Co Ltd v Philip Morris Ltd (1984) 156 CLR 414 at 444-445. [6] [1976] FSR 256 at 269. See also Reckitt & Colman Products Ltd v Borden Inc [1990] 1 WLR 491 at 499-500 per Lord Oliver of Aylmerton; Re Ping An Securities Ltd (2009) 12 HKCFAR 808 at 816-817 [17] per Gault NPJ. [7] (1915) 32 RPC 273. Lord Parker’s emphasis upon the protection of goodwill by the passing-off action derived support from Chancery cases probably beginning with Millington v Fox (1838) 3 My & Cr 338 at 352 [40 ER 956 at 961]: GE Trade Mark [1973] RPC 297 at 325-326; Tregoning “What’s In a Name? Goodwill In Early Passing-Off Cases” (2008) 34 Monash U L Rev 75 at 92-96. [8] [1973] RPC 297 at 326. [9] [1979] AC 731 at 740-743. [10] Box v The Federal Commissioner of Taxation (1952) 86 CLR 387 at 397 Dixon CJ, Williams, Fullagar, Kitto JJ. [11] [2000] RPC 669 at 701-704. [12] J Bollinger v Costa Brava Wine Co Ltd [1960] Ch 262; J Bollinger vCosta Brava Wine Co Ltd [No 2] [1961] 1 WLR 277. [13] Taittinger SA v Allbev Ltd [1993] FSR 641 at 668, 677. [14] Moorgate Tobacco Co Ltd v Philip Morris [No. 2] (1984) 156 CLR 414 at 445. This passage was adopted by Laddie J in Irvine v Talksport Ltd [2002] 1 WLR 2355 at 2364 [27]. [15] [1929] 42 CLR 352 at 368. The case concerned the conduct by the defendants of a second hand car retail business under the name “General Motors”. [16] (1912) 29 RPC 433. [17] (1929) 42 CLR 352 at 362. [18] [1902] 2 Ch 282. [19] [1973] RPC 560 at 563-564. [20] [1996] RPC 697 at 716. [21] Schechter, “The Rational Basis of Trademark Protection” (1927) 40 Harvard Law Review 813 at 831. [22] Bone “Schechter’s Ideas in Historical Context and Dilution’s Rocky Road” (2008) 24 Santa Clara Computer and High Technology Law Journal 1 at 5. [23] Carty “Dilution and Passing-Off: Cause for Concern” (1996) 112 Law Quarterly Review 632. [24] Wadlow “The Law of Passing-Off” 4th Ed, 2011, 4-048. [25] [1979] AC 731 at 743. [26] See, for example, Competition Ordinance, Cap 619, fully in force 14 December 2015. [27] (1946) 63 RPC 59 at 78. [28] For example in Southern Cross Refrigerating Co v Toowoomba Foundry Pty Ltd (1954) 91 CLR 592 at 594-595, 608. [29] (1951) 68 RPC 103 at 106. [30] [2012] FSR 555 at 574-575. [31] Hung Chan Wa v HKSAR [2005] 3 HKLRD 291 at 352 per Ma CJHC, Stuart–Moore VP, Stock JA. [32] Case C-245/02 Anheuser-Busch Inc v Budějovický Budvar, Narodni Podnik [2004] ECR 1-11018. [33] (2009) 12 HKCFAR 808 at 818 [19]. [34] [2001] RPC 293 at 299. [35] (1941) 58 RPC 147 at 162. [36] (1951) 68 RPC 103 at 105-106. [37] [2012] FSR 555 at 574 [52]. Kitchin LJ was considering a European Community trade mark and Article 9(1)(b) of a Council Regulation the terms of which were relevantly the same as Art 5(1)(b) of the EU Directive, and thus were linked to Art 16(1) of TRIPS. [38] [2001] RPC 293 at 329. 
Chief Justice Ma: 1. This appeal[1] by the defendant in the action was brought as of right,[2] the relevant matter directly in dispute (being real property, the second floor and roof of a house) having a value exceeding $1m. After hearing counsel for the appellant defendant[3] in the substantive appeal, and both counsel[4] on costs, the appeal was dismissed with costs. We indicated that the reasons for judgment would be handed down in due course. 2. I agree with the judgment of Mr Justice Tang PJ and that of Lord Phillips of Worth Matravers NPJ (which contains a summary of the legal analysis in this appeal). This was ultimately a simple appeal to resolve. The effect of the various agreements between the parties can be distilled into the following: the plaintiff contracted with the defendant to build at his own expense a three-storey house on the defendant’s land in consideration of the assignment to him (the plaintiff) of the second floor and roof of that house (“the Property”). The assignment was, however, subject to an option[5] exercisable by the defendant to purchase from the plaintiff his interest in the Property. Even if the option had been properly exercised by the defendant in the present case,[6] thereby giving rise to an obligation on his part to purchase the Property, this fell through when the defendant repudiated this obligation and the repudiation was accepted by the plaintiff. The plaintiff was then left with his contractual entitlement to have the Property assigned to him. The defendant’s suggestion that once the option to purchase was exercised by the defendant, the obligation to assign the Property to the plaintiff had somehow been extinguished, was simply wrong. The plaintiff was fully entitled to have this obligation enforced. Mr Justice Ribeiro PJ: 3. I agree with the Reasons of The Chief Justice, Mr Justice Tang PJ and Lord Phillips of Worth Matravers NPJ. Mr Justice Tang PJ: 4. The plaintiff agreed to build a village house on land[7] owned by the defendant. In return the defendant agreed to assign the 2nd floor and the roof of the house(“the Property”) to the plaintiff. [8] For that purpose, they entered into a series of agreements in Chinese. The first in time were a Joint Development House Splitting Agreement and a Supplemental Joint Development House Splitting Agreement both dated 6 June 2006.[9] However, by these agreements, the plaintiffin turn granted to the defendant a pre-emptive right to purchase the Property. The relevant clause in the 6 June agreement read: “Both parties agree that if Party A (the plaintiff) shall sell the 2nd Floor and the Roof assigned to him, Party B(the defendant) has a pre-emptive right to purchase the same by giving written notice of his intention to purchase to Party A, within a month from the date of lodging application for the Occupation Permit. The purchase price shall be the price valued by (3 named banks), whichever is highest.” 5. The pre-emptive right was modified by the 6 June supplemental agreement which provided that if the defendant was not satisfied with the valuations provided by any of the named banks: “…Party B is entitled to jointly appoint a valuer with Party A to revalue the house, …” 6. Sometime later, and it does not matter exactly when, the parties entered into a signed and witnessed by one Lau Pak On “Joint Development House Splitting Agreement” which was dated 2006[10]and a Supplemental Joint Development House Splitting Agreement dated5 August 2006.[11] 7. The 2006 agreementcontained provisions from the 6 Juneagreements[12] but only Clause III.5 of the 2006 agreement is important to this appeal. It provided: “If Party B needs to re-purchase the property title owned by Party A, the purchase price shall be the highest of the valuations made by 3 property valuers. But Party B shall communicate his decision to Party A within 30 days after the issuance of the Certificate of Compliance… (The names of three valuerswere then given.)” 8. In this action, the plaintiff claimed assignment of the Property pursuant to, inter alia, the 2006 agreement, and the defendant counter-claimed for an order that “the plaintiff do sell [the Property] to the Defendant at HK$2,440,000 …”. The Plaintiff succeeded both at first instance and in the Court of Appeal. Because the value of the Property exceeded $1,000,000, the defendant appealed to us as of right. We dismissed the appeal with costs at the conclusion of the hearing and my reasons for doing so are set out below. 9. At trial, one of the issues was whether Clause III.5of the 2006 agreement conferred on the defendant an option to purchase and not merely a pre-emptive right to purchase the Property[13]. As I shall endeavour to explain it does not matter whether the defendant had a right of pre-emption or an option. 10. The differences between an option and a right of pre-emption are well known but they do not matter in this case. It is sufficient to note that a right of pre-emption is commonly triggered by the grantor desiring to dispose of the subject property and informing the grantee of such desire.[14] In the case of an option, ordinarily, the exercise of the option does not depend on the volition of the grantor. However, as Templeman LJ (as he then was) said they share one common feature: “each prescribes circumstances in which the relationship between the owner of the property which is the subject of the right and the holder of the right will become the relationship of vendor and purchaser.” [15] 11. The 6 June agreement, required “written notice of (the defendant’s) intention to purchase” to be given “within one month from the date of lodging application for the Occupation Permit” but said nothing about any notice of desire to sell by the plaintiff. Be that as it may, it is clear that provided the plaintiff “shall sell”, the defendant was entitled to purchase at valuation. It may be it was expectedthat the plaintiff would sell and realize his profit as soon as he could[16]. Indeed, the plaintiff, using the name of the defendant, on 30 May 2007, entered into a Memorandum of Understanding to sell the Property to Fancy Spread Ltd, which was before the Certificate of Compliance was issued, and received a deposit of $320,000, being 10% of the purchase price.[17] 12. Under the 2006 agreement, the defendant had 30 days from the Certificate of Compliance to communicate his decision. The Certificate of Compliance was issued on 23 October 2007.[18] If the defendant had an option under the 2006 agreement, he could exercise it within the 30days. Even if the defendant only had a pre-emptive right, having regard to the plaintiff’s earlier decision(in May 2007) to sell the Property, it would have been triggered. 13. Thus, in my view, the defendant was entitled to purchase the Property provided he communicated his decision to do so in accordance with Clause III.5 of the 2006 agreement, whether he had an option to purchase or a right of pre-emption.[19] I turn to consider whether the defendant had exercised his right to do so. 14. The plaintiff gave a written notice dated 16 November 2007 to the defendant, notifying him that he could “reply within one month whether to re-purchase …”. The Court of Appeal noted that: “48. Although the Judge did not make any express finding, it is the plaintiff’s own evidence that the defendant’s mother told him on 16 December 2008 that the defendant would buy the Property.” On the basis of such evidence, the Court of Appeal held that there was a valid exercise of the option. With respect, this conclusion is not without problem. 15. We do not have a transcript of the plaintiff’s evidence. But from the chronology to the defendant’s skeleton arguments in the Court of Appeal, the plaintiff’s evidence was summarized as follows:[20] 16. This and other evidence showed that the plaintiff’s evidence that there was an exercise of the option on 16 December 2007 was nuanced.[21] Anyway, nothing seemed to have come of the exercise of the option.[22] 17. Then came a letter dated 8 April 2008, from the plaintiff’s solicitors to the defendant’s solicitors. After stating that under Clause III.5 of the 2006 agreement that the defendant had the “right to buy back the Property” at the highest valuation provided by 3 valuers, it went on to say that their respective clients had agreed to appoint AG Wilkinson & Associates (one of the named valuers), whose valuation would be the price payable for the Property. There was no written reply.[23] 18. Then, there was a letter dated 8 August 2008 from the plaintiff’s solicitors to the defendant, which asserted inter alia, that unless the defendant was willing to repurchase the Property at $3,200,000, which was the price which Fancy Spread Ltd, a willing purchaser ,was willing to pay on 30 May 2007, the defendant must assign the Property to the plaintiff. In this letter, the plaintiff relied on the 6 June agreements. There was no reference to the 2006 agreement[24]. 19. The defendant’s solicitors on 9 September 2008 asserted that the 6 June agreements had been superseded by the 2006 agreement and that proper valuations had been obtained from the named surveyors and the highest,by Centaline Surveyors Ltd, was $2,440,000. The letter went on to say “In the premises, we have instructions to give you notice, which we hereby do, that our client now endeavours to exercise his lawful contractual right of pre-exemption(sic) of purchase of [the Property] by way of the valuation price of HK$2,440,000.00.” 20. The fact that the defendant’s solicitors regarded the right under the 2006 agreement as a pre-emptive right does not matter. It is a matter of construction what right was conferred. However, the fact that the letter did not mention that there had been an earlier exercise of the option or pre-emptive right may bear on the Court of Appeal’s view that there had been a valid exercise of the option on 16December 2007. Further, it was not the defendant’s pleaded case that he had through his mother or at all exercised the option on 16 December 2007. The defendant’s case was not clearly pleaded. I will not go into the pleadings in detail because they are not illuminating. It was pleaded in para 34 of the Re-Re-Amended Defence “… the defendant had accepted the first offer[25] of the plaintiff, and/or exercised his pre-emptive right by accepting the plaintiff’s offer[26], to sell [the Property] in or about 28 November 2007 or 2 December 2007 …” 21. Since, it will not affect the outcome of this appeal, I will not go further into the Court of Appeal’s finding that the option had been exercised on 16 December, I will proceed on the basis that the option had been exercised on 16 December and that the price payable was $2,440,000 and that as Mr Andy Hung for the defendant contended a contract of sale thereby resulted. I will treat the letter of 9 September 2008 as a confirmation by the defendant of the exercise of the option. 22. However, by the defendant’s solicitor’s letter of 11 October 2008 addressed to the plaintiff’s solicitors, the plaintiff was informed that “our client hereby irrevocably withdraws his endeavour to exercise his right under the contract to purchase [the Property] at the valuation price of $ 2,440,000.00”. 23. By letter of 22 October 2008, the plaintiff asked for assignment of the Property whilst reserving his right to damages. 24. The Court of Appeal regarded the letter of 11 October as a repudiation[27] by the defendant of his obligation which arose from his exercise of the option to purchase and the letter of 22 October as the plaintiff acceptance of it. For that reason, the Court of Appeal dismissed the defendant’s appeal and upheld the order for specific performance in favour of the plaintiff. With respect, I agree. 25. Mr Hung, submitted rightly, in my view that upon exercise of the option to purchase, a binding contract to sell and purchase the Property thereby arose. As Hoffmann J (as he then was) put it in Spiro v Glencrown Properties Ltd [1991] Ch537 at 543: “… When the option is exercised, vendor and purchaser come under obligations to perform as if they had concluded an ordinary contract of sale.” 26. However, I cannot agree with Mr Hung that notwithstanding the purchaser’s repudiation of the contract arising from his exercise of the option (“the option contract”), the vendor only had a remedy in damages for breach of that contract and was thus limited to a monetary claim for that breach. The vendor could elect one of two alternative remedies. He could have but was not obliged to seek specific performance of the option contract. Alternatively, he could instead “proceed to claim damages for breach of the contract, both parties being discharged from further performance of the contract”.[28]Upon such discharge, the vendor would be left with his interest in the property the subject of the contract. That this is so, in a typical case where an owner granted an option to purchase, which was duly exercised by the grantee who then repudiated the contract, is beyond argument. 27. Nor should it make any difference, where as in this case, the vendor had only an equitable interest. The defendant’s option or right of pre-emption was predicated on the plaintiff having an interest to sell. 28. Here, under the 6 June agreements, which were replaced by the 2006 agreement, in consideration of his constructing the village house, the plaintiff was entitled to the Property. That is to say, the plaintiff was the purchaser of the Property from the defendant as vendor, and in due course, he would be entitled to an assignment of the Property. His right to the Property could be enforced by specific performance of the 6 June agreements as replaced by the 2006 agreement. Thus, if the option granted by the 2006 agreement had not been exercised so as to bring into existence the option contract, the plaintiff would have been entitled to the assignment of the Property by the defendant to him in due course. The Court of Appeal held that the option had been duly exercised. Upon the due exercise of the option, the option contract, being a contract of sale of the plaintiff’s beneficial interest in the Property arose “as if they had concluded an ordinary contract of sale.” However, just like an ordinary contract of sale, this contract of sale could miscarry. In this case, by the letter of 11 October 2008, the defendant repudiated the option contract. The repudiation was accepted by the plaintiff by letter of 22 October. The plaintiff was not obliged to seek specific performance of the option contract. As Lord Wilberforce made clear, he could instead opt to sue for damages for breach of the option contract (if any) and, more importantly, be discharged from further performance of that contract. The plaintiff here chose the latter course of action. Critically, though, being consequently discharged from further performance of the option contract, the plaintiff was left with his rights under the 2006 agreement, including in particular his right to specific performance thereof. 29. Mr Hung also submitted that the parties should be taken to have intended or agreed that upon the defendant exercising his option to purchase, the 2006 agreement under which the defendant agreed to assign the Property to the plaintiff was rescinded. With respect, there was no question of any rescission. As explained, upon the due exercise of the option, the option contract arose, being a contract of sale with the plaintiff as vendor and the defendant as purchaser. The formation of the option contract did not extinguish the plaintiff’s rights under the 2006 agreement. On the contrary, the option contract was predicated on the plaintiff having a beneficial interest in the Property to sell. Thus, when the option contract was discharged by reason of the defendant’s breach which was accepted by the plaintiff and brought the option contract to an end, the parties reverted to their rights under the 2006 agreement. 30. For these reasons, I dismissed the defendant’s appeal with costs. Mr Justice Fok PJ: 31. I agree with the Reasons of The Chief Justice, Mr Justice Tang PJ and Lord Phillips of Worth Matravers NPJ. Lord Phillips of Worth Matravers NPJ: 32. I agree with the judgment of Mr Justice Tang PJ. The facts of this case are, however, unusual, in that they involve two linked contracts. In these circumstances the use of terms such as “rescission”, “repudiation” and “acceptance of repudiation” are capable of giving rise to confusion. Accordingly I propose to set out briefly my own analysis of the position. 33. The two agreements dated 6 June 2006, and the “2006” agreement that replaced them conferred the following rights and obligations: a. The plaintiff had to build a house on the defendant’s land; b. On completion of the house the defendant had to transfer to the plaintiff title to the second floor and the roof of the house (“the upper part”) to the plaintiff; c. The title deeds of the property were to be held by solicitors pending completion. d. Upon completion the defendant would have an option to buy back from the plaintiff the title to the upper part at a valuation to be provided by an independent valuer. I agree with Mr Justice Tang that it is immaterial whether the option started life as a right of preemption. Nor does the precise provision for valuation of the upper part matter. 34. On completion the defendant failed to transfer title to the upper part to the plaintiff. 35. The defendant gave notice of the exercise of the option to purchase the plaintiff’s interest in the upper part for the valuation figure of $2,440,000. This gave rise to binding obligations on the defendant to buy and the plaintiff to sell the plaintiff’s interest in the upper part. Because the defendant had failed to transfer title in the upper part to the plaintiff, this was merely an equitable interest. 36. By his solicitors’ letter of 11 October 2008 the defendant repudiated his liability to buy the plaintiff’s interest in the upper part for $2,440,000. 37. By his letter of 22 October the plaintiff implicitly accepted this repudiation. This left the plaintiff with an equitable interest in the upper part, which he was entitled to enforce by requiring the defendant to transfer title to him. 38. The plaintiff also had a right to claim damages for the defendant’s failure to purchase the plaintiff’s interest in the upper part. The fact of this litigation suggests, however, that the value of the upper part exceeded the sum of $2,440,000, so that the plaintiff has suffered no damage. 39. The defendant’s case appears to have been that the parties agreed that the defendant’s obligations under the original agreement should be replaced by a simple debt of $2,440,000. That is in conflict with the clear provisions of the contract and the conduct of the parties and I can see no warrant for it. Mr Andrew Mak and Ms Carol Wong, instructed by Adrian Yeung & Cheng, for the plaintiff/respondent. Mr Andy Hung, instructed by Damien Shea & Co, for the defendant/appellant. [1] From the judgment of the Court of Appeal (Stock VP, Barma JA and Poon J) dated 20 March 2014. [2] Under s 22(1)(a) of the Hong Kong Court of Final Appeal Ordinance Cap 484. This route of appeal has been abolished by s 8(1) of the Administration of Justice (Miscellaneous Provisions) Ordinance 2014 enacted on 24 December 2014. [3] Mr Andy Hung. [4] The respondent plaintiff was represented by Mr Andrew Mak and Ms Carol Wong. [5] The lower courts dealt extensively with the question whether as a matter of law there was an option granted to the defendant or a right of pre-emption. It matters not for present purposes which it is. [6] And for reasons set out in the judgment of Mr Justice Tang PJ, there are considerable doubts as to this. [7] The remaining portion of Lot No 78 in DD 236, New Territories. [8] It seems likely, though not explored in the judgments below, that the defendant was a male indigenous villager and as such could obtain a permit to build one 3 storey village house on village land. This is known as a “Ding right”. Normally, the permit would prohibit sale of any interest in the house before a certificate of compliance is issued. Indigenous villagers who have no land sometimes enter into complicated schemes with developers so that their Ding right could be exploited. Indigenous villagers who have land but are unable or unwilling to fund their building, might enlist the help of developers who would build the houses for them in return for a share in the houses built. In this appeal, we are not concerned with the legality of any such scheme. [9] I shall refer to these agreements jointly as the 6 June agreements and separately as the 6 June agreement and 6 June supplemental agreement respectively. In these agreements the plaintiff was known as “Party A” and the defendant as “Party B”. [10] I will refer to this as the 2006 agreement. The plaintiff regarded this as a second supplemental agreement to the 6 June agreement. At trial, Deputy Judge Seagroatt regarded it as a novation agreement, and in the Court of Appeal, a replacement agreement which superseded the 6 June agreements. With respect I agree. [11] There is another copy of this agreement dated 24 July 2006 signed by the plaintiff and witnessed by Lau Pak On but not signed by the defendant. The Supplemental Agreement is not relevant. [12] There are some differences between the 6 June agreements and the 2006 agreement but they do not matter to this appeal. [13] At trial, Deputy High Court Judge Seagroatt held that the defendant had a right of pre-emption. In the Court of Appeal, Stock VP, Barma JA and Poon J, held that the defendant had an option. [14] Barnsley’s Land Options 5th ed at 6-003. [15] Pritchard v Briggs [1980] 1 Ch 338 at 418B. [16] Normally, village houses or interests in them should only be sold after Certificate of Compliance. Sale before the certificate might have consequences. [17] The memorandum expressly provided that it “does not constitute a binding agreement.” But it also provided that the deposit should be forfeited if “the Purchaser fail to complete the purchase …”. As part of the transaction, the plaintiff, personally gave a deed of indemnity in favour of the purchaser to indemnify it against loss in the event of failure by the vendor to complete the sale contemplated by the memorandum of understanding, namely within 30 days of the signing of an Agreement for Sale and Purchase, which should be signed within 14 working days of written notice that “the Certificate of Compliance has been issued …” Clause (2)(b) and (c). I will not go into further detail but this kind of transactions would not be unfamiliar to persons involved in the buying and selling of village houses before a Certificate of Compliance is issued. [18] CFI Judgment, para 48. [19] In either event, at the valuation arrived at in accordance with the 2006 agreement. [20] Plaintiff’s pleading on this point is at para 8B(c) of the Re-Amended Reply and Defence to Re-Re-Amended Counterclaim. [21] I will not go into the plaintiff’s case that the valuations obtained by the defendant were invalid because the valuers had not been jointly appointed and/or the defendant failed to disclose to the valuers the fact that there was a willing purchaser at $ 3,200,000. [22] There was no discussion on why there had been no completion within a reasonable time of 16 December 2007 or at all. [23] The defendant pleaded that he had rejected the proposal for a further valuation and re-affirmed his decision to purchase. [24] Nor to the 5 August 2006 supplemental agreement. [25] It is no clear what this referred to. [26] Nor is it clear what offer was referred to. [27] As the Court pointed out, that was also the submissions of counsel for the defendant. [28] Per Lord Wilberforce, Johnson v Agnew [1980] AC 367, 392. 
Mr Justice Fok PJ: 1. At the appellant’s trial for trafficking in a dangerous drug, inadmissible evidence infringing his right to silence was heard by the jury. The Court of Appeal, by a majority, concluded that, though irregular, this was not material to the fairness of the trial and did not render his conviction unsafe, and therefore dismissed his appeal against conviction. 2. Leave to appeal to this Court was granted by the Appeal Committee on the basis that it was reasonably arguable that there had been substantial and grave injustice.[1] At the conclusion of the hearing of the appeal, the Court allowed the appeal and made the various orders set out below. The Court indicated that it would provide its reasons for doing so in due course. These are the reasons of the Court. Background 3. The appellant was observed by plainclothes police officers in the early hours of the morning on 17 March 2012 approaching an apparently drunken man sitting asleep on the pavement in Jordan Road, Kowloon. The appellant was seen to take a paper bag from the man and was arrested by the police. The bag was found to contain a brand new Apple iPad and case worth HK$6,830. On being searched, a bag suspected to contain dangerous drugs was found in his front trouser pocket. He was arrested and cautioned for the offence of possession of a dangerous drug[2] but declined to say anything. When interviewed at the police station, he again declined to say anything. 4. The appellant was charged with one count of theft, in relation to the contents of the paper bag (Count 1). The bag of suspected drugs was found to contain 13.64 grammes of a crystalline solid containing 12.07 grammes of methamphetamine hydrochloride, or “Ice”, with a street value of HK$10,748. He was charged with trafficking in a dangerous drug (Count 2).[3] 5. At his trial, the appellant pleaded guilty to the charge of theft. In relation to Count 2, he pleaded not guilty to trafficking but guilty to the lesser charge of possession. The prosecution did not accept this plea and the trial proceeded.[4] On 24 June 2013, after a short trial, the appellant was convicted of the offence of trafficking. The judge imposed a total sentence of 7 years’ imprisonment in respect of the two offences. 6. On appeal,[5] the ground of appeal advanced by the appellant on conviction was that the inference of trafficking was not the only inference which could be drawn from his admitted possession of the drugs and that, on the evidence, the judge should have withdrawn the case of trafficking from the jury. This ground of appeal was unanimously rejected by the Court of Appeal. However, in the course of argument, the Court of Appeal drew counsel’s attention to a passage in the cross-examination of the appellant at trial (see below) and raised the issue of whether the exchange was an error of law or irregularity in the trial and, if so, whether it was material. 7. For reasons that will be examined below, Macrae JA (with whom Lunn VP agreed) concluded that the exchange was an irregularity but not material whilst McWalters JA concluded that the admission of the evidence in question was a material irregularity and an error of law. Accordingly, on this ground of appeal, the Court of Appeal dismissed the appeal by a majority. On the appeal against sentence, the Court of Appeal unanimously allowed the appeal to the extent of reducing the overall sentence to one of 6 years and 9 months’ imprisonment. The irregularity in the trial 8. During the course of the appellant’s cross-examination by prosecuting counsel, the following exchange took place: “Q. And the police, the lady officer who gave evidence, said that they didn’t find anything of significance. A. Yes. Q. So the police missed, they didn’t find the bottle and the pipes that you used to consume ‘Ice’. COURT: That’s outside the window, outside the premises. Q. Yes, outside, under the shade outside your window. COURT: Under the air-conditioner, in the air-conditioner? MR. WHITEHOUSE: Yes, yes. A. Yes. No. Q. And you didn’t think of telling them that ‘This is where I consume my ‘Ice’ and this is the equipment I use to consume it’? A. No. Q. And why not? A. Because they did not ask me. Q. I suggest to you that this is completely untrue, that there’s no equipment in your home at all. A. It’s true and equipment was there. Q. I suggest you’re not an honest person. A. What do you mean? Q. You’ve got many convictions for dishonesty. A. I did make thefts and I … COURT: ‘I did make’ what? A. I did make thefts and I did make possession and I have served sentences for those. And I did make – I did plead guilty for those offences what I had done. Q. In fact, what I was suggesting to you, I suggest your record shows that you’re a dishonest person and you are not telling the truth today in court. A. I am under oath and I am telling truth.” (Emphasis added) 9. The context of the exchange was that the prosecution was seeking to undermine the credibility of the appellant’s evidence in chief that the drugs found in his possession were for his own use and that he habitually consumed “Ice” in his bedroom at his parents’ flat where he lived. This is demonstrated by the following part of the prosecution’s subsequent closing speech to the jury: “And another question you may ask when considering the – whether or not you can accept this defendant’s evidence, why, when the police went to his flat, his home, where his mother and father live, the following day, why did the police not find any of the – any evidence of equipment to – for smoking, inhaling ‘Ice’ in his – in his room? I’ve talked about luck or bad – his bad luck. Was it just his good luck that the police wasn’t – weren’t able to find it or was it another part of his story that is not true? My submission to you, members of the jury, that the whole defendant’s account for the possession of this ‘Ice’ is inherently unbelievable.” The right to silence 10. There are two respects in which a person’s right to silence may be infringed. The first is where his silence is used to infer guilt of the particular offence of which he is suspected or charged. This may occur, for example, where a suspect is caught red-handed and, when questioned as to what he is doing, declines to answer. His silence in these circumstances may be said to give rise to an inference of guilt. 11. The second way that a person’s right to silence may be infringed is where it is suggested that a line of defence or evidence which is helpful to the suspect or accused is a late invention because it was not mentioned earlier. This suggestion of recent invention may be said to undermine the credibility of the line of defence or evidence tendered. 12. It is common ground on this appeal, as it was in the Court of Appeal, that the italicised passage in the cross-examination constituted inadmissible evidence. It did so because it clearly infringed the appellant’s right to silence in the second way referred to above. That right, being an aspect of the right against self-incrimination,[6] is a fundamental right of a person suspected of or charged with a criminal offence as recognised in this Court’s decision in Lee Fuk Hing v HKSAR.[7] 13. In Lee Fuk Hing, this Court materially held that the use of a person’s exercise of the right of silence in either way described above was impermissible. The Court agreed with the reasoning of the majority of the High Court of Australia in Petty & Another v R[8] refusing to follow English authority[9] which had drawn a distinction between the exercise of an accused of his right of silence leading to an inference of guilt and the failure of an accused to advance an explanation later relied upon at trial as relevant to his credibility. Hence, at [55], the Court held: “In the whole of these circumstances, it is inappropriate in Hong Kong to use a person’s silence against him in any way.” (Emphasisadded) 14. The Court explained the rationale for this principle in these terms at [56]: “A person’s right against self-incrimination (his right of silence) would otherwise become a possible source of entrapment. It is unfair for a person to have the right to remain silent, and usually to have been reminded of this right through the caution, and then for his silence to be put against him at trial. There are comments in some of the authorities to which we have referred which suggest that in some circumstances a trial judge may fairly inform the jury that an explanation has been made for the first time at trial provided that the judge also makes it clear that an accused has the right to remain silent and make no explanation in answer to police questions. With respect such a direction is also objectionable. It gives rise to an adverse inference against the accused.” 15. It is to be noted that the position in England in respect of a person’s failure to mention something on which he later relies in court has been clarified by s.34 of the Criminal Justice and Public Order Act 1994. This permits the court or jury, in certain circumstances, to rely on an accused’s failure to mention to the police a fact which he could reasonably have been expected to mention in order to draw such inferences as appear proper. This is reflected in the caution given by a police officer to a suspect in the United Kingdom.[10] There is no counterpart to s.34 of the 1994 Act in Hong Kong.[11] By contrast, the caution administered to suspects in Hong Kong is in unqualified terms, as the Court recognised in Lee Fuk Hing.[12] The issue on appeal and the parties’ respective contentions 16. It is common ground on this appeal, as it was in the Court of Appeal, that the italicised passage in the cross-examination constituted inadmissible evidence. Whereas the issue in the Court of Appeal was whether this was an error of law or material irregularity rendering the conviction unsafe and unsatisfactory, in this Court the issue is whether there has been substantial and grave injustice by reason of a departure from an accepted norm to the appellant’s disadvantage.[13] 17. Mr Gerard McCoy SC, leading counsel for the appellant,[14] relied squarely on the dissenting judgment of McWalters JA holding that the admission of the inadmissible evidence was an error of law and a material irregularity in the trial which was not cured by the judge’s standard direction on the right to silence in her summing-up to the jury (discussed further below). Therefore, the conviction of the appellant for the offence of trafficking in breach of his right to silence amounted, in all the circumstances, to substantial and grave injustice. 18. For the prosecution, Mr William Tam SC[15] submitted that the majority judgment of Macrae JA holding that the admission of the evidence was not a material irregularity rendering the conviction unsafe and unsatisfactory was correct and, hence, that there was no question of substantial and grave injustice having been done to the appellant. The importance of the issue of credibility in this case 19. The question of whether the right to silence has been materially infringed being fact and context sensitive, it is necessary to identify the issue to which the inadmissible evidence related and its importance in the context of the case as a whole. 20. The prosecution case, as summarised by Macrae JA in his judgment,[16] relied on the following factors: (1) the discovery of the appellant in the early hours of the morning in possession of a not insubstantial quantity of “Ice” in a public place and not his home where he said he consumed drugs; (2) the absence of the discovery of drug taking paraphernalia at his home when it was searched; (3) the fact the appellant, whose monthly income was around HK$8,400, paid just over HK$7,000 for the drugs worth over HK$10,000; (4) the fact the theft of the iPad could not have been with a view to paying for the drugs in question; and (5) the fact the quantity of “Ice” found on the appellant would have been sufficient for as much as 30 to 40 days’ personal consumption but could have supported multiple users. 21. It is quite clear that the case of trafficking was one which was properly left to the jury to determine but it was by no means a compelling one. The appellant’s defence was that he was an habitual drug abuser and that the “Ice” found on him was for his own personal consumption. No paraphernalia for drug trafficking was found either on his person or at his home and, at the conclusion of his cross-examination (although not, apparently, before that), prosecuting counsel accepted that he was a user of “Ice”. In support of the appellant’s case that he was a drug user, defence counsel at trial introduced the appellant’s criminal record which included a number of previous convictions for possession of dangerous drugs but none for trafficking. It also included previous convictions for theft, which the appellant explained related to his stealing razor blades from a shop which he then sold for money to buy drugs. As McWalters JA rightly pointed out,[17] this was a bold tactic but was designed to bolster the credibility of his evidence that the “Ice” found in his possession was for his own consumption. 22. It was common ground at trial, and on appeal, that the sole issue in the trial was the credibility of the appellant. His credibility was relevant in two respects: first, specifically, as to his claim that he was a drug abuser and that he was therefore in possession of the “Ice” for his own consumption and not for trafficking; and secondly, more generally, as to his evidence as a whole. This is an important context in which to judge the materiality of the inadmissible evidence and also the effectiveness of any remedial steps taken to cure any prejudicial effect. This context was all the more important because of the way the defence was run by reference to the introduction of the appellant’s criminal record, which would expose the appellant to the risk of being subject to unfair prejudice in his cross-examination. The introduction of the inadmissible evidence was clearly a material irregularity 23. With respect to Macrae JA’s judgment to the contrary, and notwithstanding Mr Tam’s able submissions in support of that judgment, the introduction of the inadmissible evidence was clearly a material irregularity in the context of the appellant’s trial. 24. As noted above, the sole and critical issue for the jury was whether the appellant’s claim that the drugs were for his own consumption rather than for trafficking was true. His counsel had adopted the understandably bold tactic of introducing his past criminal record to support this. That criminal record included offences of dishonesty so the credibility of his oral testimony before the jury was particularly important. Towards the end of his evidence in chief, the appellant gave evidence that he kept the equipment he used to consume “Ice” at home in the space outside his bedroom window below the air-conditioner. This was shortly before the conclusion of his evidence in chief which ended with his absolute disavowal of any intention to deliver or sell the “Ice”. 25. At the outset of the prosecution’s cross-examination of the appellant, counsel tackled the appellant’s claim to have bought the drugs for his own consumption. He put to the appellant that it was just bad luck that, on the night he was arrested, he happened to have used up his supply of drugs and had just purchased a whole month’s supply but had not had time to take this home. He then proceeded immediately to ask the appellant about the search of his home and the exchange set out in paragraph 8 above, which included the inadmissible evidence, took place. As will be apparent, the purpose of this cross-examination was in order to suggest that what the appellant had said was “completely untrue” and that he was “not an honest person” but was “a dishonest person”. 26. Counsel for the prosecution then proceeded to cross-examine the appellant on his previous convictions, emphasising that he had been disbelieved in his testimony when he had pleaded not guilty to being knowingly in possession of three forged credit cards and had been convicted of offences of theft. This culminated in the following exchange: “Q. All right. Well, I put it to you, you are a thoroughly dishonest man. What do you say to that? A. I can’t say anything. Sometime I am honest and sometime I am dishonest.” 27. In his closing speech to the jury, consistent with his identification of the sole issue being the truth of the appellant’s claim that the drugs were for his own use, counsel for the prosecution attacked the appellant’s credibility root and branch. The following extracts serve to illustrate this: “Now, bearing in mind what the prosecution say is his unlikely explanation, bearing in mind his very bad criminal record, particularly for dishonest – dishonesty, you may find it very, very difficult to believe almost anything that he told you yesterday. Certainly, in my submission, he was not telling the truth when he said that this ‘Ice’ was for his own use, for his own consumption. Now, why do I put it that way, members of the jury? Firstly, he’s been shown to be an utterly dishonest person; dishonest in the past. … Now, of course, that record doesn’t meant to say he is – it necessarily follows that he is a drug trafficker, but, to make my point, members of the jury, the prosecution’s submission is that you might find it very difficult to accept or to be impressed in any way with his evidence when he says that the ‘Ice’ was for his own use.” 28. Although he had eventually accepted that the appellant was a drug user, counsel for the prosecution addressed the question of the equipment for taking drugs in his closing speech in the manner set out in paragraph 9 above. Crucially, however, shortly before that remark in his speech, counsel made this comment about the appellant’s evidence that he had gone out on the day before his arrest with money to buy drugs: “Now, to deal with the – his defence in a little bit more detail. In evidence yesterday – for the first time, I might say, members of the jury – he claims that he went out that morning with something more than $7,000 in his possession. That was that morning; he went to work that morning. That would be the previous day, he being arrested in the early hours of the following day, of course. He went out that morning to work, with 7,000-plus dollars in his possession and you may wonder where he got – he got that money from, if it was true – and the prosecution is in doubt, so you can’t accept what he is saying. …” (Emphasis added) 29. The italicised portion in the preceding paragraph is a further instance of inadmissible material being put before the jury. It was improper for prosecuting counsel to have sought to cast doubt on the veracity of the appellant’s evidence by reference to the fact that it was only during his evidence that he first testified that he went out with sufficient money to buy the drugs found in his possession. By reason of his invocation of his right to silence, he had not made this claim previously and it was wrong for prosecuting counsel to use his exercise of that right to suggest his evidence was not to be believed. Yet, this passage in counsel’s closing speech was delivered only shortly before his rhetorical question (albeit contradicting somewhat his eventual acceptance that the appellant was a drug user) asking why the police did not find drug-taking equipment at the appellant’s home. That reference, the jury will have no doubt recalled, occurred in the cross-examination in which, similarly, prosecuting counsel had sought to cast doubt on the appellant’s evidence that the equipment was outside his window because he had not previously told anyone this. 30. The references above demonstrate that it was a consistent theme of the prosecution that the appellant was not to be believed and this was at least in part because he made things up as he went along. The latter premise was based on inadmissible evidence and material used in breach of the appellant’s right to silence. 31. For these reasons, and with respect, we disagree with the statement in Macrae JA’s judgment that: “… at no stage during prosecuting counsel’s speech to the jury was anything made of the fact that the applicant did not tell the police where his drug-taking equipment was kept.”[18] As a matter of context, there is good reason to think that the jury may very well have connected the criticism of the appellant for his failure previously to mention going out with HK$7,000 and the earlier criticism in cross-examination of his not thinking to tell the police where his drug-taking equipment was kept. 32. As a further matter of context, in this regard, it is material that the trial was a short trial lasting only 5 days. Cross-examination of the appellant took place on Day 3 of the trial (Thursday 20 June 2013) and both counsel’s closing speeches were made on Day 4 (Friday 21 June 2013), followed on the next working day (Monday 24 June 2013) by the judge’s summing-up to the jury in the morning of the last day of the trial. The jury returned their verdict in the afternoon of that day. It is reasonable to conclude that the prejudicial effect of any inadmissible evidence would therefore still have been fresh in the minds of the jury. Prejudicial effect clearly not remedied by judge’s standard direction 33. As the Court noted in Lee Fuk Hing, when an error of this sort occurs, the judge has to decide whether he can correct the error in such a way that the trial can proceed fairly.[19] There are no doubt instances when a judge may properly decide that, notwithstanding an error resulting in inadmissible material being received by the jury, “the less said the better”. Whether the error requires to be rectified and, if so, how, will depend on the facts of the individual case. 34. In her summing-up to the jury, the judge gave the standard direction to the jury on the accused’s right to silence in these terms: “The prosecution has produced, without objection from the defence, evidence that the accused said, when questioned by the police, that he had nothing to say. In other words, he exercised his right to silence, as he was entitled to do. His having done so does not advance the prosecution’s case. It is not to be held against the accused in any way whatsoever. The exercise of the right to silence does not amount to an admission of any kind, nor can it be taken to reflect a guilty conscience.” 35. Later in her summing-up, the judge did specifically focus on the evidence concerning the search of the appellant’s home and said: “The cross-examination directed to his not having told the police where he kept his drug-taking equipment does not help the prosecution since the prosecution does not dispute that the accused did consume drugs.” This statement was in the context of a summing-up that was, in general, extremely favourable to the defence case and implied that the jury ought to accept the truth of his claim that he was only in possession of the drugs for his own use and not for the purposes of trafficking.[20] 36. In his judgment, Macrae JA contrasted the standard direction given by the judge in this case with the absence of that direction in Lee Fuk Hing and noted that the judge’s statement about the inadmissible evidence was to the unqualified effect that it did not help the prosecution. For these reasons, he concluded that, viewing the summing-up as a whole, the irregularity in question was not material.[21] 37. With respect, we are unable to agree with Macrae JA’s conclusion in this regard and prefer that of McWalters JA to the contrary. 38. Whilst it is true that the judge gave the standard direction and that this was absent in Lee Fuk Hing, the standard direction referred to the appellant’s exercise of the right to silence “when questioned by the police” but did not expressly extend to the search of his home or to an occasion when he might have volunteered some relevant information, viz. that his drug-taking equipment was in fact outside the window. The cross-examination of the appellant in this regard did not identify an occasion when he was asked any question by the police. On the contrary, his asserted reason for not mentioning the drug-taking equipment to the police (and which Macrae JA considered was “quite a good answer”[22]), was that he was not asked about this, so the jury may have (wrongly) reasoned that there was no occasion to claim a right to silence. 39. More importantly, there are certain difficulties with the judge’s statement in respect of the inadmissible evidence concerning the drug-taking equipment. The whole of the passage in question is as follows: “I have to tell you, in all fairness to the accused, that his account of being only a drug user and not a drug trafficker is an account which was unshaken in cross-examination. As for the cross-examination questioning whether he was born in Hong Kong, the accused’s birth certificate showing that he was born in Hong Kong has been produced subsequently. You are also to wholly ignore the cross-examination suggesting, by reference to his previous convictions, that the accused was dishonest. He has previous convictions, but they do not help the prosecution prove trafficking against him. The cross-examination directed to his not having told the police where he kept his drug-taking equipment does not help the prosecution since the prosecution does not dispute that the accused did consume drugs.” 40. It will be immediately apparent that the comment about the drug-taking equipment is to be contrasted with the express direction that the jury are “to wholly ignore” the cross-examination suggesting dishonesty by reference to his previous convictions.[23] The jury may well have reasoned that there was a real distinction between the treatment of the different evidence: that they were wholly to ignore the evidence based on previous convictions as regards dishonesty but that, in respect of the evidence of the drug-taking equipment, the evidence, though properly to be taken into account, did not help the prosecution. Moreover, the jury may have thought that the judge’s comment on the evidence concerning the drug-taking equipment was merely her own view of that evidence, in which case they may have approached it on the basis of her direction that they were not obliged to accept that view. This is because she had earlier directed them in standard terms that: “… if in the course of my review of the evidence I appear to express my views concerning the facts or emphasise a particular aspect of the evidence, do not adopt those views unless you agree with them; and if I do not mention something which you think is important, you should have regard to it in any event and give it such weight as you think fit. When it comes to the facts of this case, it is your judgment alone that counts – remember that.” Although there is an assumption that the jury will follow the instructions which they receive from the trial judge,[24] there is a real risk here that the jury may have embarked on an impermissible line of reasoning, viz.: because the appellant did not mention something helpful to his defence at the earliest opportunity, that defence was not to be believed. That this may have happened is underscored by the fact that, notwithstanding a summing-up that “positively encouraged”[25] the jury to conclude that the drugs were for the appellant’s own consumption rather than trafficking, the jury convicted him of the latter offence. 41. A further reason the judge’s standard direction on the right to silence did not remedy the error constituted by the admission of the evidence concerning the drug-taking equipment is that, although she addressed that specific evidence in her summing-up, she made no mention at all of the inadmissible evidence introduced in prosecuting counsel’s speech that it was only during his testimony the day before that the appellant had claimed he went out that day with HK$7,000. By not referring to and directing the jury to ignore the comment that the appellant had not previously mentioned going out with the money, the jury might well have reasoned that counsel’s remark was a proper one for them to take into account. If it was proper to take that remark into account, then the same comment could, they might have reasoned, also apply to the failure previously to mention the drug-taking equipment. 42. For these reasons, I would respectfully agree with the view expressed by McWalters JA that: “What was required from the judge in respect of it was not a statement as to its probative value to the prosecution but rather a clear, unequivocal legal direction that the jury must ignore it, were not permitted to rely on it and an explanation for why this legal direction was being given. The comment of the judge left the inadmissible evidence before the jury to make of it what they would. There was no prohibition by the judge preventing them from relying on it as evidence and this meant there was no prohibition on them using it for the purpose of inferring guilt or of assessing the truthfulness of the applicant and the credibility of his testimony.”[26] 43. In his judgment, one of the factors Macrae JA referred to in reaching his conclusion that the standard direction was sufficient was that defence counsel did not raise the matter with the judge during the appellant’s evidence, or after cross-examination or before the summing-up.[27] Mr Tam also relied on this factor in his submissions in this Court. However, whilst it is true that trial counsel for the appellant did not raise with the judge any concern about the inadmissible cross-examination in question, this is not a factor of significant weight in the present case. It is by no means clear that counsel appreciated the prejudicial effect of the evidence and it was not a point that occurred to different counsel appearing for the appellant in the Court of Appeal, since it was that court which drew counsel’s attention to the point eventually taken. In any event, regardless of what counsel may have thought of it, we are of the view, for the reasons stated above, that this evidence was seriously prejudicial. Conclusion and disposition 44. By reason of his conviction in these circumstances, substantial and grave injustice was, for the reasons set out above, done to the appellant. At the conclusion of the hearing, without calling on Mr McCoy in reply, the Court indicated it would allow the appellant’s appeal. Although the respondent’s printed case sought an order for re-trial on the charge of trafficking if the appeal were allowed, Mr Tam sensibly indicated that, in light of the evidence and the lapse of time since the offence and the appellant’s detention in custody, he would not press for an order for a re-trial. As to the appropriate sentence for the lesser offence of possession, it was common ground that the sentence of imprisonment already served by the appellant was longer than any sentence that would have been imposed on him for that lesser offence, even allowing for any additional sentence for the appellant’s theft conviction. Accordingly, at both counsel’s invitation, the Court substituted a sentence which would allow for the appellant’s immediate release. 45. In the circumstances, the Court made the following orders by way of disposition of the appeal: (1) Appeal allowed; (2) The appellant’s conviction for trafficking in a dangerous drug be quashed and substituted by a conviction for possession of a dangerous drug; (3) The sentence imposed by the judge for the offence of trafficking be quashed and substituted by a sentence for possession that would allow for the appellant’s immediate release; (4) The sentence for the appellant’s conviction for theft be served concurrently with the substituted sentence for possession. Mr Gerard McCoy SC and Ms Emma TC Tsang, instructed by Littlewoods, assigned by the Director of Legal Aid, for the Appellant Mr William Tam SC, DDPP and Mr Raymond Cheng, SPP, of the Department of Justice, for the Respondent [1] FAMC 38/2015, Determination dated 3 November 2015. [2] Contrary to s.4(1)(a) and (3) of the Dangerous Drugs Ordinance (Cap.134). [3] Contrary to s.8(1)(a) and (2) of the Dangerous Drugs Ordinance (Cap.134). [4] In HCCC 348/2012, before V. Bokhary J and a jury. [5] In CACC 292/2013, an appeal against both conviction and sentence, before Lunn VP, Macrae and McWalters JJA; Judgment dated 19 March 2015 (“CA Judgment”). [6] This is a common law right as well as one guaranteed by Article 14(3)(g) of the International Covenant on Civil and Political Rights. [7] (2004) 7 HKCFAR 600 (“Lee Fuk Hing”). [8] (1991) 55 A Crim R 322 (Mason CJ, Deane, Toohey and McHugh JJ). [9] R v Littleboy [1934] 2 KB 408; R v Ryan (1966) 50 Cr App R 144. The distinction drawn in these cases was, in any event, the subject of some doubt in later English cases: see, e.g., R v Sullivan (1967) 51 Cr App R 102 and R v Gilbert (1978) 66 Cr App R 237. [10] The form of caution is in these terms: “You do not have to say anything. But it may harm your defence if you do not mention when questioned something which you later rely on in court. Anything you do say may be given in evidence” (UK Code of Practice for the Interviewing of Suspects). [11] HKSAR v Hung Chan Wa & Another (2006) 9 HKCFAR 614 per Sir Anthony Mason NPJ at [82]. [12] At [55]; the form of caution is in these terms: “You are not obliged to say anything unless you wish to do so but what you say may be put into writing and given in evidence.” [13] So Yiu Fung v HKSAR (1999) 2 HKCFAR 539 at p.543F-G. [14] Appearing in this Court with Ms Emma TC Tsang. Neither counsel appeared for the appellant in the courts below. [15] DDPP, appearing with Mr Raymond Cheng, SPP. [16] CA Judgment at [20]. [17] CA Judgment at [57]. [18] CA Judgment at [27]. [19] Lee Fuk Hing at [59]-[61]. [20] See the passages from the summing-up quoted in the CA Judgment at [17]. [21] CA Judgment at [36]-[37]. [22] CA Judgment at [33]. A better answer might have been to the effect that he was told that he did not have to say anything and that, if he did, it could be used against him in evidence. Since he had been arrested and cautioned for simple possession, there would have been no obvious advantage to him in voluntarily pointing out where his drug-taking equipment was located. [23] This express direction was repeated later in her summing-up when the Judge told the jury: “Counsel for the prosecution attack[s] the reliability of the accused’s testimony on the basis of his previous convictions. I direct you that those convictions do not amount to any reason why you should wholly reject the accused’s testimony as counsel for the prosecution urges.” [24] Montgomery v HM Advocate [2003] 1 AC 641 at p.674B per Lord Hope of Craighead, cited in HKSAR v Kissel [2014] 1 HKLRD 460 (CA) at [115]. [25] CA Judgment at [17]. [26] CA Judgment at [67]. [27] CA Judgment at [33]. 
Chief Justice Ma: 1. I agree with the judgment of Lord Phillips of Worth MatraversNPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Phillips of Worth MatraversNPJ. Mr Justice Tang PJ: 3. I agree with the judgment of Lord Phillips of Worth MatraversNPJ. Mr Justice Fok PJ: 4. I agree with the judgment of Lord Phillips of Worth MatraversNPJ. Lord Phillips of Worth Matravers NPJ: Introduction 5. This appeal raises the question of the circumstances in which a defendant should be permitted to adduce evidence of improper conduct involving police witnesses called by the prosecution that is alleged to have taken place on an occasion wholly unconnected with the offence with which the defendant is charged. Unusually that question arises in the context of a voir dire. The appellant’s case in a nutshell is that the trial judge wrongly refused to admit evidence at the voir dire which, if admitted, might have led her to exclude from the trial evidence of damaging statements alleged to have been made to the police by the Appellant at the time of his arrest, and subsequently made by him at a video-recorded interview. 6. On 19 January 2011 the Appellant was convicted after a trial before Beeson J and a jury of two counts of trafficking in a dangerous drug (Counts 2 and 3) and was acquitted of a count of manufacturing dangerous drugs (Count 1). The Appellant was sentenced to 9 year’s imprisonment on Count 2 and 23 year’s imprisonment on Count 3, to run concurrently. A significant part of the evidence adduced against the Appellant consisted of statements made by the Appellant to the police. In a voir dire held before the jury were empanelled the Appellant had unsuccessfully challenged the admissibility of these statements. At the trial he did not give evidence but relied upon parts of the statements as being exculpatory. The evidence at the trial 7. On 9 July 2009 police officers in a raiding party led by Senior Inspector K.M.Chan laid an ambush outside a flat at Parkland Villas in Tuen Mun. Three men came out, the Appellant, Hung Wai-Yip (“Hung”) and Yuen Kin-san (“Yuen”). Hung and Yuen were to be jointly charged with the Appellant. The police intercepted the three men and there was a struggle. Hung was carrying a white carrier bag containing packets of cocaine (the subject of Count 2). Upon finding these packets the police arrested the three men and cautioned them for trafficking in dangerous drugs and obstructing a police officer in the execution of his duty. The Appellant then said “I only received $1000 from someone for delivering coke for him. I know nothing else”. 8. Keys were found in the appellant’s trouser pocket, that were used to gain access to the flat. More cocaine was found inside the flat (the subject of Count 3) and equipment for manufacturing dangerous drugs (the basis of Count 1). The Appellant remarked: “I did not know there was so much coke here, Ah sir”. 9. DPC 50049 (for simplicity I shall refer to police constables by their numbers rather than by name), who was the officer who arrested the Appellant, recorded these events in his notebook. Subsequently, at Tuen Mun Police Station, in relation to each statement recorded as made by the Appellant, the Appellant endorsed the notebook with a signed declaration that the relevant part of the notebook had been read out to him by DPC 50049, that he had also read it for himself, and that he had made the statement set out in the notebook of his own free will. 10. CCTV recordings produced by the prosecution showed that the Appellant had visited the flat on a number of occasions between 3 and 9 July 2009. 11. On 10 July the Appellant submitted to an interview that was video-recorded. He declined the offer of a lawyer and confirmed that he had declined a similar offer after his arrest. He confirmed the remarks that he had made after he had been arrested and cautioned. He said that he had gone to the flat with Hung, who was a friend, looking for Yuen at his request. When the police team arrived he had not realised that they were police officers and had given one a push. When the police produced their warrant cards he no longer struggled. He said that he had borrowed the keys of the flat to go to the toilet. He said that he had asked Hung to go with him to the flat because he had been offered $1000 to help others deliver coke, but he did not know what coke was. When he discovered that it was a dangerous drug he left the flat immediately. He was only in the flat about 10 minutes. 12. At the trial the Appellant did not give evidence. His counsel, Mr Ross, relied in his submissions to the jury on that part of the interview in which the Appellant said that he did not know what the “coke” was that he had been asked to help deliver, that he had left the flat when he discovered that the “coke” was a dangerous drug and that he only had the keys to the flat because he had borrowed them to go to the toilet. The voir dire: objections to admissibility 13. In a document dated 31 December 2010 the Appellant’s counsel and solicitor set out objections to the admissibility of both the statements that the Appellant was alleged to have made after his arrest on 9 July 2009 and the video-recorded interview on the 10 July. This included the following allegations: (i) All three defendants were assaulted by police officers as they came out of the flat on 9 July. The Appellant was pushed back into the flat. He was kicked several times. He was handcuffed and hooded with the eyeholes in the hood at the back of his head, so that he could see nothing. DPC 50049 placed some keys next to him, saying “this is yours”, which he denied. DPC 50049 then hit his head and said “you’d better not talk too much”. (ii) When the police officers brought the Appellant to the police station they ordered him not to get a lawyer and to obey, otherwise he would suffer and be beaten. (iii) DPC 50049 made the Appellant endorse the contents of his notebook with the declarations and signatures without permitting him to read what those contents were. (iv) The Appellant said that he wanted a lawyer before his video-recorded interview but was told by DPC 503 that he did not need one and that he should respond in the interview to the question of whether he wanted a lawyer by “no need”, in which event he “would not be suffering”. (v) DPC 503 told the Appellant to say in the interview that he had the keys because he wanted to go to the toilet and that this was the first occasion on which he had visited the flat. He later added that the Appellant should say that he had gone there to deliver coke for $1000 because that was already in DPC 50049’s notebook. He said that if the Appellant said that as soon as he came to know that coke was cocaine he immediately refused to do it and left the flat, then he would be alright – “trust me”. The Appellant followed these instructions when answering questions in the video interview. The voir dire: the police evidence 14. DSIP Chan and four members of his team who took part in the raid on 9th July, including DPC 50049 and DPC 503 gave evidence on the voir dire that was held between 5 and 7 January 2011. Both when giving evidence in chief and under cross-examination they denied the allegations set out in the objections to admissibility. There had been an initial struggle at the time that the three men were arrested, but after that no violence had been used. A hood was put over the Appellant’s head, but only at the stage when he was escorted from the flat to the police station. The keys to the flat were not planted on him. He made the statements that were subsequently recorded by DPC 50049 in his notebook. Those statements were read to him and he read them for himself before he wrote and signed the declarations in the notebook. No threats were made before the video-recorded interview, nor was he coached as to what to say at that interview. He was not prevented or discouraged from having the assistance of a lawyer. The voir dire: the Appellant’s evidence 15. The Appellant opted to give evidence at the voir dire. He said that when he and the other two men came out of the flat at Parkland Villas a group of people rushed out and assaulted them. They were pushed back into the flat where the assault continued. He fell to the floor where he was kicked by DPC 50049 and 34311. A hood was then put over his head with the eye-holes at the back. He was punched on the head by DPC 50049. He was permitted to go to the toilet and he was given some water. Before they left the flat a police officer told them not to look for a lawyer and to obey the police otherwise they would suffer for a long time. They were then taken to the Tuen Mun Police Station where DPC 50049’s notebook was prepared and the video-recorded interview took place in the circumstances alleged in the ‘objections to admissibility’ summarised above. 16. I now come to the part of the voir dire that has given rise to this appeal. It relates to attempts to introduce evidence of a subsequent drugs raid by a team led by DSIP Chan. The raid on 26 August 2009 17. On the night of 26 August 2009 DSIP Chan led his team on a raid on a house at Ma Tin village. Two brothers were in the house, Wong Tsz Him, the elder brother, who was asleep in bed, and his younger brother Wong Tsz Pang, who was watching television. The solicitors acting for the appellant learnt of this raid when a press report of it, which included a photograph of the aftermath of the raid, was drawn to their attention. On 2 January 2011 they took a witness statement from the elder brother (“Mr Wong”). This described how the door of the house was forced open and the two brothers were confronted by a group of 6 to 8 men coming up the stairs in line. The man at the front had a crowbar. The man behind punched Mr Wong once in the chest and pushed him back onto a sofa. A second man held him down on the sofa by the neck. A third man hit his right ear and neck once. Someone else punched his back and waist three or four times. He was then handcuffed and realised that his assailants were policemen. Someone then flung a blanket over his head and he received at least 10 further punches to the back. After about 10 minutes the blanket was removed and he saw that his brother was also handcuffed. The house was searched but nothing was found. The handcuffs were then removed from the brothers. The Wongs’ parents then arrived. Mr Wong asked for a drink of water and for permission to go to the toilet and to see a lawyer. These request were refused. Subsequently, by which time a considerable number of villagers had gathered outside the house, uniformed police and an ambulance arrived and Mr Wong was taken to A and E for treatment. After this, at about 2 a.m., Mr Wong made a formal complaint that he had been assaulted by police officers to Yuen Long Police Station. The voir dire: cross-examination of police witnesses 18. In the course of cross-examination Mr Ross asked DPC 50049 if he had taken part in the raid on 26 August at Yuen Long. He replied that he had not. He was then shown newspaper photographs taken of the aftermath of that raid. He identified that two of the police officers shown in the photographs were DSIP Chan and DPC 54856, both of whom had taken part in the raid on 9 July. DPC 34311was not shown in the photographs and when he came to give evidence on the voir dire it was not suggested to him that he took part in the raid on 26 August. It is common ground that he did not. 19. When DSIP Chan came to give evidence at the voir dire Mr Hemmings for the prosecution objected to his being asked questions about the raid on 26 August on the ground that this was part of an on-going investigation involving not only DSIP Chan’s team, but other teams. Mr Ross stated that he wished to cross-examine in order to point out “any similarities in the…improper way that the operation was carried out on the latter occasion”. The judge ruled “I think you should just put it as a general question that rough tactics were used at the arrest…I do not think you can put it any more than that”. 20. Mr Ross then put to DSIP Chan that at the raid on the 26 August either he himself used violence or he witnessed his officers using violence. To this he replied simply “no”. That was the extent of the cross-examination of the inspector on this point. 21. When DPC 54856 gave evidence at the voir dire he confirmed that he had taken part in the raid on 26 August. He was asked whether he had seen anyone inside the premises with a blanket over his head and he said that he had not. He was asked whether he had seen violence used on the occupants of the premises and he said that he had not. The voir dire: The application to call Mr Wong 22. Mr Ross gave no notice that he intended to call Mr Wong to give evidence at the voir dire. Before doing so, however, he informed Mr Hemmings that he was about to call a civilian who was involved, in a way that he did not specify, in the raid of 26 August. Mr Hemmings then objected to the admission of this witness’s evidence on the ground that it went to an entirely peripheral matter. Mr Ross sought to justify calling Mr Wong on the ground that his evidence fell within “an exception” to the “finality rule when there is bad character alleged”. He said that he anticipated that Mr Wong “will give evidence of the propensity of this team to commit similar kind of raids with accompanying violence as they did in this case”. The judge ruled: “Well, I don’t see that this is relevant evidence or viable evidence and I am not allowing this witness to be called”. 23. At the end of the voir dire the judge ruled that she was satisfied that the Appellant had made the statements attributed to him at the scene of the arrest and subsequently at the video-recorded interview and that this was not a case where she should exercise her discretion to exclude the evidence in question. 24. In the following week, before the trial began, Mr Ross renewed his application to admit the evidence of Mr Wong, though it is not entirely clear to me from the transcript whether he was seeking to re-open the voir dire or seeking leave to adduce the evidence of Mr Wong at the trial itself. He stated that Mr Wong would make specific allegations against two of the police witnesses who had given evidence at the voir dire, namely that one had put a blanket over the head of Mr Wong and another had punched him in the chest on the occasion of the 26 August raid. After argument the judge refused to amend her earlier ruling. The Appeal to the Court of Appeal 25. The Appellant applied to the Court of Appeal for leave to appeal against conviction[1]. One of the grounds of appeal was that the judge wrongly refused to receive the evidence of Mr Wong at the voir dire. His application was dismissed[2]. After consideration of relevant authorities the Court explained its reasons for rejecting this ground of appeal in the following passage: “In the 2nd applicant’s case the evidential link between the 2nd applicant’s testimony and that of the prospective witness, Wong Tsz Him, as described to the judge, was weak and tenuous. Each of them had allegedly received a punch(s) and had their heads covered, one with a hood but the other with a blanket. There was no other supporting relevant evidence. It was not suggested that the police officers had tried to force admissions from Wong Tsz Him. The two events were wholly unconnected investigations, separated in time by 7 weeks. Clearly, the nature of the cross-examination of the police officers did not go to an issue in the case. Rather, it was relevant to credibility only. Its relevance was tenuous at best and not of substantial probative value. We are satisfied that the prospective evidence was not such that fell within the exceptions to the ‘finality rule’ in respect of questions as to collateral matters, including credit. Accordingly, in our judgment the judge was correct to decline to receive that evidence.” 26. The Appeal Committee of this Court[3] granted leave to appeal on the ground that the following question of the requisite importance arose: “What is the correct principle for determining when a witness may be cross-examined as to past misconduct in a different case with a view to challenging his or her credit in the instant case; and for determining when, if at all, evidence may be led to impeach answers given by such witness in the course of such cross-examination.” The relevant law 27. Two basic common law principles as to the admission of evidence are in play on this appeal. The first is that no evidence is admissible unless it is relevant to an issue in the case. The second is that evidence should not be admitted if its probative weight is insufficient to justify the complexity that it will add to the trial[4]. Whether in a criminal or a civil trial the judge should be astute to prevent the primary issues becoming obfuscated as the result of the admission of evidence of insufficient significance to justify the additional burden that it will impose on the hearing[5]. These principles apply as much to a voir dire as they do to the trial itself. 28. In applying these principles it is helpful to distinguish between primary issues and collateral issues. Whether an event upon which the prosecution relies in fact occurred is a primary issue. Evidence that such an event did or did not occur is primary evidence. A collateral issue does not bear directly on whether a primary fact in issue occurred. It bears on the reliability of the primary evidence. The second principle normally applies only in relation to evidence that goes to a collateral issue. Evidence as to credit 29. The evidence that most often gives rise to the invocation of the two principles set out above is evidence as to credit. Where a witness gives evidence that is challenged, the veracity of that witness may be put in issue. The party challenging that evidence may wish to ask questions in cross-examination about the conduct of the witness on other occasions for the sole purpose of showing that the witness is not the kind of person whose word can be believed. Such evidence of how a witness behaved on another occasion, or an issue as to this, is collateral to the primary issue of whether the evidence of the witness is truthful. 30. In a criminal trial the exploration of how a witness behaved on another occasion for the purpose of attacking his veracity is likely to distract the court, or the jury, from the primary issue of the behaviour of the defendant, to the detriment of the fairness of the trial. For this reason there are long established common law rules of practice that restrict the extent of such an exploration. In some jurisdictions these rules have been codified[6]. The codification reflects the fact that this is not an easy area of the law to apply in practice and has proved a fertile source of appeals. 31. In this jurisdiction the relevant law has not been codified. It reflects principles applied in other common-law jurisdictions, particularly England and Wales, but in recent times has not always followed the approach of the courts of those jurisdictions to the detailed application of the basic principles. The approach in this jurisdiction was clearly set out by Li CJ in HKSAR v Wong Sau Ming[7] and the general principles that he set out merit repetition: 23. In this context, the relevant general principles are as follows. First, where the veracity of the witness is challenged on cross-examination as to credit, subject to the judge's discretion to disallow improper questions, cross-examination about discreditable acts is widely permitted. See Cross and Tapper on Evidence (9th ed., 1999) p.306. Its purpose is of course to show that the witness ought not to be believed on oath. 24. Secondly, the cardinal test of relevance in the law of evidence applies to such cross-examination. The subject matter of the cross-examination must be relevant to the witness's veracity. Relevance is a matter of degree. 25. Thirdly, in applying the test of relevance, the court should in its discretion usually permit questions in cross-examination as to credit if the truth of the imputation conveyed would materially affect the court's opinion as to the witness's veracity on the subject matter of his testimony. Conversely, questions in cross-examination as to credit are usually not permissible if the truth of the imputation conveyed would not have any material impact on such opinion. The imputation may relate to matters which are so remote in time or are of such a kind that the truth of the imputation would not have any material impact on the court's assessment of the witness's veracity. cf Sankey LJ in Hobbs v CT Tinling & Co Ltd [1929] 2 KB 1 at p.51. The matters about which the witness is questioned in cross-examination as to credit must relate to his likely standing after cross-examination with the tribunal in question. See R v Sweet-Escott (1971) 55 Cr App R 316 at p.320. 26. Fourthly, as a general rule, answers to questions in cross-examination as to credit on any ground including veracity are final and evidence is not admissible to contradict them. See Harris v Tippett (1811) 2 Camp 637 at p.638; 170 ER 1277. This can conveniently be referred to as "the finality rule". The rule is conducive to a fair trial. Its rationale is that, as a matter of commonsense, a criminal trial should be kept within proper limits. The court should focus on the issues in the case. Its attention should not be diverted to collateral issues such as credit, the exploration of which may unnecessarily blur or confuse the real issues in a cloud of detail. Although the rule has sometimes been criticized, it is well-established and necessary. However, the distinction between questions going to the issues in the case and those going to credit is sometimes difficult to draw and in some cases, can be a rather thin one. 32. Li CJ goes on to refer to well established exceptions to the finality rule. He observes that at this point the Hong Kong jurisprudence does not follow the strict approach adopted in some of the English cases, so that English cases should not be regarded as having any precedent value in Hong Kong[8]. 33. Li CJ found it unnecessary to go into the exceptions to the finality rule but observed that they allow proof by independent evidence of matters which tend to affect the credibility of a witness. He added that as the common law developed in Hong Kong there might be possible room for the relaxation of the finality rule and that the list of exceptions was not closed. This statement accords well with a statement of basic principle made by Lord Bingham CJ in R v Zomparelli[9] when commending guidance given by Lord Lane CJ in R v Edwards[10], a decision which, on its facts, has been questioned both in this jurisdiction and in England[11]: “…a criminal trial should be focused on the central issue in the trial: whether the defendant is or is not proved to have committed the crime or crimes of which he is accused. Unless firm control is exercised with that principle in mind, it is only too easy for that central issue to become submerged in a mass of subsidiary and collateral issues which have only an indirect or marginal bearing on the central question…but…the judge’s overall and paramount duty is to ensure the fairness of the trial. The trial process must be fair to the prosecution; the scales of justice are not balanced if heavily over-weighted in favour of the defendant. But it must be fair also to the defendant. He is entitled to a fair trial as a matter of constitutional right. No rule of law can restrict the duty of the court to ensure a fair trial.” 34. I can summarise the position as follows. Cross-examination as to credit should be permitted in relation to any matter that, if proved, will diminish a witness’s credibility. This may be a failure to tell the truth on an earlier occasion, or some quite different type of misconduct, providing only that it bears logically on the weight that should be accorded to the witness’s word[12]. Where a prosecution witness denies, under cross-examination as to credit, that he has been guilty of misconduct of a kind that would damage his credibility, evidence that unequivocally rebuts his denial, such as a finding by a court or tribunal to this effect, should normally be admitted for that purpose. Where, however, the truth of the allegation of misconduct made against the witness cannot be determined without what is, in effect, a separate trial relating to that issue, the finality rule will normally preclude the calling of evidence to challenge his denial. Similar facts evidence 35. Evidence that goes simply to credit must be distinguished from another type of collateral evidence – similar fact evidence. Where the trial raises a primary issue as to whether a police witness, or indeed anyone else, acted in a particular way, the fact that he acted in the same way on another occasion may render it more probable that he acted as alleged. Where it is the conduct of the defendant himself that is in issue, special rules fall to be applied when deciding whether relevant similar fact evidence should be admitted. These are designed to protect the defendant against unfair prejudice. Where, however, similar fact evidence is alleged against a prosecution witness similar principles apply to those that apply to evidence as to credit, although the application of those principles will not be identical. 36. Before a witness can be cross-examined as to how he behaved on an occasion unrelated to the trial, it must be demonstrated that the cross-examination is relevant. The judge must be satisfied that the matters to be alleged in cross-examination will, if established, be relevant to the determination of a primary issue. Where similar facts are relied on this will depend, in part, on the nature of the conduct in issue and the degree of similarity between that conduct and the ‘similar fact’ conduct alleged. If this test of relevance is satisfied, the judge should permit the cross-examination. 37. It is at the second stage that the approach to similar fact evidence may differ from the approach to evidence that goes simply to credit. Evidence that goes simply to credit will almost always have limited significance. The fact that a witness is someone whose word cannot be trusted does not mean that his evidence is necessarily false. This is the justification for the application of the finality rule in respect of evidence that goes simply as to credit. Similar fact evidence may, however, be so cogent as to have substantial probative value. If conduct alleged against a witness on two different occasions has the same particular features, and there is no question of collusion between the witnesses to the two different occasions, this may be a cogent indication of the truth of what is alleged in relation to each occasion. Thus where a witness denies on cross-examination that he behaved in the manner alleged on either occasion, this will not necessarily preclude the admission of the similar fact evidence in rebuttal of his denial. The decision whether or not to permit similar fact evidence to be called will depend upon the facts of the particular case. The judge will have to weigh the impact that the similar fact evidence will have, if believed, against the dislocation that its introduction will cause to the trial process. 38. The difference between the approach to evidence that goes merely to credit and similar fact evidence is well demonstrated by the case of HKSAR v So Kam Tong[13]. On a voir dire the defendant, who was charged with murder, alleged that admissions that he had made had been induced by violence in that the police had slapped his left ear, punched his chest and abdomen and squeezed his genitals, as well as making threats and inducements to confess. The judge none the less ruled that his admissions should go before the jury. There existed a letter of complaint by another person who had been suspected of the murder to the effect that the same police team, including the same senior officers, investigating the same complaint at about the same time had squeezed his genitals in an attempt to extract a confession. This letter was not disclosed, for which reason the defendant’s conviction for murder was set aside on the grounds of procedural irregularity. The relevance of this case is the significance attached by the Court of Appeal to the matter not disclosed. “It went beyond credit. It went to an issue in the applicant’s case…The defence case was that the police were desperate to find the missing lady and that the two senior officers running the team were willing to use and countenance violence in order to locate her. The complaint which Chu had made provided potential support for the applicant to that end. It was evidence that was capable of doing more than merely damaging the credibility of the police officers, by reference, say, to their conduct on some other occasion and in some other case. It was evidence about the behaviour of one police squad and its directing officers at exactly the same time in the course of precisely the same investigation which they were undertaking in respect of the applicant. It had the potential, in other words, to bear on what truly happened to the applicant that night.”[14] The application of the law to the facts of this case 39. As I remarked at the outset the relevant legal principles apply as much to a voir dire as they do to the trial itself. On the voir dire, however, the judge is better able to control the evidence and less likely to be confused by collateral evidence than a jury, so he may well properly give the defence more latitude. 40. The wording of the question of law in respect of which the Appeal Committee gave leave suggests that this case is about the admissibility of evidence as to credit, and the basis upon which the trial judge gave permission to cross-examine suggests that she may have been under the impression that this was aimed at impugning the credibility of the police witnesses. As the Court of Appeal remarked: “Clearly, the nature of the cross-examination of the police officers did not go to an issue in the case. Rather it was relevant to credibility only. Its relevance was tenuous at best and not of substantial probative value.” 41. I agree with the last comment. The allegations made in cross-examination that violence had been used in the raid on 26 August had little bearing on the question of whether or not the police officers should be treated as truthful witnesses. If all that was in issue was credibility, the judge was unquestionably right to apply the finality rule and to refuse the application to call Mr Wong to give evidence. I would add that I do not consider it relevant that the evidence that Mr Ross wished to adduce related to events after rather than before those that gave rise to the prosecution. The relevant principles apply equally in either case. 42. Before us Mr Ross made it plain that his object in cross-examining the police witnesses and in seeking to call Mr Wong was to adduce similar facts evidence. His submission was that evidence of what took place on 26 August should have been admitted because of its similarity to what the Appellant alleged took place on 9 July. Had the judge had the benefit of this evidence she might have formed a different view as to the admissibility of the evidence of statements made, or alleged to have been made, by the Appellant. 43. As the Court of Appeal remarked[15]at no stage of the trial was the judge provided with a witness statement of the prospective evidence of Wong Tsz Him. Where defence counsel applies to adduce evidence of events that are unconnected with those that have given rise to the prosecution, whether on the basis that they are relevant as to credit or on the basis that they are relevant as similar facts evidence, and he has a witness statement setting out the evidence in question, he ought to place that statement before the judge. Counsel should adopt the procedure set out by Li CJ in relation to evidence as to credit in HKSAR v Wong Sau Ming at paras 41 and 45. All relevant material should be placed before the judge. Only in this way will the judge be in a position to make a properly informed decision as to whether the evidence should be admitted. In a jury trial this should, of course, take place in the absence of the jury. 44. What should the judge have decided had she been fully appraised of the evidence that the Appellant wished to adduce? What bearing, if any, did this have on the issues before her on the voir dire? The issues on the voir dire 45. The evidence given by the Appellant on the voir dire was of serious misconduct by the team led by DSIP Chan, in particular, DPC 50049. The most serious of these were: i) DPC 50049 recorded in his notebook incriminating statements attributed to the Appellant that were fabricated. ii) DPC 50049 made the Appellant write and sign declarations in his notebook that the statements in question had been read to him that he had read them himself and that they were true, without permitting him to read the statements. iii) DPC 50049 ‘planted’ the keys of the flat on him. iv) Police officers made threats that prevented the Appellant from seeking the assistance of a lawyer, notwithstanding that he wanted such assistance. v) DPC 503 prevailed on the Appellant to say in the video-recorded interview that he had been given the keys of the flat because he wanted to go to the toilet and that this was the first occasion on which he had visited the flat. That statement was, as DPC 503 was aware, demonstrably untrue because the police had video records of the Appellant visiting the flat on a number of occasions. 46. It was also the Appellant’s case on the voir dire that at the time that he was arrested he was kicked several times by DPC 50049 and DPC 34311, that after he was handcuffed a hood was placed over his head, with the eyeholes at the back, so that he was unable to see what was going on, that DPC 50049 then punched his head and he heard similar treatment being meted out to one of his co-defendants. 47. The similarity between these alleged events and the events alleged to have occurred on 26 August are first the use of gratuitous violence by police officers and secondly the covering of the head of those whom they were detaining. These acts were not, however, alleged to have been done by the same police officers. Indeed DPC 50049 and DPC 34311, who were alleged to have been responsible for the violence on 9 July, did not take part in the raid on 26 August. The leader of the team was, however, the same man – DSIP Chan. 48. The Judge heard the conflicting evidence of the police officers who took part in the raid on 9 July and the Appellant. That evidence did not leave her in doubt as to whether the Appellant had made the statements recorded in DPC 50049’s notebook nor as to whether those statements, and the subsequent statements made by the Appellant in the video-recorded interview were made voluntarily. Had she been fully informed as to the evidence that Mr Wong was expected to give would she, or should she, have considered that it might lead her to revise her conclusions in relation to the evidence that she had heard? I have concluded that it neither would nor should have done so. 49. The fact that gratuitous violence was, or might have been, used on the Wong brothers by two members of DSIP Chan’s team on 26 August did not bear on the likelihood that other members of that team had fabricated admissions of the Appellant on 9 July, had prevented the Appellant from having access to a lawyer and had induced him to make demonstrably false statements when taking part in the video-recorded interview. Thus Mr Wong’s evidence had no relevance to the most significant aspects of the positive case advanced by the Appellant. The matter does not, however, end there. 50. An admission that is induced by oppression should not be admitted in evidence at a trial or, if admitted, should be disregarded by the jury. The reason for this is the protection that the law gives against self-incrimination. Even though a defendant denies making the statement in question, both the judge on the voir dire and the jury at the trial must, where there is evidence that raises the issue, consider whether or not if the statement was made it was voluntary. Unless satisfied of this the judge must rule the statement inadmissible on the voir dire and the jury must be directed to disregard the statement where it has not been excluded on the voir dire[16]. 51. Thus there may be two issues in relation to a statement alleged to have been made by the defendant. First, where the defendant denies making the statement, a patent issue of whether the defendant made the statement. Secondly a latent issue – if the defendant did make the statement, may it have been induced by oppression. 52. At a trial where the defendant gives evidence denying that he made an inculpatory statement it would plainly be likely to confuse the jury if he were permitted to lead similar fact evidence of police conduct on a different occasion for the purpose of buttressing an alternative case that, if he made the statement, it was forced out of him. For this reason, a judge would normally be likely to refuse an application to adduce such evidence. Indeed, I am not aware of any case in which such an application has been made. 53. The position is not the same on the voir dire. Even though the defendant gives evidence denying making the statement in question the judge will be alive to the issue of whether, if the statement was made, it was voluntary. That is what the voir dire is all about. Defence counsel will not necessarily be inhibited from addressing the judge on the hypothesis that his client may not be telling the truth when denying making the statement. 54. In the present case Mr Ross has now made it clear why he wished to adduce Mr Wong’s evidence. This was by way of similar fact evidence to support a case that the statements alleged to have been made by the Appellant were (if made) induced by oppression. Were the contested allegations in relation to behaviour of the police during the raid on 26 August of sufficient relevance and cogency in relation to that issue to justify the admission of Mr Wong’s evidence as similar fact evidence on the voir dire? I do not consider that they were. Other than the fact that gratuitous violence was alleged to have been used on the two occasions, there were few similarities between them. First and foremost, the police officers who were alleged to have used violence on the Appellant and his co-defendant were not alleged to have used violence on Mr Wong or his brother. Indeed they were not even present on 26 August. The team had the same leader on both occasions but not all the members of the team were the same. The nature of the alleged assaults was not the same. The Appellant alleged that a hood was put on his head that prevented him seeing what was being done by the police in the flat. Mr Wong alleged that his head was covered by a blanket while he was being assaulted. 55. The judge did not elaborate on the chain of reasoning that led her to conclude that the Appellant had made the statements alleged and had done so voluntarily, but I cannot conceive that her conclusion would have been affected had she entertained evidence from Mr Wong as to what occurred on the night of August 26th. 56. The facts of this case have raised an unusual issue as to the application of the principles governing the admission of collateral evidence. I have, however, been left in no doubt that no criticism can be made of the refusal of the Judge to permit Mr Wong to give evidence on the voir dire. 	I would, accordingly, dismiss this appeal. Chief Justice Ma: 57. For the reasons contained in the judgment of Lord Phillips of Worth Matravers NPJ, this appeal is dismissed. Mr Phillip Ross, instructed by Patrick Mak & Tse, assigned by the Director of Legal Aid, for the Appellant Mr William Tam, DDPP, and Mr Ivan Cheung, PP, of the Department of Justice, for the Respondent [1] CACC 51/2011, 3 October 2013 before Hon Stock VP, Lunn JA and McWalters J. [2] Judgment 8 November 2013. [3] 4 June 2014, Mr Justice Ribeiro PJ, Mr Justice Tang PJ and Mr Justice Fok PJ. [4] O’Brien v Chief Constable of South Wales Police [2005] UKHL 26 at para 11; [2005] 2 AC 534, at 543. [5] Goldsmith v Sandilands [2002] HCA 31 at para 1, (2002) 190 ALR 370. [6] eg: in England and Wales, the Criminal Justice Act 2003 s.100; in some parts of Australia provisions under the Uniform Evidence Acts – see Goldsmith v Sandilands (supra) at para 66; in New Zealand the Evidence Act 2006 sections 37-39. [7] (2003) 6 HKCFAR 135 [8] para 52 [9] 1999/04971/Z5 (unreported: 23 March 2000) [10] [1991] 1 WLR 207 [11] See O’Brien (supra) at para 41 [12] R v Malik [2000] 2 Cr App R 8 at p 12 [13] [2010] 5 HKLRD 101. [14] per Stock V-P at para 37. [15] Para 90 [16] See HKSAR v Pang Hiu San FACC No 3 of 2014 where this area of the law is dealt with at length. 
Chief Justice Ma: 1. I agree with the Judgment of Mr Justice Cheung PJ. Mr Justice Ribeiro PJ: 2. I agree with the Judgment of Mr Justice Cheung PJ. Mr Justice Fok PJ: 3. I agree with the Judgment of Mr Justice Cheung PJ. Mr Justice Cheung PJ: 4. At the conclusion of the hearing, we dismissed this appeal. These are my reasons for doing so. The facts 5. This appeal concerns the requirement under the common law offence of attempting to pervert the course of justice that the accused’s act has to have a tendency to pervert the course of justice, in the context of an investigation by the Independent Commission Against Corruption (ICAC) against the appellant. 6. The appellant used to be a staunch supporter of Mr C Y Leung in his (successful) bid to become the Chief Executive of the Hong Kong Special Administrative Region (HKSAR) in the 2012 election. However, the relationship between the two turned sour after Mr Leung assumed office. Days following some public criticism of Mr Leung’s policies by the appellant in the local media, the appellant was arrested by the ICAC on 8 January 2013 for having conspired with others to commit an offence contrary to section 9(3) of the Prevention of Bribery Ordinance[1] and section 159A of the Criminal Procedure Ordinance.[2] The alleged offence arose out of the acquisition of an oil field in the United States by a listed company, Pearl Oriental Oil Ltd, of which the appellant was the Vice-President and Executive Director. After attending a video recorded interview, the appellant was released on bail later that night. 7. On the following day, the appellant through his personal secretary sent an email to the Chief Executive complaining about the ICAC arrest and investigation. In the email, he protested his innocence and claimed that it was a framed up charge against him. He expressed the strong suspicion that it was orchestrated by the Chief Executive in revenge for his criticisms against his policies. In the penultimate paragraph of the email, he urged the Chief Executive to: “make careful consideration from a political and from the overall perspective and urgently halt the groundless persecution against me, so as to avoid causing a huge shock to society!”[3] 8. In the final paragraph, he concluded his email by asking the Chief Executive to remind the Commissioner of the ICAC “to ask his subordinates to be extra cautious, or else a political bomb will be detonated!”[4] 9. On the same day, the appellant through his personal secretary sent another email of essentially identical content to the Commissioner. 10. On 10 January 2013, the appellant again through his personal secretary prepared a letter addressed to the Chief Executive and copied to the Commissioner, which letters were delivered to the Chief Executive’s Office and the ICAC Headquarters respectively. In the letter, the appellant accused the Chief Executive as the instigator of the “persecution” against him by the ICAC, and he once again proclaimed his innocence. He threatened that if the ICAC did not stop the groundless accusation against him and the other directors of the listed company and other people involved in the case, he would: “immediately announce the Call for Leung Chun Ying’s Resignation to domestic and overseas media and to the Central People’s Government which contains shocking insider information to perish together with the ungrateful man who returns kindness with evil … .” [5] 11. In the following paragraph, he urged the Chief Executive to “take decisive measures immediately to halt the persecution”[6] against him and the other directors and management personnel of the listed company. He added: “Please promptly instruct the Independent Commission Against Corruption to stop the unreasonable investigation.”[7] 12. On 20 August 2013, the appellant was charged with one count of “doing acts tending and intended to pervert the course of public justice” by the ICAC. The particulars of the charge stated that by his two emails and the letter, he sought by asserting his past dealings and association with the Chief Executive, by threat or by intimidation to influence the Chief Executive and/or the Commissioner to terminate an investigation being carried out by the ICAC against him and/or others. 13. In 2015, the Pearl Oriental case involving the appellant and some others went to trial in the High Court on two counts of conspiracy to defraud together with one count of dealing with the proceeds from an indictable offence.[8] On 20 May 2015, the jury unanimously found the appellant not guilty on all charges and he was acquitted accordingly, whereas two co-defendants were found guilty and convicted of some of the charges. (A third co-defendant had absconded before the trial.) The proceedings below 14. However, in relation to the charge against him of attempting to pervert the course of justice, the appellant was tried in the District Court before HH Judge Frankie Yiu and was convicted on 29 February 2016.[9] He was sentenced to 18 months’ imprisonment. 15. On 2 March 2018, the Court of Appeal dismissed the appellant’s appeal.[10] On 25 May 2018, the court granted the appellant’s application for a certificate that points of law of great and general importance were involved in its judgment which ought to be considered by this court – these became Questions 2, 3 and 4 in this appeal: “[Question 2] Given the mandatory duty to investigate alleged or suspected offences under section 12(b) of the Independent Commission Against Corruption Ordinance (Cap 204) (‘the Ordinance’)[11] and the concurrent lack of any statutory discretion as in section 12(a) of the Ordinance: (i) Does the Commissioner of the Independent Commission Against Corruption (‘the Commissioner’) have an operational discretion based on the Common Law which is identical in nature to that possessed by the UK Commissioner of Police; and (ii) If the answer to (i) is in the affirmative, is the nature of this discretion which is based on the Common Law capable of overriding his mandatory duty to investigate under section 12(b) to the extent of empowering him to stop any ongoing criminal investigation on bona fide grounds? [Question 3] Reading Article 57 of the Basic Law in conjunction with sections 5, 12, 14-17 of the Ordinance, does the Commissioner’s accountability to the Chief Executive of the Hong Kong Special Administrative Region (‘the Chief Executive’) include matters of operational issues concerning criminal investigations of individual cases, and is not merely confined only to matters of the direction and administration of the Independent Commission Against Corruption? [Question 4] If the answers to Questions [2] and [3] above are all in the affirmative, does the Chief Executive have the same operational discretion which is based on the Common Law as that of the Commissioner which is capable of overriding the Commissioner’s mandatory duty to investigate under section 12(b) of the Ordinance to the extent of empowering the Chief Executive to direct the Commissioner to stop any ongoing criminal investigation on bona fide grounds, or does the Chief Executive have any additional or other legal power to direct the Commissioner to stop any ongoing criminal investigation on bona fide grounds?”[12] 16. These three questions arose from the arguments between the prosecution and the appellant, both in the District Court and before the Court of Appeal, as to whether what the appellant had done had any tendency to pervert the course of justice, the appellant’s argument in gist being that it did not since neither the Chief Executive nor the Commissioner had the legal power to stop an ongoing ICAC investigation. In other words, despite the appellant’s threats and intimidation expressed in the emails and letter, the Chief Executive and the Commissioner simply had no legal power to do what the appellant wanted them to do. 17. Although the Court of Appeal rejected this contention of a lack of legal power,[13] it nonetheless regarded the legal issues raised as having great and general importance meriting consideration by this court. As will be seen, notwithstanding that the Court of Appeal has addressed the issues involved in Questions 2, 3 and 4 in some detail, I do not find it necessary to do so. The Appeal Committee’s certified question 18. On 2 November 2018, the Appeal Committee granted leave to appeal to the appellant on the basis of an additional certified question raising a point of law of great and general importance, which it regarded as a prior question to the three questions certified by the Court of Appeal.[14] This prior question thus became Question 1 in this appeal: “(Question 1) Where an accused seeks to cause a person to stop or otherwise interfere with a criminal investigation, is it incumbent on the prosecution, on a charge of attempting to pervert the course of justice, to establish that the person approached by the accused could, by the lawful exercise of a legal power he possesses, stop or interfere with the criminal investigation in order to prove that the accused’s act had a tendency to pervert the course of justice?” 19. The Appeal Committee pointed out that Questions 2, 3 and 4 would arise if, but only if, Question 1 was answered affirmatively. In other words, on the facts of this case, if it was unnecessary for the prosecution to establish that the Chief Executive or the Commissioner could, by the lawful exercise of a legal power that he possesses, stop or interfere with the criminal investigation in order to prove that the appellant’s acts had a tendency to pervert the course of justice, Questions 2 to 4 would not arise at all. For this reason, Question 1 assumes great importance in this appeal. The general legal principles 20. The law on the common law offence of attempting to pervert the course of justice is well established. The offence is “the doing of some act which has a tendency and is intended to pervert the administration of public justice”.[15] The actus reus is the doing of the act with the prohibited tendency and the mens rea is the intention to pervert the course of justice.[16] So far as the actus reus is concerned, as this court has pointed out: “(a) Although it is called an ‘attempt’, it is a substantive offence. But like an inchoate offence, it is committed even where the act with the prohibited tendency does not actually result in interference with the course of justice. (b) A person’s conduct has a tendency to pervert the course of justice if it has a tendency towards ‘impairing (or preventing the exercise of) the capacity of a court or competent judicial authority to do justice’. (c) The tendency must involve impairing the administration of justice in curial proceedings. Conduct with a tendency to result in unlawfulness which does not interfere with curial proceedings does not constitute the offence.”[17] 21. For an act to have a tendency to pervert the course of justice, it must have a tendency to bring about a miscarriage of justice in curial proceedings. It is not necessary for such proceedings to have been instituted at the time of the acts in question but the relevant acts must have a tendency and be intended to frustrate or deflect the course of curial or tribunal proceedings which are imminent, probable or even possible.[18] 22. Investigations by law enforcement agencies do not by themselves form part of “the course of justice” so that acts which hinder or interfere with their investigations are not sufficient in themselves to constitute a perversion of the course of justice. However, if such acts of interference carry a tendency and are intended to pervert the course of justice in relation to curial proceedings which may result from the investigation, they are capable of founding the offence.[19] The present appeal 23. On the facts of the present case, there is no dispute that even though no curial proceedings were in place at the material time, subject to the issues raised in this dispute, the appellant’s acts of asking the Chief Executive and the Commissioner to stop the investigation which might result in criminal prosecution against himself and others did fall within the ambit of the offence of attempting to pervert the course of justice. 24. The central focus of this appeal is whether the appellant’s acts had a “tendency” to pervert the course of justice, that is, to terminate or otherwise interfere with the ICAC investigation. Essentially, the appellant’s position is that there was no such tendency, or more precisely, the prosecution has not proven beyond reasonable doubt that there was such a tendency. Indeed, that was the genesis of the legal issues which eventually became the three questions of law certified by the Court of Appeal. Mr Joseph Tse SC (Ms Doris Ho and Mr Byron Tsang with him), for the appellant, pointed out that at trial, the prosecution had adduced no evidence on how the Chief Executive or the Commissioner could cause or bring about the termination of the ICAC investigation, apart from relying on various legal provisions and some common law principles to say that they have the legal powers to do so.[20] That was how these legal provisions and principles became the subject of contention before the lower courts. HKSAR v Tsui King Sing 25. In the printed case, as in the courts below, Mr Tse placed considerable emphasis on the case of HKSAR v Tsui King Sing[21]in support of his argument that it was not proven that the appellant’s acts had any tendency to pervert the course of justice. In Tsui King Sing, the defendant was a co-owner of a coffee shop which was burgled by a former employee. He approached the father of the employee and offered to either “drop” the case against her, or not attend court as a witness, for a sum of $80,000. By that time, the employee had been arrested, interviewed under caution, apparently charged and, given the time period involved, no doubt had been brought before a magistrate. The defendant was convicted after trial of doing an act tending and intended to pervert the course of public justice. His appeal to the Court of First Instance was allowed on the ground of impossibility, that is, the case was, by the time of the conversation between the defendant and the father, likely beyond any stage (if such a stage ever existed) of the defendant being able to prevent the prosecution proceeding, and that the defendant’s offer to not give evidence was entirely meaningless as he was not (and was never intended to be) a witness in the proceedings.[22] 26. Agreeing with counsel’s submission, McMahon J observed that whether an act has a tendency to pervert the course of justice is a question of fact. To pervert the course of justice means to “deflect, frustrate, impair or hinder” the ability of a court to administer justice.[23] The learned judge went on to point out: “26. Of course it was for the prosecution to prove the appellant’s offer to ‘drop’ the prosecution had a tendency to pervert the course of justice, and it seems to me there was insufficient evidence before the Magistrate so as to allow him to conclude beyond reasonable doubt that it was possible for the appellant to pervert the course of justice in the way the appellant had offered. 27. There was no evidence before the Court, as I understand it, as to whether the police would or could take such a request from the victim of an offence, such as the burglary of a coffee shop, into account in deciding whether to proceed with the case. It may be in many cases an act by a defendant will manifestly have a tendency to pervert the course of justice and little or no evidence will be required to establish that fact, but in the circumstances of this case, such evidence should have been called by the prosecution. Without such evidence it is difficult to see how the appellant's offer, regardless of his quite obvious intent, could have been found to have the required ‘clear or manifest tendency’ of perverting the course of justice: HKSAR v Wong Shing Yim & Others and R v Rogerson & Others (1991-1992) 174 CLR 268. 28. It seems to me that the Magistrate's finding, set out above, was really based on his own understanding of how police may react to a request to ‘drop’ a case and not on the evidence before him. Equally there was no evidence as to the possibility of the appellant becoming a witness in the trial of PW1's daughter. Factually, it seems to me that, on the evidence, there was some doubt as to whether in the circumstances of this case it had been proven the appellant's offer had the necessary tendency to pervert the course of justice.” My views 27. I agree that it is a question of fact in each case whether the defendant’s act had the tendency to pervert the course of justice. If the act is not proven to have such a tendency, no offence has been committed. Bribing a witness to give false evidence when the witness has already completed his evidence in court may well be one such example. Threatening a juror with unpleasant consequences if he should find the accused guilty when the jury has already returned a verdict and been discharged is another such, perhaps extreme, example. In both these examples, and also in Tsui King Sing as regards the offer to drop the case, the relevant act of the defendant may, as a matter of fact, simply have come too late to have any tendency to pervert the course of justice. 28. Lack of a tendency to pervert the course of justice could also arise in other situations. The offer not to give evidence as a witness in Tsui King Sing when the defendant was not and was never intended to be a witness in the criminal proceedings provides another example of the lack of a tendency to pervert the course of justice. McMahon J essentially found that the defendant was simply not relevantly connected with the criminal proceedings such that his offer had no bearing on the proceedings. A variation of the same theme may be where the person in question, though not totally unrelated in some sense to the proceedings, is simply, say, too junior or too remotely connected to have any possible effect on the proceedings. Thus, to bribe a security guard of the court building where the Court of Appeal sits to influence the outcome of a pending appeal before the court would be one example. 29. However, contrast that last example with the situation where the offer of a bribe is made to one of the appeal judges hearing the appeal. The tendency to pervert the course of justice in this latter example would be manifest and plain. 30. Vary this example slightly. What about a bribe or a threat made to the spouse of the judge hearing the case? In my view, while again depending on all the facts, it is obviously highly likely that the bribery of or threat to the spouse would be found to have a tendency to pervert the course of justice. 31. This brings me to the type of situation faced in this appeal, that is, approaching the Chief Executive and the Commissioner with a view to stopping or otherwise interfering with an ongoing ICAC investigation. The Chief Executive and the Commissioner are, respectively, the head of the HKSAR[24] and the head of the ICAC.[25] Moreover, the Chief Executive is the very person to whom the Commissioner is accountable in respect of his work.[26] In my view, writing to them asking them to stop an ongoing ICAC investigation, with the threat that otherwise a huge political bomb would be detonated, plainly involves a tendency to pervert the course of justice. This case is far removed from the examples I gave earlier of situations where a court may find that there was no tendency to pervert the course of justice. 32. First, the emails and letter were written at a time when the investigation was ongoing. There is no question of the matter having gone past the stage where it could be interfered with. 33. Secondly, by virtue of his position, the Commissioner was directly and most relevantly connected with the ICAC investigation. So far as the Chief Executive is concerned, he is the person to whom the Commissioner is accountable. Moreover, he is the head of the HKSAR. He is plainly a relevant person connected with the ICAC investigation. 34. To suggest, as the appellant did, that in order for the approaches to them to have a tendency to pervert the course of justice, the Chief Executive and the Commissioner must have the legal power to stop the investigation, is to overlook the reality of the situation (regardless of whether, as a matter of law, they do or do not have the legal power to terminate the investigation). Given the unique constitutional and legal position of the Chief Executive, he is undoubtedly in a position to influence or otherwise affect the Commissioner’s and, through the Commissioner, his officers’ handling of the investigation, if not effectively to stop it altogether. For instance, he may seek to persuade the Commissioner, or to put pressure on him, to take steps or adopt courses that are favourable to the appellant. That the Chief Executive may or may not be successful in what he seeks to do is not the point. In my view, what he can do (if he wants to or is forced to) would be quite sufficient to constitute a tendency to pervert the course of justice, particularly when one remembers that the attempt to pervert the course of justice need not be successful.[27] Bribing a witness who is prepared to accept a bribe, and attempting to bribe an honest and upright witness who cannot be bribed makes no difference in terms of the commission of the offence. The person offering the bribe commits the offence of attempting to pervert the course of justice in both cases. Likewise, whether the bribed witness’ evidence is accepted by the court is irrelevant. Offering the bribe to him amounts to an offence. 35. Similarly, given the position of the Commissioner as head of the ICAC,[28] there would be many things that he could do, or attempt to do, whether directly or indirectly, and whether administratively or otherwise, to frustrate or delay the investigation, or to otherwise interfere with the investigation or affect its outcome, if not to bring about its termination altogether. He may or may not be successful in his attempt. But that is not of relevant importance. The contrary position, with respect, is unrealistic and untenable. 36. In neither case, in the discussion above, is it incumbent upon the Chief Executive or the Commissioner to formally invoke any legal power to terminate the investigation as such. 37. Mr Tse argued that there are checks and balances on what the Commissioner can or cannot do regarding an investigation. In particular, there is the independent Operations Review Committee which, in short, monitors the investigation works of the ICAC and offers advice in respect of the same, and there is also the Secretary for Justice whose legal advice whether to prosecute the ICAC has to seek upon the conclusion of an investigation. 38. With respect, this argument confuses a tendency to pervert the course of justice with a certainty that the course of justice will be perverted. Counsel’s argument would mean that threatening a single justice of appeal sitting in an appeal does not amount to the offence of perverting the course of justice as he sits in a collegiate court, or that bribing one juror is no offence as a jury comprises seven or more jurors. This argument also overlooks the fact that the offence is committed not only if there is a tendency that the proceedings are terminated as a result of the prohibited act, but also where there is a tendency that they are otherwise interfered with or affected as a consequence. Answer to Question 1 39. For all these reasons, in my view, Question 1 must be answered in the negative, insofar as the present type of case is concerned. Depending on the facts, however, what is true in relation to the Chief Executive and the Commissioner may or may not be true with other office-holders. The answer in each case must depend on the facts. 40. The above answer in the present appeal effectively removes the underlying premise of the legal questions argued before the trial judge and the Court of Appeal, that is, certified Questions 2 to 4. They are simply irrelevant to the question of whether the appellant’s acts had the tendency to pervert the course of justice. Mr Tse’s further arguments 41. Mr Tse argued that the above analysis would only hold true if his client were charged with attempting to pervert the course of justice by interfering with the ongoing investigation. However, according to the particulars of the charge, the offence charged was to attempt to pervert the course of justice by stopping the investigation. Without a legal power on the part of the Chief Executive or the Commissioner to do so, the charge cannot stand. 42. I do not accept this argument. As has been pointed out by this court, whilst the particulars of a charge are undoubtedly important, the overriding test in charging an offence in Hong Kong is one of fair notice.[29] 43. In the present case, what the appellant was charged with was the common law offence of attempting to pervert the course of justice, which is, of course, not limited to the termination of the relevant investigation or curial proceedings. It is true that the particulars of the charge referred to the appellant by his emails and letter seeking to influence the Chief Executive and/or the Commissioner to terminate the ongoing investigation. It is also true that the gravamen of the complaint against him was the various threatening and intimidating matters he wrote in the emails and the letter and the ultimate goal that he sought to achieve, that is, the termination of the investigation. However, whilst the termination of the investigation was the appellant’s goal, what he attempted to achieve via his threats and intimidations must also include the Chief Executive’s and the Commissioner’s interfering with, delaying or otherwise favourably affecting the investigation or its outcome. And all this would equally amount to an attempt to pervert the course of justice with which he was charged. This could not have been lost on the appellant or his legal advisers. In my view, there is no question of their not having received fair notice of it. 44. This brings me to another point raised by Mr Tse, that is, that throughout before the trial judge and the Court of Appeal, the prosecution never raised the argument that apart from the strict legal powers that the prosecution said the Chief Executive and the Commissioner possess to terminate an ongoing investigation, a tendency to pervert the course of justice could otherwise be established. It was only at the leave stage before the Appeal Committee that for the first time it was argued that it was not necessary for the prosecution to establish such legal powers. 45. In my view, counsel’s argument had no significance, as he was not running any fair trial argument or abuse argument such that the prosecution should not be allowed to run this argument before us. 46. Nor has counsel objected to the Appeal Committee’s certifying Question 1 (referring not just to the Chief Executive and the Commissioner stopping the investigation but also interfering with it) and putting it as a prior question to Questions 2 to 4, when granting leave to appeal on that basis. Indeed, in the written case, the appellant seems to accept that the court should have regard not just to the legal position but “should extend it to include practical terms”.[30] In any event, the Court is not confined to the questions formulated by the parties but may itself raise issues necessary for the effective determination of the appeal, provided of course that no procedural unfairness is involved. 47. One last argument of Mr Tse was that leaving aside the Chief Executive’s and the Commissioner’s (supposed) legal powers to terminate the investigation, there was no or insufficient evidence adduced by the prosecution at trial to prove how they could otherwise interfere with the investigation. 48. Mr Tse was right in saying that the prosecution did not adduce any specific evidence to prove how the Chief Executive or the Commissioner could interfere with the investigation. However, for the reasons explained above, I do not think any such specific evidence was required. Given the unique and head positions that they held, both the Chief Executive and the Commissioner were obviously in a position to interfere with or affect the investigation or its outcome in a way favourable to the appellant (if they wanted to or were forced to). Disposition 49. In conclusion, I answer Question 1 in the negative (insofar as the present type of case is concerned). Questions 2 to 4 therefore do not arise, and I do not find it necessary or desirable to express any views on them in these reasons for dismissing the appeal. I wish, however, to make it clear that by not addressing the issues in these other Questions, I should not be taken as in any way agreeing with or endorsing the analysis of the lower courts on them. Mr Justice Gleeson NPJ: 50. I agree with the Judgment of Mr Justice Cheung PJ. Mr Joseph Tse SC, Ms Doris Ho and Mr Byron Tsang, instructed by Ledwin Au & Co, for the appellant Mr Johnny Mok SC, on fiat, and Mr Jonathan Lin, SPP, of the Department of Justice, for the respondent [1] Cap 201. [2] Cap 221. [3] Per the certified translation. The original Chinese reads: “從政治、全局高度考慮問題，緊急制止對我的無理迫害，避免在社會引起巨大震憾！” [4] “請他的手下慎之又慎，否則將引爆政治炸彈！” [5] “立刻向中外媒體及中央人民政府公布包含驚天內幕的《敦促梁振英辭職書》，與恩將仇報者同歸於盡！” [6] “請您立即採取果斷措施，制止對 [我] 的迫害！” [7] “迅速指示廉政公署停止不合理的調查。” [8] HCCC 561/2013. [9] DCCC 819/2013. [10] [2018] 2 HKLRD 967. [11] I will refer to this Ordinance as the ICAC Ordinance. [12] [2018] HKCA 281. [13] Para 154(II) of the main judgment. [14] [2018] HKCFA 48. [15] HKSAR v Wong Chi Wai (2013) 16 HKCFAR 539, para 30, quoting from R v Vreones [1891] 1 QB 360, 369. [16] Ibid,para 30. [17] Ibid, para 31. [18] HKSAR v Egan (2010) 13 HKCFAR 314, para 126. [19] Egan, para 127. [20] Articles 48 and 57 of the Basic Law; sections 5, 8, 9, 11, and 12 of the ICAC Ordinance; section 39(1) of the Interpretation and General Clauses Ordinance (Cap 1); and the English common law on the UK Commissioner of Police’s operational discretion. [21] [2008] 3 HKLRD 213. [22] Para 23. [23] Para 24. [24] Article 43(1) of the Basic Law. [25] Section 5(1) of the ICAC Ordinance. [26] Article 57 of the Basic Law; section 5(1) and (2) of the ICAC Ordinance. [27] See Wong Chi Wai, para 31(a). [28] See para 31 fn 25 above. Note also the powers vested in the Commissioner under section 12 of the ICAC Ordinance. [29] HKSAR v Gammon Construction Ltd (2015) 18 HKCFAR 110, paras 42 to 43; HKSAR v Ng Yee Man Yvonne (2015) 18 HKCFAR 405, para 21. [30] Appellant’s written case, para 49. 
Press Summary (English) Press Summary (Chinese) DCCC 245/2020 [2021] HKDC 38 IN THE DISTRICT COURT OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO 245 OF 2020 ------------------------------ ------------------------------ --------------------------------------- REASONS FOR VERDICT --------------------------------------- Introduction 1. The defendant, a foreign student from Kazakhstan studying at a local university who has never had any trouble with the law before, was allegedly trying to ignite a petrol bomb in the streets of Kowloon in the small hours on 18 November 2019 and arrested. 2. He denies a charge of attempted arson with intent[1] and is legally represented by Mr Lawrence Lok SC pro bono leading counsel Ms Queenie Ng. A Russian interpreter has been arranged for the benefit of the defendant throughout the trial. 3. Particulars of Offence NUKPI Abilkaiyr, on the 18th day of November 2019, near the junction of Nathan Road and Jordan Road, Yau Ma Tei, Kowloon, in Hong Kong, without lawful excuse, attempted to damage by fire property belonging to himself or another, being reckless as to whether such property would be destroyed or damaged, and being reckless as to whether the life of another would be thereby endangered. 4. The petrol bomb in question consists of a glass bottle partly filled with inflammable liquid (xylene in the present case) and a cloth wick soaked in xylene and fixed securely around the mouth of the bottle. Issues 5. This is a simple and straightforward case, i.e. a case without any independent corroborative evidence, the only evidence against the defendant (the prosecution called 2 witnesses – acting sergeant 7720 and exhibit officer DPC 10518) involves the testimony of an eyewitness only, that is the acting sergeant 7720. 6. In a nutshell, sergeant testified that he saw the defendant trying to set light to a petrol bomb with a lighter several times but failed, and in the nick of time he was able to subdue the defendant before the next attempt. 7. The defence, however, contended that it did not happen in this way. It was put to the sergeant during cross-examination that the defendant was with other protesters at the time and the petrol bomb was passed to him by one of them; no sooner had he got hold of it than he was pushed down to the ground and subdued. The sergeant categorically denied having seen anything like this and added that the defendant was all by himself at that point in time, there were no protesters around him or in his vicinity. 8. Exercising his rights as a defendant, the defence called no evidence at trial. Since there is no evidence from the defence; the whole case will then turn on the credibility/reliability of the sergeant, the question to be asked is “Is it safe to rely on sergeant’s account of events as being accurate and genuine?” 9. When dealing with a case like this, effectively a one-on-one situation, the court will as always approach it with caution. 10. Put in another way, in order to found a conviction, the prosecution, whose onus is to prove the case beyond reasonable doubt which is a very high standard, have to satisfy me so that I am sure what the sergeant had described was exactly what happened on the night in question. Anything short of that would mean the defendant’s guilt is not proven to the required standard and the defendant will be acquitted. A Summary of Sergeant’s Account of the Event 11. A. At around 2:10 a.m. on 18 November 2019, the sergeant (a member of the anti-terrorists unit) and his colleagues were dispersing a crowd of protesters of about 300 in Nathan Road. B. The crowd kept shouting and yelling and at times threw petrol bombs, bricks and hard objects at them to stall their advancing. C. When coming near the junction between Jordan Road and Nathan Road at about 2:12 a.m., he saw the defendant in black clad wearing respirator and gloves about 10m in front of him, all alone by himself; the sergeant was not at the forefront, some of his colleagues were in front of him, some behind him. D. The defendant facing the police had in his right hand a glass bottle the mouth of which was stuffed with a piece of white cloth, he estimated the bottle was half filled with liquid, and had in his left hand a lighter. E. Sergeant saw sparks flying out from the lighter every time when the defendant was trying to set light to the white cloth but wouldn’t light, this went on for 4 to 5 times (such action lasted for 2 to 3 seconds). The sergeant said if successful, he would have thought the defendant would throw the bottle at them. F. It took 5 seconds for the sergeant to run up to the defendant, first thing he did was to reveal his police identity, and then he tried to control the defendant, but a violent struggle between the two ensued. G. Eventually, the defendant was subdued on the ground face down and handcuffed behind his back. When sergeant helped him up, a bottle was lying on the ground right under his chest where the defendant fell. H. On the ground sergeant picked up 4 items, i.e. the glass bottle, a cap, a respirator and a pair of labour gloves (the last 3 items were worn by the defendant but fell off at the time). I. Sergeant did a search of the backpack carried by the defendant for 2 to 3 minutes and found two items inside, i.e. a pair of goggles and a head wrap (a piece of black cloth); he then put them back in afterwards. J. About half an hour later at 2:48 a.m., sergeant declared arrest of the defendant for the offences of unlawful assembly, possession of offensive weapon (referring to the petrol bomb) and wearing mask. K. At 2:53 a.m., sergeant passed the defendant and his backpack together with those exhibits described above on to the exhibit officer, DPC 10518, before moving off to deal with the situation ahead. Insofar as the lighter sergeant said the defendant held in his left hand is concerned, it has not been recovered (Indeed no attempt was made to search for it at scene but not to confuse with the lighter allegedly discovered by the exhibit officer in the defendant’s backpack later in the police station). Analysis 12. When it comes to analyse the evidence, I ask myself the question “whether sergeant’s testimony and his behaviour at scene as a whole was broadly consistent with other known facts?” The first known fact 13. The first known fact comes to mind is that sergeant arrested the defendant for the offence of possession of offensive weapon rather than attempted arson. Generally speaking, possession refers to those situations where the person has yet put the weapon (the petrol bomb in the instant case) into any specific use – he just had it with him at the time (simple possession). Here the defendant, according to the sergeant, went much further than possession- he was in the act of lighting the petrol bomb - only at one remove from arson – that would constitute attempted arson. 14. It was submitted by the defence that the fact that sergeant arrested the defendant for offences other than “attempted arson” showed that even sergeant himself did not regard the defendant having done anything tantamount to attempted arson. 15. The fact that the defendant was arrested for possession at least in theory gives rise to two possible scenarios; one refers to the situation where the person simply had it with him, or another like the sergeant’s version – he had done something further with it. Faced with two different factual situations, the court will go through a process of analysis of the evidence before it comes to any conclusion. 16. It may be argued that the police are not lawyers, more leeway should be allowed if the officer fails to get the charge right at the time when they declare arrest of someone for an offence. It often happens the accused is subsequently charged with an offence quite different to the one for which he was initially arrested. 17. At first glance, this may sound reasonable but when combining it with the second known fact and viewed collectively, it may lead to a different conclusion. The second known fact 18. The other known fact is that no attempt was ever made by anyone, in particular the sergeant, to look for the lighter allegedly held by the defendant at the scene. 19. Counsel for the defendant, Mr Lok SC, submitted that as regard the defendant’s purported acts of igniting the lighter 4-5 times, which is the only basis to justify the defendant’s commission of the attempted arson, there is absolutely no reason why the sergeant simply did not care about the lighter at all at the scene. The sergeant did nothing to locate the purported lighter, nor had he requested any of his colleagues to look for it. It was further submitted that the sergeant’s allegation of the defendant’s attempt to ignite the glass bottle is clearly a blatant lie. 20. Obviously, the key points in sergeant’s narrative involved a petrol bomb and a lighter. Sergeant should have known what exhibits, i.e., the lighter and the petrol bomb, to look for in support of his allegations so as to prevent them from being reduced to bare assertions. The more relevant exhibits recovered, the stronger would be the case against the defendant. 21. In the circumstances, it is only natural for sergeant to do a search for the lighter in the immediate surrounding, not to say a thorough one but at least a quick one. While on the other hand, judging from the evidence before the court, sergeant did not seem to be in a hurry to leave the spot so as to get on with some pressing matters. 22. He said having dealt with the defendant, he then moved on to assist other colleagues in dispersing the crowd; here he was talking about a period of time - no less than 41 minutes - with the defendant (for he first encountered the defendant at around 02:12 and by the time he passed the backpack to exhibit officer before moving on, it was 02:53). 23. During this time, sergeant testified to spending 2 to 3 minutes searching the backpack of the defendant and finding two more items, namely, a pair of goggles and a head wrap. Now that he had taken the trouble to search the backpack, why didn’t he look for the lighter at scene? 24. On cross-examination, sergeant acknowledged that nowhere in his first statement did he mention that he had looked for the lighter at scene. Even in his second statement which was meant to supplement further details, all that he said in relation to this matter is “no discovery of the lighter” – nothing suggests that a search was conducted at the scene for the lighter. 25. It remains a major concern for the court since sergeant had stayed at the scene for quite some time (41 minutes), if what he said about “seeing the defendant trying to light the petrol bomb with a lighter” was true, how come no attempts were made to locate an important piece of evidence, the lighter, in the immediate surrounding area where the struggling occurred. 26. Up to this stage, it has been shown that the sergeant’s behaviour at scene was not broadly consistent with what he said had happened. If the defendant was trying to light the petrol bomb several times with a lighter, he should have been arrested for attempted arson and the sergeant should have searched the scene for the lighter, but instead the contrary had happened. The defendant was arrested for possession of offensive weapon and no attempt to search for the lighter, does this lend support to the case put by the defence? 27. Before discussing the case put by the defence, however, I’d like to deal with one other aspect of sergeant’s testimony the truthfulness of which the defence strongly challenged. Additional Details 28. A last-minute addition – sparks were seen flying out from the lighter when the defendant attempted to light the petrol bomb - was made in the supplementary statement (second statement) prepared by the sergeant on 3 November 2020 (almost one year after the event and just 2 days before the trial) at the request of the Department of Justice (DOJ) after his first statement and notebook (made on the day of the event) having been reviewed by DOJ. 29. The key additions are as follows (matters not referred to in his earlier statements -the first statement and notebook): (1)- The seeing of sparks flying out 4 to 5 times from the lighter when the defendant tried to light the petrol bomb; (2)- No recovery of the lighter at scene. 30. Assuming for a moment there was no such a review by DOJ, sergeant would in the circumstances be expected to testify in accordance with his earlier statements, the main theme of his narrative would then become something like this - “he saw the defendant trying to ignite a petrol bomb with the lighter”, that would be the end of it- no description of sparks seen or anything like that would be expected. 31. In fact, we know how he described it for he was asked under cross-examination to read out the relevant part in his first statement: 「.....我見AP右手手持一個玻璃樽，而樽口用白布塞住內有液體思疑汽油彈，左手企圖用打火機點着白布，用汽油彈向我方投擲....」 “I saw AP holding a glass bottle in his right hand, and white cloth was stuffed at the mouth of the bottle which I suspect to be a petrol bomb, attempted to use the lighter to ignite the white cloth, use the petrol bomb to throw towards us.” 32. In relation to the part concerning sparks seen, sergeant was asked under cross-examination why he failed to make any reference to it in his first statement or in his notebook if it was a fact, sergeant explained that what he had recorded was to him sufficient (meaning enough for us to picture what the defendant was doing at the time), and it did not occur to him he needed to write it down until he was told by DOJ to give further details. 33. Counsel for the defendant, Mr Lok SC, submitted that this was a total fabrication because it was an important piece of evidence and not complicated at all that no reasonable person would have omitted it in the contemporaneous records (i.e. the notebook and the first witness statement) and thus called into question the credibility of sergeant. 34. The prosecuting counsel, Miss Tsoi, argued that there was no inconsistency between what was recorded in the first and second statements. If I understand her correctly, what she was saying is that the additional details relating to the sparks did not substantially change the nature or the main theme of the recorded version in the first statement and that sergeant thought it sufficient was not an unreasonable explanation. 35. First of all, the issue here does not appear to turn on one’s memory because sergeant seemed to suggest that he remembered it all along, it was his decision not to write down the further details because what was recorded initially was to him sufficient for the purposes of the case. 36. To consider this alone, it may be a matter of different opinion (not much one could criticize the sergeant’s explanation), but in the grand scheme of things, though not a clear contradiction, no reliance should be placed on the sergeant on this point. 37. Bearing in mind the inconsistent behaviour of the sergeant at scene with what he said had happened; against this background, his explanation of “sufficiency” does not seem to be convincing at all. The Case Put by The Defence 38. I must emphasize from the outset that what was put to the police witness in the instant case has not been established as evidence of the case mainly because it had not been agreed to by the witness concerned nor had the defence called any evidence in support of it - it remains only the case theory of the defence. Unless and until it is supported by evidence, the court cannot act on it but it does provide the court with an alternative perspective to consider the prosecution case. 39. It is worth mentioning that at no time did the defence imply that the defendant had had a lighter on him during the incident. Assuming that was what happened (just like the case put by the defence), it would make perfect sense for sergeant simply to arrest the defendant for possession of offensive weapon (the petrol bomb) because he was merely holding it; and as there was no lighter in his hand, that explained why sergeant did not bother to search for it at scene. The case put by defence tallied with what sergeant did. 40. In view of the foregoing analysis, the court is of the view that it is not safe to rely on the sergeant’s testimony. In other words, the court is not sure whether the event really happened in the way as described by sergeant. That being the case, the court is left with a doubt, a reasonable one, with the sergeant’s version of events. As the sergeant’s evidence being the central matter of the prosecution case, it rises or falls with it. 41. As there is one outstanding matter, though not directly relevant to the facts in issue (it does go to the credit of the witnesses concerned), the evidence of which was adduced before the court, and for the sake of completeness I am inclined to deal with. 42. It is the alleged discovery of an orange lighter in the defendant’s backpack by the exhibit officer (DPC 10518) later in the police station. The Finding of an Orange Lighter 43. First of all, one thing I like to make clear is that it was never put forward as part of the prosecution case by Miss Tsoi, the prosecuting counsel, that the orange lighter was the one sergeant said the defendant was using to ignite the petrol bomb. While, on the other hand, the defence disputed its provenance. 44. Accordingly, the orange lighter is not directly relevant to the real issue of the case but is a fact that may go to the credit of the police witnesses concerned. 45. Before looking at the evidence of the exhibit officer, it is worth revisiting what the sergeant said (or did not say) on this aspect, basically he made no reference to the discovery of an orange lighter despite his searching the defendant’s backpack for 2 to 3 minutes (What he found were a pair of goggles and a head wrap (a piece of black cloth)). 46. Now turning to the exhibit officer, he testified in court that at 02:53 hours the sergeant handed over to him the defendant together with his backpack and some exhibits for custody and later back in the police station, he found the orange lighter from the backpack for the first time at 04:00 hours and seized it as exhibit. When asked by the court where exactly in the backpack he found it, he said it was lying at the bottom of the big compartment. 47. On the face of it, there doesn’t seem to be any significant inconsistency between the testimony of the sergeant and the exhibit officer; however, what the exhibit officer said in his witness statement is quite different from his evidence on the stand. What he described in his witness statement was that when sergeant handed over the defendant and the exhibits to him at scene, the exhibits included A to J items with J being the orange lighter, so it wasn’t the case that it was found back in the police station. 48. Now the question to be asked is whether it is worthy of belief that the orange lighter originated from the defendant’s backpack? 49. If the lighter was in the backpack, I find it difficult to believe that sergeant could have missed it when he spent 2 to 3 minutes going through the backpack. I have seen the backpack in court, it was of simple design; it was not like the lighter was kept in one of the small pockets inside the backpack or anything. 50. Bearing in mind sergeant seemed not to be in a hurry to leave because of the urgency of the situation at that time, he was with the defendant for 41 minutes before handing him over to the exhibit officer, he had had plenty of time to do a thorough search of the backpack if a couple of minutes were not enough. Could the contrary be true (no lighter ever in the backpack)? Judging from the evidence of the sergeant, it was possible. 51. Now consider the evidence of the exhibit officer, he told the court that he found the lighter back in the police station at 4 a.m., giving us the impression that this was the first time he discovered the lighter. But what he described in his witness statement was quite a different version – It was written like this in Chinese: 「及後於同日02:53時ASGT 7720將AP及下述證物交予本人看管。 a…. … … j. 一個橙色打火機」 “Later at 02:53 hours the same day ASGT handed over the defendant and the following exhibits to me for custody. a… ….. ….. j. an orange lighter.” 52. There are A to J items of exhibit handed over to him by sergeant, the exhibit officer named every one of them from A to J with J being the orange lighter, which would mean the exhibit officer knew what exhibits passed to him including the orange lighter. 53. It was stated in no uncertain terms in the statement what exhibits were handed over to him by the sergeant. His explanation was that what he meant was some of them were handed over by sergeant and some seized by him. I don’t think this was borne out by the clear wording he used in his statement. 54. He either found the lighter back in the police station for the first time at 04:00 hours or it was to his knowledge handed over to him by sergeant at 02:53 hours at scene - both statements cannot be true at the same time, which strikes me that the exhibit officer is less than honest. 55. In fact, what the exhibit officer said in his witness statement suggested that both of them knew about the existence of the lighter at scene, that might cast doubt on the testimony of the sergeant for he made no mention of the lighter when he passed those exhibits to the exhibit officer. Could this be the reason why the exhibit officer gave a version on the witness stand different to the one as recorded in his witness statement? 56. Given the evidence as it is, the court is unsure of who to believe and what to make of the orange lighter; consequently, the court cannot accept that the orange lighter was found in the defendant’s backpack. Conclusion 57. For the reasons given above, the prosecution failed to prove the case against the defendant to the required standard, i.e. beyond all reasonable doubt; that being the case, the defendant is therefore acquitted. [1] Contrary to ss 60(2) and (3), 63(1) and 159G of the Crimes Ordinance 
Chief Justice Ma: 1. I agree with the judgment of Lord Walker of Gestingthorpe NPJ and with the orders contained in para 79 below. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Chan NPJ: 4. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Lord Walker of Gestingthorpe NPJ: Introduction 5. This appeal is the last stage, or almost the last stage, in protracted and contentious litigation concerned with the will of Nina Wang (who is referred to in this judgment, as she was in the judgments below, as Nina). At her death in 2007 Nina owned the Chinachem Group of companies, valued in 2012 at over HK$ 82 billion. She was reputed to be the richest woman in Asia. She left a will written in the Chinese language and dated 28 July 2002. It was what is sometimes called a homemade will - that is, made without professional advice from a lawyer. She was assisted in making it by her sister, Kung Yan Sum (“Mrs Tong”). The will named Chinachem Charitable Foundation Limited (“the Foundation”) as the principal beneficiary, but difficult and technical legal issues have arisen as to the meaning and legal effect of her will. 6. These questions of construction were not however raised in legal proceedings until 2012, when the present proceedings were commenced by an originating summons issued by the Secretary for Justice. One of the functions of the Secretary for Justice is to protect the interests of charities. Charities are a matter of general public concern because their essential characteristic is that they are for the public benefit, or the common good. 7. This delay of five years was unavoidable because there were contentious probate proceedings raising the prior issue of whether Nina’s 2002 will was indeed her last will. In those proceedings (HCAP 8/2007), commenced soon after Nina’s death, it was contended that her 2002 will had been superseded by another will allegedly made by her in 2006. On 2 February 2010 Lam J (as he then was) pronounced in solemn form for the 2002 will. The Court of Appeal dismissed an appeal on 14 February 2011 and on 24 October 2011 this Court refused leave for a further appeal. 8. Nina’s will did not contain any appointment of executors. Homemade wills seldom do. On 26 March 2012 Joseph Lo Kin Ching and Derek Lai Kar Yan were appointed as interim administrators of her estate. They were joined as the first defendant to the originating summons issued by the Secretary for Justice. The summons was amended twice following changes in the administrators and the present administrators, Chan Wai Tong Christopher, Wong Tak Wai and Jong Yat Kit are the second respondent to this appeal. The administrators have adopted a neutral position throughout the litigation. 9. The Foundation was the second defendant to the summons, and is the appellant in this Court, the Secretary for Justice being the first respondent to the appeal. The third defendant to the summons, and the third respondent to the appeal, is Nina’s very elderly mother, who would be entitled to any part of Nina’s estate as to which she died intestate. Nina’s mother entered an appearance in the proceedings but has taken no further part in them. 10. The agreed English translation of Nina’s will is set out in full in paragraph 25 below. After introductory words it contained four clauses. Clause 1 contained a gift of Nina’s property to the Foundation. Clause 2 contained provisions about the appointment of a managing organisation to supervise the Foundation, and about the funding of “a Chinese prize of worldwide significance similar to that of the Nobel Prize”. Clause 3 related to the Foundation’s management of the Chinachem Group. Clause 4 contained provisions about the Foundation providing support for members of the family of Nina’s late husband Teddy Wang (“Teddy”), and staff of the Chinachem Group and their children. 11. The principal issue in the proceedings, raised by paragraph 1 of the originating summons, is whether Nina’s net residuary estate is held by her administrators “in trust for – (a) the Foundation absolutely, or (b) the Foundation on terms that the Foundation holds the same in trust to give effect to all (or some) of the directions in clauses 2, 3 and 4 of the Will; or (c) some other person or persons and, if so, whom.” The third possibility – which might have been spelled out as devolution on intestacy – has not been adopted by any of the parties. Nina’s mother has, as already mentioned, taken no active part in the proceedings. It seems that the question was included in the originating summons because clause 4 appears to contain some sort of beneficial disposition (it is appropriate to put it in general terms in this introduction) of an unquantified part of the estate for purposes that are not charitable in the legal sense. It is settled law that the purposes of a trust must be wholly charitable if it is to be a valid charitable trust, with the well-known case of the will of Caleb Diplock (see Chichester Diocesan Fund v Simpson [1944] AC 341) as a stern warning of the possible consequences of infringing this principle. The Secretary for Justice and the Foundation are united in wishing to uphold the validity of the will, and their counsel’s submissions on clause 4 have been carefully crafted so as, one way or another, to avoid any possible intestacy. 12. What divides the Secretary for Justice and the Foundation is whether the question in paragraph 1 of the originating summons is to be answered in terms of subparagraph (a) (“for the Foundation absolutely”) or subparagraph (b) (under which the Foundation would hold as a trustee and be obliged to give effect to all or some of the directions in clauses 2, 3 and 4). Most non-lawyers, and indeed many lawyers, may find it hard to understand the intensity of the arguments on this abstract and technical issue. It is common ground that the Foundation is a company with exclusively charitable objects, recognised since its incorporation in 1988 as exempt from tax under section 88 of the Inland Revenue Ordinance, Cap 112. Its charitable objects are not narrowly restricted, but are expressed in wide terms. The whole of its capital and income, after expenses, must be used for its charitable objects, and in the event of winding-up its surplus assets would be devoted to charity. Whether these obligations arise within the framework of company law, and under the Foundation’s memorandum of association, or within the framework of trust law, with Nina’s will as the trust instrument, may seem little more than a technicality. But the issue has been argued at length in elaborate and erudite written and oral submissions, first before Poon J, then in the Court of Appeal, and finally in this Court. 13. In his judgment handed down on 22 February 2013, after a three-day hearing, Poon J answered the main question in terms of subparagraph (b), and stated that the other questions that had been argued did not arise. The general effect of his judgment was that the Foundation would hold as a trustee and be obliged to give effect to clauses 2, 3 and 4 of the will so far as possible. The Foundation appealed to the Court of Appeal (Lam VP, Cheung and Kwan JJA), and the Court handed down judgment on 11 April 2014 after a three-day hearing. The Court dismissed the appeal in a judgment of the Court given by the Vice-President. The judgment concluded (paragraph 130) that “…the Judge was correct in coming to the conclusion that the bequest to the Foundation under the Will is to settle the estate of Nina to it as a trustee holding it on a charitable trust under Clause 2 with discretionary power to make distribution during the perpetuity period under the power in Clause 4”. 14. The Court of Appeal directed that the originating summons should be restored at first instance for further directions. The further appeal to this Court has postponed that step being taken, but the restoration of the case at first instance may be necessary in order to give full effect to this Court’s judgment. That is why the proceedings before this Court may not be the very last stage in the litigation. It is to be hoped that any further proceedings will be relatively brief and non-contentious. The facts 15. The facts are not in dispute, and are fully and clearly set out in the judgment of Poon J. The summary that follows is largely derived from his judgment. 16. Nina and Teddy both came from Shanghai, where Teddy’s father had started the Chinachem business. They married in 1955, when Teddy was 20 and Nina was 18. In the following years they worked hard to expand the business of the Chinachem Group, which became one of the largest property developers in Hong Kong. In 2012 it employed about 3,300 people and had an annual operating profit of about HK$ 2.48 billion. Nina was a philanthropist who made donations to many charities in Hong Kong, mainland China and elsewhere. In 1988 she and Teddy incorporated the Foundation as a company limited by guarantee. They were the first directors. 17. The main objects of the Foundation are set out in clause 3(1) and (2) of its memorandum: “(1) To organise, promote, participate and carry out the following activities :- (a) vocational, scientific and technical education programs. (b) provision of books, equipment and student facilities to schools, colleges and universities. (c) social survey and research projects. (d) rural and community development and cultural programs. (e) religious activities for the worship of God. (f) relief of human sufferings of the poor, the aged, the disabled, the sick, weaklings, children or any other persons in need. (g) relief to victims of fire, flood, famine, war, pestilence or other calamities. (2) To grant donations, scholarships, financial or material assistance, including payment of passages, travel, living allowances and other incidental expenses and to subscribe funds to individuals and institutions for the furtherance of activities which appear to the Foundation to be of value to the community, or to attain any of the Foundation’s objects.” 18. In 2004 clause 3 was altered in two respects. In clause 3 (1) (b) the words “non-profit making” were added before “schools”. In clause 3(3) the words “In furtherance of the objects, but not otherwise the Foundation may exercise any of the following powers:-” were inserted at the beginning, and the original subclauses (3) to (17) were renumbered as paragraphs 1 to 15 of subclause (3), so that (for instance) clause 3(3) became clause 3(3.1) and clause 3(17) became clause 3(3.15). Poon J was correct in observing that these alterations did not change the substance of the objects. They clarified what had probably been the meaning of the original text. 19. Clause 5 of the memorandum provides that the Foundation’s income and property are to be applied solely in the promotion of its objects and that no income or property is to be distributed to its members. Clause 8 provides that on the winding up of the Foundation any surplus assets are not to be distributed to the members but are to be transferred to a charitable institution or institutions, to be determined by the members or in default of such determination by a Judge of the High Court. 20. Under article 36 of the Foundation’s articles of association, its business is to be managed by a board of governors. The first governors were Nina and her husband. On 2 April 1990 two employees of group companies were appointed as additional governors. On 10 April 1990 Teddy was kidnapped and was never seen again by his family or friends. On 22 September 1999 the High Court made an order that he was to be presumed to be dead. In the meantime one of the employee-governors had resigned and two other employees had been appointed. There was no further change in the board until 1 April 2007 (two days before Nina’s death) when Nina’s brother Kung Yan Sum (“Dr Kung”) and her sisters Gong Zhong Xin and Mrs Tong were appointed as governors, and one of the employee-governors resigned. 21. Since Nina’s death on 3 April 2007 there has been no further change in the board of governors. On 2 November 2007 the Foundation’s articles were amended (by articles 49(a) and (b)) to confer on the board power to invite persons to become patron and vice-patrons of the Foundation, and to set up a supervisory board to advise the board of governors on matters concerning the Foundation. 22. During Nina’s life, especially after her husband’s disappearance and presumed death, she was the moving force behind the Foundation. The written evidence of Dr Kung was that she would decide what donations were to be made, and Chan Kam Por, one of the employee-governors, would arrange for money to be distributed to the Foundation by one of the group companies, so enabling the donations to be made. Dr Kung’s evidence also shows that donations were made to a wide variety of organisations and causes, most of which were clearly charitable. But there were some gifts to political groups which could not be regarded as charities. That is however past history and is not an issue in this appeal. 23. Nina’s sister, Mrs Tong, gave written evidence as to the circumstances in which Nina made her will, with Mrs Tong’s assistance, during July 2002. There is no evidence that Nina was not then in good health. She was diagnosed with cancer early in 2004 but did not become gravely ill until 2006. Mrs Tong’s evidence is admissible under section 23B of the Wills Ordinance, Cap 30, to assist in the interpretation of meaningless or ambiguous language in Nina’s will. Paragraphs 4 to 10 of Mrs Tong’s affirmation dated 3 August 2012 are set out in full in paragraph 13 of Poon J’s judgment, which may be referred to for amplification of the summary that follows. 24. These were the salient points of Mrs Tong’s evidence. (a) Nina telephoned Mrs Tong and asked for help with the preparation of her will. Mrs Tong was surprised that she, rather than a lawyer, should be asked to help. She thought it was probably because she had in the past often helped Nina with documents to be written in Chinese. (b) Mrs Tong was not surprised that Nina wished to leave her estate to charity, since she had expressed this wish several times before. Nina wished to leave her estate to the Foundation, and said that she hoped it could develop its charitable activities “to acquire worldwide influence similar to that of the Nobel prize.” (c) Mrs Tong asked Nina “how to achieve internationality”. Nina suggested that the Premier of the People’s Republic of China and the Chief Executive of Hong Kong should be involved in the supervision of the Foundation. Mrs Tong then suggested the Secretary-General of the United Nations. Nina “said laughingly on the phone that she hoped it could be achieved.” (d) Nina hoped that the Foundation would exercise its control over the management of the Chinachem group. She was aware that the Foundation would be replacing her in the control of the group as the Foundation’s “business empire.” (e) In a later telephone call Nina referred to the need to provide for her late husband’s parents (Wang Din Shin and Madam Run Yuk Chun) and for his younger sister Wang Teh Hwa (referred to in the affirmation as Ah Hwa). Nina also mentioned her wish to provide support for other siblings of Wang Teh Hwa if they needed it. Nina also mentioned providing care and assistance for staff of Chinachem Group companies and their children. (f) Mrs Tong drafted a will. It was amended on a few points which are agreed to be immaterial. Nina executed it on 28 July 2002, and showed it to Mrs Tong a few days later. 25. The Chinese text of Nina’s will is set out in paragraph 15 of Poon J’s judgment. The certified English translation is as follows (with the insertion in square brackets of further numbering used by counsel in the course of argument for the sake of clarity): “ I, Kung Yu Sum, solemnly make my will as follows: 1. ‘Chinachem Charitable Foundation Limited’ was set up by me and my husband, Wang Teh Huei, jointly. After I pass away, all of my properties shall be bequeathed to ‘Chinachem Charitable Foundation Limited’. 2. [1] After I pass away, I wish to entrust ‘Chinachem Charitable Foundation Limited’ to the supervision of a managing organization jointly formed by the Secretary General of the United Nations; the Premier of the PRC Government as well as the Chief Executive of the Hong Kong Special Administrative Region. [2] Under its supervision, [i] not only must ‘Chinachem Charitable Foundation Limited’ continue all the projects which it has undertaken since its establishment to enable their developments continuously, but [ii] it must also continue to achieve the purpose of setting up a fund and a Chinese prize of worldwide significance similar to that of the Nobel Prize. 3. The Board of Directors of ‘Chinachem Charitable Foundation Limited’ shall practically manage the company’s businesses and capital under the supervision of the abovementioned supervising organization, to safeguard and expand the ‘Chinachem Group’ as well as all the business which we have set up, to ensure the continuous growth of the business empire of the ‘Chinachem Charitable Foundation Limited’ and with part of its profits, to continuously develop the charitable business till eternity. 4. ‘Chinachem Charitable Foundation Limited’ must continue to achieve: (1) provide for the seniors of the Wang’s family, Mr Wang Din Shin and Madam Run Yuk Chun, which provision should be in accordance with their wishes and be satisfactory to them. (2) support Wang Teh Hwa’s living and medical expenses, to look after her children as well as their needs of advanced studies. As regards the other siblings of my husband, Wang Teh Huei, if necessary, the company also has the obligation to look after them. If their children pursue university or advanced studies, the company should be responsible (for the costs). (3) ‘Chinachem Charitable Foundation Limited’ has the obligation to provide care and assistance to the staff of the ‘Chinachem Group’ and their children, to encourage them to continue studying and keep on enhancing their own value, to encourage them to work hard together towards achieving the goal of bringing benefit to mankind.” 26. The “seniors of the Wang’s family” mentioned in clause 4(1) of the will both died in 2010. As to clause 4(2), Teddy had four siblings and five nephews and nieces, none of whom was supported by the Foundation during Nina’s lifetime. As to clause 4(3), during Nina’s lifetime neither the Chinachem Group nor the Foundation had any scheme for assistance of group staff and their children. Since Nina’s death the Chinachem Group has established such a scheme for its staff, and has been consulting about an educational scheme for the children of its staff. As at 2012 it had spent about HK$3.45m on the staff scheme. The construction of wills 27. A will has two legal characteristics which have in the past tended towards the construction (that is, interpretation) of wills being regarded as different from the construction of other legal documents. First, it is a wholly unilateral instrument. It embodies, not a bilateral or multilateral commercial arrangement, but the will-maker’s personal wishes and directions as to the disposition of his or her estate. Second, it is an ambulatory instrument in the sense that it has no effect until the maker’s death, which may not occur for many years. 28. These two features are still very relevant to the construction of wills. But the modern tendency is for the court, while recognising the points of difference between wills and commercial documents, to place still more emphasis on principles that are applicable to the construction of all legal documents. This tendency is referred to in the judgments below (in particular, Poon J at paragraphs 33 and 34, and the Court of Appeal at paragraph 34) but they did not have the advantage of the important recent judgment delivered by Lord Neuberger in the Supreme Court of the United Kingdom in Marley v Rawlings [2015] AC 129. That will was concerned with an extraordinary situation in which a husband and wife, each of whom wished to make a new will, each contrived to sign the will prepared for the other, without either of them, or the witnesses, or the solicitor who prepared the wills, noticing the mistake. 29. That is not of course the position in this case, but Lord Neuberger’s general observations on the construction of wills call for citation at length: “17 Until relatively recently, there were no statutory provisions relating to the proper approach to the interpretation of wills. The interpretation of wills was a matter for the courts, who, as is so often the way, tended (at least until very recently) to approach the issue detached from, and potentially differently from the approach adopted to the interpretation of other documents. 18 During the past 40 years, the House of Lords and Supreme Court have laid down the correct approach to the interpretation, or construction, of commercial contracts in a number of cases starting with Prenn v Simmonds [1971] 1 WLR 1381 and culminating in Rainy Sky SA v Kookmin Bank [2011] 1 WLR 2900 . 19 When interpreting a contract, the court is concerned to find the intention of the party or parties, and it does this by identifying the meaning of the relevant words, (a) in the light of (i) the natural and ordinary meaning of those words, (ii) the overall purpose of the document, (iii) any other provisions of the document, (iv) the facts known or assumed by the parties at the time that the document was executed, and (v) common sense, but (b) ignoring subjective evidence of any party's intentions. In this connection, see Prenn, at pp 1384–1386 and Reardon Smith Line Ltd v Yngvar Hansen-Tangen (trading as H E Hansen-Tangen) [1976] 1 WLR 989, per Lord Wilberforce, Bank of Credit and Commerce International SA v Ali [2002] 1 AC 251, para 8, per Lord Bingham of Cornhill, and the survey of more recent authorities in Rainy Sky, per Lord Clarke of Stone-cum-Ebony JSC, at paras 21–30. 20 When it comes to interpreting wills, it seems to me that the approach should be the same. Whether the document in question is a commercial contract or a will, the aim is to identify the intention of the party or parties to the document by interpreting the words used in their documentary, factual and commercial context. As Lord Hoffmann said in Kirin-Amgen Inc v Hoechst Marion Roussel Ltd [2005] 1 All ER 667, para 64, “No one has ever made an acontextual statement. There is always some context to any utterance, however meagre.” To the same effect, Sir Thomas Bingham MR said in Arbuthnott v Fagan [1995] CLC 1396, 1400 that “courts will never construe words in a vacuum”. 21 Of course, a contract is agreed between a number of parties, whereas a will is made by a single party. However, that distinction is an unconvincing reason for adopting a different approach in principle to interpretation of wills: it is merely one of the contextual circumstances which has to be borne in mind when interpreting the document concerned. Thus, the court takes the same approach to interpretation of unilateral notices as it takes to interpretation of contracts: see Mannai Investment Co Ltd v Eagle Star Life Assurance Co Ltd [1997] AC 749, per Lord Steyn at 770C–771D, and Lord Hoffmann at 779H–780F. 22 Another example of a unilateral document which is interpreted in the same way as a contract is a patent—see the approach adopted by Lord Diplock in Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183, 243, cited with approval, expanded, and applied in Kirin-Amgen at paras 27–32 by Lord Hoffmann. A notice and a patent are both documents intended by its originator to convey information, and so, too, is a will. 23 In my view, at least subject to any statutory provision to the contrary, the approach to the interpretation of contracts as set out in the cases discussed in para 19 above is therefore just as appropriate for wills as it is for other unilateral documents. This may well not be a particularly revolutionary conclusion in the light of the currently understood approach to the interpretation of wills (see eg Theobald on Wills, 17th ed (2010), chapter 15 and the recent supplement supports such an approach as indicated in Royal Society for the Prevention of Cruelty to Animals v Sharp [2011] 1 WLR 980, paras 22, 31). Indeed, the well known suggestion of James LJ in Boyes v Cook (1880) 14 Ch D 53, 56, that, when interpreting a will, the court should “place [itself] in [the testator's] arm-chair”, is consistent with the approach of interpretation by reference to the factual context. 24 However, there is now a highly relevant statutory provision relating to the interpretation of wills, namely section 21 of the 1982 Act (“section 21”). Section 21 is headed “Interpretation of wills—general rules as to evidence”, and is in the following terms: ‘(1) This section applies to a will— (a) in so far as any part of it is meaningless; (b) in so far as the language used in any part of it is ambiguous on the face of it; (c) in so far as evidence, other than evidence of the testator's intention, shows that the language used in any part of it is ambiguous in the light of surrounding circumstances. ‘(2) In so far as this section applies to a will extrinsic evidence, including evidence of the testator's intention, may be admitted to assist in its interpretation.” Section 21 of the English Administration of Justice Act 1982 was the pattern for section 23B of the Wills Ordinance, Cap 30. Indeed, they are in identical terms. 30. This is a valuable statement of the modern law but it does, with respect, pass over one of the characteristics by which a will or codicil is still in a class of its own – that is, its ambulatory character. A will speaks from the will-maker’s death. Although the “arm-chair principle” looks to context as at the time that the will is made, any general description of a class of property or a class of beneficiaries will (in the absence of a special context) look to the position as it is at the will-maker’s death. This rule is statutory as regards property (section 19 of the Wills Ordinance, Cap 30) but is equally clear as regards a class of beneficiaries. If a married man makes a testamentary gift “to my children in equal shares” the gift is to a class ascertained as at his death, so as to include any later-born children and exclude any children who die in their father’s lifetime. This point calls for brief mention as the judgments below may possibly have gone a little too far in their enthusiasm for the “arm-chair principle” (which is limited to facts known at the time the will is made). 31. Two all-important general principles of construction are that words must be read and understood in their context, and that the will must be read as a whole. In Charter Reinsurance Co Ltd v Fagan [1997] AC 313, 384, Lord Mustill neatly combined the two principles in a single phrase, referring to the need to read words “in the landscape of the instrument as a whole”. But how, in practical terms, is the court to go about its task of reading the will as a whole? Some helpful guidance was given, again by Lord Neuberger, in his dissenting judgment (upheld by the Supreme Court) in Re Sigma Finance Corporation. In the Supreme Court Lord Mance said ([2010] 1 AER 571, paragraph 12), “Lord Neuberger was right to observe that the resolution of an issue of interpretation in a case like the present is an iterative process, involving ‘checking each of the rival meanings against the other provisions of the document and investigating its commercial consequences’.” This applies to a will also, if “commercial” is read as “practical”. The iterative process is often laborious. It may require the court to go forwards and backwards painstakingly between the various words and phrases, occurring in different parts of the document, which give rise to the problem. That is the process that must be followed in construing Nina’s will. 32. In undertaking that task the court may, and should, take account of the fact that the will was not drafted, or even considered, by a lawyer. Lord Sterndale MR said in Re Taylor [1923] 1 Ch 99, 105, “When a will is obviously not drawn by a skilled lawyer, as is the case here, the words used may be looked at less strictly than is the case where it is so drawn”. That approach can be traced back to Lord Chancellor Parker in Forth v Chapman (1719) 1 P Wm 663,666, who said that a testator who is “inops consilii” (that is, deprived of advice) may be supposed to use words in a common, and not in a legal sense. 33. Professionally drafted wills usually follow the same general pattern. First comes the appointment of executors, sometimes accompanied by directions as to the funeral. Then come minor gifts, usually in the form of specific and pecuniary legacies. There may also be gifts of annuities, and specific gifts of immoveable property. Then come the constitution and disposition of the residuary estate, accompanied by more or less elaborate administrative powers and provisions if all or part of the residuary estate is to be held in trust. The typical home-made will, by contrast, tends to begin with what is most important to the will-maker. Then other, perhaps less important, matters occur to the will-maker and are dealt with, not always in a logical order. This sort of process can easily lead to ambiguities and apparent contradictions. 34. It is appropriate to refer at this point to the principle variously known as the rule in Lassence v Tierney (1849) 1 Mac & G 551 (in which it was held that the rule did not apply, on the language of that particular will) or the rule in Hancock v Watson [1902] AC 14 (in which the rule was not only stated, but also applied). It was stated in the latter case by Lord Davey (at p22) in a passage cited in paragraph 26 of the Court of Appeal’s judgment: “It is settled law that if you find an absolute gift to a legatee in the first instance, and trusts are engrafted or imposed on that absolute interest which fail, either from lapse or invalidity or any other reason, then the absolute gift takes effect so far as the trusts have failed to the exclusion of the residuary legatee or next-of-kin as the case may be”. 35. The Court of Appeal went on to say that the rule in Hancock v Watson is not a rule of construction. That is, with great respect, a surprising view. Rules, or principles, of construction are not all of a kind. The rule that a will must be construed as a whole is obviously a different sort of rule from that which lays down the normal legal meaning of the word “money” (see Perrin v Morgan [1943] AC 399, especially Lord Romer at pp 420-421) and terminology can vary. But numerous authorities, and the leading textbooks, describe the rule in Hancock v Watson as a rule or principle of construction: see for instance Fyfe v Irwin [1939] 2 All E R 271 at pp 273, 277 and 281. Its scope is not restricted to subsequent limitations which are struck down by the rule against perpetuities or are void for uncertainty; as the citation from Lord Davey makes clear, it also covers failure by lapse or any other reason. In all those situations it might have been argued that even though the subsequent limitation has failed, its presence in the will (and the need to read the will as a whole) shows that the initial absolute gift cannot be taken at face value, as it were (that line of thought is reflected in paragraph 32 of the Court of Appeal’s judgment). But the rule says otherwise, and the court will normally be guided by the rule, unless there are compelling reasons for reaching another conclusion. Extrinsic evidence of the will-maker’s intentions, where admissible by statute, might in some cases provide such a compelling reason, but the extrinsic evidence is not decisive in this case. 36. The application of the rule would not, with great respect to the Court of Appeal, involve a blinkered approach (the expression used in paragraph 32 of its judgment). On the contrary it would, properly understood, be an aspect of reading the will as a whole. Thus Lord Romer said in Fyfe v Irwin at p281, “My Lords, it sometimes happens that a will contains two dispositions of the same property which, if literally construed, are inconsistent with one another. In such cases, the court always endeavours to reconcile the dispositions, and will, if it be possible, so construe them that neither has to be rejected altogether.” Charitable corporations and charitable trusts 37. Reference has already been made (in paragraph 12 above) to the abstract and technical nature of the main issue between the contending parties – that is, whether the Foundation is entitled to Nina’s estate but must hold it subject to particular charitable trusts set out in clauses 2 and 3 of her will, or whether the Foundation is to hold the estate for the general objects set out in clause 3 (1) and (2) of its memorandum, and not on any more particular or limited trusts. It is clear that a charitable corporation may hold some of its funds as a trustee. For instance incorporated educational charities (such as, in England, Oxford and Cambridge colleges) hold their main endowments for their general charitable purposes, but do over the years receive numerous gifts and bequests which are held in trust for particular charitable purposes such as the funding of a scholarship or prize, or the endowment of a studentship or professorial chair. The Foundation has an express power, under clause 3 (3.2) of its memorandum, to undertake the trusteeship of any particular charitable trust consistent with its main objects. 38. It may be said that an incorporated charity holds its general funds “beneficially”, but that expression may be a little misleading. Human beneficial owners of property, if of full age and sound mind, can dispose of their property as they like, but a charitable corporation may use its resources only for its charitable objects. So in practical terms it is not very different from a charitable trust, especially as the law imposes on the individuals who manage the corporation’s affairs the same fiduciary obligations (in relation to deriving personal profit from their office, and similar matters) as are imposed on trustees: see for instance Re French Protestant Hospital [1951] Ch 567. The officers of a charitable corporation are not trustees in the strict sense, but they are in “at least an analogous position”, as Buckley LJ ( a judge with great expertise in charity matters) said in Von Ernst & Cie v Inland Revenue Commissioners [1980] 1 WLR 468, 479. Slade J expressed the same view in Liverpool and District Hospital for Diseases of the Heart v Attorney General [1981] Ch 193, 209. 39. Where there is a gift to an incorporated body with wholly charitable objects, the court does not readily construe words expressing a testator’s motives or wishes as imposing a charitable trust on a gift which is already (because of the recipient’s objects) destined to be applied for charitable purposes. In Re Vernon [1972] Ch 300, 303, Buckley J, having dealt with a gift to an unincorporated body, contrasted the position where the donee is an incorporated charity: “A bequest to a corporate body, on the other hand, takes effect simply as a gift to that body beneficially, unless there are circumstances which show that the recipient is to take the gift as a trustee. There is no need in such a case to infer a trust for any particular purpose. The objects to which the corporate body can properly apply its funds may be restricted by its constitution, but this does not necessitate inferring as a matter of construction of the testator’s will a direction that the bequest is to be held in trust to be applied for those purposes: the natural construction is that the bequest is made to the corporate body as part of its general funds, that is to say, beneficially and without the imposition of any trust. That the testator’s motive in making the bequest may have undoubtedly been to assist the work of the incorporated body would be insufficient to create a trust.” Poon J quoted the whole of this passage (in paragraph 40 of his judgment) whereas the Court of Appeal (in paragraph 35 of its judgment) quoted only the second part of the first sentence. 40. Mr Green QC (appearing as leading counsel for the Foundation in this court) has referred on this point to the Canadian case of Roman Catholic Archiepiscopal Corp of Winnipeg v Ryan (1957) 12 DLR (2d) 23. In that case there was a testamentary gift to the incorporated charity, which was identified by its name, and was not referred to as a trustee. But the will stated the testator’s purpose as being to benefit a particular church in the archdiocese, or, in the corporation’s discretion, the work of the archdiocese, and thereby to promote the spiritual welfare of Roman Catholic communicants within the archdiocese. Davey JA asked himself (at page 28) whether this express purpose was “so inconsistent with an intention that the Corporation should take absolutely that a trust for such purpose must be implied” and concluded that there was no need to imply a trust. The courts’ jurisdiction to make a scheme 41. The court has an inherent jurisdiction over the administration of charitable trusts, originally based on the notion of the sovereign as parens patriae(parent of the nation). This jurisdiction could be invoked by the Attorney General (now, in Hong Kong, the Secretary for Justice) as the sovereign’s forensic representative. Where the court exercises its jurisdiction to clarify or modify the purposes of a charitable trust, or to improve the administration of a charitable trust, it does so by making a scheme – that is, a written instrument approved by the court to regulate, in whole or in part, the future management and administration of the trust. 42. The approval of a scheme may sometimes, but does not always, require the court to have recourse to the cy-près doctrine. This was explained by Maugham J in Re Robinson [1931] 2 Ch 122, 128-129: “I should point out that a scheme directed by this Court in relation to gifts for charitable purposes is not necessarily or, I think, generally a scheme for the application of the fund cy près. It is well known that a charitable gift in this country does not fail merely because there is an uncertainty as to the mode of carrying out the gift. In numerous cases of gifts for charitable purposes it is necessary to fill up a number of details in regard to which the testator or the donor has not described his wishes in clear terms. In such cases the gift does not fail, but the Court fills up the details of the donor's charitable intention by means of a scheme: there is no question of selecting objects cy près, because the objects to be benefited are the very objects pointed out by the testator, but the Court is doing no more than completing the trusts to carry out objects which, on the assumption with which I am dealing, have been indicated in sufficiently clear terms by the testator.” 43. The cy-près doctrine consists of a group of rules which enable a charitable trust to be remoulded by the court so as to preserve the trust, rather than that it should fail because the charitable purposes of the charity’s founder cannot (or can no longer) be carried out precisely in accordance with the founder’s expressed wishes. The doctrine is non-statutory in origin, although in England and some other parts of the common law world (but not in Hong Kong) its scope has been extended by statute. The origin of the Norman French phrase “cy-près” is the subject of debate, but one plausible view is that it is an abbreviation of “aussi pres que possible” – as near as possible. The doctrine was stated as follows by Lord Eldon LC in the leading case of Moggridge v Thackwell (1802) 7 Ves 31, 69 (affirmed by the House of Lords (1807) 13 Ves 416): “…if the testator has manifested a general intention to give to charity, the failure of a particular mode, in which the charity is to be effectuated, shall not destroy the charity but, if the substantial intention is charity, the law will substitute another mode of devoting the property to charitable purposes though the formal intention as to the mode cannot be accomplished.” Neither side contended that it would be necessary or appropriate to have recourse to the cy- près doctrine if the second part of the clause 2, referred to for convenience as clause 2(2) of Nina’s will declares charitable trusts which fully dispose of her estate. It is not therefore necessary, for present purposes, to explore the doctrine in more detail. 44. The court’s inherent jurisdiction over charitable trusts is also exercisable, subject to some constraints and qualifications, in relation to charitable corporations, even though their property is not held in trust in the strict sense. The constraints and qualifications arise from the variety of constitutional sources from which a charitable corporation may derive its corporate status. These including incorporation by royal charter and incorporation by a special Act of Parliament, which may go so far as to confer public law functions on the body so incorporated: see the judgment of Lightman J in Royal Society for the Prevention of Cruelty to Animals v Attorney General [2002] 1 WLR 448, especially at pp 458-459, comparing the position of the RSPCA to that of the National Trust, whose status had been considered in Scott v National Trust for Places of Historic Interest or Natural Beauty [1998] 2 All ER 705. (Both cases were concerned with these charities’ attitudes to fox-hunting and deer-hunting with hounds.) 45. In such cases it may be inappropriate for the court to interfere with administrative machinery approved by the legislature. Moreover some charitable corporations (especially universities and colleges) have a “visitor” (often the holder of some high office such as a bishopric) who has a function similar to that of a domestic arbitrator. The court will not normally intervene so as to oust the powers of a visitor. The best summaries of the position are in the judgment of Buckley LJ in Construction Industry Training Board v Attorney-General [1973] Ch 173, 185-188, and in the judgment of Slade J in Liverpool and District Hospital for Diseases of the Heart v Attorney-General [1981] Ch 193, 213-215. But again it is unnecessary, for present purposes, to explore this technical area in any more detail. Nina’s will :preliminary 46. Nina’s will must be construed as a whole in accordance with the principles already considered. In order to do that the court must undertake the “iterative” process (the expression used by Lord Neuberger in Re Sigma Finance Corporation, paragraph 31 above) of fitting together all the (sometimes conflicting) indications of the will-maker’s intentions, like pieces of a jigsaw puzzle, so as to end up with a coherent whole. In this case this is not an easy task. The indications of Nina’s intentions are a disparate mixture, some purely verbal, some contextual, some derived from established principles of construction, and some based on admissible extrinsic evidence. 47. The most important matters to be considered are as follows. (1) Matters of language and context (a) The imperative nature of the language of the will, especially in its choice of auxiliary verbs (in the agreed translation, “must” used twice in clause 2 and once in clause 4, and “shall” in clause 3, together with the imperative sense of “obligation” twice in clause 4); all these in comparison to “I wish” at the beginning of clause 2. (b) The meaning of “all the projects which [the Foundation] has undertaken since its establishment” in clause 2. (c) The meaning of “the purpose of setting up a fund and a Chinese prize of worldwide significance similar to that of the Nobel prize” in clause 2. (d) Whether the purpose identified at (c) above is within the foundation’s existing objects. (e) The significance of the references (clause 2(1), 2(2) and 3, in slightly different terms) to “the supervision of a managing organisation”. (f) The meaning and contextual significance of clause 3, especially as an indicator as to the meaning of the will as a whole. (g) The meaning and validity of clause 4. (h) Finally there is the general question of how Nina’s non-technical, layman’s language, to the extent that it is found to be imperative rather than precatory, is to be given effect by resort to the technical legal concepts of trusts, powers, charges and conditions. (2) Principles of construction The two principles of construction most in point are the general rule that a gift to a charitable company is a gift to the company beneficially unless it is sufficiently clear that the imposition of a trust was intended (Re Vernon [1972] Ch 300, 303), and the rule in Hancock v Watson [1902] AC 14, 22, as to the effect of an absolute gift followed by engrafted trusts which do not exhaust the beneficial interest. (3) Extrinsic evidence The most important extrinsic evidence, admissible under section 23B of the Wills Ordinance, Cap 30, is in the affirmation of Nina’s sister, Mrs Tong. 48. These matters are considered below. They must perforce be taken one at a time, but they must not be considered in isolation. Each is likely to have an impact on the others. But it is useful to look first at how these matters were dealt with in the courts below. The numbering below corresponds to that in the last paragraph. (1) (a) Imperative language This was the heading to section G2.1 of the judgement of Poon J. That section covers paragraphs 54 to 83. He came to the firm conclusion that by the clear and imperative language of clauses 2, 3 and 4 Nina did intend to impose a trust. The Court of Appeal came to a similar conclusion (paragraphs 37 to 42) but made an exception for the first sentence of clause 2 (referred to for convenience as clause 2(1)) as being no more than an expression of hope, which did not however detract from the binding force of clauses 2(2) and 3, despite its importation into those provisions. These views are to be found in paragraphs 45 to 52 of the Court of Appeal’s judgment, the last paragraph ending, “Reading the Will as a whole, the Judge was clearly correct in finding that Nina intended and expected her directions to be firmly binding, with or without the managing organisation”. (b) “Projects” The Judge (paragraph 70) and the Court of Appeal (paragraph 64) agreed that this expression could only sensibly apply to any activities within the Foundation’s charitable objects. They reached this conclusion despite the context of the words “all the projects which [the Foundation] has undertaken since its establishment,” apparently on the supposition that there were no such projects. (c) “The purpose of setting up a fund and a Chinese prize” The Judge considered (paragraph 75) that there was no “uncertainty or subjectivity” about this part of the will, and concluded that “Nina must have intended that the Prize should resemble the Nobel Prize in all material respects”. The Court of Appeal did not agree that the Chinese prize must closely resemble the Nobel Prize (paragraphs 121 to 126) but refrained from expressing any definite view about its precise scope. (d) Compatability of the prize with the Foundation’s objects The Judge concluded (paragraphs 45,73 and 74) that the Foundation’s objects as set out in clause 3(1)(a),(c),(d) and (g) of its memorandum, coupled with clause 3(2), were not wide enough to enable it to establish the Chinese prize. The Court of Appeal’s view on this point is not entirely clear. It concluded (paragraph 129) that there was no ultra vires problem in relation to the Chinese prize or clause 4 of the will. As regards the prize its view seems to have been that any ultra vires problem could be avoided by adjusting the precise scope of the prize. (e) The supervising organisation The Judge dealt with this at paragraphs 62 to 67. He observed (by reference to the evidence of Mrs Tong, among other matters) that Nina must have realised that the Secretary-General of the United Nations might not be able or willing to join the proposed supervisory body. He concluded (paragraph 67) that Nina’s directions were not conditional on the supervisory body being established, and that Nina “must have intended and expected the Foundation and its Directors to faithfully follow and abide by those directions, which are firmly binding, with or without the Managing Organisation.” (f) Clause 3 The Judge considered clause 3 primarily at paragraphs 78 to 81. Counsel for the Foundation had argued that its language is too uncertain to create enforceable obligations. The Judge took a different view, and held that there is a binding direction for the Foundation to retain all its shares in Chinachem Group companies indefinitely, since (paragraph 81) “The Foundation can meaningfully perform [the functions specified in clause 3] only if it remains the sole and ultimate beneficial shareholder of the Chinachem Group.” The Court of Appeal differed from the Judge as to this point (paragraphs 73 and 74), but agreed with him as to the generally imperative language of the clause. (g) Clause 4 The Judge accepted (paragraph 103) the submission of Counsel for the Secretary for Justice that clause 4(2) and (3) should be given effect as conferring on the Foundation discretionary powers exercisable in favour of the non-charitable objects of those provisions, subject to the rule against perpetuities as modified by section 8 of the Perpetuities and Accumulations Ordinance, Cap 257. The Court of Appeal must be taken to have agreed with the Judge, since his order on this point was not varied, but the Court of Appeal seems to have considered clause 4 only so far as it might be relevant to the main issue of whether a charitable trust was imposed by clauses 2 and 3 (paragraphs 78 to 91). (h) Trust, power, charge, condition? The Judge, having decided that the language of clauses 2 and 3 is imperative, concluded that they operate to impose a trust. It was not argued before him that the non-technical language, drafted by Mrs Tong, which Nina used in those clauses could or should be interpreted as imposing a personal obligation (arising not from a trust but from the implied acceptance of a condition), with or without the support of an equitable charge. Arguments on those lines were addressed to the Judge on clause 4(2), but were rejected by him, primarily on the ground of the uncertainty of the supposed obligation (paragraphs 98 to 101). In the Court of Appeal the submissions as to personal obligations were again addressed only to clause 4, and were considered only in the context of the main (trust or no trust?) issue (paragraph 83). The Court of Appeal considered the topic briefly (paragraphs 84 to 86) before concluding (paragraph 87) that Nina must have intended potential beneficiaries under clause 4 to have some recourse to her estate. In short, the argument that clause 2(2), if binding, takes effect as a personal obligation (arising from acceptance of an implied condition) has been raised for the first time in this Court. (2) Principles of construction It is possible to cover the remaining matters much more briefly. Both the Judge and the Court of Appeal accepted the principle stated in Re Vernon [1972] Ch 300, 303, but held that the language of the will was sufficiently clear and imperative to impose a charitable trust: Poon J at paragraph 44, developed at length in the following paragraphs; the Court of Appeal at paragraph 35, developed in the following paragraphs. Poon J did not refer to Hancock v Watson [1902] AC 14, and it may not have been cited to him. The Court of Appeal did refer to it at some length (paragraphs 26 and 32) but (as already noted in paragraph 35 above) did not consider it to be a principle of construction. (3) Extrinsic evidence Both the Judge (paragraph 63) and the Court of Appeal (paragraph 45) placed some reliance on the evidence of Mrs Tong, while indicating that they would in any case have been disposed to accept that Nina cannot have been confident of securing the participation of the Secretary-General of the Untied Nations in the governance of the Foundation. Both courts rejected the view that any events which occurred after Nina made her will on 18 July 2002 could be relied on as “arm-chair evidence”. Imperative words, and their effect in a home-made will 49. The Judge and the Court of Appeal reached a clear conclusion as to the imperative nature of the language used in clauses 2, 3 and 4 of Nina’s will. There is no reason to differ from this conclusion as to the binding force of the language rendered in the official translation as “must” and “shall”. The only area of real doubt (considered separately below) is whether the provisions as to a supervisory body (originally introduced in clause 2(1), but repeated in clause 2(2) and clause 3) should be understood as an essential part of Nina’s intentions. 50. The issue which has been debated at length is how the imperative nature of Nina’s intention is to be given effect in legal terms. Counsel for the Foundation has pointed out, correctly, that Nina was not a lawyer and cannot be supposed to have been familiar with the technicalities of trust law. But by the same token, she cannot be supposed to have been familiar with other legal technicalities such as personal obligations enforceable in equity under the “benefit and burden” principle, whether or not backed by an equitable charge. 51. In these circumstances a court of construction must give effect to Nina’s intentions, expressed as they were in non-technical terms, by interpreting them into the most appropriate legal terms of art. It is a commonplace that the nuances of meaning can get lost in translation. Nina’s will has already been translated into English from the original Chinese text, and the English text must now undergo a further process akin to translation, that is the process of being expressed in legal terms. The most appropriate legal terms are those that most naturally and simply give effect to Nina’s intentions. 52. The correct interpretation of clause 2(2) is that it imposes a trust for charitable purposes. The argument that clause 2(2) imposes some sort of obligation or condition which is specifically enforceable in equity, but does not amount to a trust, is complicated and unnatural. It is also unworkable, since the obligation to establish the Chinese prize is expressed in very vague and uncertain terms. That uncertainty can be cured and clarified by the law’s benevolent treatment of charitable trusts. It cannot be cured by an analysis which relies on a personal obligation that does not amount to a charitable trust. 53. Mr Green submitted, in arguments directed to clause 4(3) of the will as well as to clause 2(2), that a personal obligation arising from the “benefit and burden” principle need not be precisely quantifiable. He relied mainly on some Australian authorities including Gill v Gill (1921) 21 SRNSW 400 (approved by the High Court of Australia in Muschinski v Dodds (1985) 160 CLR 583), Bective v Federal Commissioner of Taxation (1932) 47 CLR 417 and Hourigan v The Trustees Executors and Agency Company Limited and Others (1934) 51 CLR 619. None of those cases was concerned with a gift for charitable purposes. Bective and Hourigan were concerned with provisions for the maintenance of minor children. Such provisions involve an element of discretion but they have always been regarded as being in a special category, as appears from the judgments of Dixon J in Bective at pp 420-421 and in Hourigan at pp 645-646. As Dixon J said in the latter case at p 646, “However difficult it may be to decide what is the amount to be applied, it is treated as an obligation capable of measurement, or ascertainment, and enforcement.” The Court of Chancery regarded itself as having particular expertise as to the maintenance and education of minors, because of its ancient wardship jurisdiction, and was sometimes ready to substitute its own view as to the exercise by guardians or trustees of discretionary powers of maintenance: Pitt v Holt [2013] 2 AC 108, para 64. 54. The obligations imposed by clause 2(2) of Nina’s will (and especially that relating to the Chinese prize) are of a character wholly different from the familiar obligation to maintain a minor child (especially where that obligation is imposed on a parent who is in any case subject to a legal or moral duty to look after the child). They are obligations to spend unquantifiable (but by any normal standards, very large) sums of money for purposes described (as regards the Chinese prize) in the sketchiest terms. Mr Green cited from Harvey J in Gill (at pp 406-407), “In [a case of personal obligation under the ‘benefit and burden’ principle] no higher degree of certainty is required by the Court than is required for the creation of a trust.” Mr Green deftly sought to turn this proposition round: no greater certainty is required, he said, for a personal obligation that is not a trust, than for one that arises under a trust. That may be correct if the trust in question is a private, family trust. But clause 2(2) of the will demonstrates that it is not correct if the trust in question is a charitable trust. 55. In short, the performance of the personal obligation imposed by clause 2(2) calls for mature fiduciary judgment exercised single-mindedly for the common good. It is the sort of personal obligation which the law classifies as a trust for charitable purposes, and that is how Nina’s will must be interpreted. “Projects” 56. Mr Green submitted that the Courts below erred in construing the first part of clause 2(2) as referring to all the Foundation’s charitable activities (or objects) as set out in clause 3(1) of its memorandum of association. That construction was wrong, he said, because it ignored the words “which it [the Foundation] has undertaken since its establishment.” He submitted that those words limit the obligation to the completion of projects which had been undertaken by the Foundation, but not brought to completion, at the time of Nina’s death. 57. That is a possible construction as a matter of language, if the words in the first part of clause 2(2) are taken in isolation, but it is wholly inconsistent with the general spirit and intention of the will as expressed in clauses 2 and 3. As Fok PJ pointed out in the course of the argument, clause 2(2) is expressed in the familiar double form of “not only…but also…” In that form of expression the first part of the couplet typically relates to what is already known and uncontroversial; it serves as an introduction to the second part of the couplet, on which the main emphasis falls. It seems highly improbable that Nina intended, in the first part of clause 2(2), to refer to no more than a limited number of projects that might happen to be incomplete at her death, whenever that might occur. The context is much more consonant with a reference to the charitable activities of the Foundation rather than to on-going and uncompleted projects. The Court of Appeal was correct in its view that that construction was “narrow and unreal”. In coming to that conclusion the Court of Appeal rightly attached a good deal of weight to the final words of clause 3: “… and with part of its profits, to continuously develop the charitable business till eternity.” 58. Nina was a very experienced businesswoman and she must have realized that the Foundation would, after her death, have resources far greater than those available to it during her lifetime. Her intention was that in the future the Foundation’s charitable activities should continuously increase. Moreover, as Chan NPJ has pointed out, the Chinese characters 項目, rendered in the agreed translation as “projects”, do not have a limited or technical meaning. They might have been translated by the use for a more general word such as “items” or “things”. Had Nina really intended to limit the scope of the “not only…” part of clause 2(2) in the way for which the Foundation contends, she (and Mrs Tong as her amanuensis) would surely have used more precise language to make that intention clear. 59. When clause 2(2) is understood in this way, there is no possibility that the charitable trusts which it declares will not absorb the full resources of the trust property. There is no possibility of there being a surplus of funds not disposed of by the trusts of clause 2(2). The rule in Hancock v Watson [1902] AC 14 loses any relevance. The purpose of setting up a fund and a Chinese prize of worldwide significance similar to that of the Nobel Prize 60. Before coming to the main issues it is convenient to deal first with two uncontroversial verbal points mentioned by Chan NPJ in the course of the hearing. First, the part of the will which includes the Chinese characters translated as “a fund and a Chinese prize” [設立中國的類似諾貝爾獎的具有世界性意義的獎金和基金] actually refers to “prize” before “fund”. The two words are part of a single composite expression. It is not an essential part of Nina’s intention that the Chinese prize should have its own endowment fund administered separately from the rest of the charitable trust funds, although that is something that might be decided on as a matter of administration. Second, the Chinese characters [設立中國的類似諾貝爾獎的具有世界性意義的獎金和基金] have a flexible meaning and allow for the Chinese prize to be similar to the Nobel Prize either in terms of “worldwide significance” or more generally. 61. The Court of Appeal was correct in its view that the Chinese prize need not resemble the Nobel Prize in all respects. But that leaves the intended content of the expression in a good deal of uncertainty. The Chinese prize is no doubt intended to bring further renown to the People’s Republic of China, in the same way as the Nobel Prize focuses international attention on Sweden at the time of the Nobel awards. But Nina cannot have intended that the prize should be limited to excellence in Chinese literature, or Chinese music, or any specifically Chinese cultural activity, since that would be inconsistent with her clear intention that the prize should be international in character – “of worldwide significance”. Probably Nina had in mind that the prize would be administered, and its award ceremonies held, in Beijing (or possibly Hong Kong). 62. These are all matters which the board of governors of the Foundation will need to consider very carefully, with appropriate guidance (considered in the final section of this judgment). What is essential is that the Chinese prize should be a proper charitable purpose. A prize of this sort is charitable only if it is for the public good, in encouraging the general public to strive for excellence in scientific, social and cultural activities which are beneficial to mankind. The public good which such a prize achieves is reflected, not so much in the conferment of a large financial benefit on a celebrity who may already be well provided for, as in the incentive that it gives to hundreds of thousands of people who do not win the prize, but are encouraged to strive for excellence. So the subject matter for which the prize is awarded is highly material. This is illustrated (to descend from the sublime to a very humble level) by the judgment of Vaisey J in Re Dupree [1945] Ch 16, which concerned the validity of the £5,000 endowment of an annual prize for a chess competition open to schoolboys in the English city of Portsmouth. Vaisey J said (at p 20) “One feels, perhaps, that one is on rather a slippery slope. If chess, why not draughts? If draughts, why not bezique, or so on, through to bridge and whist…” Vaisey J may have been wrong to feel any doubt about the educational value of chess, but he was right to make clear that not every prize is automatically charitable in nature. 63. Clause 3 is administrative in nature in that it is concerned (at least for the most part) with the administration and management of the trust property rather than with the application of its income or capital for charitable purposes. Its language is imperative but some of its apparent requirements must, as a matter of common sense, be taken as aspirational. No controlling shareholder, however experienced and far-sighted, can ensure the continuing and indefinite prosperity and growth, year on year, of a group of trading companies. Mr Green described the clause as “declaratory” in character. That seems to be correct in that Nina was (whether or not she was aware of it) laying down what would in any case be the most basic obligation of her trustees, to safeguard the trust property and to exercise skill and prudence in its management. 64. The Court of Appeal was correct in its view that clause 3 does not require the Foundation to remain as 100% shareholder of each of the many companies (this Court was told that there are about 200) in the Chinachem Group. Experience shows that effective control of a corporate business can often be maintained with far less than 100% voting control. But is goes further than that. Experience shows that it is prudent for trustees to achieve some degree of diversification of their investments: see the judgment of Hoffmann J (as he then was) in Steel v Wellcome Custodian Trustees Ltd [1988] 1 WLR 167. 65. That case was concerned with the Wellcome Trust, which is the second largest medical charity in the world. By his will Sir Henry Wellcome (who died in 1936) directed his trustees to retain indefinitely all his shares in Wellcome Foundation Ltd (which was, contrary to what its name might suggest, a trading company). In 1985 the trustees, with the authority of the Charity Commission, sold 25% of the shares, but the remaining 75% had by 1988 come to represent 90% of the fund. This particular investment was therefore spectacularly successful, but a degree of diversification was nevertheless appropriate. The experience of the Wellcome Trust may be contrasted with that of the Nuffield Trust, a medical charity founded by William Morris (later Viscount Nuffield), one of the pioneers of the British motor industry. His trustees were directed to retain indefinitely their shares in Morris Motors Ltd, but fortunately (and again, with the authority of the Charity Commission) the trustees diversified their investments before the British motor industry went into the steepest phase of its long-term decline. 66. For present purposes – that is, for the purposes of construing Nina’s will – the main issue on clause 3 is whether it provides guidance as to the meaning of clause 2(2). Mr Taube QC (for the Secretary for Justice) submitted that it does provide guidance, with its references to the continuous growth of the Foundation’s “business empire” and the continuous development of its “charitable business”, these being regarded as a single unit. Mr Green submitted that there is no real emphasis on the unity of the charity, and that the reference to the Foundation’s charitable business is to its objects under its memorandum on the footing that it is beneficial owner. Mr Taube’s submission is to be preferred. It provides further support, as already noted, for the wider interpretation of the first part of clause 2(2). Clause 4 67. Clause 4(1) is no longer capable of taking effect, as Teddy’s parents both died in 2010. The objects of the discretionary provisions in clause 4(2), however those provisions are characterised in legal terms, are Teddy’s two sisters and two brothers, whose present ages range from about 68 to 74 years. Between them they have five children, all adults in their thirties or forties. The discretions exercisable in favour of these beneficiaries are not identical. The younger sister Wang Teh Hwa, who has long-standing health problems, is singled out for support for her living and medical expenses, and her only child (a 38-year-old financial analyst) may come within the words “to look after her [Wang Teh Hwa’s] children as well as their needs of advanced studies”. Teddy’s other siblings are to be looked after “if necessary”. Their children qualify only for the costs of “university or advanced studies.” 68. All these discretions, however limited, are powers whose exercise the Foundation is under an obligation to consider from time to time. Each is most naturally characterised as a “power in the nature of a trust” or a “fiduciary power in the full sense” (the expression used by Chitty J in Re Somes [1896] 1 Ch 250, 255, quoted and adopted by Warner J in Mettoy Pension Trustees Ltd v Evans [1990] 1 WLR 1587, 1614. It is not a power capable of being released (that is, extinguished) in the way that a power of appointment conferred on an individual can be released by the donee’s free choice. 69. The discretions conferred by clause 4(2) are rather closer to the sort of child-maintenance provisions considered in Bective and Hourigan, but they are by no means on all fours with them. The most importance difference, which is decisive, is that clause 4(2) cannot be explained as operating to impose a personal obligation on the Foundation on the “benefit and burden” principle, since the Foundation does not take any part of the residuary estate beneficially. The Foundation takes it all in a fiduciary capacity, as trustee. 70. The court may not distort or misconstrue the language of a will so as to avoid a perpetuity difficulty, or so as to achieve some tax advantage. These points are considered here in relation to clause 4(2), and then separately below in relation to clause 4(3). It seems most unlikely that any perpetuity problem will arise under clause 4(2), as all nine possible beneficiaries now in existence are lives in being for the purposes of the modified rule. That is the effect of section 8(3), (4)(a) and (5)(b)(iii) of the Perpetuities and Accumulations Ordinance, Cap 257, coupled with the wide definition of “power of appointment” in section 2(1). 71. The needs of the clause 4(2) beneficiaries, even if met on the most ample and generous scale, cannot possibly call for more than a tiny part of the resources of the residuary estate. If there is any danger of this tiny non-charitable disposition having adverse tax consequences for the estate as a whole, consideration should be given to the appropriation of a separate (and in relative terms very small) fund which will be amply sufficient to provide (by its income and capital) for the needs of the clause 4(2) beneficiaries. It should be added that in his reply Mr Green put forward adverse tax consequences as a possibility rather than as a clear and present danger. 72. Clause 4(3) must also be treated as a fiduciary power in the full sense, for similar reasons. But analysis of the perpetuity position may be different because it may be administratively impossible to identify with sufficient certainty the thousands of staff and children of staff of Chinachem Group companies who were lives in being on Nina’s death, now more than eight years ago. So here the permissible period may be 21 years, under section 8(4)(b) of the Ordinance. Even if the task was not impossible, it would undoubtedly be inconvenient and burdensome. A better practical solution would be, as briefly discussed in the course of Mr Green’s opening, for the Foundation to achieve the clause 4(3) objective indirectly by encouraging the establishment of appropriate schemes at the corporate level. Practical steps have already been taken in that direction (see paragraph 26 above). Express authority could be given for this (and for an appropriation under clause 4(2)) by a scheme approved by the court. The issue of supervision, and the need for a scheme 73. The Court of Appeal regarded clause 2(1) as precatory only, even though it is incorporated quite prominently into the mandatory provisions of clause 2(2) and clause 3 (but not into clause 4). The Court of Appeal was no doubt correct, both as a matter of the ordinary probabilities and on the strength of Mrs Tong’s evidence, to hold that the whole of Nina’s testamentary dispositions was not to be made conditional on the acceptance of office by the President of the People’s Republic of China, the Chief Executive of Hong Kong and the Secretary-General of the United Nations. But it does not follow from that that it was not an essential part of Nina’s intentions to have some supervisory board, composed of individuals of real distinction, to guide the Foundation in the prudent management of her business empire and in the wise application for charitable purposes of the much greater distributed corporate profits that would be available after her death. As Ribeiro PJ observed in the course of argument, it does seem very likely that Nina regarded a new managing organisation as an important part of her plans for her estate. 74. The language of clause 2(1) is unusual. Nina stated that she wished “to entrust” the Foundation to the supervision of a managing organisation. The language suggests that she saw the Foundation itself as being within her gift, and that she wished (as Chan NPJ observed) to commit it to the care of a new body which was outside the Foundation, and over the Foundation, rather than being simply an organ of the Foundation itself. That is a matter that comes well within the court’s scheme-making power, as is the detailed working-out of Nina’s intention to establish the Chinese prize. 75. Paragraphs 6 to 8 of the reamended originating summons are in the following terms: “6. further or alternatively that the Court may execute any trusts contained in the Will; 7. if and so far as may be necessary that the Court may approve and direct a cy pres scheme for the application of the residuary estate of the Deceased; 8. Such further necessary or consequential directions, accounts and inquiries.” These paragraphs are not formally before this Court on this appeal, but counsel’s submissions have largely anticipated the issues which they raise. It should also be noted that paragraph 7 does not in terms refer to a scheme less far-reaching than a cy-près scheme, but the greater may be understood as including the less. 76. In the circumstances it is right for this Court to express the clear view that there should be a scheme for the administration of the charitable trusts of Nina’s will. There is a strong public interest in this important benefaction having a clear and sounder legal basis than the language of Nina’s home-made will. A scheme should be prepared and submitted to the High Court for approval after consultation (which will, it is to be hoped, be full and cooperative) between the Foundation’s board of governors and the Secretary for Justice as the guardian of the public interest. The scheme should have two principal objectives: the establishment of a supervising body, including its terms of reference and its membership; and the detailed working-out of arrangements for the Chinese prize. 77. The members of the new body should be individuals of unquestionable integrity, experience and judgment. They will no doubt bringing a variety of skills to their task, but between them they should be skilled in corporate governance and investment, and have deep knowledge of the fields in which the charity is likely to be active (such as medical and scientific research, education at all levels, disaster relief, social progress, and music, literature and fine arts). The new supervising body will also have a part to play in the detailed provisions for the Chinese prize, although one or more specialised prize committees may be thought appropriate once the subject-matter of the prize (or prizes) has been settled. The statutes of the Nobel Foundation (exhibited to the affidavit of Siu Wing Sze) provide one possible template, but there is no requirement for them to be followed closely. 78. The preparation of a scheme will also enable some other matters to be provided for. These include a possible appropriation in respect of the clause 4(2) beneficiaries (paragraph 71 above) and the possibility of the court approving provision at the corporate level for the clause 4(3) beneficiaries (paragraph 72 above). It may also be appropriate to review the Foundation’s articles, in order to consider whether the Executive Council (paragraph 54) will still have a useful function, and whether the provisions as to remunerated trustees (paragraph 55) are still appropriate, especially if the quorum for the board (paragraph 45) is still one. Disposal 79. The appeal should be dismissed. Counsel should endeavour to agree minutes of order covering all matters except costs. Written submissions as to costs should be lodged by the Foundation within 21 days after the handing down of this judgment; submissions on costs by the Secretary for Justice and the administrators within 21 days thereafter; and any submissions in reply by the Foundation within 14 days after the lodging of the Secretary for Justice’s submissions. Mr Simon Taube, QC, Mr Ambrose Ho, SC, and Mr Michael Yin, instructed by Department of Justice, for the Plaintiff/1st Respondent Mr Brian Green, QC, Mr Patrick Fung, SC and Mr Jeremy Chan, instructed by Wilkinson & Grist, for the 2nd Defendant/Appellant Mr Ashley Burns, SC and Ms Bonnie Cheng instructed by Allen & Overy for the 1st Defendant/2nd Respondent 
Chief Justice Cheung: 1. This appeal concerns the proper interpretation of section 17 of the Summary Offences Ordinance[1], involving the application of the ejusdem generis rule and the resolution of apparent differences in meaning between the English and Chinese texts of the provision. Facts 2. The facts giving rise to this criminal appeal are straightforward. On 2 November 2019, police officers were deployed in Causeway Bay to disperse an unauthorised assembly. After the dispersal, a person was seen hurling insults at the police. He was subdued by some police officers at the scene while a group of onlookers stood around. The appellant, wearing a black outfit, a black face scarf and a pair of gloves and carrying a black backpack, was amongst the onlookers. He was intercepted by a police officer and upon search, a respirator, a helmet, a pair of knuckle gloves, and a bag containing 48 pieces of 6-inch plastic cable ties were found inside his backpack. Before the magistrate 3. The appellant was convicted after trial[2] before the magistrate[3] of one count of possession of an instrument fit for unlawful purposes, contrary to section 17 of the Ordinance, namely the bag containing 48 pieces of 6-inch plastic cable ties. 4. Section 17 of the Ordinance reads : “Any person who has in his possession any wrist restraint or other instrument or article manufactured for the purpose of physically restraining a person, any handcuffs or thumbcuffs, any offensive weapon, or any crowbar, picklock, skeleton-key or other instrument fit for unlawful purposes, with intent to use the same for any unlawful purpose, shall be liable to a fine at level 2 or to imprisonment for 2 years.” 5. Holding that the “unlawful purposes” for which an instrument is fit must be related to physically restraining or injuring a person[4], or housebreaking, the magistrate found that the plastic cable ties possessed by the appellant could be easily linked together and used to physically restrain a person, and were therefore within the meaning of “other instrument fit for unlawful purposes” in section 17.[5] The magistrate also took the view that the “unlawful purpose” for which a defendant intended to use the offending instrument in the section could be any unlawful purpose and need not be related to physically restraining or injuring a person or housebreaking. On the facts, he found that the appellant had intended to use the cable ties to bind detached metal railings or other objects on the road to form barricades or other large structures for use in armed confrontations, fights, inflicting injuries or obstructing or blocking the road or traffic, which was an “unlawful purpose” within the meaning of the section[6]. The magistrate convicted the appellant accordingly and sentenced him to five months and two weeks’ imprisonment. In the Court of Appeal 6. The Court of Appeal[7] dismissed the appellant’s appeal[8] against conviction and sentence[9]. In doing so, the court departed from some previous case law which, applying the ejusdem generis rule, had placed a restrictive interpretation on section 17, on the grounds that it had since undergone amendments. Instead, placing emphasis on the subsequent Chinese version of the section which came into existence in 1993, the Court of Appeal held that the “unlawful purposes” for which an instrument is fit and the “unlawful purpose” for which a defendant intends to use the instrument are not restricted to any specific purposes. On the magistrate’s findings, the court held, the appellant was rightly convicted. 7. The Chinese text of section 17 on which the Court of Appeal placed substantial reliance reads : “任何人管有任何腕銬或其他為束縛人身而製造的工具或物件，或管有任何手銬、指銬、攻擊性武器、撬棍、撬鎖工具、百合匙或其他適合作非法用途的工具，意圖將其作任何非法用途使用，可處第2級罰款或監禁2年。” Leave to appeal 8. Upon the appellant’s application, the Court of Appeal certified four points of law of great and general importance : “(1) On the true construction of the Chinese and English texts of section 17 of the Summary Offences Ordinance (Cap 228), whether the expression ‘other instrument fit for unlawful purposes’ in that section is subject to the ejusdem generis rule? (2) If the answer to Question 1 is in the affirmative, whether the expression ‘unlawful purposes’ is confined to purposes similar to those crowbars, picklocks and skeleton-keys are fit for? (3) On the true construction of the Chinese and English texts of section 17 of the Summary Offences Ordinance (Cap 228), whether the expression ‘with intent to use the same for any unlawful purpose’ in that section is subject to the ejusdem generis rule? (4) If the answer to Question 3 is in the affirmative, whether the expression ‘any unlawful purpose’ is confined to unlawful purposes similar to those the instruments or objects referred to in section 17 are fit for?”[10] 9. On 25 February 2022, the Appeal Committee[11] granted leave to appeal in respect of the four questions of law certified by the Court of Appeal. In addition, leave to appeal was also granted on the basis that it was reasonably arguable that there had been a material departure from accepted norms, that is, when departing from earlier decisions, the Court of Appeal failed to apply the principle that any ambiguity in penal statutes should be resolved in favour of the accused, and failed properly to give effect to section 10B(3) of the Interpretation and General Clauses Ordinance[12] where a difference may exist between two versions of the text in bilingual legislation[13]. Principles of statutory interpretation 10. The rules of statutory construction are well established. Words are construed in their context and purpose. They are given their natural and ordinary meaning with context and purpose to be considered alongside the expressed wording from the start, and not merely at some later stage when an ambiguity is thought to arise. A purposive and contextual interpretation does not mean that one can disregard the actual words used in a statute. Rather, the court is to ascertain the intention of the legislature as expressed in the language of the statute. As has been repeatedly pointed out, one cannot give a provision a meaning which the language of the statute, understood in the light of its context and purpose, cannot bear[14]. 11. Context here is to be taken in its widest sense and includes other statutory provisions and the general law. The purpose of a statutory provision may be evident from the provision itself, the recommendation of a report such as that published by the Law Reform Commission, the explanatory memorandum to the relevant bill or a statement by the responsible official of the government in relation to that bill in the Legislative Council[15]. It may also be relevant in any given case to look at the history of the provision concerned[16]. The ejusdem generis rule 12. The ejusdem generis rule has been described by Lord Diplock in Quazi v Quazi[17]as follows : “As the latin words of the label attached to it suggest, the rule applies to cut down the generality of the expression ‘other’ only where it is preceded by a list of two or more expressions having more specific meanings and sharing some common characteristics from which it is possible to recognise them as being species belonging to a single genus and to identify what the essential characteristics of that genus are. The presumption then is that the draftsman’s mind was directed only to that genus and that he did not, by his addition of the word ‘other’ to the list, intend to stray beyond its boundaries, but merely to bring within the ambit of the enacting words those species which complete the genus but have been omitted from the preceding list either inadvertently or in the interests of brevity.” 13. Bennion, Bailey and Norbury on Statutory Interpretation[18] explains the rule more fully this way : “23.2 Ejusdem generis principle: description (1) The ejusdem generis principle is a principle of construction whereby wide words associated in the text with more limited words are taken to be restricted by implication to matters of the same limited character. (2) The principle may apply whatever the form of the association, but the most usual form is where a list of specific terms is followed by general words that are construed as being confined to things of the same kind as those specified ... Comment The ejusdem generis principle is a particular application of the principle noscitur a sociis ... It arises from the linguistic implication by which words having literally a wide meaning (when taken in isolation) are treated as reduced in scope by the verbal context. It may be regarded as an instance of ellipsis, or reliance on implication. The principle is presumed to apply unless there is some contrary indication. … Principle not tied to any particular formula The ejusdem generis principle is not tied to any particular formula. As the above examples show, it does not, as has been suggested, apply only where there is a string of terms that form a class followed by wide residuary or sweeping-up words (though this is a common example of its application). Thus, for example, the wider words may merely follow on from the generic words.” 14. Several observations about the rule which are relevant to the present appeal can be made : (1) The ejusdem generis rule is a general rule of construction. When it is applicable, it gives rise to a presumption. It is an aid to construction, and can be dis-applied by a contrary intention. (2) The rule only operates when there is a common genus to which the items on the list belong. By definition, it has no application when there is only one item. When there is more than one item, it can only apply if a common element or genus can be identified. (3) The application of the rule does not depend on any particular formulation of the words. Whilst, usually, it applies when specific items are followed by a general description, that is not necessary or essential. The use of the word “other” in the residuary phrase is common, but not critical. What is essential is a list of items belonging to a common genus. (4) The wording and sentence structure of the provision are nonetheless important in that the rule does not apply if there is no such list, or, where there is more than one list, the items in question belong to different, separate lists. Section 17 15. The origin of section 17 can be traced to 1844. In its present form, it is rather poorly drafted. To a significant extent, this is the result of the several amendments that it has undergone over the years. The Chinese text of section 17 came into being in 1993. As a matter of language, it reads better than its English counterpart. Ironically, as will be seen, this is where the problem lies. In any event, it should be noted that prior to 1993, the only version of the section was in English, and the English version had been subjected to important judicial interpretation. 16. The court’s task in this appeal is of course to construe the section in its present form. However, the long and rather convoluted history of the section, the previous case law on the earlier English version, and the circumstances under which the Chinese text came into being in the early 1990s, all form an important part of the context in which the section in its modern form (in English and Chinese) should be construed. This is particularly significant given that a lot of emphasis has been placed by the Court of Appeal on the Chinese text and the amendments made to the section when departing from the earlier case law and arriving at a much wider construction. I would therefore approach the task of construction in the present appeal by tracing the origin and evolution of the section into its current form. In the course of doing so, I will, where appropriate, deal with the parties’ arguments and the Court of Appeal’s reasoning. Origin and early evolution 17. Section 17 had its origin in Ordinance No 5 of 1844[19], an ordinance “for the preservation of good order and cleanliness” in Hong Kong. The Ordinance provided, among many other things, that : “if any person… shall have in his possession any spear, bludgeon, or other offensive weapon, or any crowbar, picklock, skeleton key, or other instrument fit for unlawful purposes, with intent to use the same for such unlawful purpose, …” the person so offending shall be liable to a fine. 18. The 1844 Ordinance was repealed and replaced by Ordinance No 14 of 1845[20]. A new basis for founding liability, as an alternative to proving an intention to use the offending article or instrument for unlawful purposes, was added, namely the defendant’s inability to give a satisfactory account of his possession of the article or instrument. Section 12 of the latter Ordinance provided : “And be it further enacted and ordained, that every person who shall have in his possession any spear, bludgeon, or other offensive weapon, or any crowbar, picklock, skeleton-key, or other instrument fit for unlawful purposes, with intent to use the same for any such unlawful purpose, or who shall be unable to give a satisfactory account of his possession thereof, shall be liable to a penalty of not more than ten pounds, or, in the discretion of the presiding Magistrate, to be imprisoned for any term not exceeding fourteen days.” 19. It is quite plain from both the original version and the slightly modified version in the two Ordinances that as a matter of sentence structure and use of language, the predecessor of section 17 essentially covered the possession of two categories of articles or instruments : (1) “any spear, bludgeon, or other offensive weapon”; and (2) “any crowbar, picklock, skeleton-key, or other instrument fit for unlawful purposes”. Both categories started with the word “any”, and the two were separated by a comma and the word “or” after “offensive weapon” and before “any crowbar…”. Within each category, the word “any” only appeared before the first item and the word “or” only before the last item in the category. 20. Importantly, the phrase “other instrument fit for unlawful purposes” was there in the statutory provision from day one. It was the last item included in the second category. It was preceded by the word “or”. The phrase covered “other” instruments fit for unlawful purposes, that is, other instruments besides “crowbar, picklock, [and] skeleton-key”. This has an important bearing when I come to the Chinese text which seems to suggest otherwise. 21. Although there were subsequent amendments[21], the provision remained essentially unchanged in content until 1975. By then, the section had become section 17 of the Summary Offences Ordinance[22]. Early case law 22. Before 1975, the then section 17 had already been the subject matter of important judicial interpretation. 23. In The Attorney General v Li Chu[23], a magistracy appeal, the court had to consider whether a penknife fell within the meaning of “other instrument fit for unlawful purposes” in section 17. The magistrate below[24] had held that the section applied to two categories of articles – offensive weapons and housebreaking implements, and “or other instrument fit for unlawful purposes” covered only housebreaking implements as those words had to be read “sui generis (sic)”[25] with the preceding words “any crowbar, picklock, skeleton-key”[26]. On appeal, Mills-Owens J essentially agreed with this construction and rejected the argument that “other instrument fit for unlawful purposes” constituted a free-standing, third category, having an unrestricted meaning of its own : “As I see it the section envisages two categories of article, viz. (A) any spear, bludgeon or other offensive weapon; (B) any crowbar, picklock, skeleton-key or other instrument fit for unlawful purposes. I say this because each category ends with the similar phrase ‘or other etc’, thus following a pattern of expression as it were in which specific articles are first enumerated followed by the more general expression ‘or other (offensive weapon) (instrument fit etc)’; I say this also because of the presence of a comma after ‘weapon’ and the absence of a comma both after ‘bludgeon’ and after ‘skeleton key’, and because two broadly dissimilar classes of articles are in contemplation.”[27] 24. In the course of construing the section, the judge also observed the word “such” in “with intent to use the same for such unlawful purposes” was “otiose or meaningless” : “… ‘such unlawful purposes’ is an expression which has no meaning unless a particular kind of unlawful purpose has previously been specified (to which ‘such’ can be related), which the section does not do.”[28] 25. Mills-Owens J’s construction of section 17 was endorsed by the Full Court six months later in Tang Chi-ming v R[29]. At issue was whether the defendant could be charged with possession of an instrument fit for unlawful purposes contrary to section 17 when he was found to have had half a razor blade secured to his hand by adhesive tape and to have touched the rear pockets of persons boarding a public bus. Rigby SPJ[30] agreed with Mills-Owens J that the section envisaged two, and not three, separate and distinct categories of weapons or instruments. He also agreed with the view expressed by the magistrate in the earlier case that the ejusdem generis rule applied to the construction of the second category : “But I would agree with the learned magistrate, Mr Power, that the words ‘or other instrument fit for unlawful purposes’ ought to be read, sui generis (sic)[31], as bearing some connotation or relationship to the particular words ‘crowbar, picklock, skeleton-key’ immediately preceding them, and read in conjunction with the words ‘any such unlawful purpose’ almost immediately following, should be construed to mean and intend possessing an instrument fit for an unlawful purpose of a kind such as one might expect a ‘crowbar, picklock, skeleton-key’ to be so unlawfully used. The learned magistrate, Mr Power, restricted the unlawful possession of any such instrument to its use for housebreaking purposes; that may be placing too narrow a restriction upon the meaning and intention to be attributed to the words ‘or other instrument fit for unlawful purposes’, and I would be disposed to agree with the rather wider restriction placed upon those words by my Lord the President in this case, namely, possession of such an instrument for the purpose of securing unlawful access to a building or enclosed space or object as, for example, a stationary motor-car or the boot of such a vehicle.”[32] 26. Hogan CJ was of the same view : “If the expression ‘fit for unlawful purposes’ is confined, as I am inclined to think it should be, to unlawful purposes similar to those served by crowbars, picklocks and skeleton-keys, which might briefly be described as the purpose of securing unlawful access to a building, a receptacle or an enclosed space, the charge can still be expressed in this way. It will be for the prosecution to prove that the instrument which they have specified in their charge is one which can be properly described as fit for such an unlawful purpose when in the possession of the accused.”[33] 27. Likewise, the Chief Justice took the view that as far as the mens rea requirement on the intended use of the offending instrument was concerned, the expression “unlawful purpose” should also be restricted by the ejusdem generis rule to the purpose of “securing unlawful access to something in the nature of a house, a container or an enclosed space”.[34] 28. Pausing here, I respectfully agree with the learned judges in these two cases on the construction of the then section 17. As explained, on a plain reading of the section, it covered only two, not three categories of articles or instruments. Secondly, the ejusdem generis rule obviously applied to limit “other instrument fit for unlawful purposes” to instruments fit for gaining unlawful access. The words “any crowbar, picklock, skeleton-key” followed by “or other instrument fit for unlawful purposes” plainly signified a genus consisting of articles or instruments fit for offences which involved gaining unlawful access to premises, vehicles or other private objects or spaces. 29. As for the mens rea requirement, I share Hogan CJ’s doubt as to whether the word “such” really served no real purpose as suggested by Mills-Owens J[35]. More importantly, I also take the view that “with intent to use the same for such unlawful purpose” should likewise be restrictively read to tie in with the categories of articles or instruments covered by the section. This will be more fully explained in a later part of this judgment. The 1975 amendment 30. In 1975, by the Summary Offences (Amendment) Ordinance of that year[36], the words “spear, bludgeon, or other” were removed from section 17. The first category, after that amendment, covered “any offensive weapon” only. The amendment was wholly understandable, given that by the 1970s, spears and bludgeons had become rather obsolete weapons in a modern society, and in any event, they were covered by the remaining phrase “any offensive weapon”. Tsoi Shun-hing v R 31. The significance of this amendment, however, perhaps lay in the fact that it was the first time the legislature had the occasion to look at section 17 after the two court decisions discussed above. Notably, the legislature did not take the opportunity to make any substantive changes to the section apart from deleting the words “spear, bludgeon, or other” from it. There is no need in this appeal to go into the question of whether the doctrine of “tacit legislation” exists as to which there are different views[37]. Suffice it to say, the inaction of the legislature was pointed out by the Court of Appeal as the first reason for rejecting an attempt based on the (now repealed) definition of “or”[38] in section 3 of the Interpretation and General Clauses Ordinance to revisit the number of categories of articles or instruments covered by section 17 in its decision in Tsoi Shun-hing v R[39]in 1977. That case concerned the possession of a pair of handcuffs. Rejecting the renewed argument that the word “or” in section 17 should be construed disjunctively so that the section actually covered three categories of articles or instruments, including a third, open-ended category covering “other instrument fit for unlawful purposes”, the Court of Appeal said : “We are unable to accept this submission for the following reasons. First, and perhaps most important, the amending Ordinance of 1975 did not introduce any alteration which could affect the number of categories of articles covered and it must be assumed that the Legislature was aware of the interpretation which had been placed upon the section by the Full Court : if that interpretation did not accord with the actual legislative intention no doubt an attempt would have been made to put the matter right. Secondly, we think the wording of the section, obscure though it is, does favour a ‘two category interpretation’ and we bear in mind that s. 3 of the Interpretation and General Clauses Ordinance is always subject to s. 2 and applies only where the context of that Ordinance or of some other Ordinance or instrument does not otherwise require. In our view the context of s. 17 of the Summary Offences Ordinance does exclude a disjunctive interpretation and the ejusdem generis rule does apply.”[40] The Court of Appeal concluded that the pair of handcuffs was not an instrument fit for unlawful purposes within the meaning of section 17. The 1984 amendments 32. But the matter did not stop there. Not only did the legislature not introduce any changes to section 17 in 1975 in the light of the construction placed on it by the courts, it also made amendments to the section in 1984 which were obviously premised on the correctness of that construction. By the Summary Offences (Amendment) Ordinance 1984[41], the words “any wrist restraint or other instrument or article manufactured for the purpose of physically restraining a person, any handcuffs or thumbcuffs” were added to section 17. Moreover, the word “such” was removed from the phrase “with intent to use the same for any such unlawful purposes”. With these amendments, section 17 began to resemble its modern form : “Any person who has in his possession any wrist restraint or other instrument or article manufactured for the purpose of physically restraining a person, any handcuffs or thumbcuffs, any offensive weapon, or any crowbar, picklock, skeleton-key or other instrument fit for unlawful purposes, with intent to use the same for any unlawful purpose, or being unable to give satisfactory account of his possession thereof, shall be liable to a fine of $5,000 or to imprisonment for 2 years.” 33. There is no doubt that the drafting of the 1984 addition left much to be desired. Fortunately, the reasons for the 1984 changes were fully documented. In the Memorandum for the Executive Council’s discussion held on 6 December 1983, members of the Council were clearly told the reasons and purpose for the amendments : “… The purpose of this Bill is to facilitate prosecutions in cases where persons are found in possession of handcuffs or similar articles for an unlawful purpose, by including handcuffs and similar articles in the list of instruments contained in section 17 of the Summary Offences Ordinance. Background 2. Statistics indicate an increasing use of handcuffs in the commission of crime; 10 cases in the first six months of 1982, 30 in the second six months of 1982 and 17 cases in the first six months of 1983. If an actual crime is committed, a prosecution can follow, but difficulty arises when the police stop a person in suspicious circumstances, search him, and find him to be in possession of handcuffs for which he cannot give a satisfactory explanation. In the case of Reg. v. TSOI Shun-hing, (1979 HKLR 408) the Court of Appeal held that section 17 of the Summary Offences Ordinance only extends to offensive weapons and instruments fit for house-breaking and does not extend to handcuffs or other instruments which do not fall into either of those categories. This ruling means that no prosecution under section 17 is possible for the possession of handcuffs, even if they are intended to be used for an unlawful purpose, such as robbery or kidnapping. It is therefore proposed that handcuffs should be included in the list of instruments in section 17 of the Summary Offences Ordinance. 3. The police have pointed out that although there are no precise statistics, other instruments or articles of restraint are also used for similar criminal purposes. In particular they refer to thumbcuffs and wrist restraints. There do not appear to be any legitimate reasons for the possession of such articles. Handcuffs and thumbcuffs are clearly identifiable and equally clearly intended for the purposes of restraint. Wrist restraints and other articles of restraint may not be so clearly identifiable and could also possibly consist of items designed for other purposes. To exclude such other items it is considered that in this context only items manufactured for the purpose of physical restraint should be brought within the scope of section 17. Proof of the purpose of manufacture in practice could prove a major burden, but is considered a necessary safeguard. It is proposed, therefore, that handcuffs and thumbcuffs as such, together with wrist restraints and any other articles manufactured for the purpose of physical restraint, should be included in section 17. This should enable prosecution of those cases where such items, and indeed any other novel articles manufactured for the purpose of physical restraint, are found in unexplained possession. … 5. The effect of the inclusion of handcuffs et cetera in section 17 will be that a person found in possession of these can be required to give a satisfactory account of his possession of them failing which he will be liable to prosecution. In order to clarify the meaning of the subsection, the opportunity has also been taken to remove the word ‘such’ which has been judicially held to be redundant (Attorney General v. Li Chu (1968) HKLR 242).” 34. In the second reading of the Bill before the Legislative Council, the Attorney General again explained the rationale behind the amendments and the mischief targeted : “The purpose of the Bill is to make an amendment to the Summary Offences Ordinance which is intended to assist the police in their fight against crime. In recent years in Hong Kong there have been several instances of people being found by the police in suspicious circumstances, carrying articles which could be used to immobilize or restrain a victim while he is robbed or kidnapped or his home burgled. It is thought desirable that the police should have power to arrest and charge a person found in possession of such articles without lawful excuse. If section 17 of the Summary Offences Ordinance is amended as proposed, it will mean that any person who has in his possession any wrist restraint or other instrument or article manufactured for the purpose of physically restraining another person, or in possession of handcuffs or thumbcuffs, intending to use such article for an unlawful purpose or being unable to give satisfactory account of his possession of it, will be guilty of an offence and liable to a fine of $5,000 or to imprisonment for two years. Handcuffs have been used in the course of some 57 crimes committed in 1982 and the first half of 1983 so honourable Members may well be disposed to agree that there should be some prohibition on the possession of handcuffs without lawful excuse. Thumbcuffs are a modern variation of an ancient device which have no conceivable use except for restraint. It is difficult to envisage any lawful excuse for their possession. Possession of other forms of wrist restraint will be similarly unlawful where they have been specially manufactured for the purpose. Sir, care has been taken to ensure that the amendment is not too widely drawn. Possession of a handkerchief or a piece of string has been excluded (laughter), although of course they are capable of being used for restraint. The amendments are solely directed to articles manufactured for restraint. Nor are the police required to prove the purpose of possession, which is notoriously difficult.”[42] 35. Several observations can be immediately made. First, the government and the legislature were obviously aware of the case law concerning the scope of application of section 17. Indeed, the purpose of the 1984 amendments was to enlarge its scope to cover additional items. Like the pre-existing articles or instruments which fell within two separate categories respectively, the newly introduced items, although arranged in a confusing order, all fell within a category of their own, ie articles or instruments manufactured for the purpose of physically restraining a person. Indeed, both handcuffs and thumbcuffs fell within this description of “instrument or article manufactured for the purpose of physically restraining a person”. 36. Secondly, the addition was obviously premised on the construction of section 17 placed on it by the courts that the section did not contain an independent, third category covering “other instrument fit for unlawful purposes” as such. Had there been such a category, “wrist restraint or other instrument or article manufactured for the purpose of physically restraining a person, handcuffs or thumbcuffs”, being obviously articles or instruments which are fit for unlawful purposes, would have already been covered by this third category. And the reasons for introducing the amendment would have been to put right the incorrect construction placed on section 17 by the courts. However, that was not the explanation given to the Executive Council or the Legislative Council at all. Rather, the amendment, as explained in the Memorandum and by the Attorney General, was introduced to meet a newly encountered need of the police in the fight against crime which was not met by the pre-existing law. 37. The deletion of the word “such” from section 17 was, according to the Memorandum[43], done solely in response to the comment by Mills-Owens J in Li Chu that the word was “otiose or meaningless”[44]. This simply reinforced the point that the legislature was fully aware of the relevant case law starting with Li Chu. There was no suggestion whatsoever that the deletion of the word “such” was meant to widen the scope of the mens rea requirement to any unrestricted unlawful purpose, unrelated to the offending article or instrument in question covered by section 17. The contrary submission by the respondent is therefore rejected. 38. Summarising the position immediately after the amendments in 1984, section 17 now covered three categories of articles or instruments : (1) “any wrist restraint or other instrument or article manufactured for the purpose of physically restraining a person, any handcuffs or thumbcuffs”; (2) “any offensive weapon”; and (3) “any crowbar, picklock, skeleton-key or other instrument fit for unlawful purposes”, that is, an instrument fit for the purpose of gaining unlawful access. Possession of an article or instrument falling within any of these three categories would amount to an offence if it was coupled with an intent “to use the same for any unlawful purpose” of physically restraining a person in the case of a category (1) article or instrument, of physically injuring a person for a category (2) weapon, or of gaining unlawful access for a category (3) instrument, as the case might be. Alternatively, such possession would amount to an offence if the defendant was unable to give a satisfactory account of his or her possession of the offending article or instrument. 39. Up to that point in time, the construction of section 17, as summarised above, was directly against the contention that an unrestricted interpretation should be given to the words “unlawful purpose(s)” in the section. Indeed, on a fair reading of its judgment, the Court of Appeal did not really dispute that up to 1984, the above summary represented the correct interpretation of section 17 as it then was. I now move on to deal with what happened after the 1984 amendments which led the Court of Appeal and the respondent to say that the proper construction of section 17 now should be different. Bilingualism and the Chinese translation 40. In short, there were three developments. First, the development of bilingualism in the law. As early as 1974, the Official Languages Ordinance[45] was enacted to provide for the use of Chinese in court proceedings in the magistracies. In 1987, section 4B was added to the Official Languages Ordinance : “(1) Where an Ordinance has been enacted in one official language, the Governor in Council may, by order in the Gazette made after consultation with the Bilingual Laws Advisory Committee, declare that the authentic text of that Ordinance in the other official language shall be as specified in the order. (2) Where under subsection (1) the Governor in Council has declared a text to be an authentic text of an Ordinance and it appears to him that there is any manifest error, omission or inaccuracy in that text, he may, by order in the Gazette, correct that error, omission or inaccuracy; and any such correction shall be deemed to have been incorporated in the text at the time when it was declared to be the authentic text.[46] (3) The Attorney General may, by order in the Gazette, make formal alterations to the text of an Ordinance enacted in one official language, without affecting the meaning thereof, in order to achieve harmony of expression with a text declared under subsection (1) to be the authentic text of that Ordinance in the other official language.[47] (4) No order shall be made under this section unless a draft of it has been laid before and approved by resolution of the Legislative Council, and section 34 of the Interpretation and General Clauses Ordinance (Cap 1) shall not apply in relation to any such order[48].” 41. From 1989, new legislation was enacted in both English and Chinese, and at the same time, an enormous task was undertaken by the government, in consultation with the Bilingual Law Advisory Committee, to translate all existing legislation into Chinese. It was an immense exercise involving hundreds of pieces of legislation, and took several years to complete. All this was done as Hong Kong moved towards 1997. 42. The translation of the Summary Offences Ordinance was completed in 1992. By the Official Languages (Authentic Chinese Text) (Summary Offences Ordinance) Order dated 26 October 1993, the Chinese text set out in the Annex to the Order was declared to be the authentic text in the Chinese language of the Ordinance as at 13 November 1992, pursuant to section 4B(1) of the Official Languages Ordinance. Section 17 was translated in the Chinese text as follows : “任何人管有任何腕銬或其他為束縛人身而製造的工具或物件，或管有任何手銬、指銬、攻擊性武器、撬棍、撬鎖工具、百合匙或其他適合作非法用途的工具，意圖將其作任何非法用途使用，或管有上述工具或物件而無法作出滿意的解釋，可處罰款$5,000或監禁2年。” 43. A close comparison between the English version and the Chinese one of the section would disclose differences material to the issues in the present appeal. The Chinese version, when literally translated back to English, is as follows : “Any person who has in his possession any wrist restraint or other instrument or article manufactured for the purpose of physically restraining a person, or has in his possession any handcuffs, thumbcuffs, offensive weapon, crowbar, picklock, skeleton-key or other instrument fit for unlawful purposes, with intent to use the same for any unlawful purpose, or is unable to give a satisfactory account of his possession thereof, shall be liable to a fine of $5000 or to imprisonment for 2 years.” 44. The first thing to note about the Chinese text is that a new phrase “or has in his possession”[49] is added to the section after the original phrase “in his possession”[50]. As a matter of sentence structure, the two phrases in the Chinese text separate the articles or instruments covered by the section into two categories : (1) any wrist restraint or other instrument or article manufactured for the purpose of physically restraining a person; and (2) any handcuffs, thumbcuffs, offensive weapon, crowbar, picklock, skeleton-key or other instrument fit for unlawful purposes. Whilst the first category corresponds to the first half of the items in the first category discussed above on the basis of the English text, the second category under the Chinese text is essentially a mixed bag of the remaining half of the items in the first category, together with the items in the second and third categories under the English text. 45. Moreover, the word “or” between “handcuffs” and “thumbcuffs” (which now come under the second category), and, more importantly, the one between “any offensive weapon” and “any crowbar [etc]” in the English text are both missing. Furthermore, the word “any” before “offensive weapon” and before “crowbar” is also missing from the Chinese text. The combined effect of these differences is that in the Chinese text, handcuffs, thumbcuffs, offensive weapon, crowbar, picklock, skeleton-key and “other instrument fit for unlawful purposes” are all placed on the same footing without any distinction, and grouped under category (2). 46. No doubt, these changes render the Chinese text of section 17 more readable. Much more relevant for our purpose, however, is that based on the Chinese text, there is no longer any place for the application of the ejusdem generis rule under category (2). For under that category, there is simply no genus to which the articles or instruments so grouped together commonly belong. 47. Indeed, as the Court of Appeal quite rightly pointed out[51], for a reader of the Chinese text, it is quite impossible to come to the conclusion that “other instrument fit for unlawful purposes” can and should be read ejusdem generis to mean only those instruments that are fit for gaining unlawful access, as the phrase had been interpreted to mean under the previous case law prior to 1993. Understandably, this is heavily relied on by the respondent in support of its unrestricted interpretation of the phrase. The alternative basis and the 1995 amendment 48. The second material development has to do with the alternative basis for establishing liability under section 17. Under the original section 17, a person would be convicted if the prosecution could prove that he possessed the offending article or instrument in question “with intent to use the same for any unlawful purpose”. Alternatively, he would also be convicted if he was “unable to give satisfactory account of his possession” of the offending item. This reversal of the burden of proof was not uncommon and was indeed quite acceptable before the advent of human rights law in Hong Kong. But things changed with the enactment of the Hong Kong Bill of Rights Ordinance[52] in 1991. In R v Chong Ah Choi[53], the Court of Appeal held that the second basis of liability in section 17, ie “being unable to give satisfactory account of his possession thereof” was inconsistent with the Hong Kong Bill of Rights for placing the onus on the accused to give an explanation as to the innocent possession of the article or instrument in question, and considered that it had already been repealed by section 3(2) of the Hong Kong Bill of Rights Ordinance when it was enacted in 1991. This prompted the Legislative Council to enact section 48 of the Administration of Justice (Miscellaneous Provisions) (No 2) Ordinance 1995[54] in the following year which simply said that section 17 was amended by repealing “or being unable to give satisfactory account of his possession thereof,”. 49. This development, however, has led the Court of Appeal to believe[55], and the respondent to argue, that given the enhanced protection given to an accused now that there is no onus on him to prove innocent possession and it is for the prosecution to prove intent to secure a conviction, there is no longer any good reason to place a restrictive interpretation on the phrase “other instrument fit for unlawful purposes” or the remaining mens rea requirement of an intention “to use the same for any unlawful purpose”. Later case law 50. The third development starting from the 1980s was that notwithstanding the previous case law on the proper interpretation of section 17 based on the then English text, the courts started to understand “other instrument fit for unlawful purposes” differently and give the phrase an expansive construction[56]. 51. Notably, in R v Ho Ka On[57], at issue was a roll of adhesive tape. The Court of Appeal observed that the tape was the normal small domestic variety of tape which was customarily used as either an item of stationery or to effect minor repairs in a domestic context. It went on to say : “It would be difficult to contend that it was an ‘instrument or article manufactured for the purpose of restraining a person’ or equally that it was ‘an instrument fit for unlawful purposes’. Indeed it would be surprising if it was thought that the tape could be effective in restraining a person’s hands or arms if they were secured with the tape. It would not require much force to break the tape if one’s hands were bound by it.”[58] (emphasis added) 52. This seems to suggest, at least on one reading, that the court did not have the restrictive meaning given to “other instrument fit for unlawful purposes” by the previous case law in mind. The adhesive tape was not an instrument fit for unlawful purposes, the Court of Appeal seems to be saying, not because it had nothing to do with obtaining unlawful access as per the previous case law, but because it could hardly be used to effectively restrain a person’s hands or arms in a robbery – the unlawful purpose at issue in that case. In any event, the court went on to say that Tang Chi-ming did not offer much assistance, and observed that when the case was decided the section was framed slightly differently[59]. 53. Likewise, in quite a few other cases, the courts, particularly those which heard the cases before them in Chinese, tended to give the phrase “instrument fit for unlawful purposes” a wide meaning. The Court of Appeal summarised the position in the past decades as follows[60] : “59. In sum, since the 1960s and 1970s, the Hong Kong courts seemed not to have adopted the interpretation of ‘unlawful purposes A’[61] in Tang Chi-ming. Taking the above three cases from the 1980s and 1990s to the 2000s as examples, the presiding judges were all judges of the Court of First Instance or above, and their judgments were not per incuriam, i.e. they were not made when there was a lack of due regard to Tang Chi-ming, but the results were all inconsistent with that case. In fact, based on the experience of the three judges in this Court, modern Hong Kong courts, especially courts responsible for first instance trials, have all been giving an extremely liberal interpretation to ‘unlawful purposes A’. Looking at the recent past, the so-called ‘anti-amendment bill’ cases handled with this approach are numerous. Even the defence would accept that without hesitation, and no one has ever challenged that even on appeal. Of course, while the above situation has remained the same for several decades, it does not mean that the relevant approach is correct. However, this situation shows that: firstly, section 17 per se may give rise to different interpretations; secondly, interpreting section 17 afresh would not create confusion …” The Chinese text 54. I do not believe any of these developments affect the true construction of section 17. 55. First, the Chinese translation of the original English text of section 17, although given an official authentic status under section 10B(1) of the Interpretation and General Clauses Ordinance, cannot affect the correct interpretation of the section. I accept that under the Chinese text, there is little scope for the application of the ejusdem generis rule. However, a prior question here is whether the Chinese text, having its existence as a translation of the original English text, represents a correct and faithful translation of the latter. If it is not, it does not provide a basis for changing or otherwise affecting the construction of the section under the original English text. 56. In this regard, there is no material before us to suggest that by the 1993 Official Languages (Authentic Chinese Text) (Summary Offences Ordinance) Order made pursuant to section 4B(1) of the Official Languages Ordinance, the legislature intended to effect a substantial widening of the scope of application of section 17. Apart from the tabling of the Chinese text before the Legislative Council for approval by resolution which was duly made, there was no debate in the legislature concerning section 17, nor any proposal by the government to introduce any change in meaning or scope of application of the section by the Chinese text. There was no legislative or other government paper to indicate that any such change was intended, or what the reason behind any such change was. 57. This brings me to a more general question arising from this appeal. Section 10B(1) of the Interpretation and General Clauses Ordinance provides that both the English and Chinese texts of an Ordinance shall be “equally authentic”, and the Ordinance “shall be construed accordingly”. Subsection (2) stipulates that the provisions of an Ordinance are “presumed to have the same meaning in each authentic text”. Subsection (3) then says that where a comparison of the authentic texts discloses a difference of meaning which the rules of statutory interpretation ordinarily applicable do not resolve, “the meaning which best reconciles the texts, having regard to the object and purposes of the Ordinance, shall be adopted”. 58. For a piece of legislation enacted in both English and Chinese, when a difference or an apparent difference in meaning between the English and Chinese texts appears, section 10B(3) provides the methodology to resolve the difference and arrive at the correct interpretation of the provision in question. Whether that task is straightforward or hard, of course, depends on the legislation in individual cases[62]. 59. Where, however, a piece of legislation was originally enacted in English only, and a Chinese translation in the form of an authenticated Chinese text came into being pursuant to section 4B(1) of the Official Languages Ordinance at a subsequent stage, the application of section 10B(3) of the Interpretation and General Clauses Ordinance to resolve any difference or apparent difference in meaning between the two authentic texts requires particular care. When applying the methodology mandated by section 10B(3), the court must bear in mind, as part of the relevant context, that the Chinese text is a later translation of the original English text. Moreover, one cannot ascribe to the legislature an intention to effect any change in meaning of the original English text merely by reason of the Chinese translation[63]. The translation is, after all, a translation only, and mistakes in translation are bound to be made when one is concerned with translating hundreds of pieces of legislation, despite best efforts. Indeed, until its repeal in 2011[64], section 4B(2) of the Official Languages Ordinance specifically provided that the Chief Executive in Council might, by order in the Gazette, correct any manifest error, omission, or inaccuracy in the translated text, and any such correction should be deemed to have been incorporated in the translated text at the time when it was declared to be the authentic text. All these considerations must inform the application of section 10B(3) in resolving any differences in meaning between the original English text and the subsequent translated text. 60. Where, as here, the difference in meaning between the two texts is the result of inaccuracies in translation, the original English text must be taken as better reflecting the object and purposes of the provision under construction, and the difference in meaning be resolved or otherwise reconciled accordingly. 61. Similar views were expressed by Cheung J[65] in Chan Fung Lan v Lai Wai Chuen[66] : “When the court comes to the view that the authenticated Chinese text contains inaccuracies then it should not give effect to that text but should instead rely on the original legislation. This approach is justified because under s 4B of the Official Languages Ordinance (Cap 5) where the Governor in Council has declared a text to be an authenticated text of an ordinance and it appears to him that there is any manifest error, omission or inaccuracy in that text, he may by order in the gazette correct that error and any such correction shall be deemed to have been incorporated in the text at the time when it was declared to be the authenticated text. One must bear in mind that the authenticated Chinese text started life simply as a translation of the original legislation and if there are errors in the translation, which are bound to arise in such a mammoth undertaking, such errors should not be given effect simply because under s 10B of the Interpretation Ordinance the two texts are said to be equally authentic.” 62. In HKSAR v Lau San Ching[67], Lugar-Mawson J adopted the same approach when applying section 10B(3) : “Section 10B(3) provides for a two-step approach; firstly there must be an attempt to resolve the difference of meaning by applying the rules of statutory interpretation. If this fails then the interpreter has to adopt the meaning which best reconciles the texts with regard to the object and purposes of the legislation. From that it necessarily follows that if the ordinance was initially enacted in English, the English text was the original official text from which the Chinese text was subsequently prepared and declared authentic. In ascertaining the ordinance’s legal meaning, the English text should be taken as more accurately reflecting the Legislature’s intent at the time it was originally enacted. In this case, the meaning borne by the original official English text, which was already in existence as early as 1932, should take precedence over the Chinese authentic text. The Magistrate was correct to rely on the English text.” 63. In the case of section 17, the Chinese text no doubt reads better than the English one. But that is not the point. As explained, it contains inaccuracies in translation which are highly material to the task of construction at hand. The English text is therefore preferred as better reflecting the object and purposes of the provision, which must be construed accordingly. Repeal of the alternative basis 64. As to the second development – the repeal of the alternative basis for founding liability under section 17 (“being unable to give satisfactory account of his possession thereof”), given the circumstances under which it came to be considered by the Court of Appeal in Chong Ah Choi[68] and then repealed by the legislature, one cannot discern any legislative intent beyond giving specific effect to the Court of Appeal’s view on the alternative basis. There was no material to suggest that this was meant to widen the scope of application of section 17. If anything, the Court of Appeal observed in that case: “As to the application in future of s. 17 as it now stands, our decision in this case, properly applied, will not result in any failure to charge anyone who ought to be charged under the section. Rather, it will lead to cases being investigated, presented and tried with enhanced care and an enhanced sense of relevance - which is conducive to protecting the public from criminals while at the same time protecting law-abiding people from being wrongly treated as if they were criminals.”[69] Given this sentiment, there is simply no warrant to give section 17 a sweeping meaning on the basis of the legislature’s repeal of the alternative basis following the Court of Appeal’s decision. Significance of the later case law 65. Thirdly, as regards the later case law, it is quite plain that in none of these more recent cases was there any detailed discussion of the proper interpretation of section 17 (in its modern form), or why the reasoning leading to the previous interpretation of that section should no longer be followed, apart from the fact that it had undergone amendments. Possibly, Ho Ka On, a Court of Appeal decision, has been read as casting doubt on the correctness or continued relevance of the previous restrictive interpretation. What also appears to have happened, particularly in relation to those cases heard in Chinese, was that the courts simply focused on the Chinese text and concluded that the reference to “other instrument fit for unlawful purposes” was not restricted to any type of article or instrument[70]. That was understandable for, as explained, the ejusdem generis rule simply could not apply given the way the original English text had been rendered in the Chinese translation. I would not attach any weight to these more recent authorities. The “always speaking” principle 66. In its judgment, the Court of Appeal relied on the “always speaking” rule of construction in support of its conclusion that section 17 should now be given an updated, unrestricted interpretation[71]. The respondent made a similar submission. 67. Bennion has described the rule of construction this way[72] : “Updating construction: (1) Acts are usually regarded as ‘always speaking’. Here, it is presumed that the legislature intends the court to apply a construction that allows for changes that have occurred since the Act was initially framed (an ‘updating construction’). (2) Exceptionally, an Act may be intended to be applied in the same way whatever changes might occur after its passing. An updating construction is not therefore applied to it. … Each generation lives under the law it inherits. Constant formal updating is not practicable, so an Act takes on a life of its own … The legislature may be taken to intend that an enactment (other than one whose meaning and application are, exceptionally, fixed in time) should be applied at any future time in such a way as to give effect to its original intention, making allowances for any relevant changes that have occurred since the Act’s passing. The changes that may give rise to the question of whether an updating construction is appropriate include the creation of new institutions or other bodies (eg devolved legislatures and administrations), technological or scientific developments, new natural phenomena or diseases, changes in social conditions or in the way that society views particular matters, and changes in the territory for which an Act is law. However, the categories of changes that might be relevant are not fixed.” 68. The Court of Appeal set out its reasons for adopting an updated construction of section 17 as follows : “70. The wording and structure of section 17 have been extensively discussed by this Court above. As for the original intent of the legislation, this Court believes that it cannot be clearer. Put simply, it is to punish the act of possessing instruments for criminal purposes, and the act is at the core of the offence. As the offence is preventive in nature and clearly aimed at cases involving less harm, it is therefore included in the basket of summary offences (while serious cases are covered by other ordinances). Of course, the articles and their categories listed in section 17 range from specific (such as wrist cuffs, handcuffs), relatively neutral (such as crowbars, skeleton-keys), to general in nature (other instruments fit for unlawful purposes), but it does not affect the provision’s original intent to punish possession of instruments for criminal purposes. The title of section 17 in the Summary Offences Ordinance is precisely ‘Possession of offensive weapon etc., with intent’. 71. Considering the matter from the above perspective, this Court is of the view that there is nothing wrong with applying the ‘always speaking’ construction to interpret ‘unlawful purposes A’, which can effectively deal with criminals using various or even newly invented instruments to commit crimes. Yet, if that course is taken, the interpretation of ‘unlawful purpose(s) B’ must at the same time be opened up. Otherwise, the two would not be compatible, and that would hinder the implementation of the legislative intent of section 17. That is naturally so and self-evident. Taking the baseball bat as an example, it can be used in both fights and vandalism. Gloves can be used to prevent slipping and avoid leaving fingerprints. Therefore, they are common instruments for armed attacks and burglaries. As for knives and cutters, they are weapons which robbers and sexual offenders often use to threaten their victims. There are countless other similar examples. The fact is that any instrument fit for an unlawful purpose can be used to commit more than one type of offence. Even those articles expressly referred to in section 17 are no exception. Handcuffs used for the purpose of impersonating a police officer is a good example mentioned by both the case authorities and the trial magistrate.” 69. With respect, I am not convinced that any of these propositions provide a reason for giving section 17 a liberal, open-ended interpretation on the basis of the “always speaking” rule of construction. In Royal College of Nursing v DHSS[73], Lord Wilberforce explained when it would not be appropriate to give an updated construction to a provision : “In interpreting an Act of Parliament it is proper, and indeed necessary, to have regard to the state of affairs existing, and known by Parliament to be existing, at the time. It is a fair presumption that Parliament's policy or intention is directed to that state of affairs. Leaving aside cases of omission by inadvertence, this being not such a case, when a new state of affairs, or a fresh set of facts bearing on policy, comes into existence, the courts have to consider whether they fall within the Parliamentary intention. They may be held to do so, if they fall within the same genus of facts as those to which the expressed policy has been formulated. They may also be held to do so if there can be detected a clear purpose in the legislation which can only be fulfilled if the extension is made. How liberally these principles may be applied must depend upon the nature of the enactment, and the strictness or otherwise of the words in which it has been expressed. The courts should be less willing to extend expressed meanings if it is clear that the Act in question was designed to be restrictive or circumscribed in its operation rather than liberal or permissive. They will be much less willing to do so where the subject matter is different in kind or dimension from that for which the legislation was passed. In any event there is one course which the courts cannot take, under the law of this country; they cannot fill gaps; they cannot by asking the question ‘What would Parliament have done in this current case – not being one in contemplation – if the facts had been before it?’ attempt themselves to supply the answer, if the answer is not to be found in the terms of the Act itself.” (emphasis added) 70. In HKSAR v Yuong Ho Cheung[74], Fok PJ, when applying the “always speaking” principle to the construction issue in that appeal, observed that a penal law may be applied to circumstances unforeseen at the time of its enactment, “so long as the case is fairly brought within its language” [75]. In other words, any updated construction must be constrained by the wording of the provision. It must not give the provision a meaning its language cannot bear. 71. In the case of section 17, granted that there may well be a need in society for punishing the possession of instruments other than those covered by the section (as restrictively interpreted) in the prevention of crime particularly in recent years – and the offence may be described as “preventive” in nature targeting possession before any substantive crime is attempted or committed[76], the doctrine simply does not allow the court to ignore the language of the statutory provision and give it a blanket interpretation that totally disregards an applicable rule of construction (the ejusdem generis rule). 72. As explained, despite its deficiencies in drafting, section 17 provides a classic example of the application of the ejusdem generis rule. The suggested unrestricted construction under the “always speaking” principle would, however, do violence to the language when read ejusdem generis. 73. Indeed, given this unrestricted construction of “other instrument fit for unlawful purposes”, there is no need to spell out any specific articles or instruments in the section. It simply eclipses all the specific articles or instruments mentioned in section 17 and renders them redundant. Moreover, the scope of application of the section would be extremely wide, covering many if not most items in daily use which can also be said to be fit for unlawful purposes. Coupled with the equally unrestricted meaning given to the mens rea requirement(an intent to use the instrument for any unlawful purpose), this would give section 17 the widest possible scope of application. 74. In particular, it should be noted that unlike section 33 of the Public Order Ordinance[77] which makes it an offence for any person, without lawful authority or reasonable excuse, to have with him in any public place any offensive weapon, section 17 covers possession even in private, and this makes the construction given to section 17 by the Court of Appeal and contended for by the respondent even more far-reaching in scope and impact. 75. In Tang Chi-ming, the Full Court saidthe section was not a convenient method of seeking to obtain a conviction against a suspect when there was insufficient evidence to charge him with an attempt to commit a crime[78]. The wider construction placed on section 17 by the Court of Appeal in this case would go far beyond using the offence in that way. For under this construction, almost all articles or instruments can be considered as fit for some unlawful purposes so that the actus reus requirement is essentially non-existent. In other words, under this construction, section 17 is in reality a thought crime, depending on what a defendant’s intent was at the material time (subject to proof). There is simply no warrant to suggest that this was the legislative intent. 76. Quite the contrary, from the legislative history and evolution of section 17, it can be clearly seen that when the legislature wanted to update and enlarge the scope of section 17 to cover new articles or instruments, it would do so by a specific amendment to the Ordinance, and the class of new articles or instruments to be added would be carefully defined and supported by justification. Restraint was exercised to ensure that the amendment would not cover articles or instruments not intended to be covered. Thus, in the 1984 amendment, the newly added articles or instruments were either specifically identified or when described generally, restricted by reference to the purpose of manufacture, so as not to cover other articles such as a handkerchief or a piece of string, as the Attorney General unequivocally assured the legislature. 77. To now give, as the Court of Appeal did, the section a wide meaning and scope of application, on the supposed basis of the “always speaking” doctrine and the presumed intention of the legislature, would go contrary to what had happened in the past in the evolution of section 17, and the restrained, incremental approach of the legislature. Such an “updating” would not be giving effect to the legislative intent. Overseas legislation 78. The respondent has drawn our attention to how broadly similar offences have been created in overseas jurisdictions, with a view to supporting its sweeping construction of section 17. With respect, there are no doubt different ways to enact a possession offence like section 17. What is important is to focus on section 17 and consider how best to interpret it in the light of its object, purpose, history and language. Respondent’s fallback construction 79. The respondent argues, as a fallback position, that “other instrument fit for unlawful purposes” should be read ejusdem generis to cover instruments “of a violating nature against person, property or place”.[79] 80. This construction must be rejected as it is wholly inconsistent with the English text of section 17 which covers three separate categories, and “other instrument fit for unlawful purposes” belongs to the third category only. Nor is it supported by the Chinese text as there is simply no common genus to which the various articles and instruments grouped under the second category in the Chinese text belong. The specific mens rea requirement 81. As regards the specific mens rea requirement in section 17, ie “with intent to use the same for any unlawful purpose”, as explained, the early case law has placed a restrictive construction on the requirement, so that the intended unlawful purpose must correspond with the category of article or instrument in question. 82. Section 17 refers to possession of the specific article or instrument “with intent to use the same for any unlawful purpose”. The reference to intended use is important. It occurs in section 17 after definition of the three categories of articles or instruments, each of which is defined by use or purpose. Thus, in the restraint category, the genus of the articles or instruments is defined as “instrument or article manufactured for the purpose of physically restraining a person”, that is, an instrument or article manufactured for such intended use. For the offensive weapon category, absent a statutory definition in the Summary Offences Ordinance, the commonly accepted definition of the term is found in section 2 of the Public Order Ordinance, that is, “any article made, or adapted for use, or suitable, for causing injury to the person, or intended by the person having it in his possession or under his control for such use”. Offensive weapons are therefore defined by reference to use for causing personal injury. Finally, in relation to the unlawful access category, the articles or instruments referred to are “fit for unlawful purposes”, meaning fit to be used for some form of gaining unlawful access. 83. Accordingly, the words “with intent to use the same for any unlawful purpose” should, as a matter of construction, take the meaning from the context of the uses and purposes set out in the earlier part of section 17. It lays down the mens rea of intended use corresponding to the actus reus of the intended uses of the articles and instruments. 84. For these reasons, the words “with intent to use the same for any unlawful purpose” must be understood to refer to an intent to use the offending article or instrument in a manner reflecting the stated use or purpose under the section. In other words, for an instrument or article falling within the restraint category, there must be an intent to use it for restraining a person. For an offensive weapon, the defendant must intend to use it to inflict physical injury on another person, or intend to use it to threaten physical harm to another. For an instrument or article falling within the unlawful access category, the intention must be to use it for obtaining unlawful access. A mismatch between the stated use or purpose of an article or instrument falling within section 17 and the intended use of the same by the defendant would be fatal to a prosecution under the section. The present case 85. In the present case, the 6-inch plastic cable ties do not fall within the first category of “any wrist restraint or other instrument or article manufactured for the purpose of physically restraining a person”. They were simply not manufactured for the purpose of physically restraining a person[80]. Plainly, they are not offensive weapons. Nor are they instruments fit for gaining unlawful access. They fall outside section 17. 86. For these reasons, the appellant’s possession of the bag of plastic cable ties was not caught by section 17, and he should not have been convicted under that section. Disposition 87. I would therefore allow the appeal and set aside the conviction and sentence. Mr Justice Ribeiro PJ: 88. I agree with the judgment of the Chief Justice. Mr Justice Fok PJ: 89. I agree with the judgment of the Chief Justice. Mr Justice Lam PJ: 90. I agree with the judgment of the Chief Justice. Mr Justice Gleeson NPJ: 91. I agree with the judgment of the Chief Justice. Chief Justice Cheung: 92. Accordingly, the court unanimously allows the appeal and sets aside the conviction and sentence. Mr Steven Kwan and Ms Charlotte OT Chan, instructed by O Tse & Co., for the appellant Mr Anthony Chau, DDPP (Ag) and Mr Wayne Lee, PP (attendance excused), of the Department of Justice, for the respondent [1] Cap 228. [2] ESCC 677/2020. [3] Mr Cheang Kei Hong. [4] Following 香港特別行政區 訴 梁有勝 (HKSAR v Leung Yau Sing) HCMA 293/2000, 6 May 2000, To DHCJ. [5] Reasons for findings dated 11 September 2020, [37] (iii)(c) and (d), [41]. [6] Reasons for findings, [36], [51]. [7] Poon CJHC, Derek Pang and Anthea Pang JJA. [8] HCMA 242/2020. [9] [2021] HKCA 1493. [10] [2022] HKCA 34, [5]-[6]. [11] Ribeiro, Fok and Lam PJJ. [12] Cap 1. [13] [2022] HKCFA 5. [14] Chan Ka Lam v Country and Marine Parks Authority (2020) 23 HKCFAR 414, [26]-[27]. [15] HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568, [13]-[14]. [16] Town Planning Board v Town Planning Appeal Board (2017) 20 HKCFAR 196, [29] (2). [17] [1980] AC 744, 807G - 808B. [18] 8th ed., pp 678 - 679. [19] Good Order and Cleanliness Ordinance 1844. [20] Good Order and Cleanliness Ordinance 1845. [21] The Good Order and Cleanliness Ordinance 1845 was re-structured to become the Summary Offences Ordinance 1845, and section 12 of the former became section 22 of the latter with no material change. It was placed under the Chapter “Offences against good order”. In 1933, the Summary Offences Ordinance 1933 was enacted to amend and consolidate the existing law and the former section 22 became section 16 of the 1933 Ordinance. [22] The Ordinance covered miscellaneous summary offences of all sorts – nuisances, trespasses, offences by carriers of nightsoil and pigwash, street cries, photography in court, smuggling by altering vessels, solicitation for immoral purpose, possession and importation of imitation firearms, riotous behavior, unlawful wearing of uniform, resistance or obstruction of public officer, fortune telling, loitering, drunkenness and smoking, amongst other things. [23] [1968] HKLR 242. [24] Mr Noel Power, later Power NPJ. [25] The magistrate must have meant “ejusdem generis” in the context, as “sui generis” means “of its own kind”. [26] pp 244 - 245. [27] p 247. [28] p 247. [29] [1968] HKLR 716. [30] Later Rigby CJ. [31] See footnote 25. [32] p 723. [33] p 731. [34] p 735. [35] pp 730 - 731. [36] Ordinance No 21 of 1975. [37] See Bennion, pp 773 - 774. [38] “ ‘Or’… shall be construed disjunctively and not as implying similarity, unless the word ‘similar’ or some other word of like meaning is added”. [39] [1977] HKLR 408. [40] p 411. [41] Ordinance No 2 of 1984. [42] Hansard (Official Report of Proceedings), 21 December 1983, 355. [43] [5]. [44] p 247. [45] Cap 5. [46] Repealed by section 34 of the Legislation Publication Ordinance (Cap 614) which gives the Secretary for Justice editorial and revision powers to make amendments to legislation. [47] Repealed by section 34 of the Legislation Publication Ordinance. [48] Repealed : Ordinance No 46 of 1994, s 2. Section 34 of the Interpretation and General Clauses Ordinance governs negative vetting of subsidiary legislation. [49] “或管有”. [50] “管有”. [51] [60]. [52] Cap 383. [53] [1994] 2 HKCLR 263. [54] Ordinance No 68 of 1995. [55] [52]-[53], [73]. [56] For instance, A-G v Ip Pui Leung [1989] 1 HKC 498, Michael Wong J; R v Ho Ka On [1996] 1 HKC 69, CA; HKSAR v Leung Wah Chai [1999] 4 HKC 556, Woo J; 香港特別行政區 訴 梁有勝; 香港特別行政區 訴 葉耀民[2021] HKCFI 1100, M Poon J; 香港特別行政區 訴 黃俊廸[2021] HKCFI 1319, Alex Lee J; HKSAR v Chan Tung Shing [2021] HKCFI 1669,Barnes J. [57] [1996] 1 HKC 69. [58] p 71B. [59] p 71E. [60] The court’s judgment was written in Chinese. The extracts in this judgment are taken from its certified English translation. [61] That is, “unlawful purposes” in the phrase “other instrument fit for unlawful purposes”. [62] Re Madam L [2004] 4 HKC 115 provides an example of how difficult it could be to apply section 10B(3) in a given case : [29]-[41]. [63] In HKSAR v Tam Yuk Ha [1997] HKLRD 1031, 1035F, Liu JA observed : “The Chinese language text could not have been intended to be remedial of the English language text. Any rectification would have to be effected by an amendment, not by the other language text.” [64] See footnote 46. [65] As Cheung JA then was. [66] [1997] 1 HKC 1, 8H - 9B. [67] [2004] 1 HKLRD 683, [55]. [68] See [48] above. [69] pp 271-272. [70] See the Court of Appeal’s judgment, [65]. [71] [66]-[72]. [72] pp 503 - 504. [73] [1981] AC 800, 822B-E. [74] (2020) 23 HKCFAR 311. [75] [55], quoting from Joyce v Director of Public Prosecutions [1946] AC 347, 366, per Lord Jowitt LC. [76] Court of Appeal judgment, [70]. [77] Cap 245. [78] p 728. [79] Respondent’s Case, [18](d), [108]. This construction is similar to that reached in 香港特別行政區訴梁有勝 (HKSAR v Leung Yau Sing) where the court purported to construe “other instrument fit for unlawful purposes” ejusdem generis in accordance with the Chinese text to meanthose instruments that are fit for effecting bodily restraint, inflicting physical harm or housebreaking purposes (p 5). [80] I say nothing about those much longer plastic cable zip ties which are sometimes seen to be used by law enforcement agents for restraining persons, or their purpose of manufacture. 
Mr Justice Ribeiro PJ: 1. I agree with the judgment of Mr Justice Gleeson NPJ. Mr Justice Tang PJ: 2. I agree with the judgment of Mr Justice Gleeson NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Mr Justice Gleeson NPJ. Mr Justice StockNPJ: 4. I agree with the judgment of Mr Justice Gleeson NPJ. Mr Justice GleesonNPJ: 5. The issues in this appeal concern the meaning and effect of sections 6A and 18 of the Factories and Industrial Undertakings Ordinance (Cap 59) (“the Ordinance”) and, in particular, the nature of the duty imposed, by s 6A on the proprietor of an undertaking in respect of the safety of employees, and the requirements of s 18. 6. On 6 June 2013, the appellant, Gammon Construction Limited, following a trial before Madam Kathie Cheung, Magistrate (“the Magistrate”) was convicted of two offences and fined $25,000 for each offence. The offences, which were the subject of two summonses, were: (1) Failing to provide and maintain a system of work that is, so far as is reasonably practicable, safe and without risks to health, contrary to sections 6A(1), 6A(2)(a) and 6A(3) of the Ordinance. (2) Failure to provide such information, instruction, training and supervision as is necessary to ensure, so far as is reasonably practicable, the health and safety at work of all persons employed by the proprietor, contrary to sections 6A(1), 6A(2)(c) and 6A(3) of the Ordinance; 7. Following the convictions, the appellant appealed to the Court of First Instance. Such an appeal is by way of rehearing, in the wider sense of that term as explained by Bokhary PJ in HKSAR v Chou Shih Binh[1]. The appeal was heard before Deputy High Court Judge Wright on 18 February 2014, and dismissed for reasons given by the learned judge on 28 February 2014. 8. On 26 September 2014, the appellant obtained leave to appeal to this Court. The workplace accident 9. An allegation of an offence against s 6A of the Ordinance does not depend upon the occurrence of an accident, and, if an accident occurs and an employee is injured, the prosecution, in order to make out its case, does not need to establish a causal relationship between the alleged breach and the injury. Even so, a charge alleging a failure to comply with a proprietor’s obligations as to workplace safety may be brought in circumstances where an accident has occurred, and the death or injury of an employee has resulted. In that event, as a matter of forensic reality, the accident is likely to become the focus of attention in evidence and argument about issues of safety. Those issues will be resolved according to the terms of the Ordinance, but what happened, why it happened, and what could and should have been done to prevent it are likely to be the principal matters that occupy the court. 10. It will be necessary to go into some further detail in due course, but for introductory purposes it is sufficient to state the essential facts as follows. They are taken largely from the findings of the Magistrate and of Deputy Judge Wright. 11. On 3 December 2011, the appellant, a company which has operated for many years in the construction industry in Hong Kong, was engaged in operations at a site at 2 Eastern Hospital Road. In the course of those operations an employee of the appellant (“the deceased”) was struck by a crane and killed. As is common on construction sites, there were sub-contractors participating in the operations. A mobile crane was being used to lift metal bars from below a metal platform on which it stood from an area referred to as the basement level. There were two points of access to the basement level, one on either side of the platform. Each point of access was surrounded by guard rails. The mobile crane stood between the two sets of guard rails. In order to lift materials from the basement level through one point of access or to deposit materials through the either point of access the counterweight of the crane had to turn, or slew, about a pivotal point. The rear of the body of the crane would pass close to one of the sets of guard rails. That would create what was described in the evidence as a trapping zone in which a person might be caught between the counterweight and the guard rails. That is what happened to the deceased. 12. Four persons were involved in the lifting operation. The crane was owned by a sub-contractor. The crane operator (PW1), and another witness (PW2), were employed by a sub-contractor. The deceased was employed by the appellant. The fourth person was an employee in the basement whose function was to make it known when materials were ready to be lifted. Each of those four persons was equipped with a walkie-talkie radio device by means of which he could communicate with the others or overhear the communications of others. 13. Each of the deceased and PW2 was described as a “signalman” or a “banksman”. Both the Magistrate and Deputy Judge Wright accepted a body of evidence which established that they had equivalent responsibilities. The responsibilities of a signalman, or banksman, included giving a signal to the crane operator that it was safe to lift (and therefore turn). They were also both responsible for putting the guard rails in place so as to exclude people from the danger areas. 14. Despite an attempt by counsel for the appellant to establish that there was some hierarchical relationship between the deceased and PW2, with the deceased being, as between the two of them, the superior, this was contradicted by witnesses and was rejected by both the Magistrate and the Judge. In the words of one of the witnesses, both of them were signallers; there was no superior[2]. In practice, most of the signals were given to the crane operator by PW2, but both signallers had the capacity to signal. 15. The Magistrate described what happened as follows[3]: “[I]t is clear that there was no communication / coordination between PW2 and the deceased. In other words, before the crane started to slew, there was no arrangement for PW2 to confirm with the deceased that the restricted area was clear before he gave signal to [the crane operator] for operation of the lifting. Before PW2 received instruction from [the man in the basement], he saw the deceased putting up barriers. Then, he got instruction from [the man in the basement]. Immediately after that, he gave instruction to [the crane operator] to slew the crane. He did not look or check with the deceased as to whether he had finished putting up the barriers. In fact, it was his evidence that (i) he did not need to receive instruction from the deceased as to whether it was safe to move the crane, (ii) no one had given him any signal or command not to start the crane until the deceased was safe, and (iii) he very seldom communicated with the deceased on walkie-talkie. He just had casual chat with deceased. From his evidence, the only irresistible inference is that prior to their starting the operation as a lifting team, they were not instructed that PW2 had to confirm with the deceased as to whether the restricted area was clear before the lifting operation started. Therefore, it was never PW2’s practice to confirm with the deceased on whether it was safe to start moving the crane. Both [the crane operator] and PW2 only relied on the deceased stopping them when the deceased found the situation unsafe.” 16. Exactly why it was thought necessary, or convenient, to have a lifting team that included two signalmen was not clearly explained, but there must have been a reason for it. Deputy Judge Wright referred to evidence that neither the deceased nor PW2 had a complete view of the other’s position and that may be why there were two of them. Whatever the reason, it is evident that, in the absence of proper co-ordination between the two, and between them and the crane operator, safety would be compromised. What in fact happened was that, without any communication from the deceased, upon the signal from PW2 the crane slewed and the deceased was caught in the trapping zone. The Ordinance 17. The following provisions of the Ordinance are presently material: s 6A “(1) It shall be the duty of every proprietor of an industrial undertaking to ensure, so far as is reasonably practicable, the health and safety at work of all persons employed by him at the industrial undertaking. (2) Without prejudice to the generality of a proprietor's duty under subsection (1), the matters to which that duty extends include in particular- (a) the provision and maintenance of plant and systems of work that are, so far as is reasonably practicable, safe and without risks to health; (b) arrangements for ensuring, so far as is reasonably practicable, safety and absence of risks to health in connection with the use, handling, storage and transport of articles and substances; (c) the provision of such information, instruction, training and supervision as is necessary to ensure, so far as is reasonably practicable, the health and safety at work of all persons employed by him at the industrial undertaking; (d) so far as is reasonably practicable as regards any part of the industrial undertaking under the proprietor's control, the maintenance of it in a condition that is safe and without risks to health and the provision and maintenance of means of access to and egress from it that are safe and without such risks; and (e) the provision and maintenance of a working environment for all persons employed by him at the industrial undertaking that is, so far as is reasonably practicable, safe, and without risks to health. (3) Subject to subsection (4), a proprietor of an industrial undertaking who contravenes this section commits an offence and is liable to a fine of $500,000. (4) A proprietor of an industrial undertaking who contravenes this section wilfully and without reasonable excuse commits an offence and is liable to a fine of $500,000 and to imprisonment for 6 months.” s 7A “(1) For the purpose of providing practical guidance in respect of any one or more of the requirements of this Ordinance or of regulations under this Ordinance, the Commissioner may issue such codes of practice (whether prepared by the Commissioner or not) as the Commissioner considers appropriate. (2) The Commissioner may amend or revoke a code of practice issued by him under subsection (1). (3) Where the Commissioner exercises a power under subsection (1) or (2), he shall as soon as may be reasonably practicable publish notice thereof in the Gazette and the notice shall be in such form as the Commissioner considers appropriate. (4) A failure by any person to observe a provision of an approved code shall not of itself cause him to incur any criminal liability, but where- (a) in any criminal proceedings the defendant is alleged to have committed an offence either- (i) by reason of a contravention of or a failure to comply with, whether by act or omission, this Ordinance or regulations under this Ordinance; or (ii) by reason of a failure to discharge or perform a duty imposed by this Ordinance or such regulations; and (b) the matter to which the alleged contravention or failure relates is one to which, in the opinion of the court, an approved code relates, then subsection (5) shall apply as regards the proceedings. (5) In any criminal proceedings to which this subsection applies, the following, namely- (a) compliance with a provision of an approved code found by the court to be relevant to a matter to which a contravention or failure alleged in the proceedings relates; (b) a contravention or failure to comply with, whether by act or omission, any such provision so found, may be relied on by any party to the proceedings as tending to establish or to negative any liability which is in question in the proceedings. (6) In any criminal proceedings, any document which purports to be a copy of a particular approved code shall, in the absence of evidence to the contrary, be regarded by the court as being a true copy of that code.” s 18 “(1) In a proceeding for an offence under a provision in this Ordinance consisting of a failure to comply with a duty or requirement to do something so far as is necessary, where practicable, so far as is reasonably practicable, or so far as practicable or to take all reasonable steps, all practicable steps, adequate steps or all reasonably practicable steps to do something, the onus is on the accused to prove that it was not necessary, not practicable or not reasonably practicable to do more than was in fact done to satisfy the duty or requirement, or that he has taken all reasonable steps, or practicable steps or done the appropriate thing to satisfy the duty or requirement.” s19 “This Ordinance does not- (a) confer a right of action in civil proceedings in respect of a failure to comply with section 6A, 6B or 6BA; and (b) affect the extent (if any) to which breach of any other provision is actionable.” 18. As to s 18, it was common ground in argument in this Court that the onus there referred to is a legal onus (to be discharged on the balance of probabilities) and not merely an evidentiary onus. This, as a matter of construction of the Ordinance, is correct. At trial and before Deputy Judge Wright the case was dealt with upon the assumption that the appellant carried only an evidentiary onus. Nothing in the reasoning of those courts turned on that point. The appellant’s stance in this Court brings it no practical disadvantage, and in one respect (concerning the obligation to give particulars) strengthens its position. Construction of the Ordinance 19. The Ordinance pursues the important purpose of promoting workplace safety. It does so by laying down standards with which a proprietor must comply, and imposing penal sanctions, enforced by way of prosecution for an offence, in the event of non-compliance. The consideration that what is being construed is a penal statute is to be kept in mind, and has been relied upon in argument by the appellant. The provisions of the Ordinance are to be applied according to their terms, and there is no occasion to approach them with an expectation that they will conform to standards of tortious liability imposed upon employers by the common law of negligence, or to interpret them in an attempt to secure such conformity. 20. The duty imposed upon a proprietor is non-delegable. This follows from the creation of an obligation to “ensure” (so far as is reasonably practicable) safety[4]. It is no answer to a charge to show that a particular shortcoming is the fault of a sub-contractor. 21. In a case where a worker has been injured, and the circumstances of the accident are relied upon as evidence of a breach of s 6A, negligence on the part of the worker is not of itself an answer to the charge. The duty to ensure safety, qualified as it is by reference to reasonable practicability, extends to protect workers who fail to take reasonable care for their own safety. In the present case a line of argument pursued at trial and on appeal was that the case went far beyond mere negligence, and that the worker, regrettably, was entirely the author of his own misfortune. That argument failed on the facts. 22. Section 6A(1) requires, subject to the qualification concerning reasonable practicability, the achievement of a result, that is, safety, as elaborated in sub-section (2). The Ordinance was modelled on the United Kingdom legislation considered by the House of Lords in R v ChargotLtd (trading as Contract Services)[5]. In that case Lord Hope of Craighead said[6]: “These duties [to ensure the health and safety at work of employees] are expressed in general terms … They are designed to achieve the purposes described in section 1(1)(a) and (b) . The description in section 2(2) of the matters to which the duty in section 2(1) extends does not detract from the generality of that duty. They describe a result which the employer must achieve or prevent. These duties are not, of course, absolute. They are qualified by the words ‘so far as is reasonably practicable’. If that result is not achieved the employer will be in breach of his statutory duty, unless he can show that it was not reasonably practicable for him to do more than was done to satisfy it. … It is the result that these duties prescribe, not any particular means of achieving it.” 23. To describe a place or system of work as “safe” does not necessary mean that no accident could possibly occur. It does not connote the elimination of every conceivable risk.[7] 24. In Baiada Poultry Pty Ltd v The Queen[8] the plurality said: “The words ‘so far as is reasonably practicable’ direct attention to the extent of the duty. The words ‘reasonably practicable’ indicate that the duty does not require an employer to take every possible step that could be taken. The steps that are to be taken in performance of the duty are those that are reasonably practicable for the employer to take to achieve the identified end of providing and maintaining a safe working environment. Bare demonstration that a step could have been taken and that, if taken, it might have had some effect on the safety of a working environment does not, without more, demonstrate that an employer has broken the duty imposed by s 21(1). The question remains whether the employer has so far as is reasonably practicable provided and maintained a safe working environment.” 25. There are many authorities, all substantially to the same effect, which recognise that, in considering reasonable practicability, a court may have regard to the need to balance the likelihood of risk against the cost, time and trouble necessary to avert the risk[9]. In so doing, regard will be had to what was known at the relevant time. In Marshall v Gotham Co Ltd Lord Keith[10] said that what is called for is an assessment of the situation at the material time. 26. I would stress also what Lord Reid said in the case of Marshall[11] : “I do not find it helpful to consider whether this statutory duty is in every case the same as an employer's common law duty. I think it enough to say that if a precaution is practicable it must be taken unless in the whole circumstances that would be unreasonable. And as men's lives may be at stake it should not lightly be held that to take a practicable precaution is unreasonable.” 27. The task of a court is to apply the language of the statute rather than to seek to accommodate it to principles of the common law of negligence. In a given case reasonable practicability, judged in the light of what was known at the relevant time, may turn upon considerations similar to, or, identical with considerations of reasonable foreseeability relevant to the tort of negligence. This was explained recently by the English Court of Appeal in the case of Tangerine Confectionary Ltd.[12] Where an assessment of the likelihood and gravity of risk is material to the practicability of taking precautions then reasonably foreseeability enters the equation. 28. Counsel for the appellant invited the Court to go further, and to hold (although its materiality to the outcome of the present appeal is far from clear) that reasonable foreseeability of a risk is an element in the factor of safety. It was submitted that “a critical integer in determining whether a system is ‘safe’ is the foreseeability of the particular risk”. In its application to the Ordinance that proposition is not an accurate statement of law, although it may be accepted that in deciding the reasonable practicability of safety measures foreseeability is material. 29. The reference in that submission to “the particular risk” must be a reference to the risk that has been particularised by the prosecution in respect of the offence charged. (The subject of particulars will be considered below). As noted earlier, there can be an offence against s 6A without any accident; and if there has been an accident, causation is not an element of the offence. 30. Risk assessment normally involves foresight, but standards of safety are often propounded with hindsight. It is one thing to say that safety does not require the elimination of every conceivable risk; it is another thing to say that it means an absence of reasonably foreseeable risk. 31. The decision relied upon by counsel for the appellant is Baker v Quantum Clothing Group Ltd[13]. By a majority of three to two the Supreme Court of the United Kingdom, in a civil action for damages for common law negligence and for breach of a statutory duty relating to workplace safety, held that reasonable foreseeability was relevant to the issue of safety as well as reasonable practicability. It is necessary to examine the context in which the question arose. 32. Mrs Baker was employed in a knitting factory from 1971 to 1989. Over that period she was exposed to continuous noise levels of up to 85dB(A) for eight hours a day. This led to her suffering a degree of noise-induced hearing loss. In 1972, a Code of Practice on industrial noise was published by the Department of Employment. It remained on issue until after 1989. It specified that continuous noise levels should not exceed 90dB(A) and required the use of hearing protectors above that level. Mrs Baker’s employer was operating in accordance with the Code of Practice, which in turn reflected current technical knowledge. In 1990, a lower limit, based on developing knowledge, was imposed by regulation. By then the damage to Mrs Baker’s hearing had been done. 33. Accepting that the kind of workplace safety addressed by the relevant United Kingdom statute included exposure to noise, the trial judge found that, by the standards reflected in the Code of Practice, Mrs Baker’s employer had satisfied the requirement to provide a safe place of work, so far as was reasonably practicable. The Court of Appeal disagreed. The Supreme Court restored the decision of the primary judge. The context in which the issue of compliance with the statute arose was that of developing scientific knowledge affecting standards promulgated in industry codes and, later, regulations. All the judges in Baker agreed that the Code of Practice, with which the defendant’s employer complied, was regarded by responsible opinion as acceptable at the time of Mrs Baker’s exposure to noise. The question was the relevance of later knowledge to a judgment as to whether the workplace had, so far as was reasonably practicable, been made and kept safe. 34. In Hong Kong, since 1993, when s 7A was added to the Ordinance, the Baker problem is addressed by the Ordinance, insofar as there is a Code of Practice. 35. All the judgments in Baker accepted that reasonable practicability imported consideration of what was known at the time the injury was sustained. Lord Kerr pointed out[14] it cannot be reasonable to put in place measures that are not known to be necessary. It may be practicable to do so, but it cannot be said to be reasonably practicable. This observation applies also to the Hong Kong Ordinance, and is reinforced by s 7A. There is, however, no occasion in Hong Kong to limit the concept of “safe” itself by reference to risks that were reasonably foreseeable according to knowledge available at the time the subject of the charge. 36. Furthermore, in the present case, no problem of developing technical knowledge or changes of standards arose, and the risk of injury was held to be reasonably foreseeable, and foreseen. There was a Code of Practice, which was tendered and discussed in evidence. The Code adverts to the risk associated with the absence of a clear line of authority and communication in giving signals to the operator of a mobile crane. There was also in evidence a Gammon Risk Assessment and a Lifting Plan, the latter covering the responsibility of a banksman. One of those responsibilities was ensuring sufficient space for lifting and ensuring that no person was working under the lifting area. 37. The reasoning of Deputy Judge Wright was entirely consistent with the principles expressed above. Charging an offence 38. In the course of his argument for the appellant, Mr Reynolds raised the matter of the obligations of the prosecution in proceedings for an offence against the Ordinance. The topic was the subject of consideration by the High Court of Australia in 2010 in Kirk v Industrial Court of New South Wales[15], although differences between the legislation applied in that case and the Ordinance should be kept in mind. 39. In Kirk, the manager of a farm, owned by a company whose principal director lived in the city and who left the running of the farm to the manager, was killed when an All Terrain Vehicle (ATV) he was driving overturned. The manager was using the ATV to deliver material to some fencing contractors who were working on the property. Although there was a formed road leading to the place where the contractors were working, the manager decided (for an unknown reason) to drive off the road and down a steep hill. (As its name implies, the ATV could travel off-road). The ATV overturned and the manager was killed[16]. The authorities charged the company and its principal director with a breach of the Occupational Health and Safety Act 1983 (NSW). The matter was conducted by the prosecution in the Industrial Court, and decided by that court, upon the basis that it was unnecessary for the prosecution to identify any particular act or omission on the part of the employer that constituted the contravention of the statute; it was enough that it appeared that all workplace risks had not been obviated. It was never explained by the Industrial Court how, in those circumstances, the defendant might be expected to set about establishing one of a number of defences provided by the statute. From the facts raised above, the forensic significance of proper particulars in the circumstances of the case was obvious. The High Court held that the approach to the matter taken by the Industrial Court was erroneous. The plurality said:[17] “The approach taken by the Industrial Court fails to distinguish between the content of the employer's duty, which is generally stated, and the fact of a contravention in a particular case. It is that fact, the act or omission of the employer, which constitutes the offence. Of course it is necessary for an employer to identify risks present in the workplace and to address them, in order to fulfil the obligations imposed by [the legislation]. It is also necessary for the prosecutor to identify the measures which should have been taken. If a risk was or is present, the question is -- what action on the part of the employer was or is required to address it? The answer to that question is the matter properly the subject of the charge.” 40. Two years previously in R v Chargot Ltd (trading as Contract Services)[18] the House of Lords considered the question of what was required of the prosecution in presenting a case alleging a breach of the Health and Safety at Work Act 1974 (U.K.), upon which the Ordinance was modelled. Lord Hope said, under the heading: “Prosecution practice”[19]: “As I have said, the statute prescribes the result that must be achieved. That is one thing. How the prosecution proposes to prove that this was so is another. The situation will vary from case to case. In cases such as the present, where a person sustains injury at work, the facts will speak for themselves. Prima facie, his employer … has failed to ensure his health and safety. Otherwise there would have been no accident. But a case where the alleged risk has not had this result cannot be dealt with so easily. It will be necessary to identify and prove the respects in which there was a breach of duty. This is likely to require more by way of evidence than simply an assertion that that state of affairs existed. The particular risk to which the employees … were exposed must be identified. This will require an analysis of the facts in each case. Even where an injury has occurred it may not be enough for the prosecutor simply to assert that the injury demonstrates that there was a risk. Where a prosecution is brought … it may be necessary to identify and prove the respects in which the injured person was liable to be affected by the way the defendant conducted his undertaking. Your Lordships were referred to a number of cases which show how practice varies as to the amount of detail that the prosecutor may give when the charges or counts are being formulated. The overriding test is one of fair notice.” 41. The facts of Kirk are a good example of a case where, although an injury has occurred, it is not enough simply to assert that there was a risk. Apart from the bare fact that the vehicle that the farm manager was driving overturned, nobody knew how or why the accident occurred, and, in the absence of some specified shortcoming in the system of work, it was impossible to know what case the defendant had to meet. 42. In Hong Kong, as in the United Kingdom, the overriding test is one of fair notice. This is consistent with the discussion, in Kirk, of cases such as Johnson v Miller[20]and John L Pty Ltd v Attorney-General (NSW)[21], concerning the entitlement of a defendant to proper particulars of a charge, either in the summons or information, or pursuant to an application for further and better particulars. Applications of that kind may be made before, or during, a trial. 43. Allegations of shortcomings in a system of work may be stated with varying degrees of specificity. What is sufficient to meet the requirements of fairness may depend upon the circumstances of the case. A bare statement, in a summons, of a contravention expressed in the terms of the statute is not enough and, as occurred in the present case, will ordinarily be supplemented by a statement of facts. It is open to a defendant to seek further and better particulars when that is necessary in order to give fair notice of the case the defendant has to meet. Ultimately, the trial court will need to make findings based on evidence of sufficient cogency to establish the elements of the offence charged. 44. The onus provision in s 18 reflects the practical reality that in the ordinary case it is the proprietor who has either created or is in charge of the risk, and it is not unfair to require him to show that he could not have done more to prevent or avoid it[22]. In the present case the parties exchanged written reports of safety experts before trial. As Mr Reynolds correctly submitted, the fact that a defendant bears the onus on an issue does not diminish, but rather increases, the importance of accurate identification of the issue. In a case such as Kirk it may have been very difficult for the employer to work out what steps could have been taken to avoid the risk, but in many cases the proprietor will have little difficulty in understanding and disputing the alleged deficiencies in a system of work. If there is any difficulty resulting in unfairness it may be remedied by an order for further and better particulars. If a trial is conducted in such a manner that there is a denial of procedural fairness then that will be a ground for appeal. The trial 45. There were two summonses, one of which charged an offence in terms of s 6A(2)(a) of the Ordinance and the other of which charged an offence in terms of s 6A(2)(c). Deputy Judge Wright recorded that the parties accepted that the two charges would stand or fall together. That is not disputed in this Court. 46. The Amended Brief Facts provided by the prosecution in advance of the trial said that the death of the deceased resulted from his being trapped between the bottom surface of the counterweight of the crane and the guard rails when the crane slewed in response to a signal from PW2. It referred to an assertion by the appellant’s foreman that only the deceased was supposed to give such a signal. It referred to the trapping zone that was created when the crane slewed, and to a blind spot affecting the operator’s vision. It referred to the opinion of the prosecution’s safety expert that the appellant’s own risk assessment was not properly and effectively implemented in that PW2 and not the deceased was giving signals to the crane operator, and no specific instructions were provided to the signalmen as to how to coordinate their responsibilities. 47. The prosecution’s safety expert’s report included the following: “(c) In the present case, checking the existence and locations of the blind spots was unlikely conducted before the accident. Although a risk assessment for ELS works had been made by Gammon, the assessment was too general which failed to identify and manage the hazards in relation to the blind spots, insufficient visibility from the work positions and the work of [the deceased]. In the absence of such assessment, no specific precautions, information and instructions were provided to the lifting crew … (h) The responsibilities for the banksman should be properly designed in the Plan so as to ensure that the banksman could effectively discharge his duties at work. Section 5.2 of the Plan assigned the responsibility for giving correct signal to the crane operator while stationing outside the fenced off area to stop people from entering. Such responsibilities would be very hard, if not impossible, to be discharged and implemented properly by a person/banksman on site. Reference could be made to section s 5.4.2 of the [Code of Practice] which made clear that a signaller should not be engaged in other activities while giving signal to the crane operator. (i) Pursuant to section 5.2 of the Plan and information from some persons on the site, [the deceased] was responsible for giving signal to the crane operator. However, both the crane operator and [PW2] indicated that such instruction had not been expressed to the lifting crew concerned. (j) In fact, [PW2] was the one giving signal to the crane operator in the past 1.5 months before the day of the accident. It was a strong indication that the Plan had not been properly implemented on site and the line of responsibility among the [crew] was far from clear.” 48. The appellant was plainly on notice that the prosecution was going to contend that the relevant risk was that a worker would be caught and crushed in the trapping zone between the crane and the guard rail when the crane slewed; that there was no clear and adequate line of responsibility between the signallers (the deceased and PW2); that no instruction had been given to the lifting crew as to who was responsible for signalling and as to how the trapping hazard was to be avoided; and that no provision had been made for the deceased to signal that it was safe to proceed. 49. The hearing extended over several days. The appellant set out to make a positive case. The Magistrate recorded in her reasons that it was argued that it was the deceased who was responsible for ensuring that the area was clear, and that the system required PW2 to confirm with the deceased before signalling. The appellant’s safety expert said that under a safe system PW2 would have confirmed with the deceased. However, the Magistrate found that PW2 was never instructed to confirm with the deceased. The defence case was recorded by Deputy Judge Wright, on appeal, as being that the deceased was “the safety boss” and the person in charge of the operation. Deputy Judge Wright pointed out that this was contrary to the evidence, but, if it had been true, it would have made PW2’s failure to obtain clearance from the deceased all the worse. In the appellant’s written case in this Court, the deceased was again described as “the safety boss”, in charge of overall safety. It was said that he was equipped with a radio and could (and by implication should) have given a signal to the operator to stop if any danger arose (“silence amounting to an okay”). In brief, the defence case at all stages was that the deceased was in charge of the safety of the operation and if there had been any danger he “could give a signal to the operator over the walkie-talkie to stop” (Appellant’s Case in this Court para 6.6 (xi)), or (in argument as recorded by the Magistrate) he could have required PW2 to confirm with him before proceeding. That case accepts (and in fact asserts) that the deceased had the capacity to communicate with both PW2 and the operator. 50. The theory that the deceased was “the safety boss”, that it was his responsibility to intervene, either by communicating with the operator or PW2, and that it was entirely his fault if they went ahead without him giving the all clear, could not stand with the evidence. The evidence showed that, under the system as it had operated for some weeks before the accident, there was no requirement for either PW2 or the operator to wait on a signal from the deceased before the crane slewed, and none of them had been given any instruction as to how they were meant to ensure that both PW2 and the deceased were ready for the crane to proceed. 51. It will be apparent from the above why it was common ground that the two charges would stand or fall together. That followed from the nature of the prosecution case and also the defence case. 52. As noted above, in one respect the trial was conducted on a basis unduly favourable to the appellant, in that the prosecution accepted that the onus under s 18 was evidential only. Both the Magistrate and Deputy Judge Wright found the onus to have been discharged by the prosecution. The decision of the Magistrate 53. Since the appeal to Deputy Judge Wright was by way of rehearing in the sense referred to earlier, and since it is the reasons of Deputy Judge Wright that are the subject of this appeal, it is sufficient to refer only briefly to the Magistrate’s decision. She found that there was no instruction to PW2 to seek a clearance from the deceased; she rejected the proposition that the deceased was PW2’s superior and was in charge of safety; she found that no safe system preventing or minimising the trapping hazard was provided and maintained; and she pointed out that the defence’s safety expert accepted that for the system to be safe “involves the confirmation by PW2 with the deceased as to whether it was safe to start lifting operation”. That, she said, was the critical step missing from the system adopted by the lifting team. It was practicable and could have been implemented without any additional resources. Accordingly she found absence of safety (the risk being of crushing in the trapping zone), absence of necessary instructions (to PW2 to seek a clearance from the deceased), and a reasonably practicable safety measure (to require PW2 to obtain confirmation from the deceased before giving a signal to the operator to proceed). The decision of Deputy Judge Wright 54. Deputy Judge Wright was not reviewing the magistrate’s decision for possible error. He was, as he said, considering the matter afresh, but on the basis of the evidence that had been adduced at trial (there being no application to adduce further evidence). 55. The judge declined an invitation to apply the decision in Baker, and cited a number of Hong Kong authorities on health and safety legislation. He said the question was whether there was a risk in the operations carried out; if there was, whether it was reasonably practicable to adopt any safety measures in respect to it; what precautions had been taken; and whether they were sufficient. He quoted a judgment of Patrick Chan J (as he then was) to the effect that what safety measures are required depends on reasonable foreseeability of dangers, the consequences of such dangers, and the trouble and expense involved in minimising the risk. He also referred to Marshall in respect of the test of reasonable practicability. 56. The judge then examined in detail the system of work in place. He referred to the trapping zone and the functions of the participants in the lifting operation. He referred to “the risk of being squashed between the counterweight and the guard rails”. He considered in particular the roles of the deceased and PW2, and rejected the appellant’s suggestion that the deceased was the more senior of the two and was in overall charge of safety. The evidence, he said, showed that they were of equivalent status, and that either of them could direct the crane operator. He found as a fact that the possibility of the presence of the deceased inside the safety barriers was reasonably foreseeable by the appellant, and rejected an argument that it was such extreme conduct on the part of the deceased that it was beyond the scope of what was reasonably foreseeable and that the death was the consequence, not of any shortcoming in the appellant’s system, but of conduct that was contrary to the system. 57. The judge then asked whether, on the evidence, the system was safe so far as was reasonably practicable. He referred to evidence that the view of each of the two signalmen (the deceased and PW2) as to what was going on on the opposite side of the crane was restricted, and that the view of the crane operator also was restricted. For that reason, he said, safety required each signalman to confirm to the crane operator that it was safe for him to proceed. He said that such a requirement would have been reasonably practicable. There had been no instruction that it was necessary for both signalmen to give the go ahead to the crane operator. 58. Deputy Judge Wright also (in para 44 of his reasons) said he agreed with the following passage in the reasons of the Magistrate: “14.... If the defendant had adopted the safe system by requiring PW2 to confirm with the deceased and PW2 did confirm with the deceased that the area was clear before the start of the operation, then, if for whatever reason the deceased deliberately entered the restricted area after the confirmation, I would agree with the defence submission that the defendant could do nothing to prevent it and that the incident was a departure by an employee from the safe system. 15. However, that was not the case here. From the very beginning, PW2 was not required to confirm with the deceased before start of operation. Therefore, the system is unsafe.” The appellant’s criticisms of the judge’s reasons 59. The appellant argued that the judge erred in law by failing to apply Baker in considering whether the appellant’s system was not safe. For the reasons given, on the true construction of the Ordinance, the obligations of a proprietor are to be considered in terms of the language of the Ordinance, and not as a reflex of the common law principles of negligence. The relevance of reasonable foreseeability, if it becomes an issue, will be to the question of the reasonable practicability of some measure to enhance safety. Furthermore, the problem under consideration in Baker was far removed from the issues that arose in the present case. For both of those reasons, the judge was correct in following existing Hong Kong authority, and declining to treat Baker as demonstrating the approach to be taken. 60. As to the importance of the precise identification of the relevant risk as part of the prosecution case, this may be accepted, but it was hardly a difficulty in the present case. From the outset, the risk was identified as a risk that a worker would be caught and crushed in what was referred to as the trapping zone. That risk was plainly recognised in the appellant’s own risk assessment. It was referred to in the material given to the appellant before trial, and it was referred to in evidence and argument and in the reasons of the magistrate and the judge. 61. In examining whether the system was “safe … so far as is reasonably practicable” the reasoning of Deputy Judge Wright was orthodox. He considered, and rejected, the defence case. That case is set out (no doubt in terms substantially the same as those put to Deputy Judge Wright) in the Appellant’s Written Case in this Court. In brief, it was a development of the theory that the deceased was the “safety boss”; that the deceased, PW2 and the operator were all equipped to communicate with one another; that the safety boss was responsible for ensuring that barriers sealed off the danger area; that the signaller (PW2) also checked that the barriers were sealed; that “key protective measures” included that both the deceased and PW2 oversaw the danger area and were trained in relation to the danger of crushing; and that both the deceased and PW2 could signal the operator over the walkie-talkie, or by whistle, if a danger arose. 62. The appellant’s own summary of the defence case does not sit well with the argument that there was a failure to identify the relevant risk. The appellant’s case was considered and rejected on the facts. If what was described in the appellant’s submission were intended to be the system, it was not the way it worked in practice, and nobody told the crane operator or PW2 or (so far as is known) the deceased about key aspects of it. 63. The appellant contends that Deputy Judge Wright did not state with appropriate specificity what the risk was. This is incorrect. The fact that the judge’s focus was on what happened on the day of the accident is unsurprising. What occurred on the day of the accident was that the risk of a worker being caught in the trapping zone materialised. It is also said that Deputy Judge Wright did not make complete findings on the agreed system of work or determine that the system was unsafe by reference to the appropriate tests. These criticisms are not justified. 64. Exactly how the deceased came to be in the trapping zone at the time the crane was slewing is not known, and whether it involved a failure on his part to take reasonable care for his own safety does not appear. However, the flaw in the system of operation of the crane, and the instructions (or lack of instructions) to the participants in the operation, identified by both the Magistrate and Deputy Judge Wright, was that there were two people with the capacity to give signals to the crane operator, both of whom might be in a trapping zone, but the system did not require them both to signal that all was clear. The Magistrate held that a reasonably practicable, and necessary, requirement would have been for the deceased to communicate with PW2 before PW2 gave a signal to the crane operator. Deputy Judge Wright agreed with that, but he also found that a reasonably practicable requirement would have been for both the deceased and PW2 to signal the crane operator before the crane was moved. 65. The appellant says that Deputy Judge Wright’s determination as to the requirement that both the signalmen should signal to the operator “is open to the objection that this precaution had at no stage been raised in the proceedings and was not the subject of any submission before [the judge]”. 66. The first answer to this objection is that Deputy Judge Wright also agreed with the magistrate’s finding that a reasonably practicable requirement would have been that PW2 should not signal before the deceased gave the all clear. 67. The second answer is that the appellant submitted to Deputy Judge Wright, and to this Court, that one of the “key protective measures” in the system was that the deceased could give a signal to the crane operator to stop if any danger arose (silence amounting to an okay). There was, therefore, no question about whether the deceased could have communicated with the operator. The appellant was arguing that, if there was any danger, he could and should have done so. Deputy Judge Wright took up this point and said the system should have required both PW2 and the deceased (that is to say, both signalmen) to signal. 68. The third answer is that the essence of the prosecution case was that, as the system operated in practice, there were two signallers who were responsible for seeing that the crane had clear space in which to operate, but one of them could give a signal that would put the crane in motion without the other being heard from. Both of them had limitations on what they could see. To take up the language of the appellant’s submissions, silence should not have amounted “to an okay”. Whether it is said that the answer is that the system should have required that PW2 heard from the deceased before he gave the signal, or that the operator should have heard from both of them before he moved the crane, the difference is not material. 69. The appellant argued that a requirement that both the deceased and PW2 signal the operator was not reasonably necessary or appropriate because the precautions in place in the system provided multiple protections against the risk. To an extent, this is a repetition of the appellant’s case that was rejected on the facts. The risk, it was said, although “extremely obvious” was also “extremely unlikely to materialise”, partly because it was so obvious, and because the deceased was knowledgeable and experienced. It was suggested there may have been a possible “issue of … difficulty in accepting the precaution” [referred to by Deputy Judge Wright] but, as has been observed, having regard to fact that the appellant itself was arguing that the deceased could and should have communicated with the crane operator this is unconvincing. 70. The reasoning of Deputy Judge Wright involved no error. Conclusion 71. The appeal should be dismissed. Mr Justice Ribeiro PJ: 72. The Court unanimously dismisses the appeal. Mr Guy Reynolds and Mr Raymond Pierce, instructed by Munros for the Appellant Mr Wesley WC Wong SC, DDPP, Mr Andrew HC Li, SPP and Ms Carmen Siu, PP, of the Department of Justice, for the Respondent [1] (2005) 8 HKCFAR 70 at [17] to [19]. For the various meanings of “rehearing” see Builders Licensing Board v Sperway Constructions (Syd) Pty Ltd v Anor (1976) 135 CLR 616 at 619-622. See also HKSAR v Ip Chin Kei [2012] 4 HKLRD 383. [2] Judgment of Deputy Judge Wright para [31] [3] Statement of Findings of Magistrate para [10] [4] cf Kondis v State Transport Authority (1984) 154 CLR 672 [5] [2009] 1 WLR 1 [6] At [17] [7] Baker v Quantum Clothing Group Ltd [2011] 1 WLR 1003 at [179] per Lord Kerr, referring to Sheppey v Matthew T Shaw & Co Ltd [1952] 1 TLR 1272 and Trott v W E Smith (Erectors) Ltd [1957] 1 WLR 1154. [8](2012) 246 CLR 92 at [15] [9] See the authorities cited by Gaudron J in Slivak v Lurgi (Australia) Pty Ltd (2001) 205 CLR 304 at 323 [53]. [10] [1954] AC 360 at 377 [11] [1954] AC 360 at 373 [12] (2012) 176 JP 349 [13] [2011] 1 WLR 1003 [14] [2011] WLR 1003 at 1073 [182] [15] (2010) 239 CLR 531 [16] (2010) 239 CLR 531 at 550 [17] (2010) 239 CLR 531 at 561 [18] [2009] 1 WLR 1 [19] [2009] 1 WLR 1 at 10 [20] (1937) 59 CLR 467 [21] (1987) 163 CLR 508 [22] Davies v Health and Safety Executive [2002] EWCA Crim 2949. This also was a case on the 1974 United Kingdom legislation. 
1. The Appellant, a serving Superintendent of Police at the material time, was charged with one charge of Assault Occasioning Actual Bodily Harm (“AOABH”).[1] He appeared in the Eastern Magistracy and pleaded not guilty.[2] After trial, the Principal Magistrate[3] found him guilty as charged and sentenced him to three months’ imprisonment. He appeals against the conviction and sentence. Appeal against conviction Prosecution Case 2. On 26 November 2014, which was the 60th day of the Occupy Central Movement, the Appellant was deployed as the Company Commander in charge of Tier II New Territories South Company to conduct crowd control at Argyle Street, Mong Kok. 3. It was alleged that, when the Appellant was discharging his duty, he hit with his baton a man (PW1) who was one of the persons in the crowd which the Police sought to control at the material time. Defence Case 4. At trial, there was no dispute that PW1 had been hit by the Appellant with his baton though whether actual bodily harm had been caused as alleged was in dispute. The thrust of the defence case was that the action of the Appellant was conducted as a means to control the crowd and for prevention of crime. He did not have the mens rea required for the offence and in any case he was exempted from liability by operation of certain statutory provisions. Evidence at the trial 5. At trial, the Prosecution called the man who was hit, PW1, and the doctor who attended to him, PW2. The Applicant also testified. There were two sets of Admitted Facts and exhibits were produced. Some evidence was adduced under section 65B of the Criminal Procedure Ordinance[4]. Exhibits produced included video footages produced both by the prosecution and the defence, namely: DBC video and the Ming Pao Video (exhibit P1), Apple Daily YouTube video (exhibit P2), TVB News video (exhibit D1), and Now TV video (exhibit D2).[5] These video footages were produced under relevant admitted facts. The act of the Appellant on PW1 and what happened immediately before the incident can be seen in the DBC video and the Apple Daily YouTube video. Prosecution witnesses 6. It suffices at this stage merely to outline the evidence of PW1 and PW2. 7. PW1 said when he was on the way to Admiralty on the MTR Tsuen Wan line, he received a call from a friend Miss Ma. Knowing that she was in Mong Kok, he was worried about her safety. He told Miss Ma that he would keep her company. He met Miss Ma at the Mong Kok MTR Station at around 10pm and they went out to Sai Yeung Choi Street. They then proceeded along Argyle Street and stopped near Shanghai Commercial Bank (“SCB”) as there were people gathering at the junction of Nathan Road and Argyle Street. Out of curiosity, they stayed there for about five minutes and then walked back by the original route. On the way, he saw police officers chasing a crowd of people who subsequently dispersed. The two of them were pressed closer to the wall of the Bank. At that time, he noticed he had injuries to his shin and right elbow. Then he felt a blow of medium to hard force on the back of his neck by a rod. That was the strike by the Appellant. 8. PW2 prepared a medical report on his examination of PW1 in which he said, inter alia, that “there was a patch of soft tissue swelling with tenderness over his left neck region near hairline that measured around 1 x 4 cm in dimension”. When he testified, he said it was erythema – redness on a patch on that part.[6] Defence evidence 9. The Appellant elected to testify after an unsuccessful no case submission. He described some background information, described what happened on the day, and disclosed the defence as mentioned in paragraph 4 above. His testimony may be summarized as follows.[7] 10. He had been on duty on the night of 25 November 2014, one day prior to the incident. Prior to going on duty he had watched the news and learned of the huge difficulties encountered in the clearance operation that day. 11. He described the night as being highly volatile with crowds gathered at different places trying to re-occupy the roads that had been re-opened by the Police. Many in the crowds were masked, hooded and in protective gear. Some were equipped with homemade shields. Warnings had repeatedly been issued by the Police but to no avail. Force had to be used to restore public order. Police officers were exhausted. 12. In the morning of 26 November 2014, there were still a lot of disturbance during the clearance operation and some arrests resulted. 13. He came on duty again in the evening and received a briefing at 7pm before going to the scene. His duty was to defend the junction of Argyle Street and Nathan Road to ensure the flow of traffic and prevent re‑occupation of the junction. 14. He was equipped with a gun, a helmet, pepper spray and an extendable baton. One has to be trained to use the baton for its various purposes. It might be used for “the check drill” to combat the situation when the crowd approached with hands held up in a passive manner for pushing the officers back. Its use depends on the officer’s knowledge of the baton and his confidence and skill. 15. When he arrived at the scene at 7:45pm, crowds had already gathered. There were clashes which required heavy police presence. Force was required to resume law and order. 16. One of his main duties as a commander at the scene was to monitor crowd dynamics. He described the crowd as hostile and organized. He said that apart from the hostile crowd, there were pedestrians and passers‑by around some of whom eventually took part in charging the police. He viewed them as a whole and did not distinguish them from the hostile crowd. According to his training and taking into account that there had been warnings through the public media, he considered that genuine passers‑by would not hang around at or even go to that area. 17. At about 10:30pm, shortly before the incident occurred, he was informed by a subordinate that something had happened outside the SCB. He turned and saw a large crowd across the road. He led his men to offer reinforcement. Officers took out batons, waved and used them spontaneously. He believed they were concerned about their own safety. 18. He also took out his own baton as a precaution, believing that the crowd was in “active aggression”. He described the scene as noisy and chaotic. He thought using the baton was the only effective way to combat the situation. He considered the crowds had committed various offences such as disorderly conduct, obstructing a police officer, unlawful assembly and a breach of the peace. His assessment was that use of force was the only way to disperse the crowds. 19. He therefore exercised the power pursuant to section 45 of the Police Force Ordinance[8] and section 101A of the Criminal Procedure Ordinance[9] to prevent crime happening. 20. A video footage showed he hit PW1 with his baton and he was the 4th person hitting with a baton (only the Appellant hit on PW1, other officers hit other persons, the Appellant had also done so) during the duration shown. He described the people, including PW1, as almost unwilling to walk. His opinion was that they were reluctant and resisting to disperse or not complying with police orders. He described what he was doing as “posturing”, applying pressure so that the people would leave as soon as possible. He said it was necessary to use force to compel compliance. He described the way the baton was used as whipping, and only a small part of the top of the baton came into contact with the person. It was use of a very mild force. 21. He said[10] he saw PW1 doing an aggressive act towards one of his colleagues nearby (he demonstrated the action), in his judgement, PW1 was aggressive to his colleague, and also challenging as well as being non-compliant. He therefore decided to use force to compel compliance of PW1. As there were people around PW1, he could only strike downwards, as shown in the video footage. He described the force he used as medium force, the baton would stop as soon as it came into contact. 22. He said he hit the back of PW1. He accepted that intentional strike on the neck would be lethal force. He used minimum force and would not take the risk of hitting someone on the neck as his baton is a rigid truncheon. 23. He thought he was discharging his duties by doing what he did. Findings of facts of the Magistrate 24. The Magistrate found PW1 an honest and credible witness and accepted his testimony in its entirety.[11] 25. On the other hand, she said the Appellant did not strike her as a witness of the truth and did not believe him.[12] 26. She found that[13] when PW1 was hit by the Appellant with the baton, he was amongst a group of pedestrians walking along as directed by the police officers. He did not do anything that could be described as hostile that would cause a police officer or the Appellant to hit him with a baton. 27. She also found that the injury on the back of the neck of PW1, ie, swelling and pain, amounted to actual bodily harm.[14] 28. She found the force used by the Appellant by hitting PW1 with his baton considerable and cannot be classified as minimal. She further found that such a blow did cause actual bodily harm to PW1.[15] 29. She then considered whether the use of such force in the circumstances justified under the statutes referred to by the defence, bearing in mind that the statutes only empower a police officer to use force where necessary and in any case the degree of force shall not be greater than is reasonably necessary for that purpose.[16] 30. She considered the matter in the light of her findings that at the material time PW1 was entitled to be there, he was leaving the area with Miss Ma, he was complying with police orders, and was walking steadily without exhibiting any hostility towards any police officers and had walked past the Appellant when he was hit.[17] 31. Her finding is that there was nothing to warrant the Appellant hitting PW1 with the baton.[18] 32. She found that she was satisfied beyond reasonable doubt that the Appellant did assault PW1, thereby occasioning actual bodily harm to him, and that there was no justification for him so assaulting PW1.[19] 33. On this basis, she found the Appellant guilty of the charge. 34. She did not consider sections 46(3) and 53 of the Public Order Ordinance[20] shielded the Appellant from the conviction. She said section 46 is not applicable as exercise of the power by the Appellant fell outside the scope of his duty and / or the act was an abuse of power and it was not done in good faith. She was of the view that section 53 concerned only with civil liability.[21] Grounds of Appeal 35. On appeal, the Appellant was represented by Ms Charlotte Draycott SC,[22] leading Mr Peter Pannu and Mr Benson Tsoi. The grounds of appeal put forward on his behalf are: (1) The Magistrate erred in relation to the mens rea of the Appellant. (2) She failed to adequately consider whether the Appellant was mistaken as to the factual situation. (3) She erred in her finding of fact impacting on her assessment of the Appellant’s honest belief in the necessity and lawfulness of his conduct. (4) She wrongly accepted PW1’s evidence in its entirety. (5) She erred in finding that the Appellant’s blow caused the injury. (6) She erred in her assessment as to whether the strike was justified. 36. Relying on these grounds, Ms Draycott SC submitted that the conviction of the Appellant of the offence of AOABH was wrong and not supported by evidence. She further submitted that for the same reasons, the alternative offence of Common Assault could not be established. Application to adduce new evidence 37. Before the hearing, the Appellant filed a Notice of Motion seeking leave to adduce additional evidence which is a CD-rom containing a video footage purported to be downloaded from YouTube, apparently published by Apple Daily, depicting events that occurred in Mong Kok on 26 November 2014, the day of the incident. 38. The application was objected by the Respondent on the ground that the conditions as set out in the Court of Final Appeal decision of Mahabobur Rahman[23] have not been satisfied. 39. According to the Affirmation of a legal clerk of the solicitors firm acting for the Appellant, the video footage is an uncut version of the footage produced by the Prosecution at trial (exhibit P2). It was further affirmed that: (1) He had found the footage on the internet before the Amended Perfected Grounds of Appeal was filed, but it was without sound recording. (2) After the Perfected Grounds had been filed, the Respondent was contacted to see if there was consent to produce the evidence. (3) After that, and before the Respondent replied, further search on the internet revealed the present version which is with sound recording. (4) Subsequently the Respondent replied and objected to the fresh evidence. (5) Even if the footage had been published before the trial, the Appellant failed to find it (it seems the Prosecution also did not find it). (6) The new footage is important as it shows the uninterrupted sequence of events in the minutes leading up to the strike with the baton, it shows, inter alia, the prosecution case that the strike was done long after the charge at SCB of the crowd towards police officers was wrong and that the Principal Magistrate had been misled. 40. In his objection to the application, Counsel for the Respondent, Mr Daniel Marash SC[24], submitted that the Appellant had failed to: (1) Provide a reasonable explanation as to why the new footage was not produced at trial. (2) Demonstrate its relevance as it does not show what the Appellant claims and in any case does not demonstrate PW1 was actively involved in any protest action before he was struck. (3) Produce any evidence that the new footage is unedited in support of the claimed length of the incident from the time the Appellant became involved until the time he struck PW1. Even if the Appellant can prove the prima facie authenticity of the new footage, it cannot be used to prove that the time gap was indeed 32 seconds as there is no evidence that the footage was uncut and was in sequential order. 41. Section 118(1)(b) of the Magistrates Ordinance[25] provides: “… If the judge thinks additional evidence to be necessary he may receive such evidence, and for that purpose shall have the like powers under subsections (1) and (6) to (17) of section 83V of the Criminal Procedure Ordinance (Cap. 221) that the Court of Appeal would have had if the appeal had been an appeal to which that section applied, …” 42. For the present purpose, it may only be necessary to quote section 83V(1)(a) of the Criminal Procedure Ordinance, which provides: “(1) … the Court of Appeal may, if it thinks it necessary or expedient in the interests of justice— (a) order the production of any document, exhibit or other thing connected with the proceedings, the production of which appears to it necessary for the determination of the case”. 43. In Mahabobur Rahman, the case relied upon by Mr Marash SC, the Court of Final Appeal examined its power to receive new evidence under section 17(2) of the Hong Kong Court of Final Appeal Ordinance[26] which provides that “the Court may exercise any powers of the court which the appeal lies”. In that case, before the case reached the Court of Final Appeal, the Court of Appeal dismissed an appeal against the decision of a District Judge. The power in question therefore is that provided for in the Criminal Procedure Ordinance. It follows that the criteria as set out in section 83V(2) apply. 44. Section 83V(2) of the Criminal Procedure Ordinance provides: “(2) Without prejudice to subsection (1), where evidence is tendered to the Court of Appeal thereunder the Court of Appeal shall, unless it is satisfied that the evidence, if received, would not afford any ground for allowing the appeal, exercise its powers of receiving it if— (a) it appears to it that the evidence is likely to be credible and would have been admissible in the proceedings from which the appeal lies on an issue which is the subject of the appeal; and (b) it is satisfied that it was not adduced in those proceedings but there is a reasonable explanation for the failure to adduce it.” 45. It is noted that section 83V(2) is not mentioned in section 118(1)(b) of the Magistrates Ordinance. In my judgment, it is not necessary as it is only a provision to set out the criteria to fulfil when the court is to exercise its power under section 118(1)(b) and not an enabling provision. 46. A question arose is whether a court hearing an appeal from a magistrate can only receive new evidence when the criteria in section 83V(2) are satisfied. In Mahabobur Rahman, the Court of Final Appeal said that “the conditions to be satisfied before receiving such evidence are provided in section 83V(2).” Ms Draycott SC cited a much earlier case of R v Tam Chung Shing[27] where the following approach was adopted by the Court of Appeal: “subsection (2) imposes a duty upon the court to receive further evidence, if the condition of the subsection is met, but, of itself does not restrict the discretionary power of the court under subsection (1) to receive further evidence if the interests of justice require. The conditions in subsection (2) are not imported into subsection (1) but a court should, in the exercise of its discretion, have regard to them without treating them as decisive in that exercise. The court will not receive inadmissible evidence nor will it receive evidence under subsection (1) if it is satisfied that it ‘would not afford any ground for allowing the appeal’ or it is unlikely to be credible for there must be some curb on the reception of fresh evidence to avoid the indefinite prolongation of the legal process.” 47. She also cited HKSAR v Lam Hoi,[28] which was not reported, where the Court of Appeal held: “48. We are satisfied that we should receive the new evidence under section 83V(1) of the Criminal Procedure Ordinance, Cap 221 notwithstanding that it was available at the sentencing hearing and that no reasonable explanation has been advanced as to why it was not adduced at that hearing. The latter condition is contained in section 83V(2) but it does not limit our discretion under section 83V(1) once we conclude that it is necessary or expedient in the interests of justice to receive the evidence, that the evidence is admissible and that it would afford a ground for allowing the appeal.” 48. In Mahabobur Rahman, the Court of Final Appeal was dealing with an appeal from a District Judge, the present case concerns an appeal from a magistrate. 49. In any case, section 83V(2), Mahabobur Rahman, Tam Chung Shing and Lam Hoi provide good guidance when I am to decide whether the proposed new evidence should be received. Section 83V(2) states that it is mandatory to receive the evidence if the criteria (a) and (b) are met unless it would not afford any ground for allowing the appeal. It does not say it provides for an exhaustive list of situations to receive new evidence on appeal. In any case, if production of the proposed new evidence would not afford any ground for allowing the appeal, or it is unlikely to be credible, there can unlikely be sufficient ground for me to think such evidence “to be necessary” as stipulated in section 118(1)(b) of the Magistrates Ordinance. 50. It was part of the contention of Mr Marash SC that the criteria had not been met in particular that the application relied on “bare assertion” as stated in the Affirmation. Whilst commenting the content of the Affirmation as “bare assertion”, Mr Marash SC confirmed that the Respondent did not intend to cross examine the deponent of the affirmation and would invite the court to consider the matter based on the information disclosed in the affirmation. It of course did not amount to accepting the affirmation but the basis of the court’s decision as to whether there is a reasonable explanation for the failure to adduce it at the trial has to be confined to an assessment of what deposed in the affirmation. At the appeal hearing, the deponent of the affirmation was not called. 51. A point relied upon by Ms Draycott SC is that the prosecution, who must have exercised diligence in identifying relevant recording on the internet, was not able to discover this footage earlier. 52. I am persuaded to accept the explanation for the failure to adduce the evidence in question at the time of the trial as reasonable. 53. I am also of the view that the evidence would have been admissible at the trial. 54. The primary test of admissibility of this sort of evidence is whether there is prima facie evidence that it is authentic, the most important concern is whether it has been tampered with. 55. In Choi Kit Kau v R,[29] Roberts CJ said: “The best method of proving that … recording produced in court is authentic is to show that it has been, since the time it was recorded, continuously in the custody of persons who assert that it was not tampered with. This is a course which should be followed if the authenticity of a tape is challenged and is desirable even if it is not.” 56. Authenticity may, however, be proved circumstantially. The Court of Final Appeal cited with approval in HKSAR v Yeung Ka Ho[30] the following passage in R v Murphy:[31] “Authenticity, in our view, like most facts may be proved circumstantially. In the case of a video film, the direct way is to call the cameraman who took it and the court will normally expect him to be called. But if he is not available, he need not be called; other evidence will suffice if it is logically probative that the video was authentic. That evidence may be adduced in other ways and from other sources.” 57. The proposed new evidence shows what purported to happen during a very short duration earlier than that shown in the Apple Daily YouTube video (“the exhibit”). It captures objects (distinctly the bus No 37) also shown in the exhibit. What shown appears to be smooth and continuous. I am satisfied that the authenticity requirement is met. 58. As to whether it is credible, Mr Marash SC, relying on HKSAR v Lee Chi Fai[32], submitted that although the images captured may accurately show the activities which had transpired, they may not be in their true sequence due to the cutting of the original film. Therefore, even if a prima facie authenticity of the new video footage can be proved, it cannot be used to prove the actual time span unless there is evidence that the footage was uncut and was filmed in sequential order. 59. Ms Draycott SC pointed out that a slightly shorter version of the Apple Daily footage had indeed been produced by the prosecution at trial and at that time Mr Marash SC stated expressly that “it appears to be sequential” and in any case did not take issue with it being sequential. She submitted that the court should look at the matter from the viewpoint of common sense and practical reality having regard to any circumstantial evidence, as the trial judge in R v Murphy did. 60. She also cited the Victoria Court of Criminal Appeal decision in R v Chen[33]which was also quoted with approval by the Court of Final Appeal:[34] “… The test is whether there is sufficient material before the court to allow the tribunal of fact acting reasonably to conclude that the recorded sounds reproduce those originally made by the persons identified by the evidence. … Admissibility does not depend on the party tendering the tapes having removed absolutely any chance that they are inaccurate.” 61. In the circumstances of the case, it is a matter of weight and in my judgment the proposed new evidence cannot be said to be incredible so that it should be excluded from being received as evidence. 62. I have considered whether overall it is necessary to receive the evidence. Having viewed the video footage in question and considered submission, as submitted by Ms Draycott SC what shown is relevant in deciding the circumstances surrounding the incident and in turn the mens rea issue, in particular as it shows the relevant episode which led to the strike lasted just about 32 seconds (not 1½ minutes as estimated by the Appellant when he testified) and it was a split second decision of the Appellant under a perception and belief of the situation.[35] 63. Having considered submission, perused the video footage in question, and regard to the whole circumstances, in my judgment, the proposed new evidence, though should have been received at trial, if available and if a party had sought to adduce it, would not afford any ground for allowing the appeal (as what shown in the video footage does not have the effect of altering the result). I therefore refuse to receive it since it is not necessary. Discussion and Consideration of the Grounds of Appeal Grounds 1 - 3 64. These three grounds relate to the mens rea required for the offence and whether the Magistrate had erred in her approach. As consideration of the issues involved is inter-related, I consider these three grounds at one go. 65. The term “assault” is frequently used to include both assault and battery. In the present case, the allegation against the Appellant is indeed one of battery. Battery is an intentional or reckless act of the accused by which he applies unlawful force to another person. There are circumstances in which force may be applied to another person lawfully. Obvious examples are self-defence, prevention of crime or where the other person consents to the force applied to him. If lawfulness is a live issue, it is for the prosecution to prove the unlawfulness. The mental element to be proved is an intent to apply unlawful force. 66. In the context of the present case, Mr Marash SC confirmed that there is no dispute that the requisite mens rea is, as suggested by Ms Draycott SC, an intent to apply unlawful force. 67. The issue has been conclusively decided by the Court of Final Appeal in HKSAR v Lau Shing Chung Simon[36]. 68. It is for the prosecution to prove all the elements of the offence including the mens rea and unlawfulness. In deciding the issues, as held in R v Gladstone Williams,[37] if there is credible evidence which shows that the Appellant might have been laboring under a mistake of facts, he was to be judged according to his mistaken view of the facts, whether or not that mistake was, on an objective view, reasonable or not. The Lord Chief Justice said: “The jury should be directed first of all that the prosecution have the burden or duty of proving the unlawfulness of the defendant’s actions; secondly, if the defendant may have been laboring under a mistake as to the facts, he must be judged according to his mistaken view of the facts; thirdly, that is so whether the mistake was, on an objective view, a reasonable mistake or not. In a case … where … self defence or the prevention of crime is concerned, if the jury came to the conclusion that the defendant believed, or may have believed, that he was being attacked or that a crime was being committed, and that force was necessary to protect himself or to prevent the crime, then the prosecution have not proved their case.” 69. His Lordship stated that reasonableness is a factor in deciding whether the alleged belief is to be accepted or not, but stressed that “if the belief was in fact held, its unreasonableness, so far as guilt or innocence is concerned, is neither here nor there. It is irrelevant.”[38] 70. It was also held in R v Kimber[39] that if the defendant might have been laboring under a mistake as to the facts, he must be judged according to his mistaken view of the facts. 71. The thrust of the submission of Ms Draycott SC was that the Principal Magistrate had not decided the issue of the required intent adequately or at all. She contended that in deciding the issue, there should have been a proper consideration of the chronology of events which she had set out in some details in the written submission. She further stressed that the matter should be considered from the perspective of the Appellant. She criticized the Principal Magistrate for failing to place herself in the Appellant’s shoes when she judged him, or to recognize that: (1) at the material time he was attempting to discharge his duty after having been engaged for some hours in very difficult situation; and (2) the alleged assault was a single strike delivered on PW1 in the course of dispersing the crowd. 72. She submitted that an assessment of the subjective intent of the Appellant was particularly pertinent in this case and had to be conducted in the context of his role as an experienced senior police officer of hitherto good character, whose task was to keep that area clear of protestors and to keep the roads open. The court must take into account that at the material time the Appellant was attempting to discharge his duty after having been engaged for some hours in very difficult situation, and the alleged assault was a single strike delivered on PW1 in the course of dispersing the crowd, at a time a High Court injunction was in force. 73. She also submitted that in dismissing the Appellant’s evidence of an honestly held belief in the lawfulness of his action, the Principal Magistrate mistook the actions of the crowd and placed disproportionate reliance on PW1’s evidence and the video recordings of his conduct in the seconds before the blow.[40] By so doing she failed to make the assessment from the perspective of the Appellant and judging his subjective belief against the whole circumstances. The Principal Magistrate was also criticized for not having given any, or sufficient, reasons for rejecting the evidence of the Appellant. 74. Ms Draycott SC brought to the attention of the court, inter alia, that the evidence of the Appellant revealed what had happened between 10pm and 10:30pm around which the incident took place: (1) at 10pm the junction was blocked and cordoned by the police and pedestrians were not permitted to cross, the arrangement was made clear by loudhailers; (2) after the cordoning off, emotion of the crowd was heightened and three large crowds gathered at various places nearby and had attempted to push onto the road against the police; (3) minutes before 10:30pm, there was a simultaneous uproar or commotion from the three crowds, people were gathering at SCB where there was the presence of only a few police officers; (4) at 10:30pm, the Appellant was told that something was wrong at SCB; (5) seeing that a big crowd of 500 to 600 people charging a single line of officers and succeeded in going on to the carriageway, he led about 20 officers to go there to provide reinforcement; (6) eventually the majority of the crowd dispersed quickly but there were some who lingered, including PW1. 75. It was the evidence of the Appellant that in such a situation these people who lingered were not really complying with the dispersal order. Based on his training, even if an individual had been a pedestrian originally, he might be influenced and effectively encouraged by the situation and those surrounding him into doing something which he would not normally countenance, and in his view many of those who were originally mere bystanders had joined the mob and taken part in the charging. He thought the people were one integrated group and one could not distinguish between them. It was in the midst of this chaotic situation he decided to do what he did. 76. Ms Draycott SC stressed that the court must consider whether there had been a mistake by the Appellant as to the dynamics of the factual situation, and the Principal Magistrate had not considered this matter adequately. 77. She submitted that before the required intent could be inferred, all hypotheses consistent with innocence had to be excluded, which included: (1) the Appellant’s strike was an error of judgment in all the circumstances; (2) he acted under the mistaken belief that the strike was required to keep the peace and to restore law and order; and (3) he acted under the mistaken belief that PW1 was actively engaged in the protest, was uncooperative, obstructive and verbally abusive to a fellow officer engaged in the clearance. 78. Her submission was that such exclusion was not possible on the evidence. 79. She criticized the Principal Magistrate for not having judged the Appellant’s state of mind against the accurate and full circumstances leading to his use of force before concluding that he did not hold any honest belief in the necessity or lawfulness of his actions. She submitted that the Principal Magistrate misjudged the situation in which the Appellant found himself. In this regard, she submitted that the new evidence as shown in the video footage sought to be adduced at the appeal enabled a better comprehension of the situation. She stressed that the new video footage revealed that the time between the sudden charge of the disorderly crowd and the strike on PW1 was only 32 seconds and in such a situation it was the Appellant’s belief that it was necessary for him to act as he did to restore and maintain order and to clear the junction. She pointed out that at trial, without the benefit of the new video footage, the Principal Magistrate might have been led to form an erroneous view of the situation, in particular as the Appellant thought the time gap was 1.5 minutes, which is now proved wrong by the new video footage. 80. She also criticized the Principal Magistrate for failing to take into account sufficiently what had happened in the area over the previous days in particular the number of people gathered and the violence displayed, the importance of keeping the junction open pursuant to a High Court injunction, the involvement of PW1 in the various incidents on those days, the Police overall strategy to contain the protest, the presence of “passive aggressive” protestors, the appearance of disorderly crowd of which PW1 was said to form a part at the time of the incident, and the perception and belief of the Appellant in such a situation. 81. Ms Draycott SC submitted that an examination of the events leading to the actions taken by the police was necessary for a proper understanding of the dynamics of that night, and criticized the Principal Magistrate for refusing to countenance this exercise. Probably as a result of this, the Principal Magistrate never truly appreciated the Appellant’s case. 82. She submitted that the Principal Magistrate also failed to take into account the following matters sufficiently if at all: (1) the Appellant did not know PW1; (2) the strike was not premeditated, as she accepted; (3) the Appellant knew that he was being filmed; (4) in a split second, in particular as PW1 did say something to a police officer and did move towards that officer (as shown in the new video footage), he perceived that PW1 was doing something hostile to another officer and thought he was being aggressive towards his colleagues, the Appellant made an instant decision to use force on PW1; and (5) the act was done in chaotic circumstances and yet the Appellant was careful of the force he used which was calculated so as not to cause injury. 83. She also quoted Palmer v R[41] in which Lord Morris said: “If there has been an attack so that defence is reasonably necessary it will be recognized that a person cannot weigh to a nicety the exact measure of his necessary defence action … If a jury thought that in a moment of unexpected anguish a person attacked had only done what he honestly and instinctively thought was necessary that would be most potent evidence that only defensive action had been taken.” 84. It was incumbent of the Principal Magistrate to provide reasoned analysis of how she could come to the irresistible conclusion that the Appellant had an intent to apply unlawful force on PW1 and the submission was that she failed to do so. 85. Ms Draycott SC also submitted that the Principal Magistrate’s erroneous conclusion of her assessment of the credibility of PW1 tainted her consideration of the whole case in particular in the mens rea issue as her acceptance of the evidence of PW1 “in its entirety” would have a bearing on her assessment of the belief asserted by the Appellant. 86. It was the contention of Ms Draycott SC that the Principal Magistrate never dealt with the issue of mistake of fact or error of judgment. At the end, she erred in her consideration of this matter, in particular as she failed to recognize that even if she rejected the assertion of the Appellant of his honest belief, it was not determinative of the issue of the required intent. She commented that the Principal Magistrate did not decide the issue of an intention to apply unlawful force adequately or at all. She did not discuss it, nor did she show how she came to that conclusion if indeed she did. 87. As part of her contention that the Principal Magistrate erred in her consideration, Ms Draycott SC quoted exchanges between the Principal Magistrate and Defence Counsel at the trial[42] which she said tended to show that the Principal Magistrate erred in her understanding of some of the evidence, failed to comprehend the crust of the defence case, and hindered the full and proper presentation of the defence evidence. 88. It is apparent that the Principal Magistrate did err, at least in the course of the trial, to say that there was a crowd charging from the other side towards SCB, however, I do not consider such slip so material that it has a substantial impact on her consideration of the case. It is obvious to me that the Principal Magistrate was fully aware of the overall situation and that the Appellant was in an extremely difficult crowd control situation. 89. As regards the other criticism on this front raised by Ms Draycott SC, having considered the exchanges which she brought to my attention, I do not find anything which caused me to have concern about the fairness of the trial. Despite the comment, it was not demonstrated that the defence had been unable to present any specific evidence which is of substantial importance to the decision of the case as a result of what the Principal Magistrate had said. The Principal Magistrate had a case management duty to ensure that a trial was properly and coherently conducted. There was no submission to the extent that there had been undue intervention by the Principal Magistrate. 90. Ms Draycott SC also pointed out that the Principal Magistrate unfairly criticized the evidence of the Appellant as being repetitive and quite contrived. The Principal Magistrate only said so in relation to the evidence of the Appellant regarding the statutes he thought to be relevant. There is nothing in this point which caused me concern. 91. The key question is whether the Principal Magistrate’s approach and consideration of the evidence in her decision of the relevant issues were proper or erroneous. 92. The Appellant gave detailed evidence[43] as to his perception and belief which the Principal Magistrate summarized as follows: “He said he saw PW1 doing an aggressive act towards one of his colleagues nearby. The Appellant demonstrated the act he saw, namely, the upper body moving to the right. He formed the view that PW1 was aggressive to his colleague and challenging and being non-compliant and decided to use force to compel the compliance of PW1. Because of people around him, he could only strike downwards, as he is seen doing on the video footage.”[44] 93. The above is only the gist of the evidence of the Appellant. It, together with what the Principal Magistrate said in her Statement of Findings (as set out in paragraphs 10 – 23 above) show that she had a good comprehension of the defence case. For the sake of examination, I will now set out in more details the relevant part of his testimony. He said when he was handling the situation on the other side of the road, there was a sudden uproar behind him, at the SCB side, he therefore led officers to run over there to give support. He said, having assessed the situation, he made the decision to disperse them.[45] He thought the goal of lawful dispersal would not be achieved if he did not resort to use force.[46] Officers including himself took out batons. He went to the front to do sweeping. He was the most senior officer and he had to keep assessing the dispersal momentum to ensure its completion. People dispersed quickly.[47] When he returned, he saw the first group of people, amongst whom was PW1, as if they almost did not walk.[48] There were police officers behind them but they were walking slowly when officers dispersed them with force. They were walking slowly as what happened an hour ago (which created a chaos).[49] His assessment was that they were not really complying with orders.[50] A person in black (not PW1) was resisting in the sense that he displayed very aggressive non-compliance.[51] When he turned around he saw PW1 doing an aggressive act towards a police officer.[52] He said it was a sudden act but that was his perception from his angle or from his point of view of his facial expression and body language. He thought he was challenging police officers and his decision was to use force to compel his compliance.[53] Since there were people around he could only give the strike with the motion of from up to downwards.[54] It was a kind of force where it would stop as soon as it came into contact with the object.[55] 94. Mr Marash SC argued that the Principal Magistrate did not err in her consideration. He stressed it was the finding of the Principal Magistrate that the facts immediately preceding the assault was not as alleged by the Appellant. She found that the Appellant was not a witness of truth, she did not believe the Appellant on crucial matters including his claim that he believed PW1 being actively aggressive to his colleagues and force had to be used to compel PW1’s compliance.[56] He submitted that the findings were supported by evidence and was a reasonable one so that it is not a finding that an appellate court should interfere. 95. In Gladstone Williams, the Lord Chief Justice said: “The reasonableness or unreasonableness of the appellant’s belief was material to the question whether the belief was held by him at all. … If however the defendant’s alleged belief was mistaken and if the mistake was an unreasonable one, that may be a powerful reason for coming to the conclusion that the belief was not honestly held and should be rejected.” 96. I have perused the Statement of Findings in my examination as to whether or not the Principal Magistrate had addressed the right issues, and whether or not her consideration and decision were proper. 97. I have the following observations: (1) The Principal Magistrate was fully aware that the Appellant sought to justify his actions as a means to control the crowd and for the prevention of crime.[57] (2) She was fully aware of the role played by the Appellant and the duty and task he had at the material time. (3) She was fully aware of the difficult situation the Appellant was in and facing. (4) She was aware of the inconsistencies in the evidence between PW1 and the doctor and the issue arising from the evidence whether the injury found by the doctor was a result of the blow by the Appellant.[58] (5) She demonstrated that she had examined the evidence of the Appellant, the only defence witness, and provided reasons for her major observations,[59] in doing so she had taken into account what the Appellant said under caution in an interview.[60] (6) She noted that PW1 had been supportive of the Occupy movement and had involvement in the activities in the area the night before.[61] (7) She took note that the defendant had a clear criminal record and gave herself a good character direction.[62] (8) She demonstrated she had examined with care the video footages and the still captures produced by the prosecution and the defence respectively.[63] 98. The Principal Magistrate had the following observations: (1) PW1 did turn his head to the right saying to someone perhaps a police officer (this was what the Appellant described as the hostile or aggressive act).[64] (2) PW1 was already being compliant when he was hit.[65] (3) Evidence does not support the allegation that PW1 was an active member of the unruly crowd that had tried to break police lines and cordon.[66] 99. She made specific findings of facts including the followings: (1) “At the time PW1 was hit, he was amongst a group of pedestrians walking along as directed to by the police there … he did not do anything that could be described as hostile that would give cause to the other police officers or the Appellant to hit him with a baton.”[67] (2) “The force applied by the Appellant on PW1 with his baton was neither technical nor trivial and the accompanying pain which (PW1) said he suffered was neither trifling nor transitory.”[68] (3) “… at the material time, PW1 was entitled to be where he was outside the Shanghai Commercial Bank. He was leaving the area with Miss Ma … complying with the police orders and walking steadily … and had walked past the Appellant when he was hit by the Appellant. PW1 had not exhibited any hostility towards the Appellant or any of his colleagues to warrant the Appellant hitting him with the baton … PW1 was already being compliant. There was no need to make him compliant.”[69] 100. Ms Draycott SC submitted that the Principal Magistrate erred in her assessment of the credibility of PW1. My consideration of this matter is set out later at paragraphs 136 ‑ 154. It suffices to say now that I do not consider there are sufficient grounds for me to interfere with the finding of the Principal Magistrate in this regard. 101. The question at present is: did the Principal Magistrate fall into error in her consideration of the mens rea required in the offence, in particular whether or not she failed to consider the issue having sufficient regard to the perception and belief of the Appellant to the situation, and whether her assessment of the state of mind of the Appellant was flaw as a result of her error in fact finding of the relevant circumstances. 102. The Principal Magistrate said: “I was satisfied beyond all reasonable doubt that the Appellant did not honestly believe that PW1 was being actively aggressive to his colleagues.”[70] 103. Mr Marash SC sought to support the decision of the Principal Magistrate submitting that it was sufficiently supported by evidence. He also submitted that the Principal Magistrate had given due and sufficient consideration of the issue of intention to apply unlawful force taking into account the relevant circumstances and the final triggering event. 104. Obviously, the Principal Magistrate was aware that the Appellant was a senior police officer and his role and duties in the difficult time. She must have considered the case on the basis that, at the material time, the Appellant was in the course of discharging his duties in extremely difficult and harsh situation. There is no reason to doubt that she was not aware that it was just a single strike by the Appellant. She made a finding in relation to the force applied which in my judgment is a finding she is entitled to make on the evidence. The Appellant raised the baton high and struck it at PW1. The trunk of the baton landed on an area at the upper back of PW1. Whether the tip of the baton had come into contact with PW1 is not clear from the prosecution evidence but when the Appellant testified he did say only a small part of the top of the baton came into contact with PW1. Though the resulted injury was minor and Ms Draycott SC argued strongly that it was an indication that the force used was far from being heavy, having regard to the fact that a thick clothing was placed at where the baton landed, the indication is not as convincing as she urged me to accept. Strength of the strike apart, which part of the body it aimed at could not be ignored. 105. The Principal Magistrate was entitled to take into account reasonableness in deciding whether to believe what the Appellant claimed his perception and belief were.[71] 106. There was criticism that the Principal Magistrate mistook the actions of the crowd and placed disproportionate reliance on PW1’s evidence and video of his conduct just seconds before the blow. What the Principal Magistrate could do was to decide the relevant issue by assessing the evidence adduced before her. The video footages, despite short, show a situation consistent with the finding of the Principal Magistrate. I do not consider there is sufficient reason for me to interfere with the finding of the Principal Magistrate. The relevant state of mind of the Appellant is what he had at the time he struck his baton and shortly before that. In some situation the circumstances earlier in time may be relevant, however, in the present case, the focus should be on what the Appellant perceived and believed at the immediate short duration preceding the time he decided to inflict the blow. 107. It is trite law that an appellate court would only depart from the finding of facts of the trial court, in particular when such finding depends on a witness’s credibility, if it is plainly wrong in the sense that material matters had been wrongly considered or evaluated, or that consideration or evaluation of such matter had been omitted. 108. I have perused the evidence, in particular the video footages parties invited me to view, including the proposed new evidence. The Principal Magistrate has given a summary of what shown in the video footages.[72] The DBC video and the Apple Daily YouTube video show the incident. The footages show some officers using batons to ensure the crowd keep moving forward. The Appellant was shown also using his baton. The way he used the baton was not the same. The difference was even more significant when he applied his baton on PW1. I have to stress that the action was not conclusive by itself as what the court has to do is firstly to decide whether the Appellant had the intent to apply unlawful force in the light of the perception and belief he genuinely held. The court has to consider the whole of the relevant circumstances and what I have just said are what the court may take into account. 109. I bore in mind the observation of Ms Draycott SC in this regard, in particular that PW1 did stop his movement in the course of moving, turn his head to the right shouting something (the only evidence of the content came from PW1 who said it was something along the line “we are only pedestrians walking past, don’t hit us”; on the other hand the Appellant said he thought this was an aggressive act and he did not say what he heard) and he and Miss Ma were amongst the last in the group to move forward. 110. In relation to these observations, Mr Marash SC pointed out that the video footages show the followings which must also be noted and taken into account: (1) PW1 was together with a woman (PW1 said she was Miss Ma) displaying actions in protecting her. (2) They were close to the wall. (3) The movement of them and a few people nearby were hindered by the crowd at the front. (4) Some of these people in front were hit by police officers (not with heavy force) with a view to urge them on. (5) A man fell onto the ground upon being pushed. (6) The actions of some officers might have made people feel that it was not desirable to move forward at that very moment. (7) PW1 claimed what he said to the officer was they were just passers-by (the only evidence of what had been said came from PW1), this is worth believing as it was reasonable in the circumstances. (8) It was only a very brief pause of PW1. (9) PW1 and Miss Ma were already moving forward together with others at the time he was struck. (10) They were almost at the end of the crowd (it was probably because it was their original position, there is no evidence showing that it was a result of their having moved backward) – and thus there was nobody else to clear. 111. Mr Marash SC stressed that there was nothing in the conduct of PW1 which would have caused any one to think that he was aggressive. What shown in the video footages is telling. 112. Ms Draycott SC criticized the assessment of the Principal Magistrate of the credibility of PW1 and submitted that the erroneous conclusion tainted her consideration of the mens rea issue. The Principal Magistrate said “the prosecution (relied) on the evidence of PW1 as well as the video footages to prove that PW1 had been hit by the Appellant with his baton at the material time and place”. PW1 testified according to what he recalled had happened. The video footages were objective presentation of what had happened. I have no doubt that the Principal Magistrate had the precise comprehension of what had happened and the surrounding circumstances. 113. In any case, the law is not that if a person believes, or may be believing, that the situation is required for him to use force, he can use whatever force to achieve the aim, say as in the present case, to prevent crime. Excessive force cannot be used. As Collins J said in Nimrod Owino,[73] otherwise it would justify shooting of someone who was merely threatening to throw a punch. 114. Whether the force used is excessive has to be gauged in the light of what that person believes the circumstances to be. 115. In John Scarlett,[74] Beldam LJ said: “… the accused is not to be found guilty merely because he intentionally or recklessly used force which they consider to have been excessive. They ought not to convict ‘him unless they are satisfied that the degree of force used was plainly more than was called for by the circumstances as he believed them to be. and. provided he believed the circumstances called for the degree of force used. he is not to be convicted even if his belief was unreasonable. …” 116. John Scarlett was considered by another division of the England Court of Appeal in Nimrod Owino,[75] in which Collins J warned against taking this passage out of context. It was held that: “In the context of an issue of self-defence or reasonable restraint, … a person would not be guilty of an assault unless the force used was excessive; and judging whether the force used was excessive, the jury had to take into account of the circumstances as he believed them to be. … … in judging what he believed the circumstances to be, the jury are not to decide on the basis of what was objectively reasonable; and that even if he, the defendant, was unreasonable in his belief, if it was an honest belief and honestly held, that he is not to be judged by reference to the true circumstances.” 117. The following passage in Williams[76] was cited with approval: “In a case of self-defence, where self-defence or the prevention of crime is concerned, if the jury came to the conclusion that the defendant believed, or may have believed, that he was being attacked or that a crime was being committed, and that force was necessary to protect himself or to prevent the crime, then the prosecution have not proved their case. If however the defendant’s alleged belief was mistaken and if the mistake was an unreasonable one, that may be a powerful reason for coming to the conclusion that the belief was not honestly held and should be rejected. Even if the jury come to the conclusion that the mistake was an unreasonable one, if the defendant may genuinely have been labouring under it, he is entitled to rely upon it. We have read the recommendations of the Criminal Law Revision Committee, …: ‘The common law defence of self-defence should be replaced by a statutory defence providing that a person may use such force as is reasonable in the circumstances as he believes them to be in the defence of himself or any other person.’ … In the view of this Court that represents the law …” 118. Even if there was contrary submission of Ms Draycott SC, I am of the view that Nimrod Owino represents the law. 119. In all the circumstances, I do not consider the Principal Magistrate can be said to have erred in saying: “I was satisfied beyond all reasonable doubt that the Appellant did not honestly believe that PW1 was being actively aggressive to his colleague. Even if the Appellant did so believe, I am satisfied beyond all reasonable doubt that the Appellant was not justified in all the circumstances in hitting PW1 from behind with his baton as he was seen to do.”[77] 120. Ms Draycott SC submitted that the hit was not from behind. PW1 was hit on his back. As I viewed the video footage, PW1 was clearly in the course of moving forward, having walked past, or was passing, the position where the Appellant was. This matter does not bother me. 121. Ms Draycott SC criticized the Principal Magistrate for not providing an analysed reason for rejecting the evidence of the Appellant. 122. The Principal Magistrate did not devote an itemized part in her Statement of Findings to account for her assessment on this matter. However, I do not consider the criticism justified. 123. The Principal Magistrate had these to say:[78] (1) “The Appellant said he hit PW1 to make him compliant. But PW1 was already being compliant.” (2) “The Appellant said he considered PW1 was an active member of the unruly crowd that had tried to break police lines and cordons but there was no evidence to suggest he was. The Appellant himself said he did not recognize PW1 from having seen him earlier amongst the unruly crowd and when he saw him, PW1 was amongst the pedestrians that were walking along the wall of SCB.” (3) “The Appellant said PW1 had exhibited hostility to one of his colleagues. All that PW1 did, as one can see on the video footage shown, is that he turned his head towards his right.” 124. She also took into account that PW1 was walking steadily and had walked past the Appellant when he was hit, the crowd had almost dispersed and PW1 was totally unarmed. She was entitled to do so as these can be seen on the video footages independent of the testimony of PW1. 125. As shown in the video footages, PW1 cannot objectively be said to be aggressive or not complying with order. As regards the state of mind of the Appellant, I agree with the Principal Magistrate that it was not reasonable for the Appellant to have the perception and belief he alleged to have. There is no sufficient reason for me to interfere with the finding of rejecting the asserted perception and belief of the Appellant. As held in Gladstone Williams, the issue is whether or not the defendant held the belief, not whether the belief is reasonable or not, however, reasonableness can be a factor in deciding whether the defendant did have the belief. 126. I do not accept that the Principal Magistrate had not considered the relevant issue having regard to the subjective perspective of the Appellant or failed to have considered any relevant surrounding circumstances. 127. There is no reason to suspect that she was not aware of the proper test to apply as both parties had spelled it out during final submission. 128. Having considered carefully, I see no reason to interfere with the decision of the Principal Magistrate of rejecting the alleged perception and belief of the Appellant. 129. Having rejected the assertion of the Appellant of his perception and belief, it is not the end of the matter. The Principal Magistrate should go on to consider whether the evidence is sufficient for her to be sure that the Appellant intended to apply unlawful force to PW1. 130. Ms Draycott SC pointed out that there is nothing in the Statement of Findings which shows a finding of the required intent. All the Principal Magistrate had said was “the Appellant was not justified in all the circumstances in hitting PW1 from behind with the baton as he was seen to do”.[79] The question was not whether the conduct was justified, there has to be an intent to do an unlawful act before one can be convicted for this offence. Rejection of the evidence of the Appellant is not sufficient, the court has to be sure that the Appellant had the required intent before finding him guilty. Moreover, finding the conduct unjustified indicated that the Principal Magistrate had erroneously adopted the objective test, which again was wrong. 131. I totally agree that the court has to be sure that the Appellant had the required intent. The Principal Magistrate did say she “was satisfied that the prosecution had proved each element of the charge beyond reasonable doubt”.[80] As said, both counsel had set out clearly in their submissions the elements of the offence highlighting the mens rea is an “intention to apply unlawful force” and that whether the Appellant honestly believed at the time that it was lawful for him to assault PW1 in the way he did was an issue the court had to decide. 132. Mr Marash SC submitted that a finding of the required intent was inevitable on the evidence. 133. In my judgment, the Principal Magistrate was entitled to find the required intent being the only reasonable inference supported by the whole of the evidence. 134. Grounds 1 ‑ 3 fails. Ground 4 135. This ground relates to the assessment of the Principal Magistrate in relation to the credibility of PW1. 136. Ms Draycott SC criticized the Principal Magistrate for wrongly accepting PW1’s evidence “in its entirety” as she had ignored numerous implausible elements of the narrative and placed disproportionate reliance on his demeanour. She submitted that the Principal Magistrate had given no consideration to the objective indications that he might have lied or why he might have done so. 137. In particular, Ms Draycott SC set out more than 20 matters which she alleged the Principal Magistrate had failed to recognize. These matters may be summarized into the following areas: (1) None of the injuries found on PW1 was at the part of the body where PW1 said he was hit or as shown by the video footage. (2) His claim that he had been assaulted by other police officers before was not credible, and if true it tended to show that he was not just a mere passer-by. (3) He only went to see a doctor two days after the incident and it was peculiar for him to ask the doctor to take a photograph of the injury found. (4) The assertion of PW1 that there was swelling on his neck as well as pain with stiffness lasting a month was contrary to other evidence, besides, he did not mention about the pain when he saw PW2. (5) The evidence of PW1 that other officers had assaulted his shin on the previous night and earlier on the night of the present incident was relevant as to two issues: whether he was a mere passer-by as he claimed and whether other officers held the same belief as the Appellant claimed he did have. (6) Indeed the doctor (PW2) was of the opinion that the injury to the shin of PW1 had been caused at an earlier time than claimed by him. (7) His description of the injuries was clear exaggeration at the least. (8) The evidence of PW1 about the scene in particular the behavior of the people was contradicted by what shown in the video footage. (9) Evidence of the participation and conduct of PW1 during the Occupy Movement in particular his presence at the scene of unrest on the previous night cast a light on whether he was a mere passer-by at the material time. (10) Evidence, including evidence of what he himself posted on the internet, shows that he was more active in the Occupy Movement than he asserted and his hostility towards the police. (11) The reason he gave for his presence, ie to protect Miss Ma, an acquaintance from a long ago part time job who happened to call him when he was passing Mong Kok on the MTR was implausible, so was the route he said he took when he was together with Miss Ma, as well as his returning to the scene to look for the assailant after having parted with Miss Ma, Ms Draycott SC suggested that it was only a design to conceal the fact that he was present there as a protestor. (12) PW1 refused to give the police the contact details of Miss Ma. (13) The reason PW1 gave for not doing so was that he did not want to involve Miss Ma as her boyfriend did not know of her presence but in fact Miss Ma had posted the fact on her social media. (14) When he made the report, he refused to show his injuries and said unprompted that Miss Ma would not be a witness. (15) His testimony that he was cooperative at the time just before the strike was inconsistent with what shown in the video footage: he had paused while others had continued to move forward and he had at the same time turned to the officers and shouted. 138. Ms Draycott SC submitted that these illustrate that when the Principal Magistrate said “his answers were on the whole consistent and inherently probable”,[81] she erred. Such finding was inconsistent with evidence in particular as video footage shows the behavior of PW1 which at least might have caused the Appellant to have the belief he asserted. 139. She further submitted that the error in finding PW1 credible and reliable tainted the final decision of the Principal Magistrate as she based on this finding in assessing whether she was to accept or reject the assertion of the Appellant of his perception and belief, and ultimately her final decision on the verdict. 140. She argued that the problem was compounded as the Principal Magistrate did not account for her finding that the Appellant did not have the belief he claimed. For this subject alone, I have dealt with it in paragraphs 121 – 123. 141. In the present case, there is no dispute that the Appellant did strike his baton on PW1. However, it does not mean credibility of PW1 is not relevant. It is relevant at least on the following matters which are crucial to the determination of the guilt or otherwise of the offence laid against the Appellant: (1) What PW1 had been doing prior to the strike. (2) What were the circumstances in which the strike took place. (3) Did the strike cause actual bodily harm? 142. The first two matters are relevant to the mens rea of the Appellant and the third is relevant to the actus reus. 143. Mr Marash SC urged the court not to interfere with the assessment of credibility of PW1 stressing the following matters: (1) That the back of PW1’s neck was hit by the baton was supported by video footages and the still captures. (2) The doctor’s evidence is not inconsistent with PW1’s testimony as to the location of his injury and its cause. (3) The fact that PW1 did not mention feeling pain to the doctor had been considered by the Principal Magistrate and she did so in a proper manner. (4) Whether PW1 was a protestor or a passer-by could not be determined by the injuries which PW1 said were inflicted earlier by other police officers. (5) Whether PW1 was trying to distance himself from the Occupy Movement is not material to the consideration of the case. (6) The Principal Magistrate did consider the case on the basis that PW1 was supportive of the Occupy Movement and had been in Mong Kok the night before (from the Statement of Findings, it appears to me what the Principal Magistrate said is that PW1 admitted to be supportive of the Occupy Movement and even if he had been in Mong Kok the night before, she found “he was amongst a group of pedestrians walking along as directed by the police there … and he did not do anything that could be described as hostile”).[82] (7) Whether or not PW1 was actually there to accompany Miss Ma is not material. (8) As the Appellant had not seen PW1 until a short moment prior to the strike he was not in position to distinguish whether PW1 was a protestor or a passer-by. (9) The focus should be what occurred at the location around the time of the offence. (10) The Principal Magistrate had ample time to view PW1 when he testified and assessed his credibility. (11) The portion of PW1’s evidence concerning the few seconds he was in sight of the Appellant was fully supported by the video evidence and this was the crucial period. 144. He also pointed out that whilst the Principal Magistrate said she accepted the testimony of PW1 “in its entirety”, reading the next passage in the Statement of Findings at once shows that this expression was not aptly used, as she said, with reasons, “PW1 was mistaken when he said he was hit on the right side of the back of the neck”.[83] 145. There is sense in this submission. In any case, it appears to me that there was sufficient evidence, in particular that of the video footages, for the Principal Magistrate to make the crucial parts of her findings without relying on the testimony of PW1. 146. As to the first two matters I mentioned in paragraph 141 above, which relate to the mens rea, the Principal Magistrate had these to say in relation to these matters respectively: (1) “On the Apple YouTube Video, PW1 first appears at counter 23, moving by the wall of the Shanghai Commercial Bank, and he is amongst what appears to be a group of pedestrians at that part of the pavement.”[84] (2) “… things were chaotic in Mongkok that evening, requiring a police presence to control the crowds there, … PW1 as being with his female friend and walking amongst the group near the wall of the Shanghai Commercial Bank … .”[85] (3) “Even if PW1 had been in the Mongkok area the night before and, as he himself said, was supportive of the Occupy movement and a blogger, at the material time, …, he was amongst a group of pedestrians walking along as directed to by the police there.”[86] (4) She noted the evidence of a MTR staff that no first aid post had been set up in the Mong Kok Station and there was no reports of anyone having been injured whilst PW1 said he had been offered some assistance there and she did not find this had material impact on her assessment of the credibility of PW1.[87] (5) The evidence of PC 10038 said the crowd walked slowly and some people even were standing. Repeated urge to move was required and he described the crowd as being reluctant to move. The Principal Magistrate found it inconsistent with what is shown in the video footage.[88] 147. Findings (1) ‑ (3) are supported by evidence. In relation to (4) and (5), if there was inconsistency between evidence, the Principal Magistrate did not find it material enough to affect her assessment. I do not find flaw in her approach. 148. The Principal Magistrate did find PW1 was together with a female friend. Ms Draycott SC submitted that the account of PW1 as to how they came to be together was incredible. I agree with Mr Marash SC that it is not material, in the context of the present case, what were the circumstances they met. What happened immediately before the strike is obviously what had to be focused on in consideration, rather than what happened much earlier in time, in particular as those matters had no bearing on the perception and belief of the Appellant. 149. PW1 asserted that the injuries on his shin were inflicted by other police officers on the previous night. There were indeed injuries on the shin. It was for the Magistrate to consider whether to accept or reject the allegation of PW1, having regard to the evidence of the doctor, which she did. Even accepting the allegation is true for the sake of consideration of point (2) above raised by Ms Draycott SC[89], if the Magistrate found such matter did not affect her finding of what PW1 was doing at the material time, it does not cause me concern. Firstly, what PW1 did on the previous night does not have much bearing on what he did on the night in question, in particular as what he was doing during the episode of the struck can be seen from the video footages. Secondly, the belief of other police officer(s) on the previous night does not have much bearing on the perception and belief of the Appellant on the night in question. 150. As to the third matter I mentioned in paragraph 141 above, which relates to the actus rea of the offence, the Principal Magistrate demonstrated that she had given careful thought to the evidence, including the discrepancies revealed (and her observation): (1) The discrepancies between the evidence of PW1 and the doctor[90] : (a) When was the first time the doctor saw PW1 – the Principal Magistrate observed that it was a minor discrepancy. (b) The doctor did observe an injury to the neck of PW1 (erythema, redness in a patch) but having seen the video footage showing the strike, he was of the view that the injury might not have been caused by the strike as the injury he found appeared quite superficial compared with the strike he saw – as PW1 had a sweater around his neck it was difficult to see where the strike actually landed and after having seen the reaction of PW1 to the strike, what the doctor ultimately said was he was unsure if the injury was caused by the strike.[91] (c) PW1 said the neck injury was painful, it was swelling and caused restriction in movement, whilst the doctor said when he saw PW1, he walked normally and did not appear to have any restriction to the movement of neck when examined, and PW1 did not mention pain to him.[92] (2) The immediate reaction of PW1 when he was hit, as shown in the video footage.[93] (3) She noticed that when PW1 testified he said he was hit on the right side of the back of his neck, whilst, in his witness statements, as well as the medical report, the reference is to the left side”.[94] She considered PW1 only mistaken taking into account the three years’ lapse between the time PW1 gave evidence and the incident. (4) The baton was a rigid instrument, video footage showed the Appellant raised the baton to a considerable height before lowering it to hit PW1, and the Appellant said he did apply medium force.[95] (5) As regards the testimony of PW1 that he was pressed to the wall when police officers gave chase to a crowd and he noticed injuries to his shin and on his right elbow, she said “I ignored what PW1 said about the assault on his shin and his elbow as I was only concerned with the assault to the back of PW1’s neck and the injury as a result of the assault”[96], showing she did not allow other injuries to distract her from focusing on the injury allegedly resulted from the strike of the baton. 151. The fact that PW1 did not complain pain to the doctor was a matter the Principal Magistrate had to take into account. That she did not find this matter cause her concern was not unreasonable. The evidence of PW1 that the pain lasted about a month is not necessarily contradicted by the evidence of the doctor, as was submitted by Mr Marash SC, the doctor had only seen PW1 for one time and pain is a subjective feeling. 152. Although the response of PW1 at the time of the strike is not conclusive proof that he did suffer actual bodily harm as a result of the strike, his reaction when hit was a matter the Principal Magistrate was entitled to take into account to decide whether or not PW1 was in fact hit at the part of the body he alleged and the force applied in the strike. PW1 can be seen grabbing the back of his neck on his left. It was the submission of Ms Draycott SC that since there was a limit as to where a person could reach by stretching his arm backward, this action cannot be proof of the part of the body PW1 was hit. The action of PW1 took place as soon as the strike was applied, it was for the Principal Magistrate to decide what was the use and weight of such evidence. 153. It was argued that this evidence of reaction (touching the upper left back) indeed disclosed another discrepancy of PW1’s testimony with other evidence as when he testified he said the right side of his neck was hit. The Principal Magistrate found PW1 was only mistaken when he testified, taking into account the lapse of three years. It has to be noted that in his two witness statements which PW1 gave shortly after the incident, he said the strike landed on the left side of his neck. Photographs taken by the doctor also shows redness on the left side of the neck.[97] The Principal Magistrate was entitled to make the finding she did on this aspect. 154. What the Principal Magistrate had considered and taken into account were what she should, and were entitled to, take into account. As the Court of Final Appeal recognized in Chou Shih Bin v HKSAR,[98] the appellate court does not enjoy the advantage, as the trial court does, the advantage of having received the evidence at first hand. Despite so, I have the duty to examine whether either the assessment or judgment of the Principal Magistrate is erroneous. There is no sufficient reason for me to find that she had erred in accepting the evidence of PW1 in particular the part of his evidence which is material to her decision on what the Prosecution had to prove. Ground 5 155. The Principal Magistrate found that PW1 had suffered actual bodily harm and this was caused by the strike with the baton by the Appellant. 156. Ms Draycott SC submitted that this finding was not justified in particular having regard to the following matters: (1) Video footages, evidence from the doctor and photographs depicting the injuries on PW1 show no visible injury at where the strike of the baton landed. (2) PW1 alleged he was hit on the neck whilst video footage shows the strike landed on his back. (3) The strike was delivered through a thick pullover slung around the shoulder of PW1. (4) The opinion of the doctor is inconsistent with what PW1 alleged. (5) The Principal Magistrate had placed disproportionate weight on the evidence of PW1 which was flawed as stated in the relevant passages cited under the consideration of Ground 4. 157. Ms Draycott SC further submitted that, even if the strike did occasion PW1 actual bodily harm, the Principal Magistrate failed to apply the law correctly as she failed to find there was unlawful assault on the basis either due to the existence of some sinister or improper motive for the Appellant to use his baton as he did or he had intended to cause PW1 unlawful violence. 158. I have already found the Principal Magistrate was entitled to draw the only reasonable inference that the Appellant had the required intent. It was not necessary for the Principal Magistrate to find that the Appellant had some sinister or improper motive. 159. As regards whether or not the strike did land on the body of PW1 and if so at which part of his body, in deciding whether to believe the evidence of PW1, as said earlier, the Principal Magistrate was entitled to take into account what shown in the video footage, as well as the reaction of PW1 although it cannot be conclusive proof. Indeed, quite contrary to the submission that there is inconsistent medical evidence, the doctor found tenderness near the position which PW1 grabbed at the time the strike was hit. The doctor described in his report what he found as “a patch of soft tissue swelling with tenderness over his left neck region near hairline that measured around 1 x 4 centimeters in dimension.”[99] When the doctor testified, he described what he observed on PW1’s neck as erythema, ie redness in a patch.[100] The photograph taken by the doctor shows redness on the left side of the neck of PW1.[101] 160. The doctor said the swelling in question was horizontal, not vertical.[102] Ms Draycott SC argued that this was inconsistent with the blow as shown in the video footage which should produce a vertical injury. Mr Marash SC submitted that it is not necessarily an inconsistency. He urged the court to find by viewing the video footage that the strike was not exactly vertical but angled. He also argued that this might have been a result of the possible cushion effect of the clothing put on the shoulders. I agree that this does not amount to material inconsistency. Besides, as said, the evidence is unclear whether any other part of the baton, for instance its tip, might have come into contact with the body of PW1. The reaction of PW1 can be an indication that this had happened. 161. The doctor of course was not in a position to tell how the injury found was caused. The Principal Magistrate found the evidence of PW1 credible, the video footages provide support to the finding. As submitted by Mr Marash SC the opinion of the doctor did not go to the extent of discrediting or contradicting the evidence of PW1. 162. I have no reason to doubt that the Principal Magistrate had not given sufficient consideration to the whole of the relevant circumstances including what Ms Draycott SC set out in her submission. I do not find the Principal Magistrate erred in finding that the strike did hit the left side of the neck of PW1. 163. The Principal Magistrate found such injury actual bodily harm, having considered a passage in the Archbold[103] and the case of R v Choi Wai-kwong.[104] 164. She found that “the force applied by the Appellant on PW1 with his baton was neither technical nor trivial and the accompanying pain which he, PW1, said he suffered was neither trifling nor transitory.”[105] 165. It was held by the Kings Bench Division in R v Donovan[106] that: “Bodily harm has its ordinary meaning and includes any hurt or injury calculated to interfere with the health or comfort of the victim; such hurt or injury need not be permanent but must be more than mere transient or trifling.” 166. On the evidence which the Principal Magistrate accepted, and such acceptance cannot be said to be wrong, she was entitled to make the finding she did. Ground 6 167. Ms Draycott SC criticized the Principal Magistrate for having erred in her assessment of whether the strike was justified, in that: (1) she failed to judge the reasonableness of the force by reference to the Appellant’s subjective belief of the factual circumstances as he perceived them to be; and (2) she erred in ruling that the statutory exemptions are not available to the Appellant. 168. The first point raised by Ms Draycott SC has been dealt with earlier. 169. When the Appellant testified, quite apart from the asserted perception of the circumstances and his belief, he put forward certain statutory provisions to justify his actions. 170. The statutory provisions put forward were sections 45, 46 and 53 of the Public Order Ordinance,[107] the relevant parts are as follows: (1) Section 45 provides: “Without prejudice to any other powers conferred by this Ordinance, any police officer may use such force as may be necessary— (a) to prevent the commission or continuance of any offence under this Ordinance; (b) to arrest any person committing or reasonably suspected of being about to commit or of having committed any offence under this Ordinance; or (c) to overcome any resistance to the exercise of any of the powers conferred by this Ordinance.” Section 45 confers power on a police officer to use force in relation to offences under the Public Order Ordinance or exercising powers under the same Ordinance. Indeed, generally speaking, a police officer is empowered to use force as may be necessary in similar situations as stipulated and not confined to situation where offences under the Public Order Ordinance occurs. (2) Section 46(3) provides for exemption of liability as follows: “Any person who uses such force as may be necessary for any purpose, in accordance with the provisions of this Ordinance, shall not be liable in any criminal or civil proceeding for having, by the use of such force, caused injury or death to any person or damage to or loss of any property.” (3) Section 53 provides for the so called “good faith” defence, as follows: “Without prejudice to the provisions of section 46, no person acting in good faith under the provisions of this Ordinance shall be liable in damages or otherwise for any act done by him in pursuance or exercise of any obligation, duty or power imposed or conferred, or reasonably supposed to be imposed or conferred, on him by this Ordinance, if done in good faith, and done or purported to be done in the execution of his duty or for the public safety or for the defence of Hong Kong or for the enforcement of discipline or otherwise in the public interest, if such person be a person holding office under or employed in the service of the Government in any capacity, … .” 171. Section 45(a) is relevant in the present case, I will start with it and section 46. The key words in section 45(a) are “may use such force as may be necessary” and “to prevent the commission or continuance of any offence”. 172. In R v Scarlett,[108] which was relied upon by Ms Draycott SC Beldam LJ cited a passage in Venna: “Where … an accused is justified in using some force and can only be guilty of an assault if the force used was excessive, the jury ought to be directed that he cannot be guilty of an assault unless the prosecution prove that he acted with the mental element necessary to constitute his action an assault, that is … that the defendant intentionally or recklessly applied force to the person of another.” and went on to say what I quoted in paragraph 115 above. 173. In Nimrod Owino,[109] while addressing the issue of self defence, Collins J observed: “The jury have to decide whether a defendant honestly believed that the circumstances were such as required him to use force to defend himself from an attack or threatened attack. In this respect a defendant must be judged in accordance with his honest belief, even though that belief may have been mistaken. But the jury must then decide whether the force used was reasonable in the circumstances as he believed them to be.” 174. A two-stage test was therefore set out. Indeed, the same test is stipulated in the statutory regime. The exemption in section 46(3) is subject to section 46(1) which provides: “Whenever in this Ordinance it is provided that such force as may be necessary may be used for any purpose, the degree of force which may be so used shall not be greater than is reasonably necessary for that purpose.” 175. It was the submission of Mr Marash SC that the tests required had not been satisfied. 176. The facts as found by the Principal Magistrate in the present case which led to her conclusion that the Appellant was not justified at all to use force on PW1 comprise the followings: (1) PW1 was in compliance with police orders, leaving the area, and had already walked past the Appellant; and (2) the Appellant did not honestly believe: (a) that force had to be used in order to compel the compliance of PW1; and (b) PW1 was being actively aggressive to any police officers so that using a baton on him was warranted. 177. The Principal Magistrate found that the appellant was not justified to use any force at all, it follows that the question of whether the level of force used was reasonable did not arise. 178. As examined and considered above, I find no sufficient justification to disturb these findings of the Principal Magistrate. On the evidence, as submitted by Mr Marash SC there was no room to find that the Appellant subjectively believed that PW1 was aggressive and that any force was necessary. 179. Mr Marash SC further submitted that it appears when Principal Magistrate said “even if the Appellant so believe, I am satisfied beyond all reasonable doubt that the Appellant was not justified in all the circumstances in hitting PW1 from behind with his baton as he was seen to do”,[110] she was addressing her mind to the second question stated in Orwino. This submission makes sense in the context. In any case, the important thing is that in the first place the Principal Magistrate found that “the Appellant did not honestly believe that PW1 was being actively aggressive to his colleagues.”[111] 180. Having found that the Appellant did not honestly have the belief in question, whether the force used was reasonably necessary becomes irrelevant. That said, if the Principal Magistrate had gone on to consider this issue, it does not affect the validity of the verdict if it was otherwise properly made. Her conclusion on this issue was wholly justified on evidence. 181. All in all, as the force used failed the test, I agree with the Principal Magistrate that the Appellant cannot rely on sections 45 and 46. 182. As regards section 53, the Principal Magistrate ruled that this defence provision covered civil proceedings only and not extended to criminal proceedings. 183. Ms Draycott SC submitted that this ruling is wrong, for the following reasons: (1) if the relevant statutory provision protects the Appellant from civil liability, a fortiori it protects him from criminal liability; and (2) in any event the lack of bad faith / malice on the part of the Appellant in this case was highly relevant to the question of whether he intended to apply unlawful force on PW1. 184. Ms Draycott SC relying on Bennion on Statutory Interpretation, also submitted that there is a presumption against statutory interpretation which would produce any absurd, illogical or anomalous result. 185. Mr Marash SC drew the court’s attention to the Indemnity Ordinance 1922 which provides for similar defence. Its section 3(1) provides: “No action or other legal proceedings whatsoever, whether civil or criminal, shall be instituted in any court of law for or on account of or in respect of any act, matter or thing done, whether with or without the Colony, during the war before the passing of this Ordinance, if done in good faith, and done or purported to be done in the execution of his duty, or for the defence of the realm, or for the public safety, or for the defence of the Colony, or for the enforcement of discipline, or otherwise in the public interest, by a person holding office under or employed in the service of the Crown in any capacity … or by a person holding under or employed in the service of the Government of the Colony of Hong Kong in any capacity … and if any such proceeding has been instituted before the commencement of this Ordinance, it shall be discharged and made void, subject to such order as to costs … .” 186. This provision has been repealed but Mr Marash SC submitted that it still has reference value. Although wording in the provision is not identical with section 53 it covers similar situation. It provides for immunity expressly for both civil and criminal liability, whereas section 53 is silent on criminal liability. That the provision in section 53 only expressly mentioned about “damages or otherwise” is another indication that the legislative intent is to provide a defence for civil liability only. Mr Marash SC further submitted that, as the expression used is “shall be liable in damages or otherwise”, under the ejusdem generis rule “otherwise” it should be interpreted as referring to damages and indemnity which are civil remedies. 187. It should also be noted that, in contrast, section 46(3)[112] as quoted above provides for exemption expressly for both civil and criminal liability. 188. Having considered these, I agree with the Principal Magistrate that section 53 does not extend to cover criminal liability. I do not consider such interpretation would produce absurd, illogical or anomalous result. 189. I have in any case, upon being invited, considered and examined whether or not the Appellant can successfully rely on the provision even if it extends to criminal proceedings. 190. Mr Marash SC referred to several dictionary and submitted that central to the concept of “good faith” is “honesty”. I am with him. 191. Mr Marash SC further submitted that based on the finding of the Principal Magistrate that the Appellant had no honest belief that PW1 was being actively aggressive, there is no basis to regard the Appellant as “acting in good faith”. I agree to this. 192. I therefore find that the Principal Magistrate had not erred in finding that the Appellant could not rely on any of these provisions as separate defence. Conclusion 193. For the reasons I have stated, all the grounds fail. Having considered the evidence and the reasons for the verdict, in my judgment the findings of the Principal Magistrate are reasonable and sufficiently supported by evidence. The conviction is neither unsafe nor unsatisfactory. 194. I therefore dismiss the appeal against conviction. Appeal against sentence 195. In sentencing, the Principal Magistrate took into account the following matters[113]: (1) the Appellant was a senior serving police officer, in command of a team of police officers at the scene, one of the duties being crowd control and preventing re-occupying of the road which had been cleared earlier; (2) the Community Service Order (“CSO”) suitability report[114] which content is favorable to the Appellant and CSO was recommended; (3) many mitigation letters written by various walks of people, all spoke very highly on the Appellant and his contribution to the community and the Police Force; (4) the Appellant had a long and distinguished career in the Force, and had earned many compliments and commendations; (5) the difficult situation faced by the officers and the Appellant at the material time; (6) officers were subject to insulting remarks and violent behavior from the protestors; (7) officers, including the Appellant, had worked hours prior to the incident; (8) the assault of PW1 was very short and counsel had described it as a technical breach (which the Principal Magistrate did not accept);[115] (9) the injury sustained by PW1 was not serious and he has fully recovered; (10) the submission that the Appellant was only overzealous and the chance of re-offending was almost non-existent; (11) the Appellant is a man of good character and personality; (12) since the incident, the Appellant and his family have been subjected to harassment by social media; and (13) there had been a lengthy period before the Appellant was charged for the offence. 196. She has considered the following factors which the Court of Appeal said in HKSAR v Chan Chun-tat[116] as relevant in determining the gravity of the offence and the culpability of the offender: (1) the extent to which the assault was premeditated; (2) the reason or motivation underlying the assault upon the victim; (3) the mental / emotional state of the assailant at the material time; (4) whether alcohol or drugs contributed to the action; (5) whether the assault was committed by the assailant alone; (6) the type of weapon employed; (7) the level of force or aggression; (8) the persistence with which the assault was pressed home; (9) the injuries caused to the victim; and (10) the effect of the assault upon the victim and those close to him. 197. The Principal Magistrate accepted that the circumstances of the present case were not the most serious in that the injuries sustained by PW1 were not the most serious and he had now fully recovered. She was also satisfied that the assault was not premeditated.[117] 198. She, however, highlighted the following matters which caused her to have a serious view of the case:[118] (1) PW1 was an unarmed pedestrian when he was hit; (2) he was hit from behind by the Appellant after he had walked past him (the Appellant); (3) he was walking with a woman at that time; (4) the gesture he made which the Appellant claimed had led him to believe that his colleague was at risk cannot be viewed as a threatening one; (5) PW1 did not as a matter of fact suffer more serious injury but the situation would have been different if not for the sweater hanging around his neck; (6) the baton was capable to inflict serious and permanent injuries; (7) it was used on a passing member of the public who had not displayed any hostility to the Appellant nor any police officers at the scene; and (8) the senior position the Appellant held in the Police Force. 199. Although the CSO suitability report was favourable, the Principal Magistrate did not adopt the recommendation since she considered the culpability too serious for such an order, having regard to the nature and circumstances of the offence and the Appellant.[119] 200. She was of the view that a deterrent sentence was necessary and a term of imprisonment was appropriate. 201. She adopted four months as the starting point and gave the Appellant one month discount for the following matters:[120] (1) the circumstances prevailing at the time and the great stress police officers were under; (2) the character of the Appellant and that it is his first conviction; (3) his contribution to the Police force and the community; (4) he is now retired and unlikely to re-offend; (5) the stress caused to him since the investigation of the matter and the length of time lapsed before completion of the proceedings; and (6) the stress from social media. 202. She was of the view that the circumstances of the assault were too serious to suspend the sentence. Nor did she find any exceptional circumstances warrant suspending of the sentence. 203. After the above consideration, she imposed an immediate imprisonment term of three months upon the Appellant. 204. On appeal, Ms Draycott SC stressed that the case involved a single strike the Appellant delivered on PW1 when he was discharging duty as a police officer dispersing an unruly crowd. 205. She put up the following grounds of appeal: (1) The imprisonment term was wrong in principle in the circumstances of the case in particular as the Principal Magistrate erred in not properly considered the following matters: (a) The background and context in which the Appellant committed the offence was exceptional, namely: (i) the officers including the Appellant were under unparalleled pressure; (ii) they constantly faced provocation and real risk of violence from all fronts at the junction the entire night; (iii) they had been on duty for hours trying to defend the junction; and (iv) the Appellant and his team were tasked with subduing an emotional, uncooperative and sometimes violent crowd. (b) The exemplary personal background of the Appellant, such as: (i) he has been in the Police Force for over 35 years with hitherto good character; (ii) he was four weeks away from his pre-retirement leave; (iii) he regrets what had happened; and (iv) the offence was clearly out of character. (c) The Principal Magistrate wrongly applied case authorities which involved breach of trust and aggravating factors in particular as those cases were of much more serious both in nature and facts, whereas the present case involved an instinctive decision and reaction resulting in a single strike without premeditation and the Principal Magistrate should have sentenced the Appellant on the basis of an error of judgment. (d) The option of a CSO was a reasonable one, especially as there was a report with favourable recommendation. (e) Significant delay in the case of over three years. (f) The Appellant was subjected to serious public condemnation before, during and after the trial. (g) Whether deterrence was needed. (h) The Appellant had already been remanded in custody for 16 days after conviction pending sentence. (2) The sentence is manifestly excessive and the Principal Magistrate had erred: (a) In failing to give sufficient weight to the exemplary personal background of the Appellant and the peculiar background and context in which the offence took place. (b) In considering that the present case called for deterrence, as there was no basis for either a general or a personal deterrence. (c) The sentence imposed is heavier than various cases which involved facts more serious as well as aggravating features. (3) The Principal Magistrate erred in not suspending the sentence and she had applied the wrong legal test in consideration of the matter. 206. Apart from what she set out in the written submission, Ms Draycott SC also stressed that: (1) the Appellant had already served 16 days in custody, the court should consider a “clang of the prison gate” sentence; (2) he has over the years been suffering from harassment by the press and people having political affiliation, as evidenced from what recently had happened as well as what took place outside the court building during hearing; (3) such harassment had hindered him from looking for a job even though he was physically fit after retirement and desired to do so; and (4) the matter had lingered on for almost four years now during which the Appellant could not live a reasonably private life. 207. She urged, under the very special circumstances of the case, the court to consider ordering a conditional discharge instead of the sentence now imposed. 208. Mr Marash SC sought to support the sentence imposed and in his written submission he highlighted the following matters: (1) In the present case, a heavy baton was used to strike an unsuspecting and unarmed pedestrian from behind. (2) The blow was forceful and delivered to the neck which the Appellant himself acknowledged could be lethal.[121] (3) It was only fortuitous that PW1 was not injured more seriously. (4) It was a case of a police officer committing a serious offence in the course of his duty. (5) A deterrent sentence is required in the circumstances to maintain public confidence. (6) In imposing the present sentence having given a one-month discount the Principal Magistrate had given appropriate weight to the circumstances of the case as well as the Appellant personally. (7) The Principal Magistrate did not err in not making a CSO in particular as the Appellant did not show genuine remorse. In any case, citing SJ v Wong Chi Fung,[122]it was submitted that as the circumstances were so serious which clearly calls for a deterrent sentence, even if the offender is suitable for the Order, it is not appropriate to impose one. (8) The Principal Magistrate applied the correct legal principle and had given proper consideration to all relevant factors before deciding not to suspend the sentence. 209. The Principal Magistrate must have an extremely difficult task in deciding the appropriate sentence in this case, in particular as she had said “I did have sympathy for the stress police officers were under at the time of the Occupy movement and the difficulties they faced in crowd control”.[123] 210. I do not have an easier task. 211. Having completed education at F6 level, the Appellant started his career as a police constable and after 36 years’ service he retired as a Superintendent. He must have proved himself to be an outstanding member of the Police Force. He fell from grace for a single strike with his baton, in the course of discharging his duty in a difficult situation, and at a difficult time not only for the Police but also for the whole community of Hong Kong. The effort of each member of the Police Force has to be recognised. 212. As the Principal Magistrate found, there was no premeditation. The conduct was out of character. On the other hand, the conduct cannot be regarded as a technical breach as Counsel representing the Appellant at trial suggested. 213. What the Principal Magistrate said she had considered, as outlined in paragraph 195 ‑ 202 above, are all matters she was entitled to take into account. 214. A police baton is a rigid truncheon and a potent weapon. On this occasion, it landed near the neck of PW1. As the Principal Magistrate observed, the baton was raised high and the strike was not applied without force. I agree that had PW1 not put the clothing on that part of the body the consequence would have been more harmful. 215. There is no doubt that the Appellant and his colleagues were in a very difficult situation. However, at the very moment before the strike, what I can see from the video footages were people moving steadily on the pavement and there was no sign of non-compliance, let alone any aggressive or abusive behavior either on the part of PW1 or the people around him. PW1 was in this crowd of people, together with a woman. Nothing in his conduct and behavior singled him out from the others, or could have bothered any people in his vicinity. The Principal Magistrate did not find the Appellant labouring under any honest but mistaken belief of the circumstances. 216. In the circumstances of the case, the court has to consider an appropriate sentence on the basis that the Appellant intended to inflict unlawful force on PW1. 217. There was only one strike. It was not a strike as a gesture. The strike occasioned actual bodily harm, though not serious. 218. It was an unlawful act performed by a senior police officer in the course of duty. A law abiding citizen is entitled to expect no such thing to happen on him. If that happens, public confidence is shaken. That the act was done by a senior police officer in the presence of other officers including his subordinates further aggravated the matter. There is a reasonable expectation from the community of what a senior police officer should do and not do. What the Appellant did, sadly, failed to meet the expectation. He also set a very bad example to his subordinates. 219. As Lamer CJC said in R v M,[124] the duty of those who sentence is “to draw upon the legitimate principles of sentencing to determine a just and appropriate sentence which reflects the gravity of the offence committed and the moral blameworthiness of the offender.” 220. The court is to have regard to the classical principles of sentencing in deciding how best to dispose of a case. The principles are summed up by Lawton LJ in R v Sargeant:[125] “Retribution, deterrence, prevention and rehabilitation. Any judge who comes to sentence ought always to have those four classical principles in mind and to apply them to the facts of the case to see which has the greatest importance in the case with which he is dealing.” 221. The court is under the duty to achieve the right balance in the circumstances of the particular case. 222. The Principal Magistrate obviously was aware of all the matters mentioned by Ms Draycott SC. I have no reason to doubt that she had not given appropriate and sufficient consideration of these matters. 223. Ms Draycott SC criticized the Principal Magistrate for having wrongly applied case authorities. 224. The Principal Magistrate mentioned the following cases and stated the main concern which she thought existed in the particular case: (1) AG v Chan Chi Yin:[126] culpability of an offender from a good background may be regarded as the greater particularly if he is a person to whom others look for an example; (2) AG v Tai Chin Wah:[127] professional people have a duty to uphold public standards as well as the standards of their profession; (3) R v Kwok Yee, William:[128] a breach of trust by a person in authority may require a deterrent sentence; (4) HKSAR v Leung Chun-man:[129] for a public servant who was responsible for the enforcement of law and order in Hong Kong, his involvement with these offices (corruption offences) are clearly aggravating factors; (5) HKSAR v Hui Man-tai:[130] police officers who break the laws which they are entrusted to uphold have to be made examples in terms of deterrent sentencing so that others will not be tempted to follow along similar lines and so that public confidence will be maintained; and (6) HKSAR v Yuen Ka-kui, Chris:[131] it is an aggravating factor for a serving police officer to commit serious offences while acting in the course of his duty. 225. Ms Draycott SC criticized that the Principal Magistrate had wrongly applied at least a case which involved breach of trust, probably referring to the case of Kwok Yee, William. I agree that the present case does not involve a breach of trust in the usual sense. 226. The relevance and weight of each of the cases cited by the Principal Magistrate vary. Sentence in each case must be determined by consideration of the special circumstances and facts in that case. Whilst the Principal Magistrate mentioned these cases, it is not apparent that she followed the level of sentence imposed in them. It requires no case precedent to appreciate that the community has high expectation on high ranking law enforcement officer. 227. The Principal Magistrate demonstrated she had considered the whole of the relevant circumstances. The question is whether the ultimate sentence she imposed is wrong in principle or manifestly excessive. 228. I agree with the Principal Magistrate that this case should not be viewed lightly. The court must be very cautious in deciding to impose a deterrent sentence on a person who hitherto was of impeccable character and indeed has shown to have significant contribution to the community. On the other hand, the fact that the Appellant committed the offence in the capacity of a very senior police officer and in the course of duty, and the offence was in the form of applying unlawful force on a citizen in circumstances which did not call for any use of force on him tilt the balance of consideration. 229. There is substantial element of public interest in the present case. 230. In R v Howells,[132] Lord Bingham CJ observed: “Courts should always bear in mind that criminal sentences are in almost every case intended to protect the public, whether by punishing the offender or reforming him, or deterring him and others, or all of these things. Courts cannot and should not be unmindful of the important public dimension of criminal sentencing and the importance of maintaining public confidence in the sentencing system.” 231. In all the circumstances, I am of the view that it is not wrong in principle to impose an immediate custodial sentence. 232. As regards adoption of four months as the starting point, taking into account that it was just one strike which did not result in serious injury, it is no doubt a heavy penalty. However, taking into account the whole circumstances, I am of the view that it is within the reasonable range the court may consider. 233. I am also of the view that the discount of one month is sufficient in the circumstances of the case. 234. The Principal Magistrate had considered the option of CSO and decided that it is not appropriate. She was of the view that “the nature of the present offence and the circumstances of the offence and the Appellant gave rise to a culpability too serious for a [CSO] to be imposed”.[133] 235. CSO is not a soft option to imprisonment. It stands as a sentencing option in its own right, and serves the dual purpose of punishment and rehabilitation. In SJ v Li Cheuk Ming,[134] Chan CJHC (as Chan PJ then was) held that CSO was not to be excluded as a sentencing option under exceptional circumstances even where the offences were serious. 236. Judging the Appellant as a person, he should properly be regarded as a suitable candidate for CSO. In considering whether CSO should be imposed, the court again has to turn to the four sentencing principles and engage in a weighing exercise having regard to the particular circumstances of the case. 237. The Principal Magistrate rightly observed that “a police officer has the duty to prevent the commission of crime. When he himself commits a crime when acting as a police officer, this must be viewed seriously by the court. … PW1 was unarmed and defenceless and had exhibited no hostility towards the Appellant or his colleagues to warrant his being assaulted as he was by the Appellant.”[135] 238. Besides, the Appellant did not show he has genuine remorse. He did not plead guilty. He put up an assertion of belief which was rejected by the court. I bear in mind that genuine remorse is one of the six factors mentioned in R v Brown[136]for consideration whether community service order is appropriate but subsequent cases have established that not all the factors have to be present and consideration needs not be restricted to the six factors. That said, in the recent Court of Appeal case of SJ v Wong Chi Fung,[137] the importance of genuine remorse was still stressed. 239. Having considered the whole circumstances, my judgment is that the Principal Magistrate had achieved the right balance and did not err in refusing to impose a CSO. 240. The Principal Magistrate also refused to suspend the sentence. 241. Ms Draycott SC submitted it was wrong and in particular criticized the Principal Magistrate for having applied the wrong legal test. 242. In relation to the criticism regarding the legal test, Ms Draycott SC quoted the Principal Magistrate for having considered “exceptional circumstances” whilst it had been held by the Court of Appeal in SJ v Wade[138] that it is not accurate to state “exceptional circumstances” as the criteria for suspending a sentence. Lunn VP said: “That proposition is valid in respect of certain offences only, stipulated to be such by the courts. When dealing with other offences, the courts must have regard to all the circumstances of the commission of that offence and that of the defendant in determining whether or not it is appropriate to exercise its power to suspend the operation of the sentence of imprisonment.”[139] 243. The offence in the present case is not one of those offences stipulated by the court as requiring “exceptional circumstances” to justify suspending a sentence. 244. However, as submitted by Mr Marash SC, although the Principal Magistrate did mention lack of “exceptional circumstances” at a later stage, the main reason she gave for not suspending the sentence was: “Having regard to the circumstances of the offence and of the Appellant, in my view, the circumstances of the assault was too serious to warrant the imposition of a suspended sentence.”[140] 245. I agree to the observation of Mr Marash SC that the Principal Magistrate had considered the matter by applying the proper test and only after she had found the offence too serious to suspend the sentence did she go on further to consider whether exceptional circumstances existed so that suspending of the sentence was justified. 246. The question ultimately is whether the decision not to suspend is correct. Having considered the whole of the relevant circumstances, in my judgment it is correct. 247. I cannot accept that it is correct to order conditional discharge in the circumstances no matter how much sympathy I have on the Appellant. 248. An absolute discharge or conditional discharge may be ordered upon conviction. I have such a power pursuant to sections 119(d) and 36(1) of the Magistrates Ordinance. Section 36(1) is the primary provision which stipulates: “Where … the magistrate thinks that the charge is proved but is of the opinion that, having regard to the character, antecedents, age, health or mental condition of the person charged or to the trivial nature of the offence or to the extenuating circumstances under which the offence was committed, it is inexpedient to inflict any punishment or any other than a nominal punishment …” 249. The offence cannot be said to be of trivial nature. There cannot be said to be no moral culpability. It might have been a momentary lapse and out of character but the conviction means that it was an intentional use of unlawful force by a senior police officer. I do not consider the circumstances surrounding the case and the mitigating factors available override the public interest in consideration of the proper sentence in the present case so that a conditional discharge should be ordered. I form my view having regard to the fact that the Appellant had already spent 16 days in custody. 250. For the reasons I have mentioned, I do not find the sentence imposed wrong in principle or manifestly excessive. I therefore also dismiss the appeal against sentence. Ms Charlotte Draycott SC, Mr Peter Pannu and Mr Benson Tsoi, instructed by Chong & Partners LLP, for the appellant Mr Daniel Marash SC and Ms Annie Li, PP, of the Department of Justice, for the respondent [1] Contrary to common law and punishable under section 39 of the Offences Against the Person Ordinance, Cap 212, Laws of Hong Kong. [2] At trial, the Appellant was represented by Mr Peter Pannu. [3] Ms Bina Chainrai. [4] Cap 221 of the Laws of Hong Kong. [5] Exhibits P1 and P2 were also copied into the DVD exhibit P3. [6] Page 764 R-T in the Appeal Bundle. [7] Information taken from paragraphs 58 ‑ 67 of the Statement of Findings. [8] Cap 232, Laws of Hong Kong. [9] Cap 221, Laws of Hong Kong. [10] More detailed account of this aspect of his evidence will be set out later when necessary. [11] See paragraph 86 in the Statement of Findings. [12] Paragraph 113 in the Statement of Findings. [13] See paragraph 80 in the Statement of Findings. [14] Paragraph 98 in the Statement of Findings. [15] Paragraph 102 in the Statement of Findings. [16] Paragraph 104 in the Statement of Findings. [17] Paragraph 105 in the Statement of Findings. [18] Paragraphs 105 & 106 in the Statement of Findings. [19] Paragraph 115 in the Statement of Findings. [20] Cap 245, Laws of Hong Kong. [21] Paragraph 114 in the Statement of Findings. [22] Ms Draycott, SC, did not represent the Appellant at the trial, only Mr Pannu did. [23] (2010) 13 HKCFAR 20. [24] Mr Marash SC, appeared together with Ms Annie Li, PP. [25] Cap 227, Laws of Hong Kong. [26] Cap 484, Laws of Hong Kong. [27] [1988] HKC 28, at 35B. [28] CACC 105/2016. [29] [1980] HKLR 433, at 439. [30] (2013) 16 HKCFAR 609, paragraph 52. [31] [1990] NI306, at 344. [32] [2003] 3 HKLRD 751. [33] (1993) Vic Rp 64. [34] Paragraph 53 in Yeung Ka Ho (2013) 16 HKCFAR 609. [35] A more detailed gist of her submission can be seen in paragraph 79 below. [36] FACC 6/2014. [37] (1984) 78 Cr App R 276, at 281. [38] See paragraph 95 below. [39] (1983) 77 Cr App R 225, [1983] 1 WLR 1118. [40] See paragraphs 79 ‑ 81 in the Statement of Findings. [41] [1971] AC 814, at 831. [42] Mr Peter Pannu, not Ms Draycott SC. [43] His evidence in relation to what happened at the time and place of the incident can be seen from page 841 of the Appeal Bundle. [44] Paragraph 66 in the Statement of Findings. [45] Page 848 in the Appeal Bundle. [46] Page 849 in the Appeal Bundle. [47] Page 851 in the Appeal Bundle. [48] Page 852 in the Appeal Bundle. [49] Page 853 in the Appeal Bundle. [50] Page 854 in the Appeal Bundle. [51] Page 857 in the Appeal Bundle. [52] Page 860 in the Appeal Bundle. [53] Page 861 in the Appeal Bundle. [54] Page 862 in the Appeal Bundle. [55] Page 863 in the Appeal Bundle. [56] Paragraphs 111 and 113 in the Statement of Findings. [57] Paragraph 6 of the Statement of Findings. [58] Paragraphs 86 - 93 in the Statement of Findings. [59] Paragraphs 56 - 67 and 105 of the Statement of Findings. [60] Paragraphs 101 - 102 and paragraph 113 in the Statement of Findings. [61] Paragraph 80 in the Statement of Findings. [62] Paragraph 69 of the Statement of Findings. [63] Paragraphs 73 ‑ 81 of the Statement of Findings. [64] Paragraph 77 in the Statement of Findings. [65] Paragraph 105 in the Statement of Findings. [66] Paragraph 105 in the Statement of Findings. [67] Paragraph 80 in the Statement of Findings. [68] Paragraph 100 in the Statement of Findings. [69] Paragraph 105 in the Statement of Findings. [70] Paragraph 110 in the Statement of Findings. [71] See paragraph 95 above. [72] Paragraphs 74 and 78 in the Statement of Findings. [73] [1996] 2 Cr App R 128. [74] (1994) 98 Cr App R 290, at 296. [75] [1996] 2 Cr App R 128. [76] [1984] 78 Cr App R 276. [77] Paragraph 110 in the Statement of Findings. [78] Paragraph 105 of the Statement of Findings. [79] Paragraph 110 in the Statement of Findings. [80] Paragraph 116 in the Statement of Findings. [81] Paragraph 112 in the Statement of Findings. [82] Paragraph 80 in the Statement of Findings. [83] Paragraph 87 in the Statement of Findings. [84] Paragraph 78 in the Statement of Findings. [85] Paragraph 79 in the Statement of Findings. [86] Paragraph 80 in the Statement of Findings. [87] Paragraphs 83 and 84 in the Statement of Findings. [88] Paragraph 85 in the Statement of Findings. [89] See paragraph 137 above. [90] PW2. [91] Paragraph 91 in the Statement of Findings. [92] Paragraph 92 in the Statement of Findings. [93] Paragraph 95 in the Statement of Findings. [94] Paragraph 87 in the Statement of Findings. [95] Paragraphs 95 and 96 in the Statement of Findings. [96] Paragraph 82 in the Statement of Findings. [97] Exhibit P13(2A). [98] (2005) 8 HKCFAR 70. [99] See paragraph 42 in the medical report, Exhibit P14, pages 130 and 303 in the Appeal Bundle. [100] See transcript, page 764-S. [101] Photographs 44, 90 and 91, Exhibit P13(2A), page 130 in the Appeal Bundle. [102] Page 758 P in the Appeal Bundle. [103] Paragraph 20-192 of Archbold, 2018 edition. [104] HCMA 914/1988. [105] Paragraph 100 in the Statement of Findings. [106] [1934] 2 KB 498. [107] Cap 245, Laws of Hong Kong. [108] (1994) 98 Cr App R 290, at 295. [109] [1996] 2 Cr App R 128, at 132. [110] Paragraph 110 in the Statement of Findings. [111] Paragraph 110 in the Statement of Findings. [112] See paragraph 170 above. [113] See paragraphs 118 ‑ 122 in the Statement of Findings. [114] See page 158 in the Appeal Bundle. [115] Paragraph 121 in the Statement of Findings. [116] CACC 317/2012. [117] Paragraph 127 in the Statement of Findings. [118] Paragraphs 127 ‑ 131 in the Statement of Findings. [119] Paragraphs 132 and 133 in the Statement of Findings. [120] Paragraph 134 in the Statement of Findings. [121] Pages 137 and 147 in the Appeal Bundle. [122] [2018] 2 HKLRD 699. [123] Paragraph 116 in the Statement of Findings. [124] (1996) 105 CCC (3d), 327, at 370. [125] (1974) 60 Cr App R 74, at 77. [126] [1988] HKC 44. [127] [1994] 2 HKCLR 81. [128] CACC 452/1993. [129] CACC 156/2003. [130] CACC 334/2007. [131] [2013] 5 HKC 64. [132] [1999] 1 All ER 50, at 54. [133] Paragraph 133 in the Statement of Findings. [134] [1999] 1 HKLRD 63, at 65. [135] Paragraphs 133 and 134 in the Statement of Findings. [136] (1981) 3 Cr App R (S) 294, at 295. [137] [2018] 2 HKLRD 699. [138] [2016] 3 HKC 274. [139] Paragraph 46 in the judgment. [140] Paragraph 135 in the Statement of Findings. 
Chief Justice Ma: A INTRODUCTION 1. The present appeals involve principally the determination of two important issues regarding challenges in elections for the Chief Executive under the Chief Executive Election Ordinance Cap 569 (“the CEEO”): first, the scope of election petitions under Part 6 of that Ordinance and their relationship to judicial review and other proceedings; secondly, the constitutionality of the absolute seven day time limit for lodging election petitions contained in s 34(1) of the CEEO. The first issue is predominately an exercise in statutory interpretation, the second is a constitutional issue involving the right of access to the courts under Article 35 of the Basic Law. This appeal also concerns an order for costs made in the course of the present proceedings and I shall identify that issue in due course. A.1 The procedural history 2. In the elections[1] for the Chief Executive held on 25 March 2012, Mr Leung Chun Ying (“Mr Leung”) was the returned candidate and he was publicly declared and gazetted[2] as such by the Returning Officer on 25 March 2012. The other candidates in the elections were Mr Ho Chun Yan Albert (“Mr Ho”) and Mr Henry Tang. 3. Following the election of Mr Leung, on 4 July 2012 Mr Ho lodged an election petition, followed the next day by a notice of application for leave to apply for judicial review[3], putting in issue whether Mr Leung was duly elected. The factual basis for both proceedings was the same, namely it was alleged that Mr Leung had made false or misleading statements in the course of the elections, this amounting to illegal conduct within the meaning of the Elections (Corrupt and Illegal Conduct) Ordinance Cap 554 (“ECICO”)[4] and also that this conduct amounted to his not being a “person of integrity, dedicated to his or her duties” for the purposes of Article 47(1) of the Basic Law. For reasons which will become apparent presently, it is unnecessary to elaborate on these factual aspects; it is sufficient merely to say that the allegations concerned certain unauthorized building works at the home of Mr Leung. 4. It should also be noted that on 5 July 2012, a notice of application for leave to apply for judicial review was also issued by Mr Leung Kwok Hung[5] challenging Mr Leung’s election as Chief Executive on substantially the same grounds as Mr Ho’s challenge. A.2 The judgments of Lam JA 5. The election petition and the applications for judicial review were all dealt with by Lam JA, and it is from the first three judgments that the present appeals before us emanate (but there is also a fourth judgment dated 5 October 2012, the relevance of which will become apparent below):- (1) By a judgment handed down on 30 July 2012 in the applications for leave to apply for judicial review (of Mr Ho and Mr Leung Kwok Hung)[6], Lam J[7] refused leave on the following grounds:- (a) As far as Mr Ho was concerned, it was regarded as an abuse for him to have commenced judicial review proceedings while at the same time pursuing election petition proceedings. (b) In any event, the grounds set out in s 32(1)(a) of the CEEO enabling an election to be questioned, were not available to an applicant in judicial review proceedings which sought to question the election of a candidate in Chief Executive elections. This applied to both Mr Ho and Mr Leung Kwok Hung’s challenges. (c) The challenge based on Article 47 of the Basic Law was unsustainable.[8] (2) By a judgment handed down on the 12 September 2012, Lam JA[9] considered the election petition[10] in the context of a strike-out application brought by Mr Leung on the basis that the election petition was time-barred[11] and also that in any event, on the merits, the election petition was bound to fail. On the issue of the merits, Lam JA held at that stage that Mr Ho’s case on the facts was arguable and so declined to strike-out on this basis. On the time bar, although the election petition was admittedly lodged out of time, Mr Ho contended that the seven day time limit prescribed under s 34(1) was unconstitutional as denying him access to the courts, a protected right under Article 35 of Basic Law. On this issue, the Judge agreed with Mr Ho and held that the seven day time limit contained in s 34(1) of the CEEO, if it was an absolute one, was unconstitutional. He was persuaded, however, that the correct remedy was, rather than to strike down the provision, instead to effect a remedial interpretation[12] whereby the provision was subject to the court’s discretion to extend time.[13] (3) On 28 September 2012, a third judgment was handed down dealing with the question of the costs of the hearing relating to the application for leave to institute judicial review proceedings.[14] The Judge held that there were exceptional circumstances to depart from the usual order for costs in such hearings, where the putative respondent has appeared to resist leave being granted [15]; accordingly, Mr Ho and Mr Leung Kwok Hung were ordered each to pay half of the costs of Mr Leung. No order was made regarding the costs of the Secretary for Justice (who was an intervener)[16]. This order made as to costs forms the subject matter of the third issue before this court. (4) On the 25 September 2012, Lam JA dealt with the application (which was made by Mr Ho) for an extension of time to lodge the election petition. By a judgment dated 5 October 2012, this was refused on the basis that the complaints made by him against Mr Leung did not have any real prospect of success. 6. Leave to appeal to the Court of Final Appeal was sought by all parties who were unsuccessful in the various applications I have referred to.[17] The applications for leave to appeal made by Mr Ho, Mr Leung Kwok Hung, Mr Leung and the Secretary for Justice (eight in all) were dealt with by the Appeal Committee[18] on 9 November 2012. In the Determination dated 13 November 2012, the Appeal Committee considered the question whether at the end of the day, even if the various jurisdictional issues raised in the applications were resolved in favour of Mr Ho and Mr Leung Kwok Hung, the allegations of false and misleading statements made by Mr Leung in relation to the unauthorized building works at his home were sufficiently arguable. The Appeal Committee concluded that no reasonably arguable grounds existed for appealing Lam JA’s decision dated 5 October 2012.[19] 7. The conclusion that the factual allegations raised were unarguable was sufficient to dispose of the challenges made by Mr Ho and Mr Leung Kwok Hung and thus rendered academic the other issues for which leave to appeal was sought. However, given the importance of some of the issues that were raised, the Appeal Committee decided that a sufficiently great public interest existed to grant leave even though the issues were, strictly speaking, academic.[20] Leave was accordingly granted on the following two issues:- (1) Under the CEEO, how do challenges to a CE election pursuant to the election petition procedure in section 32 relate to challenges pursuant to the judicial review procedure in section 39? (2) Does the seven-day time limit laid down by CEEO section 34 involve any infringement of the right of access to a court guaranteed by Article 35 of the Basic Law, and if so, is such time limit unconstitutional? Leave to appeal was also given on the “or otherwise” ground to Mr Ho and Mr Leung Kwok Hung in relation to the costs order made by Lam JA on 28 September 2012.[21] This would seem to follow if leave to appeal was given in relation to the two main issues. A.3 The parties before the court 8. It was indicated at the resolution of the applications for leave that given the academic nature of the appeals to this Court as concerned them, it was open to Mr Ho and Mr Leung Kwok Hung to choose not to appear at the hearing of the present appeals. In the event, Mr Ho[22], Mr Leung[23] and the Secretary for Justice[24] are before us. Mr Leung Kwok Hung does not appear, not having proceeded with his appeal. We are grateful to all counsel for their assistance. 9. Before dealing with the three issues before us, I should first set out the relevant statutory provisions. B THE RELEVANT PROVISIONS IN THE CEEO 10. The two main issues in these appeals involve the construction of relevant provisions in the CEEO. It is convenient to set out the following provisions in full (I will of course be referring to other provisions as well in the course of this judgment):- (1) Under Part 5 (headed “Elections and Polling”), s 29 states:- “A person declared under section 28 as elected at an election is presumed to be duly elected until he is ruled by the Court or the Court of Final Appeal pursuant to the determination of an election petition or otherwise as not duly elected.” (2) Under Part 6 (headed “Election Petitions”), ss 32, 33, 34, 38 and 39 are of particular relevance:- “32. Election may be questioned only by election petition made on specified grounds (1) An election may be questioned only by an election petition on the ground that – (a) the person declared by the Returning Officer under section 28 as elected was not duly elected because – (i) he was not eligible to be nominated as a candidate under section 13; (ii) he was disqualified under section 14 from being nominated as a candidate; (iii) he should have been disqualified under section 20(1) from being elected but was not so disqualified; (iv) he engaged in corrupt conduct or illegal conduct at the election; (v) another person engaged in corrupt conduct or illegal conduct in respect of him at the election in connection with his candidature; (vi) corrupt conduct or illegal conduct was generally prevalent at the election; or (vii) material irregularity occurred in relation to – (A) the election; (B) the poll at the election; or (C) the counting of votes in respect of the election; or (b) the candidate declared by the Returning Officer under section 22(1AB)(c) as not returned at the election is not returned because material irregularity occurred in relation to – (i) the election; (ii) the poll at the election; or (iii) the counting of votes in respect of the election. (2) In this section – corrupt conduct (舞弊行為) means corrupt conduct within the meaning of Part 2 of the Elections (Corrupt and Illegal Conduct) Ordinance (Cap. 554); election (選舉) includes nomination proceedings and the decisions of the Returning Officer or any Assistant Returning Officer; illegal conduct (非法行為) means illegal conduct within the meaning of Part 3 of the Elections (Corrupt and Illegal Conduct) Ordinance (Cap. 554). 33. Who may lodge election petition (1) An election petition – (a) may be lodged by any candidate in the election; or (b) may be lodged by – (i) a person who was determined under section 17 to be not validly nominated; (ii) a person the nomination of whom was not accepted by the Returning Officer; or (iii) a person who was disqualified under section 20(1) from being elected, provided that the election petition is, subject to subsection (4), subscribed to by not less than 10 members of the Election Committee in the manner specified in subsection (2). ….. 34. Period within which election petition and appeal must be lodged (1) An election petition questioning an election must be lodged within 7 working days after the day on which the result of the election is declared under section 22(1AB) or 28. ….. 38. Acts of person not invalid if ruled not to be elected A – (a) determination of the Court under section 37(1); or (b) ruling by the Court or the Court of Final Appeal, that a person who was originally declared as elected at an election was not duly elected does not invalidate acts purporting to have been done by the person as the Chief Executive before the determination or ruling, as the case may be. 39. Time limit for legal challenges (1) Notwithstanding any provision in the High Court Ordinance (Cap. 4), no – (a) application for leave to apply for judicial review under section 21K of that Ordinance; or (b) other proceedings; which put in issue – (c) whether a candidate is duly determined to be not returned at an election under section 26A(4); or (d) whether the candidate declared under section 28 as elected at an election can lawfully assume the office of the Chief Executive, shall be made or commenced more than 30 days after the publication of the declaration under section 22(1AB)(d) or the publication of the result of the election under section 28 unless the leave of the Court has been obtained. (2) The Court may upon application grant the leave to make an application for leave to apply for judicial review or commence proceedings after the expiry of the 30 days referred to in subsection (1) if it is satisfied that – (a) the person making the first-mentioned application has used his best endeavours to make the second-mentioned application or commence the proceedings within the 30 days; and (b) granting the leave applied for is in the interest of justice.” C. FIRST ISSUE: CHALLENGES IN A CHIEF EXECUTIVE ELECTION MADE BY ELECTION PETITION AND IN JUDICIAL REVIEW PROCEEDINGS 11. The issue as framed by the Appeal Committee[25] invite, first, a consideration of the election petition procedure and its ambit; and secondly, a consideration of the relationship between that procedure and the judicial review procedure envisaged under s 39 of the Ordinance.[26] It will be recalled that in the 30 July 2012 judgment[27], Lam J concluded that if any of the grounds set out in s 32(1)(a) of the CEEO was employed to challenge the election of someone as Chief Executive, this was available only in election petition proceedings and therefore unavailable as a ground of challenge in judicial review proceedings. For the reasons that follow, I am of the view that the Judge erred in this conclusion. In dealing with the First Issue, I recognize that it involves an exercise in statutory construction, the relevant provisions being those set out in Section B above. C.1 The Court’s approach to statutory interpretation: context and purpose 12. As has been reiterated recently by this Court[28], the proper starting point in statutory interpretation, as well as constitutional and contractual interpretation, is to look at the relevant words or provisions having regard to their context and purpose. 13. In the present case, the context and purpose of the relevant provisions earlier set out are principally these:- (1) We are concerned with elections, specifically with elections for the most important official post in Hong Kong, namely, that of the Chief Executive. Article 45 of the Basic Law states that the Chief Executive shall be selected by election in Hong Kong and that he or she shall be appointed by the Central People’s Government. Annex I to the Basic Law sets out the “Method for the Selection of the Chief Executive”, making reference to the election of the Chief Executive by an Election Committee comprising 1,200 members[29], intended to be broadly representative of Hong Kong people. Although the election of the Chief Executive is not by universal suffrage, his or her election is obviously of great importance and interest to Hong Kong people. (2) It follows from this that the election for the post of Chief Executive is of considerable importance in Hong Kong, thus emphasizing the need for such elections to be genuine, open, honest and fair; in short, such elections must, as far as Hong Kong residents are concerned, have integrity so as to ensure that those elected are truly representative of those whom he or she represents.[30] The existence of a satisfactory mechanism to question election results comes within this rubric. (3) Given the obvious importance of the post of Chief Executive, who fulfils the role of being the head of the HKSAR, represents the Region[31] and leads the Hong Kong Government[32], it goes without saying that there is a necessity to have certainty in the elections for the post of Chief Executive. In the context of the present appeals, where any doubts exist as to whether a person has been duly elected as Chief Executive, the sooner such doubts are resolved, the better. This Court has, in the context of elections to the Legislative Council, emphasized the need to have matters speedily determined: see Mok Charles v Tam Wai Ho[33]. C.2 The Scheme of Election Petitions and Judicial Review Proceedings under the CEEO 14. With this approach in mind, I now deal with the relevant provisions in the CEEO. 15. It is notable first that the provisions in the Ordinance regarding election petitions are very detailed[34]:- (1) Seven specific grounds are enumerated to enable challenges to be made where there have been contested elections. Where a single candidate election takes place[35], the ground stipulated in s 32(1)(b) allowing a challenge to be made against a declaration by the Returning Officer that a single candidate is not returned, is material irregularity. These grounds are presumably the most likely and serious grounds for an election result to be challenged. (2) Section 33 identifies the category of persons able to lodge an election petition. This is not restricted to candidates in the election and include persons who were not nominated, whose nomination was not accepted by the Returning Officer or who were disqualified. I shall for convenience refer to those persons identified in s 33 as the “s 33 persons”. (3) Section 34 sets out a rigorous time limit for the lodging of an election petition questioning an election result: 7 days after the declaration of an election result under s 22(1AB)[36] or s 28[37] of the CEEO. The constitutionality of the seven day limit is of course the subject matter of the second of the two main issues before this Court. (4) Sections 35 to 37 of the Ordinance also contain detailed provisions regarding who should be the respondent in election petitions, the jurisdiction of the court to determine election petitions and how the court is to determine here such petitions. 16. Given this elaborate structure in relation to election petitions, the intention must have been for this procedure to be the primary and most speedy means of enabling challenges to be made questioning Chief Executive elections:- (1) Primary because the s 33 persons will likely be that class of persons most affected by an adverse election result and therefore those most likely to take action questioning an election result. (2) That speed was of the essence can be seen from the very limited time allowed for election petitions to be lodged under s 34(1) of the Ordinance. 17. Thus far, there is perhaps little controversy but the two important questions that remain to be answered involve a discussion of, first, how exclusive the election petition procedure is and secondly, its relationship with other proceedings relevant to the questioning of elections (here the discussion will center on the judicial review procedure). As will be seen, these questions are connected. 18. On the aspect of exclusivity, although it is not in dispute between the parties that only s 33 persons may lodge election petitions (to the exclusion of all other persons), three facets need to be considered:- (1) Can an election be questioned only by an election petition to the exclusion of any other type of proceedings? (2) Can an election be questioned only on those grounds set out in s 32(1)(a) and (b) of the Ordinance? (3) Are the s 33 persons restricted only to the election petition procedure in questioning a Chief Executive election result such that, for instance, the judicial review proceedings envisaged under s 39 of the Ordinance are not open to this class? 19. Common to the consideration of these three facets is the need to construe the effect of the word “only” contained in s 32(1) of the CEEO[38]. The word “只” also appears in the Chinese text of that provision. 20. If one looked in isolation at the language of s 32 of the Ordinance without considering context and purpose or the other provisions in the CEEO, there might be some justification to conclude that an election petition was the only means by which an election could be questioned or[39] that the only grounds to challenge an election were those set out in ss 32(1)(a) and (b). However, in my view, this is clearly not the position. My conclusion on the first two facets is that (i) the election petition is not the only means of challenging an election and (ii) the grounds set out in ss 32(1)(a) and (b) are not the only grounds based on which a challenge can be made to Chief Executive elections:- (1) As to the first facet, the wording of s 39 of the Ordinance pre-supposes the availability of judicial review and other proceedings putting in issue whether a candidate who is declared to be elected “can lawfully assume the office of the Chief Executive.” It was argued on behalf of Mr Leung and the Secretary for Justice that this wording was apt only to cover post-election matters.[40] I disagree that these words should be so restricted. Nothing in the Ordinance militates towards this construction. Quite the contrary: ss 29 and 38[41] expressly envisage the situation in which challenges can be made as to whether a person was “duly elected” in proceedings other than election petition proceedings. The challenge to whether a person “can lawfully assume the office of the Chief Executive” can obviously include grounds other than the question of whether a person has been duly elected, but must in my view include that ground as well. A person who is not duly elected within the meaning of the Ordinance cannot surely be a person who can lawfully assume the office of Chief Executive. (2) As to the second facet, ss 32(1)(a) and (b) set out the grounds to question an election under the election petition procedure. These grounds relate only to election petition proceedings. Nothing is said about the grounds which may be available in judicial review or other proceedings envisaged under s 39 of the Ordinance to question whether a candidate can lawfully assume the office of Chief Executive (this as we have just seen including the right to question whether a person has been duly elected). Even in the case of the single candidate, while he or she may contest the declaration of non-return on the ground of material irregularity in an election petition[42], in the other proceedings envisaged under s 39, this non-return may be challenged by other people and this is not necessarily restricted to challenges based on material irregularity.[43] 21. Given the conclusion reached above on the first two facets, what of the remaining question whether the s 33 persons are confined to the election petition procedure in challenging elections? Mr Ho’s position is simply put: s 33 persons like Mr Ho have available to them in questioning an election both the election petition procedure as well as the benefit of other procedures (such as those envisaged under s 39); and further, that the grounds set out in s 32(1)(a) and (b) are also available to be utilized whether under the election petition procedure or in judicial review proceedings. In answer to the obvious objection as to why there should be available to s 33 persons both the unique and elaborate election petition procedure as well as judicial review (and other) proceedings based on the same grounds, Mr Lee SC pointed out that the two types of proceedings were different: if judicial review proceedings were sought to be instituted, leave was required and in obtaining leave, an intended applicant would have to demonstrate to the court why the election petition procedure was not utilized. Lam J had been of the view that judicial review and other proceedings based on the s 32(1)(a) grounds[44] would be unavailable to persons like Mr Ho (in other words, the s 33 persons) with the consequence that Mr Ho’s application for leave to institute judicial review was dismissed. 22. In my view, the Judge was right in his conclusion that judicial review (or other proceedings) were not open to Mr Ho – and therefore also unavailable to all s 33 persons – if the same grounds as set out in s 32(1)(1)(a) were used. This would also be the consequence in relation to single candidate elections in relation to the grounds set out in s 32(1)(b). The effect of ss 32 and 33 may be stated as follows:- (1) These two sections, which deal with election petitions, must be read together in order to ascertain just who can lodge election petitions and on what grounds. (2) Section 32(1) begins by stating that an election may be “questioned”. But questioned by whom? Section 33 provides the answer by identifying the s 33 persons. (3) The word “only” in s 32(1) is important and effect must be given to it. In my view, it makes clear that where an election is “questioned” by someone within the class of s 33 persons on the grounds set out in ss 32(1)(a) or (b), this can only be done by election petition. (4) Apart from the above being the proper construction of these two provisions, it also makes good sense. The election petition procedure is a unique procedure[45] open only to s 33 persons. This class of persons is, as stated earlier, that class seen to be most likely to take action to question an election. This class, although bound by a strict time limit in the institution of proceedings[46], has the considerable benefit of being able to lodge an election petition as of right, without the need to seek leave. This is not the position in judicial review proceedings where leave to apply for judicial review is required.[47] In return for this right to institute election petition proceedings as of right, the s 33 persons are, as I have said, bound by a strict time limit and also must be left with this form of proceedings as the only means of questioning an election if any of the grounds set out in ss 32(1)(a) and (b) are relied on. It makes no sense for the judicial review procedure also to be available to s 33 persons on the same grounds. It makes pointless the elaborate procedure for election petitions carefully and extensively set out in the Ordinance. The object and purpose of these provisions relating to election petitions, being the need for matters to be resolved quickly[48], would be defeated. (5) In the course of his submissions, Mr Lee SC referred the Court to extrinsic materials (mainly in the form of debates when the draft bill which became the CEEO dealing with election challenges was discussed in the Legislative Council) which refer to the availability of judicial review. I have not found these materials useful. Apart from anything else, they do not deal with the issues and facets which this Court has to consider. 23. The above analysis disposes of the first two facets set out in para 18 above, but leaves open the question of the relationship of election petition proceedings to other proceedings in relation to the questioning of elections. This is the second main question under the First Issue, to which I now turn. 24. As stated earlier[49], in considering the ambit of the other proceedings referred to in s 39 of the CEEO, the parties have focused on the position of the judicial review proceedings rather than any other proceedings. I shall likewise deal with only the position of judicial review. The reference to judicial review in s 39 assumes that such proceedings can exist where they put in issue the matters specified in ss 39(1)(c) and (d)[50]. As the parties accepted, s 39 does not found the jurisdiction to institute these other types of proceedings; it merely assumes their existence. For judicial review proceedings, the foundation of the jurisdiction is s 21K of the High Court Ordinance Cap 4. Section 39 merely states that the usual three month time limit for judicial review proceedings[51] is reduced to 30 days after the date of the publication of the declaration of the election result under s 22(1AB)(d) or s 28 of the CEEO. 25. It would be neither desirable nor appropriate to embark on a general discussion of the ambit of judicial review in the context of challenges made to election results. Relevant to the present case, however, is an analysis of the relationship between judicial review and election petition proceedings. The Judge held (as far as Mr Ho was concerned) that judicial review was unavailable to him to make a challenge based on the grounds specified in s 32(1)(a)[52]. He also held (in the case of Mr Leung Kwok Hung) that judicial review proceedings were unavailable if a s 32(1)(a) ground was relied on. Lam J’s view was that where s 32(1)(a) grounds were relied on, the only proceedings where this could take place were election petition proceedings. Judicial review was therefore excluded if a s 32(1)(a) ground was relied on. 26. For my part, I would respectfully disagree with the Judge’s views on the relationship between election petition proceedings and judicial review proceedings. I have concluded earlier that the effect of ss 32 and 33 of the CEEO is that where s 33 persons wish to question an election on any of the grounds set out in ss 32(1)(a) or (b), they can only do so by an election petition and no other proceedings are available to them if such grounds are relied on.[53] It does not follow from this construction of those two provisions, however, that persons other than s 33 persons are somehow automatically excluded from claiming that an elected person should not be permitted lawfully to assume the office of Chief Executive on the basis that he or she was not properly elected by reason of one or more of the grounds set out in s 32(1)(a) or (b). Nothing in the CEEO automatically excludes the right of non s 33 persons from so relying on the grounds set out in s 32(1)(a) or (b). I refer back to the earlier discussion regarding the first and second facets. It seems odd automatically to exclude a non s 33 person from relying on a s 32(1)(a) or (b) ground. For example (as pointed out in argument by Mr Justice Gleeson NPJ), where corrupt conduct or illegal conduct was prevalent at an election[54] in circumstances where s 33 persons would obviously be reluctant to raise the issue, it seemed unsatisfactory if no one could then raise the point in judicial review proceedings. 27. Of course, whether or not a person will be able actually to rely on one or more of the grounds set at in ss 32(1)(a) and (b), or have the necessary locus standi, to found a claim for judicial review will depend on normal judicial review considerations. 28. In reaching this conclusion, I have not ignored the argument to the effect that to allow the judicial review procedure to be made available to non s 33 persons in this way might undermine the election petition procedure set out in the CEEO and thus potentially result in duplication and prolonging of proceedings concerning elections. After all, apart from anything else, it may be said that the time limit for instituting proceedings under s 39 is 30 days[55] (in contrast to the seven day limit for election petitions), and this may be extended as well.[56] In my view, it is important to highlight the following points in this context:- (1) It should be borne in mind that the primary and most speedy form of proceedings to question an election is the election petition.[57] This is likely in practice to be the most usual form of proceedings to challenge an election result. (2) However, the election petition procedure cannot be the only form of proceedings available to question an election. The content of those provisions in the Ordinance discussed earlier make this point. Judicial review is available, although the time for instituting such proceedings is reduced to 30 days from the usual three months. (3) The availability of judicial review as a fallback procedure to deal with those situations where, for whatever reason, election petition proceedings are not instituted, constitutes an additional guarantee to enable elections for the Chief Executive to have integrity and to be genuine, open, honest and fair. The availability of judicial review should not be cut down unless this is clearly stated and justified. (4) It does not follow judicial review proceedings will necessarily prolong the challenges that may be made regarding elections. While the time limited for the lodging of election petitions may be shorter, those proceedings can be instituted and pursued as of right (subject of course to any striking out applications). Judicial review proceedings, on the other hand, require leave before they can be properly instituted. 29. In respect of the ability of s 33 persons to institute judicial review proceedings (the third facet referred to in para 18 above), it does not follow from the conclusion that they cannot institute judicial review proceedings on a s 32(1)(a) or (b) ground that they cannot under any other circumstances institute judicial review proceedings at all putting in issue whether a person elected in an election can lawfully assume the office of Chief Executive. In my view, they are able to institute such judicial review proceedings, as long as the ground relied on is not one of the grounds set out in s 32(1)(a) and (b) and provided they satisfy the usual requirements in judicial review proceedings. This conclusion also follows from the view I have earlier reached in relation to the second facet.[58] C.3 Conclusion on the First Issue 30. The views I have reached under the First Issue differ from the learned judge. They are, however, based on the true construction of the relevant provisions of the CEEO set out in para 10 above, bearing in mind the context and purpose of those provisions. Nevertheless, in terms of the actual result, the Judge was right to refuse leave to Mr Ho to commence judicial review proceedings. However, as regards Mr Leung Kwok Hung (although he has not pursued his appeal), the learned judge ought not to have, at that stage, refused leave to institute judicial review proceedings on the basis that Mr Leung Kwok Hung was not entitled to rely on a s 32(1)(a) ground to found his application for judicial review. However, as the Appeal Committee held in its Determination dated 13 November 2012, the factual assertions made by Mr Leung were unsustainable as a matter of law.[59] 31. I now turn to the constitutional issue. D SECOND ISSUE: CONSTITUTIONALITY OF S 34(1) OF THE CEEO 32. The relevant provision here is s 34(1) of the CEEO.[60] The seven day limit for the lodging of election petitions is an absolute one. There is no provision allowing the seven day limit to be extended. This is to be contrasted with the 30 day limit contained in s 39 of the Ordinance, which can be extended by the Court.[61] Mr Ho has consistently contended that this provision, if it is absolute in nature, is unconstitutional in denying the constitutional right of access to the courts. He relies on article 35 of the Basic Law (under Chapter III: Fundamental Rights and Duties of the Residents) which states that “Hong Kong residents shall have the right to … access to the courts…… for timely protection of their lawful rights and interests … and to judicial remedies.” As seen above[62], Lam JA agreed with Mr Ho’s position, but instead of striking down the provision (which would have had the effect of eliminating any time limit for the lodging of election petitions), he applied a remedial interpretation to s 34(1) so as to subject the seven day limit to the Court’s discretion to extend time. 33. In this part of the appeals, the appellants are Mr Leung and the Secretary for Justice. D.1 Is Article 35 of the Basic Law engaged? 34. At one stage in his submissions, Mr Thomas SC (for the Secretary for Justice) seemed to suggest that Article 35 gave no right of access to the courts at all, but this is plainly not the case, whether upon a simple reading of that article – it actually states there is “the right to … access to the courts” – or as a matter of substance. In Stock Exchange of Hong Kong Limited v New World Development Company Limited[63], Ribeiro PJ stated this to be the essence of Article 35 of Basic Law:- “49. What is of prime relevance to this appeal is the second dimension of art.35. As appears from its language, art.35 is also concerned with entrenching the individual’s rights in relation to ‘the courts’: individuals are to have the right of ‘access to the courts’, the right of ‘choice of lawyers … for representation in the courts’, the right ‘to judicial remedies’ and ‘the right to institute legal proceedings in the courts against the acts of the executive authorities and their personnel’. 50. This is a crucial additional feature of the constitutional architecture of the Basic Law in relation to the judicial system of the Region. Article 35 ensures that the fundamental rights conferred by the Basic Law as well as the legal rights and obligations previously in force and carried through to apply in the HKSAR are enforceable by individuals and justiciable in the courts. It gives life and practical effect to the provisions which establish the courts as the institutions charged with exercising the independent judicial power in the Region. This dimension of art.35 is therefore concerned with ensuring access to the courts for such purposes, buttressed by provisions aimed at making such access effective.” In the context of elections, I have already remarked that an effective means of making challenges enables elections to have integrity and to be genuine, open, honest and fair.[64] This is buttressed by Article 35 of Basic Law which “gives life and practical effect” to these aspects. 35. Next, it was argued both by Mr Thomas SC and Mr Mok SC that Article 35 is simply not engaged in the present case. As I understand the argument, it proceeded along these lines: the right to question Chief Executive elections by election petition proceedings was introduced by those provisions contained in Part 5 of the CEEO, which included s 34(1); that provision was one of the provisions which merely defined the jurisdiction of the election petition procedure. Accordingly, seen in this light, this definition of the jurisdictional limits of the election petition procedure could not and did not engage any right of access to the courts, any more than say a definition of rights introduced by legislation. So the argument ran, the s 34(1) restriction had only to do with jurisdictional limits. 36. I must say that I have found it difficult to follow this argument. True it is that s 34(1) can be said in a way to define the jurisdiction of the right to institute election petition proceedings, but without doubt that provision also involves the question of access to the courts. In fact, it may be said that the provision is dealing precisely with the enforcement by persons of legal rights in the courts (the right to challenge elections by election petition). This, as the passage from Stock Exchange of Hong Kong Limited v New World Development Company Limited makes clear, is what Article 35 of the Basic Law ensures. 37. In support of their submissions, reliance was placed by Mr Leung and the Secretary for Justice on a number of authorities from Australia in which the courts have discussed time provisions[65], using language to suggest that such provisions only defined the limits of jurisdiction. It is unnecessary to refer to all these authorities; it suffices just to refer to the following:- (1) In Australian Iron and Steel Limited v Hoogland[66], Windeyer J referred to such limitation provisions as merely imposing a condition “which is of the essence of a new right”. (2) In David Grant and Company Pty Limited (Receiver Appointed) v Westpac Banking Corporation[67], Gummow J (in dealing with the time constraints regarding when an application could be made to set aside a statutory demand) said that the time condition was “an essential condition of the new right conferred by s 459G”. (3) In Rudolphy v Lightfoot[68], the High Court of Australia dealt with an election petition which had been lodged with the Court of Disputed Returns beyond the time that was permitted under statute (40 days). The Court viewed this time limitation as a condition which was part of the essence of the right (given to dispute elections).[69] (4) In Hocine v Minister for Immigration and Multicultural Affairs[70], French J (then in the Federal Court) had to consider whether the time limit for filing appeals to review the decision of the Refugee Review Tribunal was in excess of legislative power. An argument was raised along the lines that the time limit went against the concept that the courts should have the ability to exercise judicial power and that this power should be real and not illusory.[71] French J, after reviewing the three cases referred to above, concluded that the time limitation was a part of the definition of the right to seek a review[72] and that such a definition of the jurisdiction did not involve any direction to the court about the manner and conduct of its exercise of the jurisdiction.[73] 38. In the course of argument, it was pointed out by Mr Justice Gleeson NPJ, who it must be noted was the former Chief Justice of the High Court of Australia, that those cases were not concerned with any consideration of a constitutionally declaredright of access to the courts. There is no equivalent in the Australian Constitution[74] to Article 35 of the Basic Law although s 75(v)[75], which directly confers on the High Court original jurisdiction to issue constitutional writs against an officer of the Commonwealth, is a mainstay of the capacity of the judicial arm of government to enforce the rule of law. Those cases were more to do with the power vested in the Australian Parliament to make laws conferring jurisdiction on the courts.[76] In Abebe v The Commonwealth of Australia[77], where the High Court of Australia had to consider the lawfulness of statutory provisions which limited the ability of the court to examine the legality of decisions (of, again, the Refugee Review Tribunal), the principal issue, as stated in the joint judgment of Gleeson CJ and McHugh J was whether Parliament had the constitutional power to do so.[78] This required a consideration of s 77(i) of the Australian Constitution. In Hocine, the court was faced with an argument that bore some resemblance to the concept of access to the courts but, as I have said, there is no such constitutional right expressed in the Australian Constitution. 39. This is the distinguishing feature in the Australian cases to which we have been referred and it is in my view a critical distinction. In Hong Kong, where Article 35 of the Basic Law articulates this right, the approach of the court will be different. Here, the approach of the court will find more similarity in the way the question of time limits was dealt with in Miller v Bull[79] where Tugendhat J tested the time provision in that case[80] against Article 6 of the European Convention for the Protection of Human Rights and Fundamental Freedoms (the right to a fair trial). 40. In my view, the right of access to the courts is engaged in the present case. But has it been infringed in the case of s 34(1) of the CEEO? D.2 Is Article 35 of the Basic Law infringed? 41. I have earlier referred to the essence of the right of access to the courts contained in Article 35 and the buttress it provides to the integrity of elections.[81] There is no doubt that s 34 of the CEEO can be regarded as placing restrictions on that right. But whether such restrictions amount to an infringement of that constitutional right depends on whether on analysis the essence of the right has been impaired. In the context of the right of access tocourt, the European Court of Human Rights has said that the right is not an absolute one and any restrictions placed on it must be examined to see whether the essence of the right has been impaired.[82] This analysis has been applied in Hong Kong in relation to the right to marry contained in Article 37 of the Basic Law.[83] 42. The question of whether s 34(1), if absolute in preventing an election petition being lodged beyond the seven day limit, impairs the essence of the right of access to the courts, must be seen in context. The context of that provision is that it is but part of a whole scheme regarding election petitions. This scheme is an elaborate one as we have seen[84], restricting the class of persons entitled to use that procedure to the s 33 persons[85] but it has the important feature of allowing election petitions to be lodged as of right without the need for leave to be obtained. 43. I have also earlier mentioned the need for any proceedings questioning an election to be dealt with speedily. This is obviously the purpose of s 34(1) and on this basis, in the context of the scheme as a whole, it does not seem to me disproportionate to impose a seven day limit. Although a tight one, given that the class of persons entitled to lodge election petitions proceedings are those who can be expected to have been intimately involved in an election right from the start and who can therefore be expected to pay close attention not only to their own election activities but also the activities of their opponents, the limit is not unduly short. Certainly, Lam JA was of the view that the seven day limit was not “so short that it would not be possible to comply with [it].”[86] I would also add that it is more or less in line with the limits imposed for similar proceedings in other jurisdictions. In his judgment, Lam JA described the position in other jurisdictions[87]:- “84. Based on that survey, for all countries where the head of state is a popularly elected office, the permissible time limits for legal challenge are specifically regulated. The shortest time limit is Poland where the time available for challenge by a protest lodged with the Supreme Court is 3 days from the date of announcement of the election result. In Ireland, leave application has to be made and the time limit is 7 days. After leave is granted, a petitioner has 3 days to lodge his petition. The position in the United States of America depends on the states or region in question: for District of Columbia the time is 7 days; for Florida it is 10 days. In France, the time limit is 10 days. No information about time limit can be found in respect of the election of the President of Germany and the mechanism for challenge appears to be quite different. In India, the time limit is 30 days. In Singapore, the primary time limit is 21 days. However, in respect of challenge based on allegations of corrupt practice or illegal practice involving payment of money, it is 28 days after the payment. In respect of challenge based on other allegations of illegal practice, the time limit is 14 days. The system in Russia also appears to be different. The first level of challenge is by way of complaints to the Election Commission and the Commission must establish the result within 10 days. After that, there is a time frame of one year for challenges to be made to the Supreme Court though the court's decision would not automatically remove the president from office. None of these countries has any statutory provisions for extension of time for such challenges.” 44. The last sentence of that passage is of note in that it is by no means unusual for time limits for the institution of proceedings questioning an election to be non-extendable, just as in the case of s 34(1). For the same reasons as indicated earlier, I do not regard as objectionable this feature of s 34(1). 45. There is also the consideration of the margin of appreciation which can be accorded by the court to the legislature. This aspect has been considered by the courts in a number of cases. In Fok Chun Wa v Hospital Authority, this Court emphasized the point that the concept of margin of appreciation reflected the different constitutional roles of the judiciary on the one hand, and the executive and legislature on the other.[88] In the context of election law, this difference in roles must be borne in mind. I have earlier discussed the role of the judiciary when dealing with the engagement of Article 35 of the Basic Law[89] in the present case. Elections, however, also involve political and policy considerations and it is in these areas where the legislature is involved. The determination that seven days is the appropriate limit for the lodging of election petitions is one that does involve considerations other than legal ones. A due margin of appreciation should be accorded in the present case. Mr Lee SC relied on those passages in Fok Chun Wa in which this Court made references to core values and fundamental concepts[90], but the right of access to the courts is not an unlimited one, particularly in the present context. 46. The main reason for the Judge making a finding that s 34(1) infringed Article 35 of the Basic Law was basically that he could not reconcile the seven day non-extendable time limit in that provision with the more generous time limit in s 39 (30 days which could be extended). He regarded this as a fundamental flaw in the legislation. Reference was made to Charles Mok 1 where, in the context of the Legislative Council election under the Legislative Council Ordinance[91], the Court remarked on the incongruity of the unavailability of an appeal mechanism for election petitions under that Ordinance contrasted with the availability of appeals in other proceedings under that Ordinance[92] to challenge the qualification of the members of the Legislative Council.[93] Given this fundamental flaw, according to the Judge, any margin of appreciation that might otherwise have been accorded was of no significance. 47. I am, with respect, unable to agree with the Judge’s reasoning:- (1) It is important to have regard to the difference between election petition and judicial review proceedings under the CEEO. As explained above, they are quite different, although admittedly overlaps do exist. The election petition procedure can be regarded as the primary and most speedy means of resolving questions regarding elections. The judicial review procedure enables the question of whether an elected person can lawfully assume the office of Chief Executive to be raised. It can be regarded as a residual means of challenge. The fundamental difference between the two procedures is that one that has already been noted: election petition proceedings can be instituted as of right whereas the judicial review procedure requires leave to be obtained. (2) Devising the scheme of challenges to elections in the way that exists in the CEEO, with different time limits for different proceedings, represents an attempt by the legislature to balance on the one hand the need to resolve any questions about the legality of elected persons to become the Chief Executive as speedily as possible and due respect for the integrity of elections on the other.[94] A tight time limit governs the election petition procedure, which is for the benefit of those persons who have been intimately involved with the relevant election. For those who have not been so intimately involved, the time limit is more relaxed for the residual means of challenge (judicial review), even though the time limit is actually considerably less than the usual three months. (3) No useful guidance can be obtained from Charles Mok 1. In that case, the court was faced with a total absence of any appeal in election petition proceedings and this was contrasted with the availability of avenues of appeal in comparable proceedings. This is not the same situation in the present case where a comparison had to be made between time limits. 48. Mr Lee SC also made a submission to the effect that the inflexible time limit in s34(1) had the potential of causing injustice where (and he used the present case as an illustration), the facts supporting one or more of the grounds in s 32(1)(a) of the CEEO, were not discovered until after the seven day time limit had expired. He gave as an example the ground set out in s 32(1)(a)(iv), being illegal conduct based on election expenses.[95] Since returns for election expenses in Chief Executive elections do not have to be lodged until 30 days after the publication of the election result[96], it may be well past the seven day deadline before any such illegal conduct could be discovered. 49. Admittedly, it is possible for these situations to arise but in my view it does not follow from this that the seven day limit becomes then objectionable from a constitutional point of view. First, where a line is drawn, it is inevitable that there may be hard cases that would arise when persons fall within the wrong side of the line.[97] Secondly, it must be borne in mind in the present context that if a situation were to arise where one or more of the grounds in s 32(1)(a) or (b) only came to light after the seven day period, the election petition procedure does not provide the only means of redress. The existence of judicial review proceedings (although not open to s 33 persons on the s 32(1)(a) or (b) grounds); the possibility of criminal proceedings under, say, ECICO; proceedings under Article 73(9) of the Basic Law; or simple political realities, are all relevant to be considered in this context. D.3 Conclusion on the Second Issue 50. The appeals by Mr Leung and the Secretary for Justice must accordingly be allowed. The election petition proceedings instituted by Mr Ho ought to have been struck-out on the basis that they were barred by s 34(1) of the CEEO. E THIRD ISSUE: THE COSTS ORDER OF 28 SEPTEMBER 2012 51. Both the main protagonists in relation to this part of the case, Mr Ho and Mr Leung, were agreed that the principles to be applied were those contained in Sky Wide Development Limited.[98] Although Mr Thomas SC for the Secretary for Justice in his written Case questioned whether the applicable principles regarding costs in contested applications for leave to institute judicial review proceedings were those as stated in Sky Wide Development Limited, I do not regard the present case as an appropriate occasion to go into this question. Apart from the agreed position of Mr Ho and Mr Leung, I am in some doubt as to whether the Secretary for Justice has locus standi to argue in this appeal at all, he having all along taken a neutral view on costs. 52. This appeal can be quickly disposed of. We have differed from the reasoning of the learned judge in the 30 July 2012 judgment. Although the effect is the same in that it was correct that leave to institute judicial review proceedings should not have been given to Mr Ho, I have not fully accepted the submissions made by Mr Leung or the Secretary for Justice. In the circumstances, the correct order for costs should be that no order for costs be made. F CONCLUSIONS ON THE APPEALS 53. The appeals giving rise to the two main issues are, as stated earlier, academic but leave to appeal was given on them by the Appeal Committee in view of their importance. It followed from the giving of leave to appeal on the two main issues that leave to appeal should also be given in relation to the Third Issue. 54. The formal orders following the resolution of the three issues should be as follows:- (1) The appeals of Mr Leung in FACV 24 and 25 of 2012[99] are allowed. Paras 1 to 3 of the Order of Lam JA dated 5 October 2012 are set aside. (2) The appeal of the Secretary for Justice in FACV 27 of 2012[100] is allowed. Paras 1 to 3 of the Order of Lam JA dated 5 October 2012 are set aside. (3) The appeal of Mr Ho in FACV 1 of 2013[101] is dismissed insofar as the Order of Lam J dated 30 July 2012 is concerned. The appeal is allowed insofar as the Order for costs made by Lam JA dated 28 September 2012 is concerned, and there is to be substituted in its place no order as to costs. 55. As for the costs of these appeals, both Mr Leung and Mr Ho have to an extent succeeded but also failed in relation both to outcome as well as in their submissions on various issues. The Secretary for Justice maintained a neutral position in relation to costs in these appeals. In the circumstances, I would make an order nisi that there be no order as to costs in these appeals. If any party wishes to have a different order for costs, written submissions should be served on the other parties and lodged with the court within 14 days of the handing down of this judgment, with liberty on the other parties to lodge written submissions within 14 days thereafter. In the absence of such written submissions, the order nisi will stand absolute at the expiry of the time limited for such submissions. Mr Justice Chan PJ: 56. I agree with the judgment of the Chief Justice. Mr Justice Ribeiro PJ: 57. I agree with the judgment of the Chief Justice. Mr Justice Tang PJ: 58. I respectfully agree with the judgment of the Chief Justice. Mr Justice Gleeson NPJ: 59. I agree with the judgment of the Chief Justice. Mr Johnny Mok SC and Mr Abraham Chan instructed by Sit Fung, Kwong & Shum for Mr Leung Chun Ying, the Appellant in FACV 24, 25 of 2012, the 1st Respondent in FACV 1 of 2013 Mr Michael Thomas SC, Mr Stewart KM Wong SC and Mr Jin Pao instructed by the Department of Justice for the Secretary for Justice, the Appellant in FACV27 of 2012 Mr Martin Lee SC, Mr Hectar Pun, Mr Jeffrey Tam and Mr Carter Chim instructed by Lam and Lai for Mr Albert Ho, the Respondent in FACV24, 25 and 27 of 2012, the Appellant in FACV 1 of 2013 [1] Under the CEEO, the Chief Executive is elected by an Election Committee, the constitution of which is set out in Part 3 of the Ordinance. Voting by members of the Election Committee is governed under Part 5. [2] Pursuant to s 28(2) of the CEEO. [3] Mr CY Leung was named as the 1st Respondent in both proceedings. The 2nd Respondent was the Returning Officer for the relevant election. It was the declaration made on the 25 March 2012 that was the relevant decision challenged in both proceedings. The Returning Officer is a required party in election petitions: s 35 of the CEEO. The Returning Officer has effectively taken no part in the proceedings to date and did not participate in the present appeals. [4] Such conduct is a ground by which an election of the Chief Executive may be questioned: ss 32(1)(a)(iv) and (2) of the CEEO. [5] Mr Leung Kwok Hung is a member of the Legislative Council and was a member of the Election Committee for 2012 Chief Executive Elections. Though participating in the hearings in the court below, for reasons that will presently appear, he took no part in the appeals before us. [6] Although such applications for leave are usually dealt with on an ex parte basis, the judge heard submissions made on behalf of Mr Leung and also from the Secretary for Justice who had intervened in the proceedings. [7] As he then was. [8] This issue is no longer before the Court. [9] As he had become. [10] The Secretary for Justice had also intervened in these proceedings in view of the importance of the legal issues raised. [11] Under s 34(1) of the CEEO, an election petition questioning an election has to be lodged within seven working days after the day on which the result of an election is declared under s 22(1AB) or s 28. Mr Ho had lodged the election petition only on 4 July 2012, well after the seven day limit. [12] See HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574, at paras 67-69 (608D-611E). [13] Accordingly, on this basis, the Judge indicated that Mr Albert Ho should be given an opportunity to apply for an extension of time to lodge the election petition. [14] See sub-para (1) above. [15] The rule appears to be that usually no order as to costs is made even where a putative respondent (in the present case Mr CY Leung) successfully resists leave being granted, unless there are good reasons or unusual circumstances to suggest the contrary: see Sky Wide Development Ltd v Building Authority [2011] 5 HKLRD 202. [16] The Secretary for Justice had undertaken when he intervened that he would at no stage be seeking costs orders. [17] Under s 22(1)(c) of the Hong Kong Court of Final Appeal Ordinance Cap 484. [18] The Chief Justice, Ribeiro and Tang PJJ. [19] Para 5(4) above. [20] See Secretary for Security v Sakthevel Prabakar [2003] 6 HKCFAR 397, Yeung Chun Pong v Secretary for Justice, FAMC 101 of 2005, 2 March 2006 and Chit Fai Motors Company Limited v Commissioner for Transport [2004] 1 HKC 465. [21] Para 5(3) above. [22] Represented by Mr Martin Lee SC, Mr Hector Pun, Mr Jeffrey Tam and Mr Carter Chim. [23] Represented by Mr Johnny Mok SC and Mr Abraham Chan. [24] Represented by Mr Michael Thomas SC, Mr Stewart Wong SC and Mr Jin Pao. [25] Para 7(1) above. [26] Section 39 of the CEEO refers to both judicial review and “other proceedings” but no party in the present appeals has addressed the Court on any type of proceedings other than judicial review. [27] Para 5(1) above. [28] See Vallejos Evangeline Banao v Commissioner of Registration, FACV Nos 19 and 20 of 2012, 23 March 2013, at paras 76 and 77; Fully Profit (Asia) Limited v The Secretary for Justice, FACV 17 of 2012, 13 May 2013, at paras 15 and 16. [29] Annex I to the Basic Law originally made reference to the Election Committee comprising 800 members. By an Amendment to Annex I approved by the Standing Committee of the National People’s Congress on 28 August 2010, the number of members was increased to 1,200 persons. [30] In the context of elections to the Legislative Council making this point, see Mok Charles Peter v Tam Wai Ho [2012] 3 HKC 398 at paras 17 and 18 (408D-G). I shall refer to this decision as Charles Mok 2. [31] Article 43 of the Basic Law. [32] Article 48(1) of the Basic Law. [33] (2010) 13 HKCFAR 762, at para 51(2) (786). I shall refer to this case as Charles Mok 1. [34] Part 5 of the CEEO is headed “Election Petitions” although other proceedings are also referred to. [35] See ss 22(1AB), 23, 26A and 28(1) of the Ordinance. [36] Where a single candidate is not returned at an election. [37] Where, in the case of multiple candidate elections, the winning candidate is returned. [38] The word “only” is also used in the heading to s 32 but headings have no legislative effect: s 18(3) of the Interpretation and General Clauses Ordinance Cap 1. [39] As Lam J held in the 30 July 2012 judgment. [40] Such as a failure publicly to declare that he or she is not a member of a political party and to give an appropriate undertaking to this effect (s 31 of the Ordinance). Or, subsequent to the election, he or she acquires a right of abode in a foreign country (s 13(c) of the CEEO; Article 44 of the Basic Law). [41] Set out in paras 10(1) and (2) above. [42] Under s 32(1)(b). [43] Section 39 is silent on the available grounds. [44] The s 32(1)(b) ground was obviously unavailable since the 2012 election was a multi-candidate election. [45] Although it has similarities to the election petition procedure under the Legislative Council Ordinance Cap 542, the District Council Ordinance Cap 547 and the Village Representative Election Ordinance Cap 576. [46] Section 34(1) of the CEEO. [47] And the threshold for obtaining leave is by no means an easy threshold to overcome since there has to be a reasonably arguable claim which enjoys a realistic prospect of success: Po Fun Chan v Winnie Cheung (2007) 10 HKCFAR 676. [48] See para 13(3) above. [49] Para 11 footnote 26 above. [50] Section 39(1) is set out in para 10(2) above. [51] See Rules of the High Court Order 53 Rule 4(1). [52] It will be recalled that Mr Albert Ho’s challenge was based on s 32(1)(a)(iv) of the CEEO: see para 3 footnote 4 above. [53] See para 22 above. [54] Section 32(1)(a)(vi) of the CEEO. [55] Section 39(1) of the CEEO. [56] Section 39(2) of the Ordinance. [57] See para 16 above. [58] Para 20(2) above. [59] Para 6 above. [60] This provision is set out in para 10(2) above. [61] See s 39(2) of the CEEO. [62] Para 5(2) above. [63] (2006) 9 HKCFAR 234, at paras 49-50 (255G-256B). [64] Para 28(3) above. [65] Such as s 34(1) where time limits were prescribed for the institution of proceedings or for some other court process to be commenced. [66] (1962) 108 CLR 471, at 488. [67] (1995) 184 CLR 265, at 277. [68] (1999) 197 CLR 500. [69] At para 11. Reference was made to that passage in Australian Iron and Steel Limited v Hoogland referred to in sub-para (1) above. [70] (2000) 99 FCR 269. [71] At para 43(3). [72] At para 45. [73] At para 46. [74] This is a reference to the Commonwealth of Australia Constitution Act 1900. [75] Section 75(v) states:- “75. Original jurisdiction of High Court In all matters: …… (v) in which a writ of Mandamus or prohibition or an injunction is sought against an officer of the Commonwealth; the High Court shall have original jurisdiction.” [76] This power vested in Parliament is referred to in ss 76 and 77 of the Constitution, which states:- 76. Additional original jurisdiction The Parliament may make laws conferring original jurisdiction on the High Court in any matter: (i) arising under this Constitution, or involving its interpretation; (ii) arising under any laws made by the Parliament; (iii) of Admiralty and maritime jurisdiction; (iv) relating to the same subject-matter claimed under the laws of different States. 77. Power to define jurisdiction With respect to any of the matters mentioned in the last two sections the Parliament may make laws: (i) defining the jurisdiction of any federal court other than the High Court; (ii) defining the extent to which the jurisdiction of any federal court shall be exclusive of that which belongs to or is invested in the courts of the States; (iii) investing any court of a State with federal jurisdiction. [77] (1999) 197 CLR 510. [78] At para 1. [79] [2010] 1 WLR 1861. [80] The time to serve a notice of the amount a nature of the security given in relation to an election petition. [81] Para 34 above. [82] 		Zwiazek Nauczycielstwa Polskiego v Poland (2005) 41 EHRR 21, at para 29. See also Ashingdane v United Kingdom (1985) 7 EHRR 528 at para 57; Tolstoy Miloslavsky v United Kingdom (1995) 20 EHRR 442, at para 59 (both these cases were referred to for this purpose in Ng Yat Chi v Max Share Limited (2005) 8 HKCFAR 1, at para 75). [83] 		W v The Registrar of Marriages, FACV 4 of 2012, 13 May 2013 at paras 68,69 and 108. [84] Paras 15 and 16 above. [85] Being the class of persons most likely to be affected by an adverse election result and therefore most likely to take action: para 16(1) above. [86] The 12 September 2012 judgment at para 109. [87] In para 84 of the judgment. [88] [2012] 2 HKC 413, at para 64. [89] Para 36 above. [90] Fok Chun Wa at paras 77 to 79. [91] Cap 542. [92] Namely s 73 of the Legislative Council Ordinance. [93] Para 68 of the judgment in Charles Mok 1. [94] In this latter respect, see para 28(3) above. [95] Contrary to s 24(1) of ECICO. [96] Under s 37(2)(a)(ii)(A) of ECICO. [97] See Fok Chun Wa at para 71. [98] See para 5(3) footnote 15 above. [99] Relating to the Second Issue. [100] Relating to the Second Issue. [101] Relating to the First and Third Issues. 
Chief Justice Ma: 1. I agree with the judgement of Lord Hoffmann NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgement of Lord Hoffmann NPJ. Mr Justice Fok PJ: 3. I agree with the judgement of Lord Hoffmann NPJ. Mr Justice Cheung PJ: 4. I agree with the judgement of Lord Hoffmann NPJ. Lord Hoffmann NPJ: 5. On 12 October 2014 the appellant took part in a Distributed Denial of Service (“DDoS”) attack on the website of the Shanghai Commercial Bank. When you call up a website, your computer sends a “request” to the website server, i.e. the computer which supports the website, calling up the website page. If it is available, the website server will respond to the request by transmitting the page. When it appears on your screen, you may click on some service you want to access, for example, your bank account, and your computer will send another request, and so on. The capacity of the server to deal with requests at any given time (its “bandwidth”) is finite. The method of a DDoS attack is for a number of co-ordinated computers to send a very large number of requests at more or less the same time to exhaust the server’s bandwith, thereby denying access to persons wishing to transact their ordinary business through the website and possibly causing the overloaded system to crash. In the present case the server received 504,592 requests within the space of an hour, of which 6,652 came within a space of 16 seconds from the appellant’s computer. But the attack was a failure because the server had enough surplus capacity to prevent the attack from having any effect upon its other operations. The Charge 6. The appellant was charged with criminal damage, contrary to section 60(1) of the Crimes Ordinance Cap 200, namely that he without lawful excuse damaged property belonging to another intending to damage such property or being reckless as to whether such property would be damaged. It is an offence punishable by a maximum of ten years imprisonment. The meaning of “damage any property” was given an extended meaning by a definition inserted into section 59 by the Computer Crimes Ordinance 1993 (Ordinance No 23 of 1993). A new subsection 59(1A) provided that in relation to a computer, damage to property included “misuse of a computer”. This phrase was in turn defined to mean: “(a) To cause a computer to function other than as it has been established to function by or on behalf of its owner, notwithstanding that the misuse may not impair the operation of the computer or a program held in the computer or the reliability of data held in the computer; (b) to alter or erase any program or data held in a computer or in a computer storage medium; (c) to add any program or data to the contents of a computer or of a computer storage medium…” The Issue 7. The principal issue in this appeal is whether the appellant had caused the computer “to function other than as it has been established to function by or on behalf of its owner” within the meaning of paragraph (a) of section 59(1A). The prosecution says that the bank, as owner, established its server to receive and respond to genuine requests for its services. Causing the computer to respond to requests which were sent only for the purpose of using up its bandwith was causing it to function other than as the bank had established it to function. The appellant, on the other hand, said that the computer was established to receive and respond to requests. During the attack it responded to the appellant’s and other requests exactly as it had been programmed to do. The attack caused no difference to the way it functioned. The 1993 Ordinance 8. The Computer Crimes Ordinance 1993 created a number of crimes under various Ordinances in relation to computers. The simplest was the offence of unauthorised access to a computer by telecommunications (“hacking”), punishable by a fine and inserted into the Telecommunications Ordinance Cap 106 as section 27A. The offence is committed by a person who (1) by telecommunications (2) knowingly causes a computer to perform any function (3) to obtain unauthorised access to any program or data held in a computer. The offence is committed simply by obtaining unauthorised access through the internet (“by telecommunications”), whether on account of curiosity, malice or dishonest intent. It protects the privacy of ordinary computers. Websites, on the other hand, invite the public to access them. There may of course still be unauthorised access to a particular “program or data” on the websites contrary to section 27A but websites also require a different form of protection. 9. The new section 161 of the Crimes Ordinance dealt with obtaining access with intent to commit an offence, with a dishonest intent to deceive or with a view to dishonest gain or with a dishonest intent to cause loss. 10. Against this background, one can see that the extension of the concept of criminal damage in the Crimes Ordinance to misuse of computers was particularly important for the protection of websites. The definition of the offence does not require access as such to have been unauthorised. So it applies to computers which, through their websites, offer open access to the world. If the access was with dishonest intent, that is covered by section 161. The offence under section 60 is committed when, having obtained access to the computer through the website, one causes it to “function other than as it has been established to function by or on behalf of its owner”. The question therefore is how one describes the way in which it has relevantly been established to function by its owner. Mr Randy Shek, for the appellant, says it was established to, among other things, receive and respond to requests. Mr Ned Lai, for the prosecution, says it was established to provide banking services. Proceedings in the lower courts 11. The principal issues before the magistrate (Mr Raymond Wong) and on appeal to the judge (Deputy High Court Judge Stanley Chan) were whether the appellant was the user of the computer at the time, and whether the appellant had participated in the attack intentionally or by accident. Both tribunals found that he was the user of the computer and had done so intentionally. These findings are not challenged. The magistrate dealt briefly with the construction of section 59(1A). He said (paragraph 70) that the appellant had added information to the bank’s computer within the meaning of paragraph (c). As to the question of intent, he said (paragraph 72) that the appellant had the intent to damage the bank’s computer, and also held (paragraph 73) that the appellant was at least “reckless” as to whether the bank’s computer would be damaged. 12. The reasons of the judge are likewise mainly given over to the question of whether the appellant intended to participate in the attack. He also dealt briefly with the question of construction. He said of paragraph (a) of section 59(1A) “obviously, over 500,000 ‘requests’ within 11 minutes would not be the established way of functioning of the computer under attack”. As to paragraph (c), he said that the requests had added data “even when they were of such minute amount”. Paragraph (a) 13. I do not think that it is sufficient to say that the computer functioned as it had been established to do because it dealt with the attackers’ requests in accordance with what it had been programmed to do. Generally speaking, computers can only do what they have been programmed to do and such a narrow construction would deprive paragraph (a) of any effect. This conclusion is reinforced by the words “notwithstanding that the misuse may not impair the operation of the computer or a program held in the computer or the reliability of data held in the computer”. Mr Shek was hard put to offer an example of an act which came within his construction of “other than as it has been established to function by … its owner” which was not excluded by the following words. In my opinion the functions for which the computer is established to do are not so much concerned with the way it works (or fails to work) but what it was intended to do. The way it works depends upon how it was constructed by its manufacturer. But the statute is concerned with what the owner has set it up to do. The website and its server were established to provide banking services, not to deal with a multitude of requests made for no purpose except to inconvenience the bank and its customers and generate publicity for the attackers. 14. Although it is a decision on a different statute, there is some analogy with the decision of the English Divisional Court in DPP v Lennon [2006] EWHC 1201 (Admin) on the question of whether the owner of a computer consented to receive the equivalent of a DDoS attack in the form of a torrent of e-mails. The court said that his general consent to receiving emails - “plainly does not cover emails which are not sent for the purpose of communication with the owner, but are sent for the purpose of interrupting the proper operation and use of his system.” 15. In my view, a DDoS attack is very appropriately described as a misuse of the bank’s computer. Mr Shek said that it was wrong to have regard to the use of the term “misuse of a computer” in construing the meaning of paragraph (a). “Misuse of a computer” is a defined expression. It means, and means only, what is stated in paragraphs (a), (b) and (c). I agree that the defined term cannot be used to enlarge the meanings of paragraphs (a), (b) and (c) but there is no reason why it cannot be considered when construing them. A defined expression is not usually an arbitrarily chosen word but intended as a broad description of a concept more precisely defined in the ensuing paragraphs. (See Chartbrook v Persimmon [2009] 1 AC 1101, 1112-1113). In such case, it may be used to elucidate any ambiguities in those paragraphs. Furthermore, paragraph (a) itself describes the offence as a “misuse” of the computer. 16. Mr Shek suggested that so broad a definition of the offence could cause injustice. But one must take into account the general defences to a charge of criminal damage and in particular the requirement that the “damage” must have been caused “without lawful excuse”. In particular, section 64(2)(a) of the Crimes Ordinance provides that it is a lawful excuse honestly to believe (whether reasonably or not) that the owner had consented or, if he had known, would have consented, to the act in question. It seems to me that the Legislature was content to rely upon such general provisions to prevent any injustice being caused by the breadth of the definition of misuse of a computer. 17. I would therefore uphold the decision of the judge on paragraph (a). Paragraph (c) 18. The appellant submitted that prosecution based its case before the magistrate solely upon paragraph (a). (Appellant’s case paragraphs 86-87). There was no mention of paragraph (c) and the appellant adduced no evidence upon the question of whether data was added to the bank’s computer or whether he could be said to have intended or been reckless as to whether this would happen. This objection appears to me well founded (see HKSAR v Hau Tung Ying (2011) 14 HKCFAR 453, 470). I therefore consider that reliance upon paragraph (c) would be unjust to the appellant. But for the reasons I have given in relation to paragraph (a), I would dismiss the appeal. Chief Justice Ma: 19. For the above reasons, the appeal is dismissed. Mr Randy Shek and Mr Geoffrey Yeung, instructed by Eric Yu & Company, for the Appellant Mr Ned Lai, SADPP, Ms Mickey Fung, SPP and Mr Ivan Cheung, SPP, of the Department of Justice, for the Respondent 
Chief Justice Cheung: 1. I agree with the judgment of Mr Justice Ribeiro PJ on Question 1 and with that of Mr Justice Lam PJ on Question 2. Mr Justice Ribeiro PJ: 2. The first question which arises on these appeals is whether a restraint order (“RO”) freezing certain assets made under the Organized and Serious Crimes Ordinance[1] (“OSCO”) falls to be discharged as a matter of law where the court decides not to grant an application for a confiscation order (“CO”). The second question is whether an appeal relating to the discharge of such an RO is to be characterised for procedural and jurisdictional appellate purposes as civil or criminal. 3. Leave to appeal was granted by the Appeal Committee[2] on those two questions formulated as follows: “What on the true construction of the provisions of [OSCO] referred to herein, was the consequence of the Judge’s decision in HCMP 1207/2014 (dated 10 August 2016) not to make a confiscation order on the basis of her conclusion that the respondent had not ‘absconded’ within the meaning of section 8(1)(a)(ii)(B)? In particular, did this in law result in the application for a confiscation order being ‘concluded’ within the meaning of section 2(16A)(a) with the consequence that the restraint order dated 30 March 2011 was discharged by operation of section 15(5)(b)?” (Question 1) “Where a person against whom proceedings under OSCO have been instituted seeks to bring an appeal against a Judge’s refusal to discharge a restraint order, is that appeal criminal or civil in nature for jurisdictional and procedural purposes?” (Question 2) 4. This judgment deals with Question 1, while Mr Justice Lam PJ deals in his judgment with Question 2. I have had the advantage of reading that judgment in draft and respectfully agree with his reasoning and conclusion. A. The transactions giving rise to this case 5. The present matter is connected with the case of Ao Man-long (“Ao”), formerly Secretary for Transport and Public Works in Macau, who was convicted there on 90 counts of corruption, money laundering and related offences, ultimately resulting in a sentence of 29 years’ imprisonment. 6. The respondent is Madam Tam Kit-I (“Tam”), who was employed as financial controller of a group of companies in Macau owned by Mr Ng Fok, a works contractor who had frequent business dealings with Ao. Tam became the focus of attention because of transfers between her account with the Wing Hang Bank in Hong Kong which she had opened on 1 March 2006 (“the Wing Hang Account”) and certain accounts maintained or controlled by Ao, the allegation being that the funds transferred were the proceeds of corrupt transactions involving Ao. (a) Thus, on 21 March 2006, Tam transferred $25 million from her account with Banco Weng Hang SA in Macau to the Wing Hang Account, and that sum was further transferred on 17 June 2006 to an account with Hang Seng Bank held by Ao’s father, of which Ao was the sole authorised signatory. Those funds were eventually confiscated as a result of proceedings brought in Hong Kong by the Macau government. (b) Then between 18 and 25 August 2006, a total of $25 million was transferred to Tam’s Wing Hang Account from an account held in the name of Ecoline Property Limited at ICBC (Asia) Ltd, Hong Kong, of which Ao was an authorised signatory. (c) Of that sum, on 4 December 2006, $2.6 million was transferred to the account of Best Choice Ltd with Hang Seng Bank, Hong Kong, of which Ao was the sole authorised signatory. A balance of $22,400,000 thus remained in the Wing Hang Account. Two days later, Ao was arrested and in his safe were found pre-signed cheques totalling $22,400,000 drawn on the Wing Hang Account. (d) Restraint on dealings with and confiscation of that $22.4 million balance are the subject-matter of Question 1. B. The procedural history B.1 Contemplated POBO proceedings 7. After Ao was arrested, the authorities in Hong Kong initially contemplated proceedings against Tam under the Prevention of Bribery Ordinance[3] (“POBO”). On 13 February 2007, they obtained an ex parte restraining order from V Bokhary J under section 14C of that Ordinance freezing the aforesaid balance in the Wing Hang Account. That order was served by ICAC officers on Tam in Macau on 23 February 2007. 8. POBO section 14C empowers the court to make an RO in relation to property in the possession or under the control of or due to a person who is “the subject of an investigation in respect of an offence alleged or suspected to have been committed by him under” the POBO.[4] The person whose property is subject to such an order may be outside Hong Kong[5] and it is provided that the RO shall continue in force for 12 months but, on application by the Commissioner, may be extended by the court “for periods of 12 months at a time”.[6] 9. The RO against Tam was extended three times, with each fresh order being served on Tam by ICAC officers who visited her in Macau. After the first extended RO was served on Tam on 4 February 2008, she visited Hong Kong on 6 June 2008 and returned to Macau on the same day. She was placed on the Hong Kong Immigration Department’s stop list on 11 July 2008. 10. On 4 February 2009 and again on 5 June 2009, ICAC officers invited Tam to come to Hong Kong to be interviewed regarding offences concerning the Wing Hang Account but, unsurprisingly, she declined. B.2 A shift to proceedings under OSCO 11. Having been advised that there was insufficient evidence to charge Tam under POBO but that a viable case existed under OSCO, on 12 January 2011, prosecutors obtained a warrant from the magistrate for Tam’s arrest for money laundering contrary to OSCO section 25.[7] In the paragraphs of my judgment which follow, unless otherwise indicated, references to sections are to provisions in OSCO. Section 2(15) provides that the issue of such a warrant is one means by which proceedings for an offence under that Ordinance are “instituted”. 12. Two days later, on 14 January 2011, on the application of the Secretary for Justice (“SJ”), V Bokhary J issued an RO under section 15[8] freezing the Wing Hang Account. On 30 March 2011, her Ladyship extended that Order, providing that it would “remain in force unless it is varied or discharged by a further order of the Court”. These appeals focus on that RO. 13. An OSCO RO which is issued after proceedings have been instituted is not subject to any expiry period. Nor is it necessary to apply for extensions. The power to issue an RO and its continued validity are premised on the prosecution proceedings or the application for a confiscation order not having been concluded.[9] C. The confiscation application and the Judge’s decision 14. Some three years later, on 20 May 2014, the SJ applied under section 8(1)(a)(ii)(B) for the Wing Hang Account funds to be confiscated on the basis that Tam had absconded. Sections 8(1)(a)(ii) and 8(1)(b) materially provide: “Where ... proceedings for one or more specified offences have been instituted against a person but have not been concluded because the person (A) has died; or (B) has absconded; and … an application is made by or on behalf of the Secretary for Justice for a confiscation order, the Court of First Instance ... shall act as follows.” Sections 8(3) to 11 then go on to prescribe that the court “shall act” by making certain inquiries and determinations in deciding whether to grant the application. 15. On 10 August 2016, Campbell-Moffat J[10] rejected the SJ’s application for a CO, holding that Tam had not been shown to have absconded and that section 8(1)(a)(ii)(B) was not applicable. On that basis, she did not embark upon the prescribed process of inquiry and determination. However, she held that “the [RO] made under HCMP 54/2011 remains extant”.[11] 16. The Judge’s decision was not subject to any appeal. However, since the question of whether a person has “absconded” for the purposes of section 8(1)(a)(ii)(B) has some relevance to the disposition of these appeals, her decision is reviewed in Section K of this judgment. D. The application to discharge the RO 17. Tam waited some two-and-a-half years before applying on 31 January 2019 to discharge the RO, the matter again coming before Campbell-Moffat J. Tam’s contention was that upon the Judge’s rejection of the SJ’s application for confiscation, the RO was automatically discharged. 18. The argument ran as follows: (a) Section 15(5)(b) provides that an RO “shall be discharged on the conclusion of the proceedings or application concerned”. The “application concerned” is the SJ’s application for a CO. (b) Section 2(16A)(a) states that “An application for a confiscation order made in respect of a defendant where section 8(1)(a)(ii) or (7A) is applicable is concluded ... if the Court of First Instance ... decides not to make such an order, when it makes that decision.” (c) Since the Judge had decided in 2016 not to make the CO sought, the relevant application was thereupon concluded. Thus, it was argued, by virtue of section 15(5)(b), the RO was automatically discharged. 19. Campbell-Moffat J rejected that argument.[12] She held that section 2(16A) relates only to an application for a confiscation order made in respect of a defendant “where section 8(1)(a)(ii) ... is applicable”, in other words, in respect of a defendant found to have absconded (or to have died). Since she had decided in 2016 that Tam had not been shown to have absconded, she held that section 8(1)(a)(ii) was not applicable, so that section 2(16A) did not operate and neither the proceedings nor the confiscation application were “concluded” for the purposes of discharging the RO pursuant to section 15(5)(b). E. The Court of Appeal’s decision 20. The Court of Appeal[13] allowed Tam’s appeal and discharged the RO on the sole ground[14] that the Judge’s refusal of the confiscation order automatically led to its discharge. The Court’s reasoning is adopted by Tam in its entirety on the present appeals. 21. Citing section 19(2)[15] and SJ v Tan Lam Chuan,[16]Chow JA pointed out that an RO is a temporary measure designed to freeze assets until a CO is made and is not intended to apply indefinitely.[17] He noted that the Judge had decided that Tam was not an absconder within the meaning of section 8(1)(a)(ii)(B), but he reasoned that this merely meant that a necessary precondition for the grant of a CO had not been met. The fact remained that she had decided not to make the CO so that, Chow JA held, by virtue of section 2(16A) the relevant application had “concluded”. Consequently, the RO was discharged by virtue of section 15(5)(b).[18] 22. His Lordship stated: “In our view, it does not follow from the fact that the condition in 8(1)(a)(ii)(B) was not satisfied, or, in the Judge’s words, ‘section 8(1)(a)(ii)(B) was not applicable’, that the Confiscation Application was not concluded. It does not seem to us to matter what may be the reason why the Confiscation Application failed. The fact remains that the Judge decided not to make the confiscation order sought by the Secretary for Justice, thereby concluding the application. There was no appeal against the 2016 Decision of the Judge. We consider it to be clear that there was ‘conclusion’ of the Confiscation Application and, by the operation of s 15(5)(b), the OSCO Restraint Order was accordingly discharged.”[19] F. The true construction of the relevant OSCO provisions 23. I am, with respect, unable to agree with the Court of Appeal. Their Lordships’ judgment rests on their construction of section 2(16A) in combination with section 15(5) which state as follows: Section 2(16A) “An application for a confiscation order made in respect of a defendant where section 8(1)(a)(ii) or (7A) is applicable is concluded— (a) if the Court of First Instance or the District Court decides not to make such an order, when it makes that decision; or (b) if such an order is made as a result of that application, when the order is satisfied.” Section 15(5) “A restraint order— (a) may be discharged or varied in relation to any property; and (b) shall be discharged on the conclusion of the proceedings or application concerned.” 24. It was not in dispute that the “application concerned” was the SJ’s application for a CO. Crucially the Court of Appeal held that the fact that the Judge had decided not to make the CO was itself sufficient to conclude the application and to trigger discharge of the RO and that the reason for the Judge reaching her decision was immaterial. 25. In my view, that conclusion was erroneous. For the reasons developed below, on a proper contextual and purposive construction of section 2(16A), such an application is concluded and the RO consequently discharged only when the purpose of the RO is spent. Thus, it is essential to consider the basis and consequences of the judge’s decision not to make the order. The purpose of the RO may, for instance, be exhausted (subject to any appeal) when the court completes the process prescribed by sections 8(3) to 11 and either decides to refuse the application or decides to make the CO and the CO is then satisfied.[20] In either case, there would no longer be any extant or prospective CO and thus no point in continuing to freeze the affected assets with a view to making them available for enforcement of such an order. F.1 The power to confiscate the proceeds of crime 26. A central objective of OSCO is to authorise the confiscation of property representing the value of the proceeds of crimes which are “specified offences” or “organized crimes”.[21] The persons targeted are those who benefited from such crimes by receiving payment or some other reward in connection with the commission of such offences.[22] COs confiscating such persons’ property are issued to implement that objective and ROs are made in support of existing or prospective COs. The provisions which govern the making and discharge of COs set the context in which the provisions governing the making and discharge of ROs fall to be construed. F.2 The provisions governing the making of COs 27. Section 8(1) provides as follows: “(1) Where— (a) either— (i) in proceedings before the Court of First Instance or the District Court a person is to be sentenced in respect of one or more specified offences and has not previously been sentenced in respect of his conviction for the offence or, as the case may be, any of the offences concerned; or (ii) proceedings for one or more specified offences have been instituted against a person but have not been concluded because the person— (A) has died; or (B) has absconded; and (b) an application is made by or on behalf of the Secretary for Justice for a confiscation order, the Court of First Instance or the District Court, as the case may be, shall act as follows.” 28. Sections 8(3) to 11 then detail how the court “shall act”, setting out the inquiries and determinations that the court must make in deciding whether to make or refuse the CO applied for. The “application” therefore comprises that process and concludes when that process ends and the court disposes of the application. This bears directly on the meaning of the phrase “the conclusion of the ... application concerned” in section 15(5)(b). 29. Section 8(1)(b) enables the SJ to apply for a CO in two situations: (i) where a defendant has been convicted of a specified offence and is about to be sentenced for it;[23] and (ii) where proceedings for a specified offence have been instituted against a person but have not been concluded because the person has died or has absconded. 30. In cases where the defendant has been convicted of the specified offence, the court “shall act” as follows. First, the defendant should be sentenced.[24] The court should then determine whether he or she benefited from that offence[25] before deciding whether to make a CO. If it decides to make an order, it then decides upon the quantum of recovery. 31. Additional steps have to be taken by the court where proceedings have been instituted but have not run their course because the person has died or absconded as provided by section 8(1)(a)(ii). The person in question will not be before the court, so the court must first be satisfied that he or she has indeed died[26] or absconded.[27] Where the person is said to have absconded, the court must moreover be satisfied that at least 6 months have elapsed since the putative date of absconding.[28] Since the court will not be able to rely on an actual conviction, it must be satisfied on the basis of specified categories of evidence[29] that the person concerned could have been convicted if he or she had not died or absconded.[30] It must also be satisfied that such person has benefited from the putative crime to an extent exceeding $100,000.[31] Due process concerns must also be taken into account in absconding cases. Reasonable efforts must have been made to bring the person in question before the court and sufficient notice of the proceedings given to enable them to be defended.[32] In death cases, the personal representative of the deceased person is given standing to resist a CO.[33] 32. Where the required steps have been taken and the court decides to make a CO, it determines the amount which the person should be ordered to pay, reflecting the value of the defendant’s proceeds of the offence.[34] Enforcement procedures may then follow with the court inter alia fixing the time for payment and setting a term of imprisonment in default.[35] 33. The statutory scheme therefore only permits confiscation of a person’s property after the aforesaid process of inquiry and determination. It establishes the context for addressing the question whether, as a matter of construction, the application concerned was “concluded” within the meaning of section 2(16A) and the RO consequentially discharged by virtue of section 15(5)(b). F.3 The power to make ROs and their purpose 34. The power to make an RO is conferred by section 15(1) which states: “The Court of First Instance may by order (referred to in this Ordinance as a restraint order) prohibit any person from dealing with any realisable property, subject to such conditions and exceptions as may be specified in the order.” 35. For present purposes, by section 14, that power is exercisable where proceedings have been instituted in Hong Kong against the defendant for a specified offence;[36] where the proceedings have not been concluded;[37] and where the court is satisfied that there is reasonable cause to believe that the defendant has benefited from that specified offence.[38] 36. Section 19 spells out the purpose of an RO, stating: “(1) This section applies to the powers conferred on the Court of First Instance by [section 15] ... (2) Subject to subsections (3), (4), (5) and (6) [not presently relevant], the powers shall be exercised with a view to making available for satisfying the confiscation order or, as the case may be, any confiscation order that may be made in the defendant’s case the value for the time being of realisable property held by any person by the realisation of such property.” 37. ROs are therefore made in aid of COs, securing property with a view to making it available to satisfy such orders should they eventually be made. The statutory intent is evidently for an RO to continue in vigour so long as a CO may prospectively be made in respect of the seized assets. F.4 The provisions regarding the discharge of ROs 38. The pertinent provisions governing discharge of ROs consist of section 15(5)(b), read together with sections 2(16) and 2(16A). They relevantly state as follows: Section 15(5)(b) “A restraint order ... shall be discharged on the conclusion of the proceedings or application concerned.” Section 2(16) “Proceedings for an offence are concluded on the occurrence of one of the following events— (a) the discontinuance of the proceedings whether by entry of a nolle prosequi or otherwise; (b) an order or verdict acquitting the defendant, not being an order or verdict which is subject to appeal or review within the meaning of subsection (17); (c) the quashing of his conviction for the offence except where, under section 83E of the Criminal Procedure Ordinance (Cap. 221), an order is made that he be retried; (d) the grant of the Chief Executive’s pardon in respect of the conviction for the offence; (e) the court or magistrate sentencing or otherwise dealing with him in respect of his conviction for the offence where the Secretary for Justice either does not apply for a confiscation order, or applies for a confiscation order and the order is not made; or (f) the satisfaction of a confiscation order made in the proceedings (whether by payment of the amount due under the order or by the defendant serving imprisonment in default).” Section 2(16A) “An application for a confiscation order made in respect of a defendant where section 8(1)(a)(ii) ... is applicable is concluded— (a) if the Court of First Instance or the District Court decides not to make such an order, when it makes that decision; or (b) if such an order is made as a result of that application, when the order is satisfied.” F.5 Discharge on conclusion of the proceedings 39. As we have seen, where the prosecution results in the defendant’s conviction for a specified offence, the court proceeds to sentence the offender and then determines whether he or she benefited from the offence in question and if so, the value of the property to be confiscated.[39] It may be noted that even where the court decides to issue a CO after making the requisite determinations, the proceedings are not yet “concluded” for the purposes of triggering section 15(5)(b). Section 2(16)(f) stipulates that the proceedings are only concluded for that purpose upon “the satisfaction of a confiscation order made in the proceedings (whether by payment of the amount due under the order or by the defendant serving imprisonment in default)”. Until that occurs, the RO is not discharged and retains its purpose of securing the assets against which the CO may eventually be enforced. 40. That statutory purpose is reinforced by the other provisions of section 2(16). The instances enumerating when proceedings for an offence are concluded all involve situations where there is no longer any prospect of a CO being made, either because the defendant has essentially been acquitted or he or she has been convicted and there is no longer any justification for continuing to restrain dealings with the affected assets. 41. Thus, paragraphs (a) to (d) in section 2(16) deem proceedings to be concluded where they are discontinued by a nolle prosequi or otherwise; where the defendant is acquitted and the acquittal is not subject to appeal or review; where the conviction is quashed on appeal without any order for a retrial, or where the defendant has been pardoned. In all these instances, the defendant has “got off” and there is no prospect of a CO being made as part of a sentence. Paragraphs (e) and (f) cover cases where the defendant is convicted and either a CO is not sought or the order is made and has been satisfied. The purpose of the RO is exhausted in all these cases and the order is discharged as provided for by section 15(5)(b). F.6 Discharge on conclusion of the application for a CO 42. How the court “shall act” upon the SJ’s applying for a CO has previously been noted.[40] It is only at the conclusion of the prescribed process that the court decides either to make or refuse a CO. It is against that background that section 2(16A), which is at the heart of these appeals, falls to be construed. As we have seen, it provides as follows: Section 2(16A) “An application for a confiscation order made in respect of a defendant where section 8(1)(a)(ii) or (7A) is applicable is concluded— (a) if the Court of First Instance or the District Court decides not to make such an order, when it makes that decision; or (b) if such an order is made as a result of that application, when the order is satisfied.” 43. Section 2(16A) thus specifies that the application for a CO concludes, first where the court decides not to make the order and secondly, where it does decide to issue the CO. Read in the context of the provisions discussed above, the court’s decision either way must be understood to be a decision on the merits of the application, taken at the culmination of the prescribed process of inquiry and determination. Until that process has been gone through, the RO continues to perform its function of securing the frozen assets and the application cannot be regarded as concluded so as to trigger the RO’s discharge. This is reflected in section 2(16A)(b) which, mirroring section 2(16)(f) in respect of concluded proceedings, stipulates that the application is concluded in cases where the CO is made only when the order is satisfied, ie, at the end of the prescribed process and when the purpose of the RO is spent. G. The provisions applied in the present case 44. In the present case, the Judge appears to have decided not to embark at all upon the prescribed process since she held that section 8(1)(a)(ii) was not applicable on the basis of her finding that Tam had not absconded. That gave rise to a somewhat odd situation. If a defendant has not absconded, one might generally expect that to mean that he or she is susceptible to arrest and prosecution within the jurisdiction, so that the possible grant of a CO is deferred to await substantive proceedings and the RO meanwhile remains in place. But that was obviously not the position in the present case. Nevertheless, although declining to make a CO, her Ladyship expressly held that “the [RO] made under HCMP 54/2011 remains extant”.[41] Her decision thus did not result in the elimination of any prospect of the making of a CO and she evidently recognised that the purpose of the RO was not spent. Furthermore, under section 8(1)(a)(ii)(A), it still remains open to the SJ to apply for a CO on the basis that Tam has died, should that eventuality arise. 45. Contrary to the Court of Appeal’s view, the reason or basis for the court’s decision not to grant the application for a CO and the consequences of that decision are important. The following illustrations may be considered. Suppose that the SJ applies for a CO alleging that the defendant has absconded but that person then turns up at the hearing. The Judge would no doubt refuse the SJ’s application on the basis that section 8(1)(a)(ii)(B) is not applicable, ie, that this is not in fact an “absconding” case. Prosecution proceedings are likely then to be taken up against the defendant and the [RO] would remain in place since it would continue to play its role of securing the frozen property pending their conclusion. The SJ’s application, which has not been decided on its merits in accordance with the prescribed statutory regime, is not concluded and discharge of the RO is not triggered. Since the process has not begun, it is inapt to speak of it being “concluded”. A similar illustration might involve an application for a CO being launched on the basis that the defendant has died, which allegation is disproved by that person showing up at the hearing. 46. To take a further example, suppose that the court is not satisfied that the defendant was given notice sufficient to enable the proceedings to be defended. The court might therefore decide not to make the CO sought, either adjourning the application to a later date or dismissing it without prejudice to a further application being made after proper notice is given to the defendant. Again, the merits of the application would not have been considered and the court’s decision plainly would not justify an automatic discharge of the RO. The mere fact that the court decided not to make the CO on the SJ’s application is not of itself sufficient to trigger discharge of the RO. H. The SJ’s argument 47. Mr William Tam SC, appearing for the SJ, while also advocating a contextual and purposive construction, proposed a simple argument for disposing of these appeals. 48. That argument runs as follows: (a) In the present case, the RO would only be discharged under section 15(5)(b) on the conclusion of the SJ’s application for a CO. (b) That application is only deemed concluded if section 2(16A) applies. And that provision only applies “where section 8(1)(a)(ii) ... is applicable”. (c) Since the Judge held that section 8(1)(a)(ii) was not applicable, section 2(16A) is not engaged and the SJ’s application is not “concluded” so that discharge under section 15(5)(b) is not triggered. 49. While it has the attraction of simplicity, I am unable to accept that argument as conveying the true construction of the relevant provisions. Section 2(16A) must be construed together with section 2(16) which it immediately follows. They refer respectively to (i) the conclusion of proceedings (section 2(16)) and (ii) the conclusion of an application for a CO (section 2(16A)). In other words, they identify two categories of matters being “concluded” for the purposes of triggering discharge of an RO under section 15(5)(b). 50. When section 2(16A) refers to an application for a CO being concluded “where section 8(1)(a)(ii) ... is applicable”, it is making it clear that it is addressing the second category of “conclusion”, in contradistinction to the first. It is not concerned with examining how the judge actually dealt with any particular application, eg, as to whether, on the facts, the defendant was shown to have absconded. On its true construction, section 2(16A) is engaged where the SJ’s application is made under section 8(1)(a)(ii). The question which remains is what the phrase “decides not to make such an order” in section 2(16A)(a) means. As explained in the foregoing sections of this judgment, that meaning is to be ascertained by construing that phrase contextually and purposively in connection with the question when discharge of an RO under section 15(5)(b) is triggered. I. Answer to Question 1 51. For the foregoing reasons, I would answer Question 1 as follows: On the true construction of the relevant OSCO provisions, the Judge’s decision not to make a CO on the basis of her conclusion that Tam had not absconded did not result in the application for a CO being concluded within the meaning of section 2(16A)(a) and did not carry the consequence of discharging the RO dated 30 March 2011 by operation of section 15(5)(b). 52. With respect, the Court of Appeal fell into error in holding that the reason for a court’s decision not to make a CO does not matter. If such a decision is taken for a reason which does not involve the court determining the merits of the application and without eliminating the prospect of a CO eventually being made, the RO retains its intended purpose and remains in place. J. Abuse of process 53. Mr Nigel Kat SC, appearing for Tam, advanced an additional argument (in several guises) submitting that retaining the RO amounts to an abuse of process. He was at pains to argue that an RO is intended to be a temporary measure and that obtaining such an order when there is no prospect of Tam coming to Hong Kong to be tried in substantive criminal proceedings means that the order is intended to operate abusively, not as a temporary measure, but indefinitely with a confiscatory effect. Attorney General v Blake (Jonathan Cape Ltd Third Party)[42] was cited in support.[43] Mr Kat SC also sought to rely on section 14, submitting that its provisions show that the statutory intent is for ROs to be of a limited duration, militating against indefinite or long-term orders freezing assets. 54. He submitted that: “... if the person is a foreigner, the RO continues unless and until he or she chooses voluntarily to come to Hong Kong to be tried in criminal proceedings, irrespective of whether those criminal proceedings relate to an extraditable offence or whether he or she is extraditable in law. If the foreigner does not come to Hong Kong to be tried, then the restraint of his or her assets (often worldwide) can continue until death when the SJ can apply for confiscation under s.8.”[44] 55. In other words, he argued, “It is an abuse to operate these provisions extra-territorially to force a person who is foreign either to come to HK to defend those proceedings or eventually, on death, to lose the property”.[45] 56. The essence of the abuse argument is therefore that: (1) there are no prospects of Tam voluntarily coming to Hong Kong to stand trial; (2) thus no proceedings will ever be pending to justify the RO’s freezing of her assets; (3) to maintain the RO in such circumstances is to use the RO abusively as an indefinite restraint which would operate as a confiscatory measure instead of a temporary measure pending the outcome of the proceedings; (4) moreover, an aspect of the abuse is that it seeks to force Tam to submit to the criminal jurisdiction as the price of challenging the RO. 57. That argument must be rejected. Points (3) and (4) above are unfounded. The SJ did not seek to keep the RO in place indefinitely. On the contrary, an application for a CO was made on 20 May 2014. If that application had run its course, the RO would have been brought to an end one way or the other: either upon the CO being refused on its merits or being granted and satisfied by recourse to the frozen bank funds. As it turned out, the Judge declined to embark upon the prescribed statutory process on the basis of her finding that Tam had not “absconded”. 58. Nor is there any question of Tam being forced to submit to the criminal jurisdiction as the price of challenging the RO. OSCO section 15(6) caters for persons affected to mount such a challenge. Tam did so with success in the Court of Appeal, without having to come to stand trial. 59. In practice, the SJ cannot abuse the process with impunity by keeping an RO in place indefinitely. A person affected by an RO does not have to sit by passively but is entitled under section 15(6) to apply for a discharge or variation of that order. It might, for instance, be contended that the SJ cannot show on the balance of probabilities that the person affected could have been convicted or did in fact benefit from the offence and that there is accordingly no realistic possibility of a CO eventually being made. In other words, a person affected by an RO could precipitate an inquiry along the lines set out in sections 8(3) to 11 to bring the matter to a conclusion. Other possible grounds for challenge may also exist. 60. Attorney General v Blake is not a relevant authority. It concerns the propriety of granting an interlocutory injunction freezing assets where no substantive adjudication is anticipated. Mr Kat SC relied[46] on the propositions stated in part of the headnote to the report which reads as follows: “[Held] that, although the injunction restraining the defendant from receiving royalties was interlocutory in character and was expressed to freeze rather than to extinguish the defendant's title to the royalties, there was, in the absence of a private law claim, and in the absence of any realistic prospect of the defendant returning to the United Kingdom to face criminal proceedings, no event pending which that money was to be held; that, therefore, the injunction was in substance a confiscatory order in that it was envisaged that the defendant would be deprived indefinitely of his unpaid royalties; that the making of confiscatory orders was strictly controlled by statute, and the court had no common law power to confiscate property without compensation; that on the facts of the defendant's case there was no statutory power to confiscate his property; and that, accordingly, the court had acted outside its jurisdiction in granting the injunction.” 61. The position under OSCO is obviously different. We are not concerned with an interlocutory injunction freezing assets when there is no event pending which that money is to be held. OSCO expressly authorises the confiscation of property representing the proceeds of certain offences and ROs are authorised to be made in support of COs. The making of those orders is not dependent on the owner of the assets facing pending proceedings as in Attorney General v Blake. Section 8(1)(a)(ii) expressly caters for such orders to be made where there will be no concluded proceedings because the person concerned has died or absconded, that being the basis of the SJ’s application in the present case. 62. Neither does section 14 assist the abuse argument. Section 14 sets out the requirements for the making of ROs, which requirements vary in certain cases. The present case involves an RO made on the footing that proceedings had been instituted in Hong Kong against Tam for a specified offence;[47] those proceedings had not been concluded;[48] and the court was satisfied that there was reasonable cause to believe that Tam had benefited from that specified offence.[49] Those are the only qualifying conditions for issuing the RO in the present case. There is nothing in those conditions that suggests any time limit on the RO so obtained. 63. Mr Kat SC referred to other subsections of section 14 which do lay down time limits on the operation of an RO but which have no application to this case. Thus, section 14(1A) is concerned with an RO obtained under section 14(1)(ba) which refers back to section 2(15)(aa) and which therefore addresses the situation where a person has been arrested and released on bail or has refused bail but an RO has been issued on the basis that there is reasonable cause to believe that the defendant may be charged with the offence after further investigation is carried out. Section 14(1A) provides that in such a case, the RO is to expire after giving the time reasonably necessary for the investigation to take place, with an outside limit (subject to applications for extension[50]) of 6 months. 64. Similarly, section 14(4) (referring back to section 14(2)) provides for the discharge of an RO “if proceedings in respect of the offence are not instituted with such time as the [court] considers reasonable”. This caters for a situation where the RO is granted before the institution of proceedings but in the expectation that “whether by the laying of an information or otherwise a person is to be charged”. Thus, if proceedings are not in fact instituted within a reasonable time, the RO is liable to be discharged. Again, this deals with a special situation and no such time limit applies in a case like the present, where proceedings have been instituted and have not been concluded and where the court is satisfied that there is reasonable cause to believe that the person has benefited from the specified offence. K. The Judge’s ruling on “absconding” 65. As noted above, Campbell-Moffat J[51] rejected the SJ’s application for a CO, holding that Tam had not been shown to have absconded and accordingly that section 8(1)(a)(ii)(B) was not applicable. That ruling was not the subject of appeal and strictly does not now arise for consideration. However, her Ladyship’s decision not to make the CO gave rise to the issues canvassed on these appeals. It is also relevant to the argument that continuation of the RO after that decision constitutes an abuse of process. It furthermore bears on the fate of the RO upon disposal of these appeals. As serious doubts exist as to the correctness of the Judge’s approach to “absconding”, it is appropriate that her reasoning should be reviewed for future reference. 66. The SJ’s application for a CO was brought on the basis that Tam had “absconded” making section 8(1)(a)(ii)(B) applicable. Section 2(1) states: “absconded, in relation to a person, includes absconded for any reason whatsoever, and whether or not, before absconding, the person had been (a) taken into custody; or (b) released on bail”. 67. The Judge fell into error first by holding that that provision compels the conclusion that a person can only be held to have absconded if he or she did so after being arrested or charged. Her Ladyship stated: “Ms Chan [for the SJ] submits that it is not necessary that an absconder be charged (and therefore remanded or bailed), before she is regarded as absconding. I do not agree. A simple reading of the plain language of section 2 indicates that proceedings have commenced because they refer to custody or bail, both of which can only happen after arrest or charge. Ms Chan’s interpretation of that section suggests that it was intended to cover situations where neither custody nor bail were in place i.e. no steps of prosecution have been taken. My reading of the section suggests that it is not necessary that the absconder be in custody, she could also abscond if on bail. In other words, some steps towards prosecution must have taken place even if those steps are only at the stage of police bail rather than court bail.”[52] 68. Section 2(1) does indeed “refer to custody or bail”, but with a meaning and effect which is the opposite of the reading adopted by the Judge. Section 2(1) says that a person may be found to have absconded “whether or not, before absconding, the person had been (a) taken into custody; or (b) released on bail”. Thus, as counsel had submitted and contrary to the Judge’s understanding, one may have “absconded” even where one has not been taken into custody or released on bail. 69. Secondly, the Judge held that proceedings had to have been instituted against Tam before 6 June 2008, which was when she might be taken to have absconded. That was the date of Tam’s last visit to Hong Kong, returning to Macau on the same day. The Judge relied on section 8(1)(a)(ii) in support, stating: “... section 8 requires that proceedings have been instituted at section 8(1)(a)(ii). I do not read that requirement to mean ‘have been instituted’ at the time of the application for confiscation rather than at the time of the absconding. There was no warrant issued on 6 June 2008.”[53] 70. But section 8(1)(a)(ii) addresses the making of a CO application by the SJ and the steps which the court must then take in deciding such application in cases of death or absconding. It provides that such an application can only be made where proceedings have been instituted (as defined by section 2(15)). There is nothing in section 8(1)(a)(ii) to suggest that proceedings must have been instituted prior to the person’s act of absconding. 71. It is hard to see any sensible reason for requiring proceedings to have been instituted before the defendant absconded. There will frequently be cases where a person absconds before the authorities manage to arrest or charge or otherwise institute proceedings against him or her.[54] The objective of an RO and an eventual CO is to deprive a person who absconds, but who the court finds (after due inquiry) could have been convicted, of the benefits of the offence which may be susceptible to seizure in Hong Kong. The policy is not to permit the absconder free access to such assets simply because he or she managed to abscond before the issue of an arrest warrant or charge. 72. Thirdly, the Judge adopted a dictionary definition of “absconding” to the effect that it means “to go away suddenly and secretly, especially in order to avoid arrest”. She held also that it was necessary to show that Tam had left in anticipation of imminent arrest and that there was no basis for such a finding: “... the [SJ] would still need to show that the respondent went away ‘suddenly and secretly’ in anticipation of imminent action by the prosecuting authorities. Was that the situation on 6 June 2008? I think not. Hong Kong is not the respondent’s place of residence. She travelled here for the day. She did not spend the night here. We do not know why she travelled here at that time but we know she arrived and later she left. There is no evidence that anything occurred during that time which would put the respondent on notice of imminent arrest as opposed to her state of mind at any other later time of which there is some evidence before the court.”[55] 73. As we have not heard argument on the proper construction of “absconding”, it would not be appropriate to attempt to define that term in this judgment. I would however observe that there is no obvious reason for requiring “abscond” to involve an act which is sudden and secretive.[56] The definition of “absconded” in section 2(1) invites a broad approach to that term, stating that it “includes absconded for any reason whatsoever, and whether or not, before absconding, the person had been (a) taken into custody; or (b) released on bail”. Considered contextually and purposively, it seems to me that the essence of “absconded” in section 8(1)(a)(ii) involves the evasion of apprehension to face criminal proceedings for the offence concerned. 74. In any case, even on the Judge’s approach, requiring it to be shown that “the respondent went away ‘suddenly and secretly’ in anticipation of imminent action by the prosecuting authorities”, her Ladyship appears to have taken an untenable view of Tam’s position. The Wing Hang Account was first frozen on 13 February 2007, the RO being served on Tam in Macau on 23 February 2007. That order stated that Tam and Ao were “the subjects of an investigation in respect of offences alleged or suspected to have been committed by them” under POBO and froze the Wing Hang Account. Ao was convicted of corruption and money laundering in Macau on 30 January 2008 and two days later, on 1 February 2008, the POBO RO was extended for 12 months. It was served on Tam on 4 February 2008. It was against that background that Tam visited Hong Kong on 6 June 2008 and returned to Macau without staying overnight. She then declined invitations to come to Hong Kong and did not challenge the RO in respect of the Wing Hang Account until 31 January 2019, 12 years after the funds were first frozen. On those facts, it is hard to avoid the conclusion that Tam well knew that she was subject to imminent arrest if she remained in Hong Kong and had absconded in order to evade such arrest and thereafter avoided coming here, so that she came within section 8(1)(a)(ii)(B) for the purposes of justifying an RO and entertaining a CO application. 75. The Judge’s reasoning in her 2016 judgment[57] refusing a CO stands in sharp contrast with her reasoning in the 2019 judgment[58] in which she sustained the RO, stating: “There is, in fact, overwhelming evidence to suggest that these funds were and are the proceeds of crime and that a crime has been committed in the jurisdiction of Hong Kong. ... It is a matter for the respondent if she wishes to submit to the jurisdiction or not, but it certainly is not to be to her advantage that she chooses not to do so where these monies are concerned.”[59] Her Ladyship added: “There came a point in time, after 6 June 2008, when she no longer wished to cooperate, albeit there has been continued contact through her legal representatives. Proceedings are not instituted based upon the whim of the alleged offender. They are instituted when there is cogent evidence to indicate an offence has taken place within the jurisdiction, as was the case here. There has been no misuse or abuse of the system in the manner by which these proceedings were instigated or maintained. The fact that they cannot move forward at present because of the stance taken by the respondent, should not be seen to indicate that the proceedings are automatically drawn to a close or, as in this case, discharged. lf there is a lacuna in the law in this regard, it will be for the legislature to close it.”[60] 76. That is an approach more consonant with a finding that Tam was an absconder. If the errors referred to above had been avoided, her Ladyship might well have reached a different conclusion on whether Tam had “absconded”. L. Conclusion 77. For the foregoing reasons, I would allow the appeal and set aside the Court of Appeal’s order discharging the RO. If disposal of the SJ’s application for a CO had not miscarried as indicated above and had instead been pursued on its merits, its outcome would have determined the fate of the RO, one way or the other. That would have been a more normal course of events. In the circumstances which have occurred, the alternative basis upon which an application for a CO may prospectively be made rests on section 8(1)(a)(ii)(A) which depends on the SJ alleging that the person concerned has died. That is of course not something one wishes upon that person as, quite apart from the legal merits, one would generally be disposed to wish everyone a long and fruitful life. Nonetheless, in the events which have happened, as a matter of law, the RO retains its purpose of securing the frozen assets against a prospective CO made pursuant to an application under section 8(1)(a)(ii)(A). It does not fall to be discharged. Mr Justice Fok PJ: 78. I agree with the judgment of Mr Justice Ribeiro PJ on Question 1 and with that of Mr Justice Lam PJ on Question 2. Mr Justice Lam PJ: 79. I agree with the judgment of Mr Justice Ribeiro PJ on Question 1. In the following judgment, I shall address Question 2. 80. At the time when the SJ sought leave to appeal from the Appeal Committee, two applications for leave were made[61]: one following the civil route and the other one following the criminal route. Apart from the procedural avenues for bringing an appeal to this Court, there are other more fundamental implications arising from the nature of the proceedings. If the matter was criminal in nature, the judgment of the Court of Appeal of 24 March 2022 would be a nullity because: (a) The appeal did not fall within the scope of Section 13(3) of the High Court Ordinance (Cap 4) (“HCO”) which circumscribes the jurisdiction of the Court of Appeal in entertaining criminal appeals; (b) Section 34B(2) of the HCO requires a criminal appeal to be heard by not less than three judges whilst the appeal was processed at the court below on the papers and the judgment of 24 March 2022 was delivered by two judges. 81. These concerns were canvassed at the leave hearing and Question 2 was formulated to enable this Court to give a definitive ruling on the nature of appeals brought in respect of decisions made under the OSCO. 82. In the meantime, the parties took the following steps since the grant of leave on 16 September 2022[62] to cater for different scenarios: (a) To guard against the contingency that the judgment of the Court of Appeal was a nullity due to the criminal nature of the appeal, Tam sought a certification of Question 1 from Campbell-Moffat J (“the Judge”) on 8 December 2022. Based on such certification, the Appeal Committee granted leave to Tam to appeal against the judgment of the Judge of 2 December 2019. Pursuant to such leave, she brought a fresh criminal appeal[63]; (b) To guard against the contingency that the judgment of the Court of Appeal was a nullity due to the composition of the court in giving the judgment of 24 March 2022, the parties adopted the procedure suggested by this Court in Hip Hing Timber Co Ltd v Tang Man Kit[64]and invited the Court of Appeal to consider the redelivery of the judgment de bene esse after they filed a consent in accordance with Section 34B(4)(c) of the HCO. The Court of Appeal acceded to that request and re-issued its judgment on 12 December 2022. The Appeal Committee granted leave to the SJ to appeal against the re-issued judgment and the SJ brought such appeal as a fresh civil appeal[65]. The civil and criminal jurisdiction of the Court of Appeal 83. In Hong Kong, the Court of Appeal was first established under the Supreme Court Ordinance 1975[66]. Though the title of the ordinance was changed to the HCO in 1998, the civil jurisdiction of the Court of Appeal continues to be governed by Section 13(2). For present purposes, the relevant provision is Section 13(2)(a) which reads: “The civil jurisdiction of the Court of Appeal shall consist of appeals from any judgment or order of the Court of First Instance in any civil cause or matter.”[67] 84. The criminal jurisdiction of the Court of Appeal is governed by Section 13(3) which primarily cross-references the appeal avenues under the Criminal Procedure Ordinance (Cap 221) (“CPO”). Under the CPO, subject to some exceptions which are not relevant for present purposes[68], appeals are restricted to appeals against convictions and sentences. The restriction reflects a policy against the fragmentation of the criminal process, see HKSAR v Chen Keen (No 2)[69]. 85. Though it is acknowledged in Sections 13(2)(c) and 13(3)(e) that the Court of Appeal also has jurisdiction conferred on it by other statutes, there is no special provision for appeals in the OSCO except Section 8(8A) (“Subsection (8A)”) of that ordinance. That subsection deems a confiscation order made against a person (including a confiscation order made against an absconded person or a deceased person) to be a sentence passed on that person for the purposes of any ordinance conferring rights of appeal in criminal cases. Reading this subsection together with Section 83G of the CPO, an appeal against a confiscation order falls under the criminal jurisdiction of the Court of Appeal under Section 13(3)(a). 86. However, the deeming effect of Subsection (8A) has no application in respect of other decisions made under the OSCO (including the decision of the Judge which is the subject matter of these appeals). Civil or criminal cause or matter 87. Notwithstanding the differences in the relevant statutory provisions governing appeals, Hong Kong courts have referred to the UK authorities in determining whether our Court of Appeal has jurisdiction to entertain an appeal as a civil cause or matter[70]. Prior to George Tan Soon-gin v His Honour Judge Cameron[71]there were conflicting decisions in Hong Kong on the approach to be adopted[72]. George Tan was an appeal in respect of an application for judicial review of a refusal by a District Judge to grant a stay of criminal proceedings in the District Court. Barnett J dismissed the judicial review application in the High Court. When the matter came before the Court of Appeal, it was held that the matter in question was criminal in nature and therefore the court did not have jurisdiction to entertain the appeal. The court drew a dichotomy between civil and criminal causes or matters by reference to the guidance of Lord Wright in Amand v Home Secretary and Minister of Defence of Royal Netherlands Government[73]. 88. The approach of the Court of Appeal was upheld in the Privy Council. Lord Mustill made the following general observations[74]: “The language of the Ordinance directs attention, not to the proceedings which led to the order from which the appeal is brought, but to the nature of the cause or matter ‘in’ which the appeal is brought. If the cause or matter is properly characterised as criminal, it cannot lose that character simply because at one stage it is carried forward by techniques which closely resemble those employed in civil matters, or which lead to relief often granted in civil matters, or which are available in civil or criminal matters alike; any more than, having gained this new character by the employment of such techniques, it would revert to its former status when the deployment of the techniques came to an end.” 89. Reverting to the subject matter of that appeal, Lord Mustill was able to determine the character of the cause or matter by a simple analysis[75]: “Nobody could doubt that the applications made by the applicant to the district judge were applications in a criminal cause, for their purpose was to determine the way in which the prosecution should proceed. The purpose of the judicial review was to dispose of the district judge’s order so as to permit the substitution by the reviewing court of a different order, still directed to the way in which the matter should proceed. … everything happening in the present case has been no more than one stage in a continuing contest between the prosecutor and the applicant in a matter which from the outset has been exclusively criminal in nature.” (my emphases) 90. After George Tan, the law in Hong Kong was changed as the legislature deemed it unsatisfactory that no appeal could be brought to the Court of Appeal in respect of a judicial review relating to a criminal cause or matter. Subsection (3)(aa) was added to Section 13 and a new Section 14A was added to the Supreme Court Ordinance in 1993 to confer criminal jurisdiction on our Court of Appeal to hear an appeal from a judgment or order of the Court of First Instance (then known as the High Court) given or made in the exercise of the powers conferred on it under Sections 21I(1) and 21K “relating to a criminal cause or matter”[76]. So Wing Keung v Sing Tao Ltd 91. The enactment of these provisions did not resolve all issues regarding the jurisdiction of the Court of Appeal relating to a criminal cause or matter. In 2005, the Court of Appeal had to examine this question in So Wing Keung v Sing Tao Ltd[77]. The appeal in that case was brought against the decision of Hartmann J who set aside an ex parte order granted by Stone J to issue search warrants under Part XII of the Interpretation and General Clauses Ordinance (Cap 1) in respect of journalistic material. Since the proceedings were not brought by way of an application for judicial review, Sections 13(3)(aa) and 14A were not applicable. 92. Ma CJHC (as he then was) gave the leading judgment of the Court of Appeal. His Lordship came to the conclusion that the cause or matter was criminal in nature and the Court of Appeal had no jurisdiction to entertain that appeal. The primary reasons of His Lordship can be found at [31], in particular [31](7) to (13). After alluding to Amand and George Tan, His Lordship said at [31](8): “The above statement of the principles … do not, however, answer the fundamental question … as to how one is to determine whether the ‘nature’ of a cause or matter is civil or criminal.” 93. Guided by the decision of the English Court of Appeal in Carr v Atkins[78], the Chief Judge held that the matter was criminal in nature. Even though criminal proceedings had not yet begun, His Lordship emphasized that the whole point of the exercise was to further investigations into criminal offences. 94. Carr v Atkins was a case on the production of “special procedure material” under the Police and Criminal Evidence Act 1984 for a criminal investigation and the judgment was given by Sir John Donaldson MR who subsequently also penned the judgment in In re O[79]which I discuss below. In the course of his judgment, the Master of the Rolls considered an earlier Court of Appeal decision in R v Southampton Justices, Ex parte Green[80] and distinguished it as follows: “[It] is really a decision that questions of the estreatment of bail are so collateral to a criminal trial that they do not themselves constitute a criminal cause or matter.”[81] (my emphasis) 95. A similar criterion was adopted by His Lordship in In re O. Subsequently the English Court of Appeal[82] as well as the Supreme Court[83] alluded to this criterion in drawing the distinction between civil and criminal causes or matters. The nature of the cause or matter which underlies the making of restraint order 96. In the present case, the decision of the Judge was given in a set of proceedings brought under the OSCO instead of an application for judicial review. Her Ladyship exercised an original jurisdiction under the statute. Hence, as in So Wing Keung, Sections 13(3)(aa) and 14A of the HCO cannot be relied upon to confer jurisdiction on the Court of Appeal to entertain appeals from the decision. The only statutory provision which Mr Kat SC relies upon is Section 13(2)(a) which governs civil appeals. The relevant question is whether the decision of the Judge was made in a civil cause or matter. 97. Though appeals in respect of restraint orders under OSCO and other similar statutory schemes[84] in Hong Kong have always been processed through the civil route, the nature of the underlying causes or matters has never been addressed in the judgments[85]. Whilst the Court of Appeal had addressed this question in Secretary for Justice v Lin Xin Nian[86], the subject matter in question was a forfeiture order made under Section 24D of the Drug Trafficking (Recovery of Proceeds) Ordinance (Cap 405)[87] which is different in nature from a confiscation order and a restraint order. Understandably, Mr Kat referred this Court to English authorities on similar statutory regimes to support his contention that the cause or matter is civil in nature. 98. The provisions of the OSCO when it was first enacted in 1994 were largely modelled on similar provisions under Part VI of the Criminal Justice Act 1988 in England (“1988 Act”). Under the 1988 Act, the powers in relation to restraint orders were vested in the High Court whilst the power to make a confiscation order was vested in the Crown Court and the magistrates’ courts hearing criminal trials. As the making of a confiscation order was regarded as part of the sentencing process, naturally an appeal against such order was held to be criminal in nature, see R v Benjafield[88], R v Zinga[89]. 99. But this did not preclude the English courts from concluding that decisions relating to restraint orders were made in civil causes or matters. Recently, in R (Belhaj) v Director of Public Prosecutions (No 1)[90] Lord Sumption JSC held that the approach of Lord Wright in Amand had its limits and that one needed to distinguish between a decision which was an integral part of the criminal process (and was thus made in a criminal cause or matter) and an application collateral to the exercise of criminal jurisdiction (which could itself be a decision made in a civil cause or matter). As observed by the English Court of Appeal in R (McAtee) v Secretary of State for Justice[91]: “It certainly is not the law that just because the underlying proceedings are criminal in nature then any decision or step thereafter taken which has some sort of connection with those criminal proceedings is necessarily of itself a criminal cause or matter.”[92] 100. It depends on whether the decision bears a sufficiently close connection with the criminal process to treat it as being part and parcel of that process. This test is consistent with the approach of Lord Mustill in George Tan. It also accords with the limits set by Lord Sales JSC in In re McGuinness[93] on criminal cause or matter by examining if the order in question is directly consequential upon the outcome of the criminal process. 101. Mr Kat invited this Court to adopt the reasoning in two cases decided under the 1988 Act, In re O[94]and Government of the United States of America v Montgomery[95]. 102. In re O was an appeal in respect of an order for disclosure in the context of an application for variation of a restraint order. Lord Donaldson of Lymington MR drew the distinction between a judgment in a criminal cause or matter and a judgment collateral to a criminal cause or matter[96]. His Lordship cited Gooch v Ewing[97](which concerned the enforcement of a compensation order made in criminal proceedings) as an example of the latter having a civil character. Alluding to the clear dichotomy in the 1988 Act between the jurisdiction to make confiscation orders and the jurisdiction to make restraint orders and charging orders as enforcement procedures[98], His Lordship held that the enforcement procedures were wholly collateral to the criminal regime and they were civil in character[99]. His Lordship was mindful that if it were held otherwise, third parties such as banks and alleged donees of gifts would have no avenue of appeal[100]. 103. Subsequently, in Montgomery, the House of Lords endorsed this approach. It was an appeal against the discharge of a restraint order which had been made under the 1988 Act to facilitate the enforcement of a confiscation order made in criminal proceedings in the United States. Lord Hoffmann held that the nature of the proceedings in which the confiscation order was made did not necessarily render the machinery of enforcement through the courts a criminal cause or matter[101]. His Lordship went on to observe: “Modern legislation … confers powers upon criminal courts to make orders which may affect rights of property, create civil debts or disqualify people from pursuing occupations or holding office. Such orders may affect the property or obligations not only of the person against whom they are made but of third parties as well. Thus the consequences of an order in criminal proceedings may be a claim or dispute which is essentially civil in character. There is no reason why the nature of the order which gave rise to the claim or dispute should necessarily determine the nature of the proceedings in which the claim is enforced or the dispute determined.” 104. Like Lord Donaldson in In re O, Lord Hoffmann also attached significance to the unlikelihood of Parliament depriving the persons affected (including the prosecution and the defendant) any right of appeal[102]. 105. According to the analysis in these authorities, the underlying cause or matter for a decision relating to a restraint order is the regime for the enforcement of a confiscation order. Notwithstanding the criminal nature of a confiscation order, the enforcement process under the statutory regime is collateral to the criminal process and it is civil in character due to the issues that may arise and the parties potentially affected. 106. In Hong Kong, the power to make restraint orders is vested in the Court of First Instance whilst the power to make confiscation orders is vested in the Court of First Instance and the District Court. Amendments were made to the OSCO in 1995 empowering the court to make a confiscation order in respect of a person who has died or absconded. As mentioned, Subsection (8A) was added to deem a confiscation order to be a sentence passed on a person for the purposes of any statute conferring rights of appeal in criminal cases. These features did not appear in the 1988 Act[103]. 107. Notwithstanding the differences between the 1988 Act and the OSCO, the UK authorities discussed above are highly persuasive. In my view, they provide valuable insights for present purposes. Subsection (8A) does not affect the character of the cause or matter in which a decision concerning a restraint order is made. The cause or matter underlying a restraint order is not the criminal proceeding which underlies a confiscation order. As discussed above, the underlying cause or matter of a restraint order is the civil regime for the enforcement of a confiscation order. 108. There is no conflict between this analysis and the judgments in George Tan and So Wing Keung. As we have seen, in the latter case Ma CJHC had been guided by Carr v Atkins in which the distinction between proceedings collateral to the criminal process and those in a criminal cause or matter was drawn. Soon after the delivery of the judgment in So Wing Keung, the learned judge heard an appeal in Secretary for Justice v Tan Lam Chuan[104] concerning the variation of a restraint order for funds to pay legal expenses incurred in criminal proceedings. In the latter case, His Lordship did not doubt the court’s jurisdiction to hear the appeal as a civil cause or matter notwithstanding the criminal nature of the pending proceedings against the defendants. 109. The SJ accepted that a restraint order and proceedings for the same are civil in nature[105]. It was also accepted that orders made at the enforcement stage are civil in nature[106]. However, Mr Tam SC submitted that as the Judge’s refusal to grant an automatic discharge in the present case depended upon Her Ladyship’s rejection of the application for a confiscation order, the relevant cause or matter was to be determined by the nature of the proceedings for a confiscation order which is a criminal cause or matter[107]. 110. I do not accept this submission. The nature and character of a cause or matter in which a decision was given has to be determined generically. It is unhelpful to draw a distinction by reference to the issue that the court has to address in a particular application. The authority cited by Mr Tam, Secretary for Justice v Cheung Kai Yin[108], provides no support to such an approach. 111. In light of Subsection (8A), proceedings for a confiscation order are statutorily identified as criminal in nature. The underlying matter is the proceedings for the criminal conviction or notional conviction of a defendant. Since the decision on the applicability of Section 8(1)(a)(ii)(B) is made in such proceedings, it is also a decision in a criminal cause or matter. 112. On the other hand, proceedings related to a restraint order are distinct from the proceedings for a confiscation order. As discussed above, the making of a restraint order and other decisions to be made under the enforcement regime in Part IV of the OSCO are decisions made in a civil cause or matter. This would include the decision of the Judge refusing to discharge the restraint order in the present case. The importance of access to justice at the appellate level 113. Since the answer to Question 2 depends on the construction of Section 13(2) and (3) of the HCO which set out the jurisdiction of the Court of Appeal, the rights of appeal under our legal system are part of the relevant context. Under Section 13, there is a general avenue of appeal to the Court of Appeal in civil causes or matters whilst the right is more circumscribed in criminal causes or matters consistent with the policy against fragmentation. The rectification of errors at first instance in individual cases arising from a trial should be addressed by an appeal against conviction and/or sentence pursuant to Sections 82 and 83G of the CPO. 114. Further, with the enactment of Sections 13(3)(aa) and 14A of the HCO, there is a general right of appeal to the Court of Appeal in applications for judicial review related to a criminal cause or matter. In this respect, Hong Kong’s appellate regime is more generous than the regime in the UK. Notwithstanding this, there are cases (including decisions made under the OSCO) that fall outside the scope of these sections which have some connections with the criminal process. The viability of appealing to the Court of Appeal in those cases depends on the meaning of “civil cause or matter” in Section 13(2)(a). 115. Moreover, this Court has no jurisdiction to hear any appeal from a decision of the Court of First Instance under Section 31(b) of the Hong Kong Court of Final Appeal Ordinance (Cap 484) which is not final in nature, see HKSAR v Chen Keen (No 2)[109]. Since a restraint order is not a final order[110], there is no avenue of appeal under the Hong Kong regime if an appeal in respect of a decision concerning a restraint order is characterized as a decision in a criminal cause or matter[111]. 116. In In re McGuinness[112], the Supreme Court had to address the nature of the cause or matter in issue in an appeal from Northern Ireland because the parties proceeded on the basis that it was an appeal against a judicial review decision in a criminal cause or matter. Under the relevant statutory regime, like that in England and Wales, such appeal could not be brought to the Court of Appeal of Northern Ireland. Lord Sales JSC made the forceful observation that an overly expansive interpretation of the phrase “criminal cause or matter” would unacceptably reduce access to justice at appellate level[113] and that any restriction on such access beyond the paradigm type of criminal proceedings had to be justified[114]. 117. These observations are equally apposite in the present context. As mentioned, Lord Donaldson and Lord Hoffmann also attached great weight to access to appellate justice by those affected by the enforcement regime under the 1988 Act. 118. I respectfully agree with the sentiment expressed by Lord Hoffmann in Montgomery: “… it seems to me very improbable that Parliament intended that there should be no right of appeal from orders made in the High Court under Part VI. … It seems clear to me that Parliament made no special provision for appeals because it considered that there was an appeal under the general jurisdiction of the Court of Appeal to hear appeals from the High Court. Of course Parliament may have been under a misapprehension, but this conclusion would produce such an unfortunate result that I would not accept it unless I felt compelled to do so.”[115] 119. I can find nothing in the legislative materials and the text of the OSCO indicating an intention on the part of our legislature to forestall appeals to the Court of Appeal in respect of orders made under Part IV of the OSCO for the enforcement of a confiscation order. In light of the generous appellate regime for judicial review relating to criminal causes or matters in Hong Kong, I cannot see any justification for treating persons affected by a restraint order differently by depriving them of the potential means of seeking judicial redress through appellate avenues. Since proceedings on restraint orders have a different character and are separate from the criminal trial, the availability of appeals would not disrupt the criminal process and the policy against fragmentation which justifies the restriction on appeals in criminal causes or matters is not engaged. Answer to Question 2 120. My answer to Question 2 is that an appeal against a judge’s refusal to discharge a restraint order under the OSCO is an appeal against a decision in a civil cause or matter. The Court of Appeal has jurisdiction to entertain the same under Section 13(2)(a) of the HCO. As it was an appeal against an interlocutory decision, it could be heard by a two-member court. In giving the judgment of 24 March 2022, the Court of Appeal was properly constituted. It follows that this Court should decide Question 1 in the context of the appeal in FACV 12/2022 and there is no need to make any order in the other appeals referred to above. Mr Justice French NPJ: 121. I agree with the judgment of Mr Justice Ribeiro PJ on Question 1 and with that of Mr Justice Lam PJ on Question 2. Chief Justice Cheung: 122. The appeal in FACV 12/2022 is accordingly unanimously allowed and the Court of Appeal’s Order discharging the RO is set aside. The Court makes an order nisi that the costs of this appeal and the proceedings below be paid by Tam to the Secretary for Justice. Absent an application to vary, the order nisi shall become absolute after the lapse of 14 days from the date hereof. 123. The appeals in FACC 4/2022, FACC 1/2023 and FACV 2/2023 are dismissed with no order as to costs. Mr William Tam SC, DDPP, Ms Denise Chan, SADPP and Mr Douglas Lau, PP, of the Department of Justice, for the Appellant in FACC 4/2022, FACV 12/2022 and FACV 2/2023 Mr Nigel Kat SC and Mr Benson Tsoi, instructed by H M Tsang & Co, for the Appellant in FACC 1/2023 [1] Cap 455. [2] Ribeiro and Lam PJJ and Tang NPJ [2022] HKCFA 19 (16 September 2022). [3] Cap 201. [4] POBO section 14C(1)(a). [5] POBO section 14C(3). [6] POBO section 14C(4)(a). [7] OSCO section 25(1) provides: “Subject to section 25A, a person commits an offence if, knowing or having reasonable grounds to believe that any property in whole or in part directly or indirectly represents any person’s proceeds of an indictable offence, he deals with that property.” [8] OSCO section 15(1) provides: “The Court of First Instance may by order (referred to in this Ordinance as a restraint order) prohibit any person from dealing with any realisable property, subject to such conditions and exceptions as may be specified in the order.” [9] Sections 14(1)(a) and (b) and 15(5)(b). [10] HCMP 1207/2014. [11] Ibid at §18. [12] [2019] HKCFI 2931 (2 December 2019). [13] Cheung and Chow JJA [2022] HKCA 454 (24 March 2022), Chow JA writing for the Court. [14] Considering it unnecessary to deal with other arguments advanced. [15] Section 19(2) is set out in Section F.3 of this judgment. [16] [2005] 3 HKC 545 at §15. [17] [2022] HKCA 454 at §§25-27. [18] Ibid at §32. [19] Ibid. [20] Section 2(16A)(b). [21] Section 2(1) and Schedules 1 and 2. An offence contrary to section 25 of OSCO is one such listed offence. [22] Section 2(8). [23] Section 8(1)(a)(i). [24] Section 8(3)(a)(ii). [25] Section 8(4)(a). [26] Sections 8(1)(a)(ii)(A) and 8(3). [27] Sections 8(1)(a)(ii)(B) and 8(3). [28] Section 8(3)(c)(i)(A). [29] Sections 8(8), 8(9) and 10. [30] Sections 8(3)(b)(i)(B) and 8(3)(c)(i)(C). [31] Section 8(4)(b). [32] Section 8(3)(c)(i)(B). [33] Section 8(8D). [34] Sections 8(6), 8(7), 9 and 11(1). [35] Section 13. [36] Section 14(1)(a). By section 2(15) one means by which proceedings are instituted involves the issue by a magistrate of an arrest warrant under section 72 of the Magistrates Ordinance (Cap 227) as occurred in respect of Tam on 12 January 2011. [37] Section 14(1)(b). [38] Section 14(1)(c)(ii). [39] Section 8(4)(a). [40] In Section F.2 above. [41] HCMP 1207/2014 at §18. [42] [2001] 1 AC 268 (HL). [43] Respondent’s case at §§36-38. [44] Ibid at §48(6). Italics in the original. [45] Ibid at §49. [46] Respondent’s case at §37. [47] Section 14(1)(a). Section 2(15) provides that the issue of an arrest warrant is one means by which proceedings for an offence under that Ordinance are “instituted” as occurred here. [48] Section 14(1)(b). [49] Section 14(1)(c)(ii). [50] Section 14(1B). [51] HCMP 1207/2014 (10 August 2016). [52] Ibid at §11. [53] Ibid. [54] The SJ states in his written submissions (at §18) that in 2020, 70.8% and in 2021, 58.9% of CO applications involved “absconder” cases. [55] HCMP 1207/2014 at §12. [56] The Shorter Oxford English Dictionary gives a number of meanings: “Hide (oneself); leave hurriedly and secretly; flee from justice.” [57] HCMP 1207/2014 (10 August 2016). [58] [2019] HKCFI 2931 (2 December 2019). [59] Ibid at §11. [60] Ibid at §15. [61] FAMC 13/2022 and FAMV 367/2022. [62] For the criminal appeal in FACC 4/2022 and the civil appeal in FACV 12/2022. [63] In FACC 1/2023. [64] (2004) 7 HKCFAR 212 at [27]. [65] In FACV 2/2023. [66] Ordinance No 92 of 1975. [67] Since its enactment in 1975, this subsection has not been changed except that the reference to the High Court was amended to the Court of First Instance due to the change in nomenclature of the court. [68] Section 81E provides for SJ’s appeal against a discharge under Section 16 or 79G of the CPO or Section 22 of the Complex Commercial Crimes Ordinance (Cap 394). Section 81F provides for SJ’s appeal against an order quashing an indictment. Section 83J provides for appeal by the defendant against a verdict of not guilty by reason of insanity. Section 83M provides for appeal by the defendant against a finding of disability. There are also provisions for reservation or reference of question of law to the Court of Appeal under Sections 81 and 81D and provisions for review of sentence by the SJ under Sections 81A to 81C. [69] (2022) 25 HKCFAR 34 at [17]. [70] The relevant statutory provisions discussed in the UK authorities are Section 18(1) of the Senior Courts Act 1981 and its equivalent which provide that no appeal shall lie to the Court of Appeal from any judgment of the High Court in “any criminal cause or matter”. The discussion in the authorities focused on whether a decision of the lower court was made in a criminal cause or matter. [71] [1992] 1 HKLR 149 (CA) and [1992] 2 AC 205 (PC). [72] In an application by the Attorney General for orders of certiorari and mandamus [1972] HKLR 336; In re an application for judicial review by a firm of Solicitors [1990] 2 HKLR 146; Attorney General v Alick, Au Shui-yuen [1992] 1 HKLR 88. [73] [1943] AC 147. The relevant part of Lord Wright’s judgment was cited by Yang CJ in [1992] 1 HKLR 149 at p.153. See also the judgment of Silke V-P at p.171. [74] [1992] 2 AC 205 at p.221. [75] Ibid. [76] In this respect, the position in England is different. An appeal against a decision of the Divisional Court in a judicial review concerning a criminal cause or matter is subject to stringent criteria and has to be brought directly to the Supreme Court, see Section 1(1)(a) of the Administration of Justice Act 1960. See also [42] of the Court of Appeal’s judgment in R (McAtee) v Secretary of State for Justice [2019] 1 WLR 3766 which was expressly endorsed by Lord Sales JSC in In re McGuinness (Attorney General for Northern Ireland and others intervening) [2021] AC 392 at [90]. [77] [2005] 2 HKLRD 11. Previously, the Court of Appeal in Apple Daily Ltd v Commissioner of the Independent Commission Against Corruption (No 2) [2000] 1 HKLRD 647 had proceeded on the basis that an application pertaining to a search warrant issued under Cap 1 was civil in nature notwithstanding that there were underlying criminal investigations. However, notwithstanding a tentative view expressed in that regard at p.658, Keith JA ultimately chose to dispose of the case on the merits without reaching a final conclusion on the nature of the cause or matter. Hence, his view was obiter and Ma CJHC in So Wing Keung was entitled to depart from such view in light of fuller arguments in the latter case, see [2005] 2 HKLRD 11 at [31](14). [78] [1987] 1 QB 963. [79] [1991] 2 QB 520. [80] [1976] 1 QB 11 in which the court held that an order estreating the recognizance provided by a third party was not criminal in nature. Although this was challenged in a subsequent case, the House of Lords did not hold it to be incorrect as to its outcome, see In re Smalley [1985] 1 AC 622 at p.634. [81] [1987] 1 QB 963 at p.970G. The view was first expressed by Sir John Donaldson MR in Day v Grant (Note) [1987] 1 QB 972 at p.976H after due consideration of Amand. [82] R (Guardian News & Media Ltd) v City of Westminster Magistrates’ Court [2011] 1 WLR 3253 at [29] and [36]. [83] In re McGuinness [2021] AC 392 at [75]. [84] Drug Trafficking (Recovery of Proceeds) Ordinance (Cap 405); Mutual Legal Assistance in Criminal Matters Ordinance (Cap 525). [85] Secretary for Justice v Guardecade Ltd (CACV 47/2004, 25 May 2004); Secretary for Justice v Tan Lam Chuan [2005] 3 HKC 545; Secretary for Justice v Hui Yat Fai (CACV 187/2004, 8 February 2005); Secretary for Justice v Zhang Hongbao (CACV 120/2008, 24 December 2010); Secretary for Justice v Wu Li Hui [2009] 1 HKLRD 78; Secretary for Justice v Zhenly Ye Gon [2012] 2 HKC 523; Re Yeung Ka Sing Carson [2012] 3 HKLRD 491; Secretary for Justice v Male Ye Zhiqiang (CACV 195/2012, 12 July 2013); Ortmann v Secretary for Justice [2015] 1 HKLRD 240; Re Kim Dotcom (CAMP 123/2019, 6 November 2019). [86] [2001] 2 HKLRD 851. [87] It is well settled in Hong Kong that forfeiture proceedings have a civil character, see Wong Hon Sun v HKSAR (2009) 12 HKCFAR 877. The issue in that case was whether articles 10 and 11(1) of the Hong Kong Bill of Rights were engaged in such proceedings. This Court held that the proceedings did not comprise the determination of a criminal charge and therefore those articles were not engaged. [88] [2003] 1 AC 1099. [89] [2014] 1 WLR 2228. [90] [2019] AC 593 at p.633A to D. [91] [2019] 1 WLR 3766 at [41]. [92] See also R (Guardian News & Media Ltd) v City of Westminster Magistrates’ Court [2011] 1 WLR 3253 and R v Southampton Justices, Ex parte Green [1976] QB 11 where the underlying criminal proceedings were not determinative of the nature of the cause or matter in question. [93] [2021] AC 392 at [76]. [94] [1991] 2 QB 520. [95] [2001] 1 WLR 196. [96] [1991] 2 QB 520 at p.527B. [97] [1986] 1 QB 791. [98] [1991] 2 QB 520 at p.527H to 528B. [99] Ibid at p.528B. [100] Ibid at p.528C. [101] [2001] 1 WLR 196 at [19]. [102] Ibid at [23]. [103] The regime under Part VI of the 1988 Act was replaced in the UK by the Proceeds of Crime Act in 2002. Whilst the 2002 Act has provisions for the making of confiscation orders against absconded or deceased persons, it also contains express provisions for appeals (including appeals in respect of restraint orders) to be brought to the criminal division of the Court of Appeal in England, see the Proceeds of Crime Act 2002 (Appeals under Part 2) Order 2003 and the Criminal Procedure Rules 2020 Part 42. In light of that, the position under the 2002 Act is not germane for our purposes. [104] [2005] 3 HKC 545. [105] SJ’s Supplemental Case at para 3. [106] SJ’s Supplemental Case at para 6. [107] SJ’s Supplemental Case at paras 8 and 9. [108] [2016] 4 HKLRD 367. [109] (2022) 25 HKCFAR 34. [110] Ortmann v Secretary for Justice [2015] 1 HKLRD 240. [111] In the UK, an appeal from any decision of the High Court in a criminal cause or matter shall be brought to the Supreme Court, thus not confining to a “final decision”, see Section 41(1) of the Judicature (Northern Ireland) Act 1978 and Section 1(1)(a) of the Administration of Justice Act 1960. [112] [2021] AC 392. [113] Ibid at [68]. [114] Ibid at [45]. [115] [2001] 1 WLR 196 at [23]. A similar sentiment was expressed by Lord Bridge of Harwich in In re Smalley [1985] 1 AC 622 at p.643A to C in the construction of Sections 28(1) and 29(3) of the Supreme Court Act 1981 which precluded judicial review of a judgment or decision of the Crown Court relating to trial on indictment. 
Chief Justice Cheung: 1. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Mr Justice Ribeiro PJ and Mr Justice Fok PJ: A. Introduction 2. Every resident over the age of 11 is required to register for a Hong Kong Identity Card (“HKID card”).[1] Failure to produce a HKID card when required to do so by a police officer or member of the Immigration Service is an offence.[2] Production and inspection of ID cards is ubiquitous, not just in dealings with government officials but routinely in a wide variety of everyday transactions as a means of verifying a person’s identity. Thus, one may be asked to produce one’s ID card when entering a building, applying for a job, using banking services, signing up for a mobile phone plan or making a medical appointment, just to take a few examples. 3. On its face, the ID card contains the holder’s photograph and states the holder’s name and date of birth. It also indicates whether the holder is male or female (the “gender marker”), a feature intended to function as an “identifier”, helping to verify the identity of the person producing the card as its holder. It should be noted that the gender marker does not signify recognition of the holder’s sex as a matter of law.[3] It merely operates as an element of an identification document. 4. The two appellants are female to male (“FtM”) transgender persons who were diagnosed as suffering from gender dysphoria, a medical condition involving much distress and discomfort arising out of the discordance they experienced between the (female) sex assigned to them at birth and the (male) gender with which they intrinsically identified. Having undergone a lengthy course of medical and surgical treatment designed to affirm their male gender identity resulting in conforming bodily changes, the gender dysphoria of each of the appellants has been medically certified to have been sufficiently attenuated to enable their social integration and psychological well-being without the need for additional surgical procedures. For medical purposes they may thus be regarded as having transitioned from their assigned female gender to the acquired male gender. In this judgment the appellants are referred to using masculine pronouns and the focus of discussion is on FtM cases, for the most part leaving aside male to female (“MtF”) cases for present purposes. 5. The appellants applied to the Commissioner of Registration (“the Commissioner”) to have the gender markers on their HKID cards amended to reflect their acquired gender. Their evidence is that unamended gender markers cause them to suffer discrimination, humiliation, violation of their dignity and invasion of their privacy resulting from having to reveal to third parties their transgender status when producing their HKID cards. 6. The Commissioner refused their applications on the basis that they had not undergone certain surgical procedures required under published guidelines[4] to qualify them for a change to the gender markers on their ID cards. The appellants consequently brought judicial review proceedings to challenge that decision, contending that the Commissioner’s refusal violates their constitutional right to privacy under Article 14 of the Bill of Rights (“BOR 14”).[5] 7. Their applications before the Court of First Instance[6] and the Court of Appeal[7] were both dismissed. Leave to appeal to this Court was granted by the Court of Appeal.[8] B. The diagnosis and treatments for gender dysphoria 8. The medical evidence[9] is to the following effect. While the precise causes of gender dysphoria are not presently known, the consensus is that it is a biological condition and not a lifestyle choice. Dr Winter explains that, “... under all the diagnostic criteria under ICD-10, ICD-11 and DSM-5, the core feature of diagnosis is a misalignment between experienced gender and assigned sex”.[10] Gender dysphoria is the discomfort or distress that arises out of and is related to such gender incongruence. 9. The distress and discomfort may be of varying degrees and may result from the incongruence between the individual’s gender identity and the birth-assigned gender as recognised by others in social interactions (“social dysphoria”), and/or from the mismatch between the person’s gender identity and bodily sex characteristics (“physical dysphoria”). 10. Social dysphoria involves distress and discomfort resulting from transgender persons being marked by a social stigma often leading to harassment, abuse and discrimination across many areas of everyday life. Gender markers on identification documents are an important factor in this context. 11. Physical dysphoria involves transgender persons experiencing distress regarding their bodies and a desire to bring about a physical change. In acute cases, if left untreated, this may lead to associated self-harm and suicidal behaviour. 12. The clinical condition of gender dysphoria patients will vary and individualised treatments are required. The Hospital Authority has, since about 1980, been providing health care in Hong Kong for persons with gender identity issues.[11] In line with practice elsewhere, cases are approached on a multi-disciplinary basis. As Dr Ho explains,[12] teams comprising psychiatrists, clinical psychologists, surgeons, gynaecologists, physicians, endocrinologists, occupational therapists and medical social workers work collaboratively, with treatments fashioned to meet individual patients’ needs. 13. The “treatment pathway” generally adopted involves an initial assessment by psychiatrists and clinical psychologists and, upon confirmation of a diagnosis of gender dysphoria, a 12-month period of “real life experience”, ie, living life as a member of the experienced gender with support and guidance from mental health professionals. If the real life experience is deemed successful and if the patient is assessed to be psychiatrically ready for hormonal treatment, he is referred to an endocrinologist to start treatment. 14. In FtM cases, testosterone is prescribed and the therapy can be expected to produce marked bodily changes involving the development of male characteristics. As Dr Ng explains, during the first three months there is normally: “... cessation of menses, increased libido, increased facial and body hair, increased oiliness of skin, increased muscle and redistribution of fat mass.” After a year, “changes including deepening of voice and clitoromegaly [enlargement of the clitoris] are expected to occur, and some individuals may experience male pattern hair loss.”[13] 15. If the patient is assessed to have used the hormones continuously and responsibly for 12 months and if surgical treatment is desired, the psychiatrist may refer the patient to the surgical team for assessment. 16. A range of surgical procedures may be considered. As Dr Safer explains, the treatment involves “changing outward appearance to match gender identity to the extent required by the transperson”, the goal being “to ensure that they have successfully transitioned their appearance to align with their gender identity.”[14] 17. Prof Monstrey points out that “[the] first (and, arguably, most important) surgery performed in FTM is the creation of a male chest by means of subcutaneous mastectomy [removal of the breasts] which allows the patient to live more easily in the preferred male gender role (ie external physical appearance in day-to-day social settings)...”[15] The evidence shows that “a significant number of transgender persons find that hormones, and/or breast surgery, are sufficiently effective to physically alter their body so as to alleviate their feelings of discomfort or distress about their body (their physical dysphoria).”[16] As Dr Winter observes: “It is only when the bodily dysphoria results in distress which cannot be resolved by less intrusive methods will a clinician assess and recommend more intrusive surgical options as they are medically necessary.”[17] 18. Those options may involve, as the ultimate surgical intervention, full sex reassignment surgery (“SRS”), ie, genital surgery for transgender men which “comprises hysterectomy (removal of the uterus), oophorectomy (removal of the ovaries), vaginectomy (removal of the vagina), and phallus construction. Phallus construction may be by way of phalloplasty or metoidioplasty. Phalloplasty is a particularly complicated set of procedures aimed at creating a realistic phallus, with best results not only allowing urination while standing, but also the enjoyment of sexual sensation, and erectile function (through insertion of a prosthesis). Metoidioplasty is a somewhat easier surgical procedure involving the unhooking of the hood covering the clitoris, the latter enlarged through testosterone hormone therapy. Both types of surgery may also involve construction of a scrotal sac containing testicular prostheses.”[18] This is major surgery and full SRS generally takes place in stages and carries certain post-operative risks and possible urologic complications.[19] 19. The foregoing medical evidence provides indispensable background for discussion of the issues arising on these appeals. It must however be borne in mind that the Court is presently concerned with a judicial review challenge to the Commissioner’s refusal to change a gender marker on an identification document and not with determining the sex of each of the appellants (in the light of their treatment and medically certified transition) as a matter of law. 20. The medical witnesses provide valuable explanations of the diagnosis and treatment for gender dysphoria, but with a different orientation. Thus, Dr Chiu evidently regards successful treatment as in some sense measurable by reference to completion of “change of sex”, stating: “From the point of view of the patients’ own desires and well-being, change of sex may be considered completed when the patients’ own dysphoria is attenuated enough for their social integration and psychological well-being ...”[20] Similarly, Dr Winter states: “... there is a consensus among contemporary transgender healthcare providers that a transgender person’s change of sex is complete when their gender dysphoria is reduced to such an extent that enables them to live and be accepted as a member of their experienced gender.”[21] 21. A discussion of the criteria which should be applied for determining whether individuals should be regarded in law as having changed their sex may be of great relevance in the context of considering a potential gender recognition scheme or otherwise determining whether a transgender person qualifies as a member of his preferred sex for other legal purposes. But such issues do not arise on these appeals. 22. The Commissioner’s policy which is pivotally in issue involves his insistence upon full SRS before a transgender individual is permitted an amendment to the ID card gender marker (“the Policy”). The foregoing medical evidence is of great importance in the discussion of its constitutionality. It provides an essential appreciation of the nature and clinical implications of the Policy and locates full SRS at the most invasive end of the spectrum of treatments for gender dysphoria. C. The circumstances of Q and Tse C.1 Q 23. The appellant Q was born in Hong Kong in 1992. Q’s sex at birth was registered as “female” and, when aged 11, Q was issued with a HKID card which stated the holder’s gender as “female”. Q had a desire to be male from an early age, understood about gender dysphoria from the age of 16 and identified as transgendered at the age of 19. He has lived as a male since he went abroad in 2011 to study, initially in the UK and later in Canada. 24. In August 2012, Q was assessed for gender dysphoria by Dr Mak Kai Lok at the Prince of Wales Hospital and was prescribed testosterone pills which he took daily. After returning to the UK in 2013, he continued to receive psychiatric treatment under Dr Richard James Curtis and to live as a male. He continued to receive testosterone injections. In March 2014, Dr Curtis confirmed to the UK Passport Office that Q was under his care for the treatment of female to male transsexualism and that this change was likely to be permanent. Q was issued with a new passport in 2014 showing his gender as male. 25. Having been issued with an adult HKID card as a permanent resident of Hong Kong in 2010 in his birth name in Chinese and English (with a feminine English forename), he changed his name by Deed Poll in 2013 adopting a masculine forename. On his application to the Commissioner to amend the registered particulars on his HKID card, Q was issued with a replacement HKID card in July 2013 in which his photograph was updated and his English forename was changed to a masculine forename as per the Deed Poll. 26. In August 2014, Q again consulted Dr Mak, who certified that Q had masculine features, had adapted well in a male role socially and psychologically and was psychiatrically fit to proceed to surgical assessment for female to male sex reassignment surgery. 27. In addition to receiving continuous testosterone treatment, as a result of which he is medically sterile and does not menstruate, Q underwent an irreversible mastectomy in August 2015 to remove all breast tissue. However, he has made an informed decision not to undergo further sex reassignment surgery including a hysterectomy, to remove the uterus, and an oophorectomy, to remove the ovaries, because of the health risks involved and possible pain and complications. His evidence confirms that, post-mastectomy, he is comfortable with his body as it now is and does not feel any psychological need to undergo medical sterilisation or reconstructive organ surgery to feel comfortable with his male identity. 28. Q has formed new relationships with friends who accept him as male and, save for members of his family, all his friends and acquaintances know him only as male. C.2 Tse 29. The appellant Tse was born in Hong Kong in 1991. Like Q, Tse’s sex at birth was registered as “female” and, when aged 11, Tse was issued with an HKID card which stated the holder’s sex as “female”. Tse completed the HKCEE and GCSE at a well-known all girls school in Hong Kong in 2008 and then completed secondary school at a co-educational school in the UK before matriculating at the University of Warwick in 2011, graduating in 2015. 30. Tse had gender dysphoria from a young age and identified as male, despite attempts by his family to force him to act and dress as a girl. After moving to study in the UK, Tse was referred by his general practitioner to a psychologist and psychiatrist at a gender identity clinic in London. Tse was assessed by Dr Penny Lenihan, a consultant psychologist, to have a history of gender dysphoria, with a presentation consistent with likely transsexualism, in April 2012. This assessment was confirmed in September 2012 by Dr Stuart Lorimer, a consultant psychiatrist. Tse was referred to an endocrinologist who further referred him to a general practitioner to prescribe androgen hormone treatment, which he took for two years from 2012 to 2014. This caused physical changes including growing an Adam’s apple and facial hair and developing a more muscular physique. 31. Tse changed his name by Deed Poll in the UK in August 2012 to his current male name. The UK Passport Office accepted his application for a new passport, which was issued to him in his new name and with a male gender marker. In the summer of 2013, Tse changed his name in Hong Kong by Deed Poll. 32. Tse was issued an adult HKID card as a permanent resident of Hong Kong in his birth names in Chinese and English (with a female English forename) in 2009. In July 2013, on his application to the Commissioner to amend the registered particulars on his HKID card, Tse was issued with a replacement HKID card in which his photograph was updated and his names in Chinese and English changed (that in English being changed to Henry Edward Tse in accordance with the change effected by Deed Poll). 33. On returning to the UK, Tse underwent a bilateral mastectomy to remove both breasts, since the breasts he had developed were a cause of discomfort and embarrassment. He interrupted his androgen hormone treatment in order that his eggs could be harvested for possible use by a future female spouse and has since then resumed hormone treatment which he intends to continue indefinitely. 34. In July 2016, after living as a male for four years, Tse was issued in the UK with a Gender Recognition Certificate as a male. By a letter dated 16 November 2016, Tse’s treating physician, Dr Malik Saoudi, confirmed his opinion that further medical surgery was not necessary to treat his gender dysphoria. D. Appellants’ application to amend their HKID cards 35. Regulation 18(1)(a) of the RPR places a duty on holders to make a report to a registration officer whenever ID card particulars previously submitted “have become incorrect”. Failure to do so without reasonable excuse is an offence.[22] 36. In the light of the medical treatment that they had each received, the appellants applied to the Commissioner to alter the gender markers on their respective ID cards. Their applications were made under RPR 14 which empowers a registration officer to issue a replacement ID card with such alterations “only ... after the production of such evidence, under oath or otherwise as he may require; and after such investigation as he may consider necessary.” Both applications were refused. 37. Regulation 18 is evidently intended to buttress the function of HKID cards as a means of verifying the holder’s identity. If particulars on the document have become incorrect, that function is obviously compromised so that corrective alterations should be made. The amendment process is not mechanistic, but requires the registration officer to assess the evidence in support of the proposed alteration and to pursue any further investigations considered necessary. This bears on the Commissioner’s guidelines and the objective of having “bright line” criteria discussed below. E. The Commissioner’s Guidelines 38. In refusing the appellants’ application, the Commissioner applied the Policy contained in guidelines issued on 5 April 2012 (“the Guidelines”) which relevantly provide as follows: “Generally speaking, persons who have received different forms of treatments by professional psychiatrists and clinical psychologists, including psychotherapy, hormonal treatment and real-life experience of the chosen gender role for a period of time may be recommended for sex re-assignment surgery (SRS). Persons who have undergone the above treatments and have completed SRS should follow the below procedures and submit application together with the relevant supporting documents to reflect their change of sex on their identity cards: (a) produce a medical proof which should indicate that the following criteria for the completion of SRS are met: (i) for sex change from female to male • removal of the uterus and ovaries; and • construction of a penis or some form of a penis; (ii) for sex change from male to female • removal of the penis and testes; and • construction of a vagina; (b) In general, the medical proof should be produced by the doctor who performed the SRS in accordance with the criteria as set out above; (c) If the SRS was performed outside Hong Kong, the medical proof should carry the doctor’s medical qualification, place where the medical qualification is obtained and other contact information of the doctor; (d) Where there are difficulties in obtaining the relevant medical proof from the doctor who performed the SRS outside Hong Kong, the applicant may request a Hong Kong registered doctor to give an assessment on the SRS that has been undergone; ... (f) Upon receipt of the relevant documents, consideration will be given to whether to allow amendment of the personal particular sought by an applicant having regard to the particular circumstances of the case.” 39. The Guidelines therefore make it a condition that a FtM transgender applicant seeking an amendment to the ID card gender marker must have “completed SRS”, that is, he must have undergone surgery effecting removal of the uterus and ovaries and construction of “a penis or some form of a penis”. In other words, he must have had a hysterectomy and surgical genital reconstruction to qualify for alteration of the gender marker. As we have seen, such surgical procedures are at the most invasive end of the treatment spectrum for gender dysphoria and, as the medical evidence shows, a full SRS is not medically required by many transgender persons (including the appellants) whose gender dysphoria has been effectively treated, and who are successfully living in their acquired gender. F. The challenge by way of judicial review and BOR 14 40. In the case of a constitutional challenge alleging a violation of a constitutional right or freedom, it is first necessary to identify whether a constitutional right is engaged. The next question is to ask whether the impugned provision or conduct amounts to an encroachment on such right, being an interference with, or restriction of, that right. If so, unless the constitutional right is absolute, a proportionality assessment must then be undertaken to determine whether such interference with the right can be justified.[23] 41. The appellants’ applications for judicial review challenges the Policy insofar as it lays down the aforesaid condition for changing the gender marker on the basis that it is an unlawful interference with their constitutional rights under BOR 14 which provides: “(1) No one shall be subjected to arbitrary or unlawful interference with his privacy, family, home or correspondence, nor to unlawful attacks on his honour and reputation. (2) Everyone has the right to the protection of the law against such interference or attacks.” 42. BOR 14 is in the same terms as Article 17 of the International Covenant on Civil and Political Rights (“ICCPR”). By virtue of Article 39 of the Basic Law, the rights and freedoms in the ICCPR as applied to Hong Kong and incorporated via the BOR are protected and given constitutional effect.[24] The rights under BOR 14 are not absolute and may be restricted as prescribed by law. 43. In the present case, there is no dispute between the parties that the rights protected under BOR 14 include the right to gender identity and the right to physical integrity. This was accepted by the Court of Appeal,[25] which rightly observed that gender identity is one of the most crucial identities of a person since it concerns who people are and what sort of people they identify with, directs their personal development and behaviour, governs their relationships and interaction with others and underpins most of their societal arrangements.[26] 44. In so holding, the Court of Appeal were correctly adopting an interpretation of the concept of privacy under BOR 14 consistent with the materially equivalent concept of respect for private life in Article 8 of the European Convention on Human Rights (“ECHR” and “ECHR 8”).[27] In this context, the ECtHR has held that “… ‘private life’ is a broad term not susceptible to exhaustive definition. It includes not only a person’s physical and psychological integrity …, but can sometimes also embrace aspects of an individual’s physical and social identity …. Elements such as gender identification, names, sexual orientation and sexual life fall within the personal sphere protected by [ECHR 8] … [It] also protects the right to personal development and the right to establish and develop relationships.” Accordingly, the ECtHR has held: “… that, as the very essence of the Convention is respect for human dignity and human freedom, the right of transgender persons to personal development and to physical and moral security is guaranteed.”[28] 45. In AP, Garçon and Nicot v France,[29] the ECtHR re-stated the above propositions and concluded that: “The right to respect for private life under [ECHR 8] applies fully to gender identity, as a component of personal identity. This holds true for all individuals.”[30] 46. BOR 14 is clearly engaged in the present case. Privacy is a concept inherently linked to a person’s dignity. The Policy concerns the appellants’ eligibility for an altered ID card gender marker which reflects their acquired gender to enable them to conduct their lives and affairs consistently with their experienced gender. The refusal to allow an amendment to the gender marker involves, as the evidence discloses, humiliation, distress and loss of dignity in routine activities involving the inspection of their HKID cards. Furthermore, the Policy’s condition that they undergo full SRS requires them to make a choice between accepting frequent infringements of their BOR14 rights to privacy when using unamended ID cards and undergoing major invasive and medically unnecessary surgery. 47. The constitutional right under BOR14 being clearly engaged and the Policy constituting an encroachment upon these rights, it falls to be determined whether the Policy can be justified as satisfying the test of proportionality, the burden being on the Commissioner to provide such justification.[31] 48. The established proportionality assessment in Hong Kong involves a structured four-step inquiry which asks: (1) whether the encroachment pursues a legitimate aim; (2) whether such encroachment is rationally connected with achieving that aim; (3) whether the encroachment represents a proportionate means of achieving that aim; and (4) whether a reasonable balance has been struck between the societal benefits of the impugned measure and the individual’s constitutionally protected right or freedom.[32] G. Legitimate aim and rational connection 49. The aim of the Policy, as formulated by the Commissioner, is “to establish a fair, clear, consistent, certain and objective administrative guideline to decide when a change of the sex entry on the identity card is to be accepted”. That aim was held by the Judge to be legitimate,[33] a holding endorsed by the Court of Appeal[34] and not disputed by the appellants. It was also accepted by the appellants that the Guidelines setting out the Policy were rationally connected to that legitimate aim.[35] 50. It is of course generally desirable and legitimate that clear guidelines should be drawn up to give direction to those administering a policy and to inform those affected by it. And (leaving aside the word “fair” and subject to our comments on the Policy’s exemption) the requirement for certification of completion of SRS is clear enough to suggest a rational connection. 51. However, the case is not resolved simply by considering whether the Policy requirements are clear and thus conducive to administrative certainty. The focus in the Courts below has rightly been on the content of the Policy itself – ie, the full SRS condition for amending the gender marker – and as to whether its incursion into constitutional privacy rights can be justified as proportionate. H. The margin of discretion 52. In a proportionality analysis, the margin of discretion available to the decision maker is sometimes a matter of debate. Where a wide margin of discretion is called for, the “manifestly without reasonable foundation” threshold may be appropriate. In cases where a narrow margin of appreciation is available, the correct test may be one of “reasonable necessity”.[36] But as previously recognised, these two tests are not wholly independent concepts but instead points on a continuous “reasonableness” spectrum by which the court determines the intensity of judicial scrutiny.[37] 53. In the Court of First Instance, Au J rejected the Commissioner’s submission that the proportionality of the Policy should be examined on the “manifestly without reasonable foundation” standard of scrutiny. He held that even though the issue in question also concerns public and social interests, since the right to gender identity and physical integrity are essential fundamental human rights and core values, the court should accord a narrow margin of discretion to the Commissioner and approach the question of proportionality on the “no more than reasonably necessary” standard of scrutiny.[38] 54. The Court of Appeal agreed, dismissing the Commissioner’s respondent’s notice contending that Au J had erred and arguing instead for a “manifestly without reasonable foundation” standard. The Court of Appeal held that the Policy engaged “core values relating to personal or human characteristics in terms of gender identity and physical integrity” so that it must be subject to the court’s vigilant scrutiny by the more stringent standard.[39] 55. In doing so, both Au J and the Court of Appeal applied the approach laid down by the Court in Fok Chun Wa v Hospital Authority,[40] a case concerning the allocation of public funds and limited financial resources in the context of the provision of subsidised obstetric services in public hospitals. There, a distinction was drawn between “core-values relating to personal or human characteristics (such as race, colour, gender, sexual orientation, religion, politics, or social origin)” which “involve the respect and dignity that society accords to a human being” and are “fundamental societal values”, where the more stringent standard would apply, and a question of general, social or economic policy, where more leeway would be permitted.[41] The Courts below in this case accepted that the Policy engaged such core values and therefore required that the proportionality analysis be conducted on the “no more than reasonably necessary” basis.[42] 56. Before this Court, the Commissioner renewed his submission that the appropriate standard of review was not the more stringent standard, inviting the Court to clarify, for future reference, the correct approach in deciding where on the spectrum the standard should be set. 57. The Policy concerns the question of gender identity and as such it clearly addresses matters relating to personal or human characteristics, or what were referred to as “core values”. It was the submission of Ms Carss-Frisk KC, for the Commissioner,[43] however, that it was wrong to apply the stringent standard whenever such values or characteristics are involved since this would be to apply a mechanical and inflexible approach. Instead, she submitted, a more flexible approach was called for.[44] Since the right to privacy was not absolute, respect for an individual’s privacy should give way when brought into contact with public life or in conflict with other protected interests[45] as was the case here. The impact on the public of a change of gender marker on a HKID card for someone who had not completed full SRS was an important matter involving social policy making and was morally and ethically sensitive, hence a wide margin of discretion should be accorded the Commissioner. 58. We do not accept this contention and conclude that the Courts below were correct in applying the “no more than reasonably necessary” standard of scrutiny. 59. As the Court held in Hysan, factors relevant to choosing the basis for assessing an impugned provision include the significance and extent of interference with the right in question[46] and the identity of the decision-maker as well as the measure’s content and features relevant to the margin of discretion.[47] As already stated, the Policy concerns the expression of an individual’s gender identity on a HKID card and a requirement to undergo extensive surgical intervention as a condition of a change of gender marker. These are clearly core values which engage a narrower standard of scrutiny. 60. The Commissioner contended that a wider margin of discretion is appropriate because of the lack of consensus in different jurisdictions and the involvement of sensitive moral and ethical issues.[48] However, it is important to recognise the distinguishing features in the authorities relied upon by the Commissioner in support of his contention. 61. The present case is not one where the issue of a person’s sexual status for all legal purposes is involved. The challenge to the Policy concerns merely the correction of a gender marker on an identification document which does not affect legal status. The issues in these appeals do not engage the need to have regard to any relevant consensus across different jurisdictions. Nor do they give rise to complications about the relationship of inter-linked legislation across different contexts as would arguably be the case if the appeal concerned the question of gender recognition generally,[49] which was addressed in the consultation paper of the Inter-departmental Working Group on Gender Recognition (“IWG”).[50] 62. Accordingly, it is appropriate to conduct Step Three of the proportionality analysis applying the standard of scrutiny of “no more than reasonably necessary”. I. Is the Policy no more than reasonably necessary? I.1 The Commissioner’s three justifications 63. As the Court of Appeal noted,[51] the Commissioner puts forward three main reasons why the Policy, drawing the line at full SRS, is justified. First, he argues that “a full SRS is the only workable, objective and verifiable criterion to enable a registration officer to determine the application”.[52] Secondly, he submits that practical administrative problems due to incongruence between the external physical appearance of the holder and the gender marker would arise if some other line was drawn.[53] And thirdly, he argues that “hormonal and psychiatric treatments that precede full SRS are not absolutely irreversible”, giving rise to a risk that a “FtM pre-operative transgender person, whose sex entry on the identity card has been changed to male, stops hormonal treatment, recovers fertility, becomes pregnant, and gives birth”.[54] I.2 Choice of an invasive surgical intervention as the criterion 64. Before proceeding with an analysis of those justifications, a striking feature of the Policy may be noted. As observed above, the function and purpose of the gender marker in HKID cards is to help verify the identity of the holder. It does not signify recognition of the holder’s sexual status as a matter of law. Thus, it would have been rational to adopt a policy accepting an amendment to that marker if its verification function is impaired because of an incongruence between the holder’s appearance and the contents of the ID card, as is likely to occur in the case of a transgender person. Indeed, such amendments are prescribed by RPR 18(1)(a) whenever the ID card particulars previously submitted “have become incorrect”. RPR 14 empowers the registration officer to decide whether there should be a change after examining the evidence and after such investigation as he may consider necessary. 65. However, the Commissioner has taken an entirely different approach. The Policy adopts as the criterion for alteration of the gender marker, completion of surgery to remove the uterus and ovaries and to construct “a penis or some form of a penis”. As previously noted, this makes amendment conditional on undergoing the most invasive surgical intervention in the range of treatments for gender dysphoria – treatment which may be medically unnecessary for many transgender persons (including the appellants). Such a criterion might logically be put forward (although increasingly rejected in many jurisdictions[55]) when debating the appropriate basis for recognising a person’s change of sex for all legal purposes. But that is not in issue in the present case. As the evidence indicates, some transgender persons feel pressured to undergo such surgery just to get a replacement ID card in order to avoid the frequent experience of discrimination, humiliation, violation of their dignity and invasion of their privacy. 66. Such pressure is objectionable in principle. As the Strasbourg Court pointed out in AP, Garçon and Nicot v France: “...medical treatment must always be administered in the best interests of the individual and adjusted to her/his specific needs and situation”.[56] It should not be prescribed merely to promote administrative convenience or clarity. The Court added: “Medical treatment cannot be considered to be the subject of genuine consent when the fact of not submitting to it deprives the person concerned of the full exercise of his or her right to gender identity and personal development, which, as previously stated, is a fundamental aspect of the right to respect for private life...”[57] 67. Accordingly, in our view, the adoption of such a criterion weighs significantly against the Policy in assessing its proportionality. I.3 Is full SRS the only workable, objective and verifiable criterion? 68. The Commissioner elaborates upon his proposition that “a full SRS is the only workable, objective and verifiable criterion”[58] making two points, namely: (a) That “[anything] less than that may amount to self-declaration which cannot be accepted”; and (b) That the decision would otherwise “... be left to the judgment of individual medical practitioners involved in different applications to certify if the change of sex had been completed” giving rise to “... certificates based on varying standards, [resulting] in arbitrariness, inconsistency in treatment and unfairness.”[59] 69. As to the first point, it is, with respect, quite untenable to suggest that a line drawn at requiring full SRS is the sole workable line and that the only alternative would involve self-declaration. 70. The Guidelines operate by laying down what must be medically certified. They presently require certification of full SRS. However, as is acknowledged in the Commissioner’s own evidence and as noted by the Court of Appeal,[60] an exception exists permitting certification of different medical reasons and consideration of applications by the registration officer on a case-by-case basis. Thus, after setting out the Guidelines, a Sub-Divisional Instruction dated 3 April 2012,[61] states: “In regard to individual cases with justifiable medical reasons that the SRS cannot be completed, the case officer (IO) may require the applicant to provide further information on a case-by-case basis. The authority for approving these applications rest with SIO while for applications to be refused, they should be referred to CIO for decision.” 71. There is also the evidence of Tsui Yat, Assistant Secretary for Security,[62] who traces the genesis of the Guidelines to a lunch meeting discussion on 25 February 2011.[63] The “Brief note” of that discussion records agreement as to the “new criteria” for accepting “change of sex on HKIC” later reflected in the Guidelines. However, while FtM cases generally require proof of both a hysterectomy and “some form of male genital reconstruction”, it was agreed that the second requirement admits of possible discretionary exceptions: “If the applicant have removed all the original sex organs without reconstruction of opposite sex organs due to special reasons (eg health reasons), case will be considered exceptionally with expert advices from competent authorities.” 72. These exceptions show that requiring full SRS is not the only line that can be drawn for deciding whether alterations to gender markers should be made. Other criteria, short of requiring full SRS, are plainly workable. As the exceptions also demonstrate (and as RPR 14 acknowledges), such decisions need not be made mechanistically, applying some one-size-fits-all criterion, but can be approached on a case-by-case basis. The administrative burden is not great since the evidence suggests that there are likely to be relatively few applications from transgender persons.[64] 73. The abovementioned conclusion is supported by evidence which shows that in numerous other jurisdictions, criteria short of full SRS are regularly applied in deciding whether gender markers on identification documents should be changed. Such criteria are also applied, with more far-reaching implications, in determining whether a transgender person should be recognised to have changed his or her sex for all legal purposes. Referring to the IWG Consultation Paper, examples of different policies which do not require full SRS adopted in other jurisdictions (as at June 2017) are given in the Appellants’ Case[65] as follows: “(1) The Australian States, the Canadian States, Belgium, Bolivia, Croatia, Ecuador, Estonia, France, Iceland, Italy, Netherlands, Norway, Germany, Portugal, Spain, Sweden, Ukraine, United Kingdom, around half of the United States of America, Uruguay, and Vietnam require a declaration plus medical evidence to be submitted to a Court or to the relevant government department or administrative authority without the need for SRS. (2) Finland and Poland require a declaration, and medical evidence of sterilisation through hormone treatment but not SRS. (3) Courts in Austria, Colombia, Greece, Hungary, Luxembourg, Romania, South Korea, Switzerland and Taiwan have ruled that applications to amend identity documents can be made to either to the courts or to the relevant administrative authority without need for proof of SRS.” 74. An illustration of a scheme of certification which does not require SRS can be found in the United Kingdom’s Gender Recognition Act 2004. This provides for a Gender Recognition Panel comprising legal and medical members reviewing specified medical evidence and granting a Gender Recognition Certificate if satisfied that the applicant is over the age of 18 and (a) has or has had gender dysphoria, (b) has lived in the acquired gender throughout the period of two years ending with the date on which the application is made, (c) intends to continue to live in the acquired gender until death, and (d) complies with the evidential requirements imposed by the Act. 75. The IWG Consultation Paper[66] records that since 2006/2007 the number of applicants for Gender Recognition Certificates has been steady, comprising approximately 300 cases per annum in the UK. As the appellants also point out, there is no evidence that such a model has caused administrative difficulty in the aforementioned jurisdictions, “far less difficulties of a sufficient degree to lead any of those countries to amend their legislation or policies to require SRS”.[67] 76. It should incidentally be noted that the appellants have made it clear that they are not suggesting that the Policy should be replaced by a scheme of self-certification. Such a possibility is not an issue under consideration by the Court. Both appellants have submitted certificates from specialist medical practitioners both in Hong Kong and the United Kingdom, providing detailed accounts of their psychiatric, real life experience, hormonal and surgical treatments short of SRS. They have been medically certified as having effectively transitioned to living lives in their acquired gender without the need for further surgery. They are not advocating self-certification. 77. The Commissioner’s second point seeks to justify the full SRS Policy on the basis that drawing a different line might involve having to deal with medical certificates based on varying and inconsistent standards. We do not consider that a point of any substance. 78. The possibility of questionable certification arises whatever the Policy may require to be medically certified. The Guidelines set out in Section E of this judgment cater for this in relation to certification of completion of full SRS. They could obviously make similar provision in respect of such other certification as may be required. Thus, the present Guidelines provide that the medical proof should normally be produced by the doctor who performed the SRS and if the operation was done outside Hong Kong, certification “should carry the doctor’s medical qualification, place where the medical qualification is obtained and other contact information of the doctor”. The Guidelines go on to state that where there are difficulties in obtaining the proof from the doctor who performed the SRS outside Hong Kong, “the applicant may request a Hong Kong registered doctor to give an assessment on the SRS that has been undergone”.[68] Plainly, if the line were to be differently drawn, the Commissioner could stipulate what constitutes acceptable certification and could in case of doubt, for instance, require endorsement by government doctors after their own examination of the applicant. I.4 Practical administrative problems as justification for the Policy 79. The second main justification of the Policy offered by the Commissioner involves administrative problems thought likely to arise if a line short of full SRS were to be drawn. The argument, as summarised by the Court of Appeal, runs as follows: “The sex entry on identity cards are used by law enforcement agencies, operators and frontline staff of various organizations and service providers, including government departments, as an indicator of the gender of the holder in many aspects of everyday life. Such aspects range from law enforcement, emergency responses, publicly funded social and residential services, social hygiene services, in-patient services in hospitals, access to sex-specific public toilets, sports, and enrolment in single-sex schools. The requirement of full SRS is necessary to avoid the practical difficulties which would be caused if the external physical appearance of the holder is incongruent with the sex entry thereon.”[69] (footnote omitted) Elaboration is provided by Wong Him Yu, Government Counsel, who elicited the views of various government bureaux.[70] 80. With respect, for the reasons given below, the argument is very much over-stated and unconvincing as a justification for insisting on full SRS. I.4a Unreal examples 81. First, we are bound to say that several of the instances of purported practical problems relied on by the Commissioner are highly contrived and have no realistic bearing on the issues at hand. 82. Thus, under the heading “Emergency Responses” the deponent[71] postulates a “chemical, biological, radiological and nuclear attack” and envisages officers assisting victims of the same sex and the use of decontaminating water spray facilities which require victims to strip off their clothes in sexually segregated locations. The suggestion is that confusion as to the victim’s sex or gender “may bring embarrassment to the subject as well as the officers and other patients/victims and even complaints of misconduct of a sexual nature against various persons concerned”. It is fanciful that in such dire circumstances, the niceties of sexual modesty would have any bearing on the urgent emergency responses required. It is in any event hard to imagine that victims would first be asked to produce their ID cards before being given emergency assistance. Even in less far-fetched cases involving, for instance, ambulance service officers responding to a serious accident, the argument is unconvincing. No one would expect ambulance men to refuse urgent assistance to a badly injured female victim and to wait for a female officer to arrive. In such urgent situations any incongruence between the victim’s apparent sex and the ID card gender marker is rendered insignificant. 83. Another instance relied on by the Commissioner involves single-sex schools. The deponent reports that the system “for allocation of students to Primary 1 and Secondary 1 respectively is sex-specific” and asserts: “… if there is incongruence between the sex identified on the HKIC and the external physical attributes of the person concerned, it will bring confusion to the allocation systems and hence grave embarrassment to teachers and students who are young in age and particularly sensitive to gender differences, specifically the physical attributes.”[72] 84. That suggestion loses touch with the present issues. Children allocated Primary 1 and Secondary 1 school places are aged about 6 and 12 respectively. Children under the age of 11 are exempt from registering for a HKID card. Moreover, as Dr Ho states, SRS will not be performed on anyone under the age of 18 and it is “... very uncommon that [transgender patients can] go through all stages of assessment for SRS before the age of 21”.[73] Accordingly, whatever other problems candidates for Primary 1 and Secondary 1 school places may pose, such problems do not include issues arising out of incongruence between a transgender person’s appearance and the gender marker and the Policy requiring full SRS. 85. Another questionable example involves recruitment for the disciplined services. The deponent states: “While the sex of an applicant is not a relevant consideration for recruitment in civilian grades, certain disciplined service grades set different physique requirements for male and female candidates.” It is suggested that practical difficulties “including the fairness of the physical recruitment examinations” might arise “if there is incongruence between the [gender marker] and the external physical attributes” of the applicant.[74] The logic is not easy to follow. While a candidate’s ID card would no doubt be inspected (with possible external incongruence difficulties, as discussed below), any special physique or fitness requirements would surely be subject to physical tests rather than mere reliance on the gender marker before a decision on recruitment is taken. It is hard to see how any of this justifies the Policy. I.4b External incongruence 86. Another unsatisfactory aspect of the Commissioner’s “practical problem” justification of the Policy involves a somewhat unfocussed approach to “incongruence”. He asserts that the requirement of full SRS should be adhered to in order to avoid practical problems which otherwise arise due to an incongruence between the physical appearance of the transgender person and the ID card gender marker. However, this assertion fails to distinguish between what may be called “external incongruence” and any incongruence arising out of a FtM man’s retention of female genital organs and his lack of a surgical male genital reconstruction. 87. The incongruence which regularly exposes transgender persons to violation of their dignity and invasion of their privacy most commonly involves the discordance between their outward appearance (rather than the appearance of their genital area) and the unamended gender marker, when their ID cards are produced for inspection. This is what we call “external incongruence”. A FtM person who has undergone hormonal treatment and is living as a male will generally present himself and be regarded by others as a male. He may have facial hair, an Adam’s apple, a deeper voice and a male physique with increased muscle and redistributed fat, as well as a male hairstyle, clothing and demeanour, giving rise to possible external incongruence due to an unaltered gender marker. If a gender marker amendment had been made so that his external appearance was in line with his gender marker, such incongruence and any associated problems would be far less likely to arise, if at all. It is misplaced to suggest that “practical problems” involving external incongruence are somehow avoided by adhering to the full SRS Policy and refusing alterations to the gender marker on that basis. 88. As Lord Pannick KC pointed out, it is only rarely that exposure of a person’s genital area is required. Indeed, as the appellants have pointed out in their evidence, they go to great lengths to avoid being placed in that position as a matter of everyday experience. 89. In the great majority of cases of possible external incongruence, leaving the gender marker unamended produces greater confusion or embarrassment. For example, if a transgender man who had not had full SRS but whose external appearance was in every respect male was to enter a women’s public lavatory, the reaction of the women using the facility would almost certainly be one of consternation. It would be of little relevance for them to be told that he had entered because his gender marker stated that he was female and that he was using the women’s lavatory (he believed) to avoid being prosecuted.[75] Yet the Commissioner’s evidence suggests that lawful use of such public conveniences should be in accordance with the gender marker’s designation: “According to FHB [Food and Health Bureau], should the situation so warrant, the [lavatory] attendant may report the case to the Police for assistance and inspection of the subject's HKIC. If the HKIC sex entry cannot be reliably referred to in ascertaining the actual physical sex of a subject, there would be practical difficulties in enforcing the law when there is incongruence between the sex identified on the HKIC and the external physical attributes of the person concerned, as well as grave embarrassment for, and even complaints of misconduct of a sexual nature by or against, other users of the facility.”[76] 90. The implication that the gender marker establishes the holder’s “actual sex” as a matter of law is unsound. The construction and application of the Regulations governing conduct in public conveniences are not issues arising on the present appeals. Nevertheless, in practical terms, it seems clear that in the aforesaid example, if, instead of entering the women’s lavatory, the transgender person, appearing in all external respects to be a man, walked into the men’s facility, no one would have raised an eyebrow. After much experience of living in his acquired gender, he could be relied on to deal with his own transgender needs in a sensible and discreet way, such as by using a cubicle to ensure privacy. 91. Similar considerations would arise in connection with the various sex-appropriate accommodation or residential arrangements referred to by the Commissioner by way of justification. Thus, for instance, external incongruence would be far more likely to raise concerns if a transgender man with a male appearance was admitted to a female hospital ward because his gender marker said “female” than if he was admitted to a male ward. It is true that he might well receive a medical examination in the male ward which would reveal that he retained female genitalia and lacked any surgically reconstructed male organs. However, in practical terms such an examination is unlikely to concern other men in the ward as it is likely to be conducted in privacy behind screens, and one would expect his attending doctors and nurses to understand and accommodate his transgender status. One might add that in a situation where consent is given to receiving such medical care, privacy rights are likely to that extent to have been waived. This also applies to the Commissioner’s example of “practical problems” relating to a transgender person’s attendance at a Social Hygiene Clinic providing treatment for sexually transmitted diseases. 92. To take a final example of possible problems due to external incongruence being exacerbated by unamended gender markers, one may consider police officers making routine ID checks. If a transgender person whose external appearance is male were to produce an ID card which stated that he was female, this might cause the officer to ask numerous questions and perhaps to doubt whether he was the lawful holder of the document, leading to the embarrassment, humiliation, violation of dignity and invasion of privacy complained of in this appeal. In such a case, the unamended gender marker’s function as an identifier would be deficient. The officer, in conducting the stop and search operation, might properly be interested in checking whether there was any outstanding warrant for the holder’s arrest or whether he was an illegal immigrant, and so forth. But the officer would have no legitimate interest in delving into the holder’s transgender history. If, on the other hand, the gender marker had been amended to correspond to the holder’s external appearance, the external incongruence issue would not arise without compromising the ability of the officer to carry out his lawful duties. I.4c Problems that do not bear on gender markers 93. Nothing in the foregoing discussion is meant to deny that there are many areas of society where genuine and difficult issues concerning the appropriate treatment of transgender persons arise. However, it is frequently the case that amendment of the ID card gender marker and the Policy demanding full SRS as a condition of amendment are irrelevant to resolving such difficulties. Instead, the problems call for appropriate social arrangements to be made and measures devised to accommodate both the legitimate needs of transgender persons and the needs of other members of the public. It therefore does not follow that pointing to the existence of such issues provides a justification for the Policy’s interference with the constitutional rights engaged. 94. To take one example, it is undeniable that difficult problems may arise as to whether a transgender person sentenced to imprisonment should be incarcerated in a male or female prison. Since intrusive and intimate bodily searches are likely to be required, should such searches be conducted by male or female prison officers? Would a male or female prison be more appropriate, taking into account the need to protect the transgender inmate or other inmates from possible sexual abuse? These are real problems, but their existence does not bear on the Policy’s justification. 95. A factual illustration may be found in Navarro Luigi Recasa v Commissioner of Correctional Services[77] which involved a MtF transgender person who had undergone hormonal treatment and breast augmentation surgery and had acquired a female external appearance with a feminine physique, although her male genitalia remained intact. She was convicted of trafficking in a dangerous drug and sentenced to 20 months’ imprisonment. She was held in a male prison but, because of her female appearance and the risk of sexual harassment, she was housed in the prison’s Vulnerable Prisoners Unit. The decision to do so, involving a pragmatic protective measure, was upheld by the Court. However, her complaint about the practice of male prison officers conducting strip or cavity searches, which the Judge regarded as involving a discretionary decision by the prison authorities, failed on the facts. Plainly, it would have made little difference to those problems whether her ID card gender marker stated that she was male or female. The same difficult issues which arose out of her transgender status and her sentence of imprisonment would have had to be faced. 96. Similar considerations apply to the Commissioner’s reference to fairness in sport. There are undoubtedly controversial issues regarding fairness in the classification of transgender athletes who compete in sport. They are issues addressed by various international sporting organisations but their resolution has little to do with the ID card gender marker of the person concerned. I.5 Reversibility 97. The Commissioner’s third main justification of the Policy is that the treatment received by transgender persons is “not absolutely irreversible”, so that there is a risk that a “FtM pre-operative transgender person, whose sex entry on the identity card has been changed to male, stops hormonal treatment, recovers fertility, becomes pregnant, and gives birth”.[78] The argument is therefore that the Policy justifiably requires irreversible removal of the female reproductive organs to exclude the possibility of a post-transition pregnancy. 98. Dr Safer acknowledges that the concern expressed by the Commissioner cannot be ruled out but states that “such occurrences would be truly exceptional”, adding “... in my years of practice, I have never personally come across a patient who under proper care and guidance of a qualified endocrinologist, and conscientious with medication has conceived while on testosterone therapy at the recommended levels”.[79] 99. The exceptional rarity of such a pregnancy is unsurprising. This is because in the great majority of cases, a FtM transgender person’s commitment to achieving a permanent transition to the male gender is plain and obvious, even if full SRS is not performed. The lengthy treatment pathway undertaken has been described. And as Dr Safer points out, elements of FtM hormonal treatment including facial hair growth, lowered vocal pitch and changed larynx and midline structures, are irreversible. A bilateral mastectomy is also obviously irreversible. Thus, as pointed out in W v Registrar of Marriages,[80] transgender persons who have undergone such a course of treatment have shown themselves “willing to endure such a long and painful ordeal to acquire a body which conforms as far as possible with their self-perception and to struggle for social recognition”. Such persons are highly unlikely to decide to revert to their assigned gender, much less to do so with a view to becoming pregnant. 100. What such an exceptional course may have to involve is illustrated by R (McConnell) v Registrar General for England and Wales (AIRE Centre intervening),[81] an extremely unusual case. The applicant, Mr McConnell, had been registered as female at birth and, when 22 years of age, transitioned to live in the male gender, undergoing hormonal treatment and a double mastectomy, but not full SRS. He obtained a Gender Recognition Certificate confirming that he was male, having declared that he “intend[ed] to continue to live in the acquired gender until death”. However, he then suspended testosterone treatment and commenced fertility treatment intending to fertilise an egg in his womb. He subsequently underwent intrauterine insemination during which donor sperm was placed inside his uterus. The process was successful and he became pregnant, carrying the pregnancy to full-term and giving birth to a son. He brought legal proceedings to challenge the Registrar’s decision to register him as the child’s “mother”, contending that he had a right to be registered as the boy’s “father” or, if not, as his “parent”. His application failed for reasons that do not require discussion here. 101. What the McConnell case illustrates is how an extremely elaborate and medically-assisted course of action had to be followed for a transitioned FtM person to achieve pregnancy. Indeed, in claiming to be registered as the child’s “father”, Mr McConnell appears to have wanted it both ways: to affirm his transition to the male gender and thus to be called “father” while having his eggs fertilised and giving birth to a child. Such cases must obviously be extremely rare. There is no evidence of there having been such a case in Hong Kong. 102. In our view, it would be wholly disproportionate to regard the risk of a rare and exceptional post-transition FtM pregnancy as a justification for the Policy, thereby requiring all FtM transgender persons to have full SRS as a necessary condition for acquiring a change to their HKID card gender markers. I.6 Conclusion as to reasonable necessity 103. For the reasons developed above, we are unable to accept any of the three reasons advanced by the Commissioner by way of justification for the Policy. We do not accept that it represents the only workable, objective and verifiable criterion for altering the gender marker. Nor do we accept that the Policy’s full SRS criterion is justified by a need to avoid the alleged practical problems discussed above. And as we have just stated, the exceedingly small risk of post-transition reversibility leading to pregnancy cannot justify adherence to the Policy. Moreover, as pointed out in Section I.2 above, it is objectionable in principle to adopt as the criterion for amending a gender marker, a requirement of undergoing a highly invasive surgical intervention which may be medically unnecessary. 104. The Commissioner has failed to demonstrate that the Policy on which his decision to refuse the appellants’ application for an amendment to their gender markers is based, is no more than reasonably necessary to accommodate his legitimate concerns and to justify interference with the appellants’ BOR 14 rights. It is therefore our view, respectfully differing from the Courts below, that the Policy fails the test of reasonable necessity and is disproportionate. J. Step Four: striking a reasonable balance 105. Since the Policy and the Commissioner’s decision have failed the proportionality test, it is strictly unnecessary to go on to consider the fourth step of the proportionality analysis. However, as was noted in Hysan,[82] the four requirements inevitably overlap because the same facts are likely to be relevant to more than one of them. 106. If it had been necessary to proceed to Step Four and to ask “whether a reasonable balance has been struck between the societal benefits of the encroachment and the inroads made into the constitutionally protected rights of the individual, asking in particular whether pursuit of the societal interest results in an unacceptably harsh burden on the individual”[83] we would have answered “No” to first part of that question and “Yes” to its second part. 107. The foregoing analysis leads to the conclusion that the societal benefits of the Policy are in many respects illusory and are at best relatively slim. The Policy’s consequence is to place persons like the appellants in the dilemma of having to choose whether to suffer regular violations of their privacy rights or to undergo highly invasive and medically unnecessary surgery, infringing their right to bodily integrity. Clearly this does not reflect a reasonable balance. The Policy imposes an unacceptably harsh burden on the individuals concerned. K. Conclusion and disposition 108. For the foregoing reasons, we would allow the appeals. We would hold that each of the appellants is entitled to an Order quashing the Commissioner’s decision refusing their applications for alteration of the gender markers on their HKID cards. 109. We would also grant a Declaration that the aforesaid decisions and the underlying Policy requiring FtM transgender persons to undergo full SRS as set out in the Guidelines as a necessary condition for altering the gender markers on their HKID cards, violate the appellants’ BOR 14 rights and are unconstitutional. 110. As Lord Pannick KC acknowledged, it is not for the Court to re-write the Commissioner’s Policy. As indicated above, there are various models and approaches that might be considered for re-formulating the Policy in a manner consistent with the rights protected under BOR 14. 111. We would also make an Order nisi that the costs of these appeals and of the proceedings below be paid by the respondent to the appellants with liberty to the parties, if so advised, to lodge written submissions as to costs within 14 days of the date of this judgment. Mr Justice Lam PJ: 112. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Lord Sumption NPJ: 113. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Chief Justice Cheung: 114. The Court unanimously allows the appeals and quashes the Commissioner’s decisions refusing the appellants’ respective applications for alteration of the gender markers on their Hong Kong Identity Cards. 115. We grant a Declaration that the aforesaid decisions and the underlying Policy requiring Female to Male transgender persons to undergo full Sex Reassignment Surgery as set out in the Guidelines as a necessary condition for altering gender markers on Hong Kong Identity Cards, violate the appellants’ rights under Article 14 of the Bill of Rights and are unconstitutional. 116. We make an Order nisi that the costs of these appeals and the proceedings below be paid by the respondent to the appellants with liberty to the parties, if so advised, to lodge written submissions as to costs within 14 days of the date of this judgment. Lord Pannick KC, Mr Hectar Pun SC and Mr Earl Deng, instructed by Haldanes, assigned by the Director of Legal Aid, for the Applicants (Appellants) Ms Monica Carss-Frisk KC, Mr Stewart Wong SC and Ms Bonnie Y.K. Cheng, instructed by the Department of Justice, for the Respondent [1] Registration of Persons Ordinance (Cap 177) (“RPO”) section 3; Registration of Persons Regulations (Cap 177A) (“RPR”) regulation 25(g). [2] RPR 11. [3] RPO sections 9 and 7(2)(j); RPR 11A; Tsui Yat (Security Bureau) Affirmation 26 July 2016, §5: “[The gender marker] does not connote any general or formal or legal recognition of the person’s sex or gender as such. An HKIC serves the fundamental purpose of identifying the holder as a particular individual, with certain details or particulars put in as identification features of the individual such as the name, date of birth and sex.” [4] See Section E of this judgment. [5] Claims initially made under Article 3 (cruel, inhuman or degrading treatment) and Article 22 (discrimination) of the Bill of Rights as well as under the Sex Discrimination Ordinance (Cap 480) are no longer pursued. [6] Au J [2019] 1 HKLRD 1244; [2019] HKCFI 295. [7] Poon CJHC, Kwan VP and Barma JA [2022] 1 HKLRD 803; [2022] HKCA 172. [8] [2022] HKCA 675. [9] Dr Ho Pui Tat (psychiatrist, “Dr Ho”) Affirmation 25 July 2016; Dr Ng Wan Sze Vanessa (endocrinologist, “Dr Ng”) Affirmation 26 July 2016; Dr Chiu Tor Wo (plastic surgeon, “Dr Chiu”) Affirmations 26 July 2016 and 14 September 2017 giving evidence for the Commissioner. Dr Stephen John Winter (psychologist, “Dr Winter”) Affidavit 21 December 2016; Dr Joshua David Safer (endocrinologist, “Dr Safer”) Affirmation 22 December 2016; Professor Dr Stanislas Jozef Maria Monstrey (plastic surgeon, “Prof Monstrey”) Affirmation 31 December 2016 giving evidence for the appellants. While certain differences of opinion are expressed, those differences are largely immaterial for present purposes. [10] Dr Winter at §43. To similar effect, Dr Ho at §7. “ICD-10, ICD-11” refer to the 10th and 11th revisions of the International Statistical Classification of Diseases and Related Health Problems issued by the World Health Organisation and “DSM-5” refers to the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders published by the American Psychiatric Association. The editions cited were current at the time the evidence was filed. [11] W v Registrar of Marriages (2013) 16 HKCFAR 112 at §15. [12] Dr Ho at §15. [13] Dr Ng at §7. [14] Dr Safer at §§19 and 21. [15] Prof Monstrey at §35. [16] Dr Winter at §56. [17] Ibid at §57. [18] Ibid at §69. [19] Prof Monstrey at §§69 and 73. [20] Dr Chiu Affirmation 26 July 2016 at §30. [21] Dr Winter at §60. [22] RPR 19(1)(a). [23] Catholic Diocese of Hong Kong v Secretary for Justice (2011) 14 HKCFAR 754 at §65. [24] Comilang Milagros Tecson v Director of Immigration (2019) 22 HKCFAR 59 at §§24-25. [25] CA at §30. [26] Ibid at §31, citing the European Court of Human Rights (“ECtHR”) in YY v Turkey, Application No. 14793/08, Judgment dated 10 March 2015. [27] See, in this context, ZN v Secretary for Justice [2019] HKCFA 53; (2020) 23 HKCFAR 15 at §60. [28] YY v Turkey, Application No. 14793/08, Judgment dated 10 March 2015 at §§56-60. [29] Application Nos 79885/12, 52471/13 and 52596/13, Judgment dated 6 April 2017. [30] Ibid at §95. [31] Secretary for Justice v Yau Yuk Lung (2007) 10 HKCFAR 335 at §21; Fok Chun Wa v Hospital Authority (2012) 15 HKCFAR 409 at §§56 and 60. [32] Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372 at §§134-135. [33] Au J at §§17 and 28. [34] CA at §§6(1) and 46. [35] Appellants’ Case (“AC”) §53. [36] Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372 at §106. [37] Ibid at §119-122. [38] Au J at §§46-47. [39] CA at §§49-50. [40] (2012) 15 HKCFAR 409. [41] Ibid at §§77-78. [42] Respondent’s Case at §§8-9. [43] Appearing with Mr Stewart Wong SC and Ms Bonnie Y.K. Cheng. [44] Kwok Cheuk Kin v Secretary for Constitutional and Mainland Affairs (2017) 20 HKCFAR 353 at §37; R (SC) v Secretary of State for Work and Pensions [2022] AC 223 at §195. [45] Democratic Party v Secretary for Justice [2007] 2 HKLRD 804 at §§59, 64-65. [46] (2016) 19 HKCFAR 372 at §§105-107, 108-113. [47] Ibid at §§114-118. [48] Hamalainen v Finland (2014) 37 BHRC 55, Application No. 37359/09, Judgment dated 16 July 2014 at §67; R (McConnell) v Registrar General for England and Wales (AIRE Centre intervening) [2021] Fam 77 at §§81-82; R (Elan-Cane) v Secretary of State for the Home Department [2022] 2 WLR 133 at §62. [49] Cf R (McConnell) v Registrar General for England and Wales (AIRE Centre intervening) [2021] Fam 77 at §§62 and 82. [50] Dated June 2017. [51] CA at §51. [52] Ibid at §52. [53] Ibid at §53. [54] Ibid at §54. [55] As listed (as at June 2017) in the IWG Consultation Paper at §§6.55-6.57. [56] Application Nos 79885/12, 52471/13 and 52596/13, 6 April 2017 at §73 citing the Commissioner for Human Rights of the Council of Europe. [57] Ibid at §130. [58] CA at §52; see Tsui Yat (Security Bureau) Affirmation 26 July 2016, §28. [59] Ibid. [60] CA at §1. [61] Extract of Registration of Persons Sub-Divisional Instruction No 1/2012, §5. [62] Affirmation 26 July 2016. [63] Attended by three doctors and four civil servants. [64] Dr Ho affirms (at §15) that in 2013-14, 121 individuals received services under Psychiatric SOPCs [Specialist Out-Patient Clinics], 12 of whom underwent SRS; Dr Ng states (Affirmation at §1) that she took care of about 60 patients in the first 7 months of 2016; and Dr Chiu (Affirmation §1) deposes in 2016 to having treated 20 patients since 2014. Citing a government Press Release dated 9 December 2015, the IWG reported as follows: “The HA [Hospital Authority] estimated that around 30 new cases with gender identity disorder or gender dysphoria would be referred for psychiatric assessment per year, and that around one in 10 of these would require assessment for SRS. According to the HA, the number of gender identity disorder/gender dysphoria patients who underwent partial or full SRS in each of the five years from 2010/11 to 2014/15 is, respectively, 4, 2, 6, 12 and 16.” [65] At §77. [66] At §3.86. [67] AC at §77. [68] Guidelines §§(b)-(d). [69] CA at §53. [70] Affirmation 26 July 2016 at §§5-22. [71] Ibid at §7. [72] Ibid at §22. [73] Dr Ho at §4(c)(ii). [74] Wong Him Yu Affirmation at §21. [75] Section 7 of the Public Conveniences (Conduct and Behaviour) Regulation, Cap l32BL, provides that no male or female person shall enter any part of the public convenience which is allocated for the use of the persons of the opposite sex. [76] Wong Him Yu Affirmation 26 July 2016, §19. [77] [2018] 4 HKLRD 38 (Au J). The case proceeded on grounds other than BOR 14 and so is not referred to here for its legal content and we do not express any view on the correctness or otherwise of the learned Judge’s decision. [78] CA at §54. [79] Dr Safer, Affirmation at §§36 and 39. He also casts doubt on an article referred to by Dr Ng regarding a survey of transgender men who became pregnant as having “inherent weaknesses” pointing out (at §38) that: “It is a simple survey conducted without any use of objective hormonal baseline level screening to confirm that the participants of the survey were able to ovulate and conceive while on testosterone therapy.” [80] (2013) 16 HKCFAR 112 at §102. [81] [2021] Fam 77. [82] Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372 at §62, citing Lord Sumption JSC in Bank Mellat v Her Majesty’s Treasury (No 2) [2014] AC 700 at §20. [83] Ibid at §135. 
Mr Justice Tang PJ: 1. On 13 January 2011, the appellants were convicted after trial in the High Court and were sentenced to substantial terms of imprisonment.[1] Their applications for leave to appeal against conviction were unsuccessful and ended on 17 December 2013.[2] 2. On 22 January 2013, 8 February 2013 and 21 January 2014, the 3rd, 2nd and 1st appellants respectively, filed applications for an extension of time to give Notice of Application for Leave to Appeal against Sentence. The applications were dismissed by the Court of Appeal[3] on 25 June 2014 when the Court also ordered under s 83W(1)[4] that two months of the time during which the applicants were in custody pending the determination of their applications should not be reckoned as part of their respective terms of sentence.[5] The joint Reasons for Judgment was delivered on 15 September 2014. However, on that occasion, in an addendum McWalters JA expressed his concern: “that the Court of Appeal may not be empowered to order loss of time under section 83W(1) on a prisoner who is applying for an extension of time in which to seek leave to appeal, simply because he would not be, at that time, an applicant for leave to appeal.” 3. On 30 March 2016, the Appeal Committee granted leave to appeal so that we could decide whether the Court has power under s 83W(1) of the Ordinance to make a loss of time order in such circumstances. After hearing submissions from Leading Counsel for the appellants, the Court dismissed the appeal with Reasons for Judgment to be provided. 4. Under s 82,[6] a person convicted on indictment[7] may appeal to the Court of Appeal as of right on any ground which involves a point of law alone;[8] and with the leave of the Court of Appeal, on any ground which involves a question of fact alone, or a question of mixed law and fact, or on any other ground which appears to the Court of Appeal to be a sufficient ground of appeal. 5. Section 83Q provides that a person who wishes to appeal or to obtain leave to appeal should give “… notice of appeal or, as the case may be, notice of application for leave to appeal, in such manner as may be provided by rules and orders made under section 9.”[9] Section 83Q(2) states the time limit within which such notices should be given. However, the time limit “may be extended, either before or after it expires, by the Court of Appeal.”[10] 6. Section 83W is the provision under consideration in this appeal and provides: “(1) The time during which an appellant is in custody pending the determination of his appeal shall, subject to any direction which the Court of Appeal may give to the contrary, be reckoned as part of the term of any sentence to which he is for the time being subject. (2) Where the Court of Appeal gives a contrary direction under subsection (1), it shall state its reasons for doing so; and it shall not give any such direction where – (a) leave to appeal has been granted; or (b) a certificate has been given by the judge of the court of trial under section 82; or (c) the case has been referred to it by the Chief Executive under section 83P.” 7. Mr Bruce SC, for the appellants relied on the language of s 83W. He submitted that the plain language of the section shows loss of time could only be ordered in respect of an appellant and only for the time during which his appeal was pending. That being the case, s 83W plainly covers a person who is entitled to appeal as of right. But, it would not cover a person who needs leave to appeal. An applicant for leave to appeal would not be an appellant before he was granted leave to appeal, and that is why “appellant” is defined in s 2 to include “a person who has given notice of application for leave to appeal.” Absent such an inclusive definition, s 83W would not cover an applicant for leave to appeal. The fact that the definition was not exhaustive does not matter. It extended the word appellant to include an applicant for leave to appeal. Absent an equivalent extension, “appellant” would not include an applicant for an extension of time. 8. Mr Bruce also relied on Hau Kin & Others[11] v HKSAR[12] where at para 4, Li CJ said: “On a plain reading of [s 83W(1)], the time is the period between the filing of his application for leave to appeal and the determination of his appeal. The determination of an appellant’s appeal is only pending after he has filed his application for leave to appeal.” 9. In Hau Kin, this Court was concerned with three appeals which were heard at the same time. One of them, namely, the appeal by Ho Fook Shing,[13] was in respect of an application for leave to appeal against sentence out of time.[14] Ho, who had pleaded guilty was sentenced on 6 May 1998, and his application was made on 9 November 1999. Hence, Wong JA said he was out of time for nearly 1½ years,[15] and refused leave to appeal out of time. Also, because his application was completely devoid of merit, the Court ordered a three months’ loss of time. However, there were only two months and three days between 9 November 1999 when Ho’s application was lodged and 13 January 2000, when the application was dismissed, for that reason, this Court reduced the loss of time to two months and three days.[16] So, in Hau Kin, albeit without the benefit of any argument to the contrary, this Court proceeded on the basis that loss of time could be ordered in respect of an applicant for an extension of time. 10. Mr Bruce submitted, rightly so, in our view, that this appeal turns on the proper construction of s 83W. The approach to statutory construction is well settled. The statutory language is construed according to its context and purpose.[17] 11. The purpose of s 83W is clear. In Chau Ching Kay v HKSAR,[18] this Court explained that the purpose of the power to order loss of time was to deter unmeritorious applications which would otherwise take up time which could be better spent on meritorious cases. 12. Section 83W is identical to s 29 of the English Criminal Appeal Act 1968 (“the Act”) on which it was based. There is also an identical definition of “appellant” in s 51(1) of the Act. The English Courts also regarded the purpose of s 29 to be the deterrence of unmeritorious appeals to avoid delay in the hearing of meritorious appeals.[19] 13. Sections 83Q and 83W provide the context: the timely disposition of appeals or applications for leave to appeal and the deterrence of unmeritorious appeals or applications. An appeal is a process which begins with its initiation and ends at its termination. Whether the process is initiated by an appeal or application for leave to appeal within time, or an application for an extension of time to do so, the Court would examine the proposed ground of appeal for merits, as Lunn VP said in the joint Reasons for Judgment below “so as to ensure that should it refuse the application for an extension of time it would not be ‘shutting out a substantial and plainly arguable ground of appeal.’”[20] 14. Mr Bruce also fairly accepted that in reality, in terms of wasted time, an unmeritorious application for an extension of time to appeal or for leave to appeal may well take up more time than an application made within time. 15. That being the case, it is obvious and, Mr Bruce could not contend otherwise, that there is no discernible policy reason why the power to order loss of time should not extend to an application to appeal or for leave to appeal out of time. 16. Mr Bruce submitted, however, that conceptually, an application for an extension of time to appeal or for leave to appeal, is not an appeal or an application for leave to appeal. So, even if it was an oversight that the definition was not extended to cover an application for an extension of time, we should not give the word “appellant” a meaning which it cannot bear. 17. As long ago as 1990, in R v Lau Kam,[21] the Court of Appeal held that an applicant for leave to appeal out of time was no less an appellant who made his application within time. Macdougall JA giving the judgment of the Court of Appeal said: “The application he makes, whether it is within time or out of time, is still an application for leave to appeal.” 18. In England, where the statutory provisions are identical, loss of time orders have been made in respect of persons who applied for leave to appeal out of time. For example, R v Anthony Brind & Others[22] where loss of time orders were made in respect of five applicants[23] who had applied for an extension of time. 19. Having regard to the context and purpose of s 83W, with respect, Macdougall JA was right when he said an application for leave to appeal out of time is still an application for leave to appeal. That is the reality of the situation. As stated above, an application for time would entail a consideration of the merits of the intended appeal, and as Mr Bruce recognized, would take as much if not more court time than an appeal or application for leave to appeal made within time. Nor should reality give way to the so-called conceptual difference between an application for time and an appeal or application for leave to appeal. 20. Common sense has a place in interpretation.[24] As the learned editor of Bennionput it “When a particular matter is not expressly dealt with in the enactment this may simply be because the drafter thought that as a matter of common sense it went without saying.”[25] As stated above, there is no conceivable policy reason why an applicant for an extension of time should not be covered by s 83W.[26] We do not believe the absence of an explicit inclusion in the definition of appellant was an oversight. Rather, we believe the drafters and all those who were responsible for the enactments, in England as well as in Hong Kong, thought it went without saying that the word appellant in s 83W includes a person who appeals or applies for leave to appeal within time as well as one who applies out of time. 21. Mr Bruce referred us to R v Adams,[27] a decision of the Australian High Court. But with respect, that decision does not help. There the High Court held as a matter of statutory interpretation, the penal sections of Part XIV of the Bankruptcy Act covered only actual bankrupts and not a person who was deemed to be a bankrupt for certain purposes. 22. Mr Martin Hui SC for the respondent has in his helpful and comprehensive written submissions pointed out that the statutory form prescribed for an application for an extension of time, namely, Form XI, shows quite clearly that it is a composite application for an extension of time and for leave to appeal or appeal as the case may be. Form XI shows quite clearly that an application for an extension of time to appeal or for leave to appeal is regarded as in effect an appeal or application for leave to appeal although an explanation for the delay is also required. Form XI was made under the Criminal Appeal Rules (“CAR”) Cap 221A.[28] CAR 3 should also be noted. It reads “A notice of appeal, notice of application for leave to appeal or notice of application for extension of time within which such notice shall be given, shall be signed by the appellant himself, …” (emphasis added). Mr Bruce accepted that s 83W has long been regarded as applicable to an applicant for an extension of time and that the Criminal Procedure Rules, Form XI which was made under them as well as the decisions of the Courts, in Hong Kong as well as in England, including the decision under appeal, all support the view that s 83W was applicable to an applicant for extension of time. However, Mr Bruce contended that the true construction of s 83W requires a different answer. For the reasons given above, we do not agree. Mr Andrew Bruce SC, instructed by Tang and Lee, assigned by Director of Legal Aid, for the 1st to 3rd appellants Mr Martin Hui SC, DDPP and Mr David Chan, ADPP, of the Department of Justice, for the respondent [1] They have served their respective sentences. [2] When leave to appeal to this Court was refused. [3] Lunn VP and McWalters JA. [4] Criminal Procedure Ordinance (“the Ordinance”) Cap 221. [5] At para 7. [6] All references are to the Ordinance, unless otherwise stated. [7] This and other kindred provisions were also made applicable to appeals from the District Court. Section 83 District Court Ordinance, Cap 336. [8] Or if the trial judge grants a certificate that the case is fit for appeal on a ground which involves a question of fact, or a question of mixed law and fact. [9] Section 83Q(1). [10] Section 83Q(3). [11] There were three appellants whose appeals were heard together. [12] (2005) 8 HKCFAR 63. [13] CACC 564/1999, judgment 13 January 2000, Stuart-Moore VP, Leong JA and Wong JA. [14] In the other two appeals, the applications were made within time. [15] At page 2. [16] At para 20. [17] Li CJ in HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at para 12. [18] (2002) 5 HKCFAR 540. [19] See, for example, R v Herbert K [2005] EWCA Crim 955, a decision of the English Court of Appeal. [20] At para 19. [21] CACC 428/1989 (unreported, 11 May 1990), Yang CJ, Silke VP and Macdougall JA. [22] [2008] EWCA Crim 934, a decision of the Court of Appeal. [23] In separate appeals. [24] Bennion on Statutory Interpretation 6th edition, s 197, at page 511. [25] At page 512. [26] Indeed, as Lunn VP and Macrae JA suggested at para 16 in their joint judgment dated 13 March 2015 refusing leave to appeal to this Court, were that the case, it would provide a perverse advantage to applications out of time. [27] (1935) 53 CLR 563. [28] Which were made under power conferred by s 9 of the Ordinance. 
Mr Justice Ribeiro and Mr Justice Fok PJJ and Mr Justice Chan NPJ: 1. We have read in draft the judgments of Mr Justice Tang PJ and Sir Anthony Mason NPJ and agree that the appeal must be dismissed on the factual basis identified by Tang PJ. We agree with the judgment of Sir Anthony Mason NPJ and in particular with the statement of the law which it contains. We are respectfully unable to agree with Tang PJ’s views regarding Seldon v Davidson[1] but agree with his Lordship’s views in relation to the appellant’s case on unjust enrichment. Mr Justice Tang PJ: 2. The plaintiff Big Island Construction (Hong Kong) Limited (“BIC”) is a Hong Kong company, which is owned as to 99.99% by Big Island Asia Limited (“BIA”) which in turn is owned as to 99.99% by Mr Ben P Lee (“Mr Lee”). Mr Lee’s sister Ms Li Hung (“Ms Li”) owns the balance of 0.01%. 3. The defendants are the Wu Yi Development Company Limited (“WYD”), the first defendant, and Wu Yi Construction Company Limited (“WYC”), the second defendant. WYD and WYC belong to the Wu Yi group of companies which are ultimately owned by the Fujian Province People’s Government.[2] 4. This appeal is concerned with only one of 4 actions tried before Poon J, namely, HCA 1957/2005. In HCA 1957/2005,[3] BIC claimed against WYD and WYC for money lent pursuant to an oral agreement allegedly made in around October 1999 between Mr Lee on behalf of BIC and Mr Xu Zhong Hua (“Mr ZH Xu”) on behalf of WYD and WYC: “to enable [WYDand WYC] to finance, inter alia, the operational expenses of the Wu Yi Group in Hong Kong and to settle accrued interest … (the ‘Loans’) and that the Loans shall be repayable by [WYD and WYC] a year from the date on which the relevant sums were advanced by the plaintiff to [WYD and WYC] (‘the Loan Agreement’).”[4] 5. The claim against WYD was in respect of 18 loans totalling HK$100,676,120 said to have been made to WYD and the claim against WYC was for a loan of HK$1,793,700. Interest on these sums was claimed: “… at such rate and for such period as the Court thinks fit…” 6. An alternative claim was added in the Re-Amended Statement of Claim[5] for these sums against WYD and WYC respectively as money had and received by them from the plaintiff without consideration. 7. WYD and WYC admitted the receipt of the sums particularised in the claim but denied any loan agreement. Their case was that the payments were made pursuant to various Fund Exchange Agreements (“FEAs”), 9 of which were carried out, under which WYD: “agreed to remit or cause or procure the remittance of RMB to [BIC] in the [Mainland]in return for [BIC] agreeing to remit or cause or procure the remittance of the equivalent amount of Hong Kong or US dollars to [WYD] in Hong Kong.”[6] They alleged that cashier orders totaling RMB 109,836,700 were provided to BIC’s agent in the mainland following which, their equivalent in Hong Kong dollars, totalling HK$100,676,120 were provided to WYD in Hong Kong.[7] In respect of HK$1,793,700 claimed against WYC, WYC’s case was that a cheque in that amount was given to it pursuant to the FEA dated 25 May 2000 in return for RMB 2,000,000 delivered to BIC’s agent.[8] 8. In other words, it was WYD and WYC’s case that the moneys received from BIC were paid to them in return for earlier payments of RMB to its account,and in reply to the claim for money had and received, that the payments were made pursuant to the various FEAs and there was consideration for the payments.[9] 9. In their defence, WYD and WYC gave particulars of the various FEAs signed by both BIC and WYD, as well as the receipts signed by WYD and WYC respectively in respect of moneys received from BIC[10] and BIC’s receipts acknowledging the receipt of RMB totaling RMB 57,248,844.60[11]. They also pleaded and relied on a Settlement Agreement dated 28 August 2001signed by Mr Lee on behalf of BIC and WYD which acknowledged that: “Each of [WYD] and [BIC] has returned all funds which they lent to each other. The settlement of accounts has been completed, and neither one of the two parties owes to the other party anything.” 10. In the Re-Re-Re Amended Reply (“reply”), the plaintiff admitted that they had entered into the FEAs with the defendants but asserted that they were irrelevant to any of the loans. As for the Settlement Agreement, “the Plaintiff and its directors, officers or employees have no recollection of seeing, signing or otherwise entering into” it and that in any event it did not relate to the loans.[12] As for the receipts, those signed by WYD and WYC were admitted but the receipts said to have been provided by the plaintiff as well as the payments[13] they evidenced were denied[14]. 11. The trial took place before Poon J. The parties were represented by leading counsel and after a 43 day trial BIC’s claims were dismissed. Mr Ben Lee and his sister Ms Li were the only witnesses for BIC. A total of nine witnesses were called on behalf of the defendants, including Mr ZH Xu. 12. In respect of the loan agreement the learned judge concluded: “121. In my judgment, BIC has failed miserably to discharge the burden of proving its primary claims based on the Loan Agreement. I find that the Loan Agreement did not exist and that the Sums advanced to WYD and WYC were not loans as alleged. BIC’s primary claims are wholly incredible, which is determinative. Those claims must fail, even if I were to find that WYD and WYC’s defence based on the nine FEAs are equally incredible.” 13. As for the claim based on unjust enrichment, Poon J said: “129. … Applying the doctrine of failure of consideration, BIC has to prove (a) the purpose of the payments to WYD andWYC and (b) the purpose has failed. Its case is that the payments were loans and the failure of the purpose lied in the non-repayment. If BIC failed to prove the Loan Agreement, which must be the case in light of my earlier finding, there is no applicable unjust factor to support the claim of total failure of consideration. Mr Ho[15] is correct when he submitted that the alternative case based on restitution adds nothing to BIC’s claims. It must fall together with its primary case on the Loan Agreement.” 14. Although Poon J took the view that it was strictly not necessary for him to deal with the defendants’ case on the nine FEAs, he went on to do so and concluded: “162. … that when taken in its totality the evidence presented by WYD and WYC on the FEAs is so poor that, even on a balance of probabilities, I must reject it as improbable. I find that WYD and WYC have failed to discharge their burden of proving that the purported fund exchanges pursuant to any of the FEAs took place as alleged.” 15. The learned judge had the impression that he had not been told the whole truth: “164. … Something more than meets the eyes is there, which they have chosen, for reasons best known to them, to conceal from the court.” 16. He then dismissed BIC’s claims: “165. … whether it is the primary claim based on the Loan Agreement or the alternative claim based on money had and received. …” 17. BIC appealed to the Court of Appeal, contending,[16] first, that the learned judge erred in rejecting BIC’s case on the Loan Agreement and ought to have found in favour of BIC: “Given that there are only two version of events put forward by the parties, the learned judge ought to have considered but failed to consider the relative probabilities of the two versions,”[17] 18. Secondly, relying on Seldon v Davidson [1968] 1 WLR 1083 (CA), that given that the learned judge had rejected the parties’ respective cases, and given that receipt of the payments was admitted, the burden of proof was on the defence to show on a balance of probabilities that the payments were gifts, in settlement of an existing debt, in return for cash or something of the sort.[18] Accordingly, judgment should have been entered in favour of BIC. 19. Thirdly, that in relation to the claim for money had and received, the learned judge ought to have held that in the absence of proof by WYD and WYC of justification, it would be unjust to allow WYD and WYC to retain the money. 20. By aRespondent’s Notice, WYD and WYC asked that the judgment be affirmed on the ground that they had successfully proved their defence. Ground 1 21. In the Court of Appeal, the plaintiff contended that the judge ought to have found on balance that the plaintiff’s case on the oral loan agreement was more probable than the defendant’s case based on FEA’s and asked that judgment should be entered in its favour accordingly. The plaintiff’s appeal against the rejection of his claim was dismissed by the Court of Appeal. The plaintiff was entitled to appeal as of right[19], and in its written case, again asked for judgment on the same basis.[20] 22. In the Court of Appeal, the defendants asked in their Respondents’ Notice that the judgment be affirmed on the additional ground that the defendants had successfully proved their defence on the FEAs. The Court of Appeal dismissed the plaintiff’s appeal without dealing expressly with the Respondents’ Notice. 23. BIC’s argument is that the trial judge and the Court of Appeal were in error in dismissing its case because BIC had not discharged its burden of proving the oral agreement, without examining the probabilities. BIC submits once the probabilities are examined, they establish that it was more probable than not that the parties entered into the loan agreement. However when the probabilities are examined, as I shall shortly do, it is clear that the probabilities support the view that the parties did not enter into the loan agreement. Indeed, when the probabilities are examined, it becomes apparent that the trial judge should have found on the evidence that the defence case was established on a balance of probabilities. 24. There are concurrent findings against the plaintiff’s appeal against the rejection of the loan agreement. This court does not ordinarily entertain appeals against concurrent findings.[21] This is not a case where we should make an exception. The concurrent findings are plainly right. 25. I will deal with the facts briefly. BIC relied on an oral agreement. Both the trial judge and the Court of Appeal regarded an oral agreement to be improbable given the commercial context and the amount involved. On the other hand, the defence is well documented and the judge made important findings about the documents. On the basis of such findings, he ought to have found the defence proved on a balance of probabilities. 26. The first of such documents are the minutes dated 1 March 1999 of a meeting of the top management of the Wu Yi group in Hong Kong attended by, amongst others, Mr ZH Xu. This document is set out in para 39 of the judgment. It recorded that the defendants needed US dollars in Hong Kong and that a consortium introduced by BIC urgently needed RMB in the mainland. And that for such reasons, “it is agreed to negotiate with (BIC) and adopt the method of mutual loan of funds, namely, lending the corresponding value of funds to each other, (interest free)” BIC challenged the minutes’ authenticity. The judge held that they were authentic.[22] This document supports the defence. After finding that the minutes were authentic the judge gave no further thought to them. 27. Foreign Exchange Agreements (“FEAs”). It is confusing that the parties had signed a total of 20 FEA’s. The defendants’ case is that only nine of them were used by the parties for the fund exchanges in question, namely, the 9 FEAs dated between 25 May 2000 and 15 June 2001. Neither party dealt with the first seven FEA’s which were dated between 10 August 1997 and 28 November 1999 in their witness statements. It was common ground that the last four FEA’s were never carried out. As for the nine FEAs relied on by the defendants, it emerged during the defence evidence that they were signed by Mr Lee in March 2001 and were backdated.[23] With respect, the judge overlooked the significance of these nine FEAs. I take the first of the nine as an example. It recorded that in return for a temporary loan of RMB 8,000,000 to be made available to BIC by WYD,BIC would deposit their equivalent in HK$/US$ in a bank account designated by WYD. The evidence showed that cashier orders dated 28 June 2000 in the sum of RMB 1,800,000 and cashier orders both dated 29 June 2000 in the sums of RMB 4,000,000 and RMB 4,200,000 respectively were obtained by the defendants’ mainland associates[24] and that on 30 June 2000 2 sums totaling HK$8,927,000 which represented the RMB 10,000,000[25] were paid into WYD’s account by BIC. I will not go into details but this was repeated in respect of the other eight FEAs, namely, there were cashier orders obtained which corresponded in amount and time with the plaintiff’s payments into WYD’s account.[26] The nine FEAs which were admittedly signed by BIC strongly support the defendants’ case. 28. The next set of documents are receipts signed by the defendants (17 by WYD and 1 by WYC) respectively and given to BIC. Of these receipts, 12 described the payment by the plaintiff as “互借款 (mutual loan)”, 1 as “往來款 (mutual remittance)”, 2 as “暫借款 (temporary loan)” and 3 had no description.[27] Mr Lee’s evidence was that the descriptions “mutual loan” and “mutual remittance” were untrue and incorrect[28] but he didn’t object because he only read the first receipt which used the description 暫借款 (temporary loan).[29] That evidence was rejected by the judge.[30] These receipts are significant because they clearly supported the defendants’ case of Foreign Exchange Agreements. “Mutual loan” and “mutual remittance” suggest a two way movement of funds. The authenticity of these receipts was never in doubt. Unfortunately, the judge failed to give them any regard. They strongly supported the defence. 29. There were also 11 receipts issued by BIC acknowledging receipts of RMB. If genuine, they strongly supported the defence. Their authenticity was denied by BIC. The receipts stated that BIC had received RMB as “mutual loans”. This is what the judge said: “141. … Mr J Chen said he received all the receipts in one go in the second half of August 2001 from Ms Li, after the signing of the Settlement Agreement. Although Ms Li denied that, I accept Mr J Chen's evidence. I find that the receipts are authentic and are contemporaneous evidence capable of supporting the Wu Yi parties’ case.” 30. Mr Benjamin Yu SC, appearing for the plaintiff, challenged the judge’s finding that the receipts acknowledging receipts of RMB by BIC were genuine. 2 of the 11 receipts were in respect of amounts down to 2 decimal points. The evidence was that these figures tallied with the figures in the internal accounting documents of the defendants. Mr Yu relies on the fact that it was not explained how BIC might have come by these figures. That being so, Mr Yu submits these receipts must be forgeries. I do not see why that should be the only reasonable explanation. Moreover, the judge had had the benefit of this argument before he concluded that the receipts were genuine. Mr Yu also submits that the Court of Appeal in rejecting the same argument wrongly thought that the point had not been made before the judge. Even so, it does not matter, the argument is in any event unsound. 31. These receipts only acknowledged the receipt of RMB 57,248,844.60 and not all the RMB supplied to BIC. The judgment was silent about this. The receipts are nevertheless important. First, they are BIC’s acknowledgment that it had received RMB from WYD as “mutual loans”. However, as the particulars given in para 7 of the defence made clear these receipts tallied with the cashier orders given to Mr Li, the plaintiff’s agent. So they support the defence case the cashiers orders were given to BIC’s agent. The 2 receipts with decimal points remain a mystery, e.g. the receipt for RMB 4,421,209.77 does not tally with the cashier orders which totaled RMB 4,421,200. And the receipt for RMB 5,817,234.84 does not tally with the cashier orders which were for RMB 5,600,000. In the scheme of things I do not believe they matter, there is a wealth of documentary evidence in favour of the defence case that the parties were engaged in money exchange. 32. Lastly, there was the Settlement Agreement. Like the other documents, it was in Chinese. There were three clauses each of one sentence. The learned judge said: “163. What remains is the Settlement Agreement. As noted, Mr Lee initially said that his signature on the Settlement Agreement was forged. He eventually accepted that it was his but alleged that he signed it without having read the contents. I reject Mr Lee’s allegation, which I find incredible. That said, the Settlement Agreement only showed that the parties had treated the twenty FEAs as having been executed. It does not show that the purported fund exchange arrangements pursuant to the FEAs, as alleged, took place. I do not think WYD and WYC can derive much assistance from it.” 33. The Court of Appeal disagreed. Yuen JA said[31] it was a fatal blow to BIC’s case[32]. With respect, I agree. Using the simple language of the Settlement Agreement, it stated that all agreements about lending funds to each other had been performed andneither party owed the other anything. The error in the trial judge’s understanding of the Settlement Agreement further undermined his finding that the defence was not made out. I appreciate that the trial judge gave many reasons why he was dissatisfied with the defence evidence. With respect, he had been distracted by details and overlooked the big picture, burdened no doubt by the fact that the action was bitterly fought, had taken 43 days and as he said: “12. … (counsel)(took) every conceivable point, big and small.” 34. With respect, the trial judge failed to consider the inherent probabilities and concentrated instead on the details of the oral evidence of the witnesses who may be confused, forgetful and self-serving. Here, the defence was well supported by documentary evidence, which were either not disputed or if disputed found to be genuine by the judge. They should have been carefully weighed and the trial judge ought to have had regard to them in his assessment of the credibility of the defence witnesses and the inherent probabilities of the parties’ respective cases. Had he considered the big picture, I believe he would have concluded that given his rejection of BIC’s case and the strong and clear evidence[33] that the parties were engaged in the exchange of HK$ for RMB, he would have found the defence established. 35. I mentioned earlier that the Court of Appeal had not dealt with the Respondents’ Notice expressly. I believe the proper conclusion to draw given its express finding that the Settlement Agreement was fatal to BIC’s claim is that had they thought it necessary to do so they would have accepted the defence case and affirmed the judgment on this basis too. This conclusion is not inconsistent with the general rule of practice that this court does not ordinarily entertain appeals against concurrent findings of fact. Yuen JA’s finding on the Settlement Agreement differentiated the Court of Appeal’s findings on the defence case from those of the trial judge. 36. Therefore, on the facts, not only was the judge right in rejecting BIC’s case, he ought to have found in favour of the defence based on the FEAs. For this reason, the appeal must be dismissed. Ground 2 Seldon v Davidson 37. This is an ex tempore judgment of the English Court of Appeal (Willmer and Edmund-Davies LJJ) on an interlocutory appeal from the County Court. There, the plaintiff claimed the return of money lent to the defendant who was her chauffeur and handyman. The money (by 2 cheques) was paid to the defendant’s solicitors and used in the purchase of a house. The receipt of the money was admitted but the defence was that they were gifts, alternatively that they were not immediately repayable but were repayable as and when the defendant was able to do so.[34] On such pleadings, the County Court judge ordered the defendant to begin, being of the view that the legal burden was on the defendant to prove one or other of his defences. The trial was adjourned, however, to enable the defendant to appeal. On appeal, the plaintiff contended that no appeal lay from a mere ruling. Neither member of the Court of Appeal Willmer LJ, and Edmund-Davies LJ expressed a concluded view on this point. However, both of their Lordships affirmed the decision of the County Court judge that the burden of proof was on the defendant. 38. Seldon is often cited in our courts and is regarded in Chitty on Contracts, 31st edition, volume 2, specific contracts at 38- 259 as authority that: “If money is proved, or admitted, to have been paid by A to B, then in the absence of any circumstances suggesting a presumption of advancement, there is prima facie an obligation to repay the money; accordingly if B claims that the money was intended as a gift, the onus is on him to prove this fact.” 39. In Seldon, the defendant relied on Cary v Gerrish (1801) 4 Esp 9, and submitted that the burden was on the plaintiff to prove a loan of money. In Cary the claim was for money lent to the defendant during the lifetime of the testator. Payment was proved. On a plea of non assumpsit, Lord Kenyon said, payment by a draft made out in the name of the defendant and paid: “… is no evidence to establish a debt. No evidence is offered of the circumstances under which the draft was given; it might be in payment of a debt due by the testator; or the defendant might have given cash for it at the time … ” 40. Willmer LJ distinguished Cary and said there was no suggestion in Seldon that the money: “… was paid in settlement of an existing debt, or that it was given in return for cash, or anything of that sort. In the absence of any such circumstances, money paid by the plaintiff in circumstances such as these is prima facie repayable on demand. If the defendant seeks to evade repayment of the money which was paid to him, it seems to me that the judge was right in placing the onus upon him to prove the facts which he alleges show that the money was not repayable.” (1088F) 41. Edmund-Davies LJ agreed with Willmer LJ and said: “when the simple payment of money is proved or admitted between strangers … on that bald state of affairs, proof of payment imports a prima facie obligation to repay the advancement in the absence of circumstances from which presumption of advancement can or may arise.” (1090F) 42. Both of their lordships said that no presumption of advancement could arise[35]. Willmer LJ said since counsel accepted that the house which was bought with the money supplied by the plaintiff would be held on a resulting trust for her, it would be strange “if the same consideration did not apply to the money paid by the plaintiff to the defendant to assist him in the purchase of the house.” 43. As Lord Browne-Wilkinson explained in Westdeutsche Landesbank Girozentrale v Islington LBC [1996] AC 669 a resulting trust is not imposed by law against the intentions of the trustees but gives effect to them and it could arise: “(A) where A makes a voluntary payment[36] to B or pays (wholly or in part) for the purchase of property which is vested either in B alone or in the joint names of A and B, there is a presumption that A did not intend to make a gift to B: the money or property is held on trust for A (if he is the sole provider of the money) or in the case of a joint purchase by A and B in shares proportionate to their contributions. It is important to stress that this is only a presumption, which presumption is easily rebutted either by the counter-presumption of advancement or by direct evidence of A’s intention to make an outright transfer…” 44. Thus, it is only when a payment is voluntary that a resulting trust may arise. If money is provided as a loan, no resulting trust will arise.[37] Since the payments would be at the disposal of the borrower and there would be no identifiable trust fund.[38] 45. But what is the nature of a presumption of resulting trust or the corresponding presumption of advancement? Lord Diplock[39] in Pettitt v Pettitt [1970] AC 777 at 823G said such presumptions are used to impute: “… an intention to a person wherever the intention with which an act is done affects its legal consequences and the evidence does not disclose what was the actual intention with which he did it. This situation commonly occurs when the actor is deceased. When the act is of a kind to which this technique has frequently to be applied by the courts the imputed intention may acquire the description of a ‘presumption’ – but presumptions of this type are not immutable. A presumption of fact is no more than a consensus of judicial opinion disclosed by reported cases as to the most likely inference of fact to be drawn in the absence of any evidence to the contrary… ” 46. Lord Upjoin in Pettitt at 814F said a presumption of advancement “is no more than a circumstance of evidence which may rebut the presumption of resulting trust”. As Snell’s Equity25-007 put it : “… the approach in recent cases has been to strive to determine the real intentions of the parties. It may only resort to the formal presumptions where the direct evidence of those intentions is absent and a default rule is needed.” 47. Here, the plaintiff case that the moneys were provided as loans (thus at the disposal of the defendants) was incompatible with any resulting trust. They were not voluntary payments in the sense used by Lord Browne-Wilkinson so no resulting trust could arise. Nor was it the plaintiff’s case that there was a resulting trust. 48. Mr Yu has referred us to a number of decisions in England where Seldon was considered. In all but one, the court was concerned with domestic disputes where the dispute was over whether the money paid was a loan or a gift. The most recent and a case which was not concerned with gift is Patel v Mirza[2015] 1 All ER 326. It concerned a claim for repayment of money paid pursuant to an oral contract to profit from illegal insider dealing. But the insider information never transpired and the agreement was not and could not be performed. The trial judge dismissed the plaintiff’s action because it was barred by illegality. His appeal was allowed by the Court of Appeal, by Rimer and Vos LJJ on the ground that since the illegal contract remained executory, he could repudiate and withdraw from the agreement and recover what he had paid or transferred. However, Mr Yu relies on the fact that Gloster LJ said Seldon was instructive and that: “90. Even in the absence of any trust, or retention of a proprietary interest in the money, once [the defendant] had admitted receipt of the money for the purpose of speculating on the IG Index, in my view the evidential and legal burden shifted to him to rebut the presumption that he was bound to repay [the plaintiff]. Whilst [the plaintiff] was required to demonstrate that the money had not been transferred by way of gift or pursuant to an enforceable contract that entitled [the defendant] to retain the funds, absent those factors, proof of payment imported a prima facie obligation to repay.” 49. Since Gloster LJ said the plaintiff was required to demonstrate that the money had not been transferred by way of gift, I do not believe her observations assist Mr Yu. 50. In Chapman v Jaume[2012] EWCA Civ 476, the parties were unmarried co-habitees. The plaintiff claimed the return of £162,589.42 which he alleged were loans subject to express terms about the time at which they were repayable. The defendant’s case was that the money had been paid in lieu of contribution to the running of the household. The judge rejected the respondent’s case but dismissed the claim since the plaintiff had failed to prove the precise conditions about the time at which the money would be repaid. The appeal was allowed. Lewison LJ[40]referred to Seldon and in particular, the passages from Willmer LJ’s judgment and Edmund-Davies LJ quoted above and said it was unfortunate that the judge’s attention was not drawn to this authority. He then rejected the respondent’s argument on presumption of advancement, saying that there was no presumption between cohabitants and then he said: “25. … on the facts found by the judge, he ought to have drawn the inference that the money was repayable within a reasonable time after demand.” So the actual decision did not turn on Seldon. 51. Clark v Mandoj (unreported, 19 March 1998). The plaintiff who cohabited with the defendant sued her for debt. Her defence was that they were gifts. The judge: “… rejected the plaintiff’s evidence and case that there was an agreement for a loan on the terms put forward by the plaintiff.[41] He also rejected the case of the defendant that the plaintiff had intended that the whole of that which he had contributed should be a gift.” (p 6) and decided that only a limited amount need to be repaid. The plaintiff appealed. In the Court of Appeal, Hobhouse LJ said Seldon: “concerns the burden of proof and the presumption which exists in the absence of adducing of evidence” However, Hobhouse LJ was of the view that the judge had decided as a matter of fact by way of inference that there were certain limited funds which should be repaid and that the remainder shall not be repaid and dismissed the appeal. Swinton Thomas LJ who agreed with Hobhouse LJ added at p10, that if the trial judge had not made any findings he had no doubt following Seldon and the other cases mentioned in Hobhouse LJ’s judgment, there would be a clear presumption in favour of a loan together with an obligation to repay it. 52. In Freeman v Tems, unreported, 27 Jan 1993 (which was relied on by Hobhouse LJ in Clark), the defence to a claim for money lent was that they were gifts. The Court of Appeal ordered a re-trial. Russell LJ pointed out that the judge had not had cited to him Seldon, “an important case” and said (with the concurrence of Simon Brown LJ), after quoting the passages from the judgments of Willmer and Edmund-Davies LJJ quoted above, that the County Court judge having failed to direct himself upon the principles set out in Seldon, there was a risk that he approached the problem of whether this is a loan or a gift from the wrong standpoint. 53. The English cases cited to us show that Seldon is accepted in England as authority that, when receipt is not in issue, and the defence is that the money received was a gift, in the absence of circumstances giving rise to a presumption of advancement, the burden is on the defence to prove the gift. That might be explained on the basis that in such context, the money was a voluntary payment in the sense used by Lord Browne-Wilkinson, and the defence that it was a gift is akin to a defence of confession and avoidance, namely, that the payment was voluntary but that the defendant was entitled to retain the payment because it was a gift. Whether that is the basis of the decision in Seldon does not matter, I will proceed on the basis that ever since Seldon was decided, in such situations in England, the burden of proof has been on the defendant to prove that the payment was a gift. Even so, there is no support for Mr Yu’s submission that upon proof of mere payment, the burden is on the defence to prove, for example, that the payments were in repayment of an earlier debt or in return for cash. I am firmly of the view that the burden is not on the defence to prove, for example, that the payments were in repayment of an earlier debt or in return for cash. Of course, if the loan was admitted and the defence is that it had been repaid, the burden is on the defence to prove payment.[42] 54. Seldon has often been cited in our courts. In Mak Ka Hing v Pang Ming Chung [2011] 1 HKLRD 347, counsel conceded and the Court of Appeal thought it was proper for him to do so that Seldon was authority that when receipt was admitted: “… the legal burden was on the defendant to prove either that there was a gift or, if it was a loan, that it was not repayable at the date of the issue of the writ.” 55. However, in that case, the trial judge had rejected the defence case that the money paid was a commission and accepted the plaintiff’s case that it was an advance. So the decision did not turn on Seldon at all. 56. In El Vince Ltd v Wu Wen Sheng [2011] 4 HKLRD 541, the plaintiff case was that there was a loan of $2 million, the defence was that it was part repayment of an earlier loan of $4 million. The judge accepted the defence and disbelieved the plaintiff. The plaintiff appealed against such findings of fact, which had no merit. However, the plaintiff also relied on Seldon and submitted that the judge was wrong to hold that the burden of proof was on the plaintiff. The Court of Appeal said Seldon did not help the plaintiff because, the defendant’s pleaded case was that the $2million was a part repayment of$4million paid by the defendant to the plaintiff. In other words, the defence was not that the money was a gift and as such distinguishable. 57. Lui Fai Yeung v Chui Kin Man (2012) 15 HKCFAR 803 is a decision of this court. There, the plaintiff sued the defendant for money lent. There were 178 payments. Receipt of the money was admitted and the relevant defence was that the monies were gifts. The trial judge was not satisfied with the evidence of either the plaintiff or the defendant and took the view the monies might have been gifts, joint investment or loans and dismissed the claim because the plaintiff failed to prove how much of the payments fell into which category. The plaintiff relied on Seldon in the Court of Appeal and before this court. This court did not find it necessary to decide whether Seldon was concerned with a persuasive or evidential burden, and held that in the circumstances of that case, the proper inference to draw was that the payments were gifts. 58. In the present case as in the other cases in Hong Kong where Seldon has been cited, the decision did not turn on Seldon. Here, the plaintiff’s claim was based on a loan agreement under which the loans were repayable within a year. The defence was not that they were gifts. The defence was that the payments were made pursuant to various FEAs. There was never any possibility of any resulting trust. Cary is clear authority that the fact of mere payment is not evidence of a loan. Nor does it stand alone. 59. In Aubert v Walsh and Another (1812) 4 Taunt 293, in a claim for money had and received, the defendant sought to set off the claim by proving payment to the plaintiff by cheques. Mansfied CJ said: “I am sure I remember a case before Lord Mansfield CJ in which a check given was produced as evidence of a debt, and his lordship held that that alone was not sufficient.” Chambre J was of the same view. 60. In Welch and Another, Assignees of Evan Evans, a Bankrupt, v Seaborn (1816) 1 Stark 474 it was reported that: “Lord Ellenborough was of opinion that (mere proof of payment) was not sufficient evidence to leave it to the jury, whether this money had been advanced … , by way of loan, since the presumption of law was, that money when paid is paid in liquidation of an antecedent debt.” 61. Seldon does not represent the law in Australia. In Heydon v Perpetual Executors, Trustees and Agency Co (WA) Ltd (1930) 45 CLR 111, the claim was by executors for money lent by the deceased and the defence was that the money was a gift. Gavan Duffy CJ said with the concurrence of the other members of the High Court of Australia that: “… the burden of proving the facts in support of either one or other cause of action … lies on the plaintiff.” Dixon J mentioned in his judgment Aubert v Walsh as authority that the burden of proof was on the payer. Earlier, during the argument, Dixon J said in response to counsel’s submission that a payment to a stranger raises a presumption of a resulting trust: “Godefroi, 3rd ed, at p 195, 4th ed, at p 145 says chattels which pass by delivery are not within the rule, and the presumption arising from a voluntary delivery of them is that a gift was intended, in the absence of circumstances; and in George v Howard(1819) 7 Price 646, Richards C B says: ‘If I deliver over money … to another, even although he should be a stranger, it would be prima facie a gift.’”[43] 62. In Joaquin v Hall [1976] VR 788, a decision of the Supreme Court of Victoria, Jenkinson J refused to follow Seldon. He said the court in Seldon appeared to have treated the defence pleaded as a confession and avoidance but the defence was in fact a denial of an essential ingredient in the cause of action. His lordship followed Heydon. 63. In New Zealand, Zheng v Qiu (unreported, CIV 2006-404-5720, 11 June 2007), Stevens J, on an appeal from the District Court of Auckland, was of the view that “the Court of Appeal in Seldon was speaking about a factual presumption that it was prepared to apply in the particular circumstances of that case.” at (44) and that “the Court may or may not decide to apply them depending upon the particular circumstances of the case …” at (46). 64. As the discussion of the authorities show, although Seldon was often cited few if any decisions turned on it. Disposal of a claim on the burden of proof should be rare and exceptional. I believe a judge should resolve conflicting versions of fact by deciding which is more probable uninfluenced by any consideration of who has the burden of proof. One should look to the burden for help as a last resort. In practice, it is difficult to conceive a case where there is no evidence apart from the payment. Indeed, when a defence of gift is pleaded it suggests that evidence is available to prove it[44]. Ditto, a claim that it was a loan. In a case, where the contest is between a loan and a gift, it is difficult to conceive a case where the court has nothing more than the evidence of payment and receipt. Given the relatively short limitation period one or more of the protagonist is likely to be available. One would expect evidence to explain why the payment was made from the parties. Payments and receipts do not normally happen spontaneously. Moreover, the relationship of the protagonists, their evidence, the circumstances of the payment, its size, the alleged purpose of the paymentor the use to which the money was put, are all matters which might help the court to decide the character and effect of the payment. Here, in my opinion, the trial judge should have found in favour of the defendants on the FEAs and this is not a case which turns on the burden of proof. But in a case where there was nothing more than the fact of payment which is proved and the defence is silent on the reason of the payment, I am of the view that following Cary, the claim should be dismissed. 65. What then is the position in Hong Kong where the defence to a claim on debt is that it was a gift? Such cases should rarely if ever turn on the burden of proof. I have had the advantage of reading Sir Anthony Mason’s judgment in draft, he has taken a different view on Seldon, a view which is shared by other members of the court, thus henceforth, Seldon will no longer be followed in Hong Kong. I welcome that result. Given my view on the first ground of appeal, Seldon has no application at all, but I am glad that this appeal has provided the occasion to clarify a question which has much exercised our courts. Unjust enrichment 66. BIC’s pleaded claim on restitution was based on failure of consideration, a claim on this basis will fall with the claim based on the loan agreement. However, in the Court of Appeal as in this court, BIC contended for restitution on a new basis, namely, there should be restitution in the absence of a basis for the payment. Presumably, Mr Yu believes that given the trial judge’s decision that there was no apparent reason for the payments made by BIC, BIC was entitled to judgment on this basis. Our decision on the first ground of appeal has made this argument academic. 67. The leading authority on unjust enrichment is Shanghai Tongji Science and Technology Industrial Company Limited v Casil Clearing Limited (2004) 7 HKCFAR 79. There Ribeiro PJ after giving close consideration to the matter said[45] that following the traditional common-law approach four questions should be asked, namely: (a) was the defendant enriched ? (b) was the enrichment at the plaintiff’s expense? (c) was the enrichment unjust? (d) are any of the defences applicable? 68. The question which is relevant to the present appeal is “(c) was the enrichment unjust?” As Lord Hope of Craighead explained in Kleinwort [46]: “The approach of the common law is to look for an unjust factor, something which makes it unjust to allow the payee to retain the benefit … It is the mistake by the payer which, as in the case of failure of consideration and compulsion, renders the enrichment of the payee unjust. The common law accepts that the payee is enriched when the sum was not due to be paid to him, but it requires the payer to show that this was unjust.” 69. Mr Yu submitted that we should follow the approach adopted by the Canadian Supreme in Garland v Consumers’ Gas Co (2004) 237 DLR (4th) 385 and adopt “absence of basis” as the test. With respect, as academic writings show the Canadian approach has its own complications.[47] 70. Shanghai Tongji was followed and applied in Takahashi v Cheng Zhen Shu (2011) 14 HKCFAR 558. Mr Yu submitted that in between these cases, in Cheong Shing Ltd v Yu Kwan (2008) 11 HKCFAR 594, this court adopted the “absence of basis” approach. I cannot agree. There was no reference to Shanghai Tongji or any discussion of “the absence of basis” approach in Cheong Shing. The decision in Cheong Shing is explicable on the basis that the court was satisfied that there was a total failure of consideration, “a well-recognized category of unjust enrichment.”[48] 71. There is no reason why we should reconsider Shanghai Tongji and we will not do so. Disposition 72. For the above reasons, I would dismiss the appeal and make an order nisi that BIC pays the costs of the appeal such costs to be taxed unless agreed. Sir Anthony Mason NPJ: 73. I would dismiss this appeal. I agree with the reasons for judgment given by Tang PJ except in so far as they relate to the issues concerning the onus of proof and Seldon v Davidson[49]. On these issues my reasons are set out below. Onus of proof – the pleadings 74. At the forefront of the appellant’s case is the submission that the trial judge and the Court of Appeal’s were in error in holding the onus of proof rested with the appellant (plaintiff) of establishing its case that it agreed to advance loans to the respondents (defendants), that the loans were made and not repaid. Mr Benjamin Yu SC for the appellant based his argument on this point largely on Seldon v Davidson, a decision of the English Court of Appeal. 75. First, however, it is necessary to ascertain what were the issues as they arose on the pleadings and determine on which party the onus of proof rested in the context of these issues. As Phipson notes: “[i]n all but the simplest cases, the burden of the issues will be divided, each party having one or more cast upon him”[50] 76. The central matter pleaded by the appellant in its re-amended statement of claim was the oral Loans Agreement whereby the appellant would advance loans to the respondents to enable them to finance the operations of the Wu Yi Group in Hong Kong and to settle accrued interest owed to the banks, the loans being repayable one year from the date on which they were made (para 2A). The appellant then pleaded that loans were made (paras 3 and 4), of which particulars were given, as recorded in the reasons for judgment of Tang PJ, that the loans had not been repaid (para 6) and that the respondents were in breach of the Loan Agreement (para 7). Alternatively, the appellant pleaded a case of money had and received without consideration, based on the same averments (para 7A). 77. In this court and the courts below the claim for money had and received was the vehicle for an alternative claim of unjust enrichment; there was no suggestion that the claim for money had and received relieved the appellant from establishing whatever it needed to establish under its pleaded case for breach of contract, except in so far as the adoption of the Canadian approach to the law of unjust enrichment might independently bring about such a result. 78. By their re-re-re-amended defence the respondents denied the making of the Loans Agreement (para 2A) and asserted that they received the sums referred to in the re-amended statement of claim pursuant to “written Fund Exchange Agreements” and not otherwise (para 3(b)). The written Fund Exchange Agreements are described in the reasons for judgment of Tang PJ and there is no occasion for me to set them out. The respondents then denied breach of the Loans Agreement (para 17) and the claim for money had and received without consideration (para 17A). In the same paragraph of the re-re-re-amended defence the respondents averred the giving of consideration in the form of payments under the “Fund Exchange Agreements” and other matters of defence that are of no relevance to the question now under consideration. 79. By its re-re-re-amended reply the appellant in effect joined issue with the respondents’ denials of the matters alleged by the appellant in its re-amended statement of claim. The appellant went on to admit the making of nine “Fund Exchange Agreements” but alleged they were unrelated to the “Loans” (para 4) and denied receiving any funds pursuant to those Agreements (paras 5 and 9). 80. One matter, if no other, emerges with dazzling clarity from the chaotic confusion of the pleadings. It is that the respondents denied the making of the Loans Agreement and breach of the Loans Agreement. Indeed, that denial was the very core of the respondents’ case. The consequence was that the appellant carried the burden of proving the elements of its case on the pleadings i.e. the making of the Loans Agreement, the making of the alleged “Loans” payments and breach by the respondents, namely failure to repay. Nothing could be clearer. 81. Had the respondents pleaded payment as an answer to the appellant’s case of breach of the Loans Agreement, without denying the making of the Loans Agreement, the plea of payment would have operated as a plea by way of confession and avoidance, that is, it would have admitted the making of Loans Agreement and avoided liability on the pleaded cause of action by asserting payment resulting in discharge. See the illuminating discussion by Sir Owen Dixon CJ in Young v Queensland Trustees Ltd[51] where his Honour said: “The ‘[Regulae Generales]’ of 1834 provided that in every species of assumpsit, all matters in confession and avoidance including those not only those by way of discharge but those that show the transaction to be either void or voidable … shall be specially pleaded. It is significant that in doing so the rules expressly include payment as an example.”[52] 82. His Honour went on to say: “The law was and is that speaking generally, the defendant must allege and prove payment by way of discharge as a defence to an action for indebtedness in respect of an executed consideration… it was soon settled that upon a plea of payment the defendant had the right to begin at the trial.”[53] 83. The principles of pleading stated by Sir Owen Dixon CJ were applied by Jenkinson J in Joaquin v Hall[54],a case where the plaintiff’s cause of action as pleaded was an agreement for a loan repayable on demand and breach of that agreement by failure to repay after demand. The defence was a denial of an essential ingredient in the cause of action, thereby casting the burden of proof on the plaintiff who failed to discharge it. 84. There was in that case, as in this case, no plea of payment, the effect of which would be to confess the elements in the appellant’s cause of action and thrust the burden of proof on the respondents. The respondents’ burden would then have been to prove the payments. 85. Mr Benjamin Yu SC for the appellant attempted to suggest that the respondents’ plea that payments had been made under the Fund Exchange Agreements was to be equated with a plea of payment. The suggestion is insupportable for a number of reasons. First, there was no plea of payment under the Loans Agreement. Instead, the respondents pleaded payment under the Funds Exchange Agreement and not otherwise. Secondly, the respondents specifically traversed the elements in the appellant’s cause of action as pleaded, thereby casting the onus of proof of those matters on the appellant. Thirdly, the main issue thrown up by the pleadings was whether the parties entered into the Loans Agreement as alleged by the appellant or only the Fund Exchange Agreements as alleged by the respondents. In this situation, there was no basis whatsoever for suggesting that the respondents bore the onus of establishing that there was no Loans Agreement. Seldon v Davidson 86. This brings me to a consideration of Seldon v Davidson. I begin by saying that, in my view, that case was wrongly decided by the English Court of Appeal and should not be followed in Hong Kong. My reasons for this conclusion follow an examination of what the case decided. 87. The case concerned two cheques totalling £1,550 paid by S to D, her chauffeur and handyman, which were used by D to buy a house. The parties then fell out. S claimed the money back, asserting it was a loan. D admitted receipt of the moneys but pleaded that they were a gift and, if not, a loan which was not repayable until D was able to do so, an event that had not come about at the date of the writ. So the primary issue was whether the payments were a loan or a gift. The appeal arose from a contested ruling by the trial judge that, given admission of receipt, D carried the legal burden of proving the gift and had to begin. The Court of Appeal (Willmer and Edmund Davies LJJ) in ex tempore judgments held on this point that the trial judge was right in placing the onus on D to prove the facts which he alleged to show that the moneys were not repayable. 88. Willmer LJ stated: “Payment of the money having been admitted, prima facie that payment imported an obligation to repay in the absence of any circumstances tending to show anything in the nature of a presumption of advancement. This is not a case of father and child, or husband and wife, or any other such blood relationship which could have given rise to a presumption of advancement. Mr Sears was constrained to admit that the house which had been bought with the aid of the money paid by the plaintiff was no doubt prima facie subject to a resulting trust in favour of the plaintiff. That being so, it would be strange indeed if the same considerations did not apply to the money paid by the plaintiff to the defendant to assist him in the purchase of the house.”[55] 89. Willmer LJ went on to say: “…we have from the defendant in this case a clear admission of the payment of the money, and no suggestion that it was paid in settlement of an existing debt, or that it was given in return for cash, or anything of that sort. In the absence of any such circumstances, money paid by the plaintiff in circumstances such as these is prima facie repayable on demand.”[56] 90. Edmund Davies LJ stated that in the circumstances which gave rise to no presumption of advancement, counsel for D was forced to concede that the house must be regarded prima facie as being held by D by way of resulting trust for S[57]. Later, his Lordship said, after considering two earlier cases, which he sought to distinguish: “…one is really driven back to consider this matter without the assistance of authority and, being so unassisted, I ask myself what is to be inferred as to the nature of the transaction when the simple payment of money is proved or admitted between strangers. I entirely agree with my Lord that, on that bald state of affairs, proof of payment imports a prima facie obligation to repay the advancement in the absence of circumstances from which presumption of advancement can or may arise.[58] (emphasis supplied) 91. The Court of Appeal’s reasoning, based on the prima facie proposition that D held the house property on a resulting trust for the benefit of S, leading to the conclusion that the moneys paid by S to D should be repaid, is difficult to explain. The imposition of a resulting trust was designed evidently to justify an order for the repayment of the money paid on the footing that it was a loan. The relationship of resulting trust and loan in the circumstances of the case was inconsistent, as appears from the discussion below. What is more S did not claim that there was a resulting trust. And to add to these difficulties, the Court of Appeal was mistaken in thinking that the defence pleaded amounted to a confession and avoidance. As Jenkinson J pointed out in Joaquin v Hall[59], the defence in Seldon v Davidson denied the loan alleged by S which was an essential ingredient in her cause of action as pleaded. The fact that the defence pleaded was gift or, in the alternative, a loan on different terms cannot alter the onus of proof arising from the denial of the loan alleged by S. 92. As the Editors of “Jacobs’ Law of Trusts in Australia” point out the presumption of resulting trust does not apply where A lends money to B and B uses the money to buy property from X.[60] As Jacobs goes on to state: “B is not an implied trustee of the land for A because A has not acted as purchaser.”[61] In Australia, these principles are well established by a series of cases in the High Court of Australia; Charles Marshall Pty Ltd v Grimsley[62]; Napier v Public Trustee (WA)[63] and Calverley v Green[64]. 93. In England, the law with respect to purchase price resulting trusts (the first category) is no different. Browne-Wilkinson J in Re Sharpe[65] said: “In my judgment, if, as in this case, moneys are advanced by way of a loan there can be no question of the lender being entitled to an interest in the property under a resulting trust. If he were to take such an interest, he would get his money twice: once on repayment of the loan and once on taking his share of the proceeds of the sale of the property.” 94. Lord Browne-Wilkinson (with whom Lord Slynn agreed) in Westdeutsche Landesbank Girozentralev Islington LBC[66] affirmed the proposition that when A makes a voluntary payment to B or pays (wholly or in part) for the purchase of property which is vested in B alone or in the joint names of A and B, there is a presumption A did not intend to make a gift to B; the money or property is held on trust for A (if he provides all the money) or in the case of a joint purchase by A and B in shares proportionate to their contributions. 95. Lord Browne-Wilkinson, with reference to this first category of resulting trusts, stressed: “…this is only a presumption, which presumption is easily rebutted either by the counter-presumption of advancement or by direct evidence of A’s intention to make an outright transfer.”[67] 96. His Lordship, having referred to the second category of resulting trusts, namely express trusts where the trusts declared do not exhaust the whole beneficial interest, then said: “Both types of resulting trust are traditionally regarded as examples of trusts giving effect to the common intention of the parties. A resulting trust is not imposed by law against the intentions of the trustee (as is a constructive trust) but gives effect to his presumed intention.”[68] His Lordship then added: “As the article by William Swadling, ‘A new role for resulting trusts?’ 16 Legal Studies 133 demonstrates, the presumption of resulting trust is rebutted by evidence of any intention inconsistent with such a trust…”[69] 97. Thus the presumption will be displaced if the person who provides the money intended only to make a loan of money which was to be repaid. In such a case there would be no resulting trust because the common intention is inconsistent with the person who provides the money taking an equitable interest[70]. 98. It follows that the imposition of a resulting trust in Seldon v Davidson was unjustified by reference to principle and authority in both England and Australia. 99. There remains for consideration the correctness of the proposition affirmed in that case that the payment of money prima facie imported an obligation to repay it. It will be recalled that Willmer LJ was of the view that in the absence of any suggestion that the money was paid in settlement of a debt, or given in return for cash, or anything of that sort, the presumption applies. On the other hand, Edmund Davies LJ expressed himself more cautiously referring to a “simple payment” and “that bald state of affairs”. Consistently with those expressions it would be possible to so confine the implied obligation to repay so that it has no application at all to commercial transaction, such as the transaction in this case. If Seldon v Davidson is to be regarded as having authority in Hong Kong that is the effect I would give to it, with the result that it provides no support to the appellant’s case. 100. As I see it however, in the light of the earlier discussion, it is preferable that this Court should determine the basic question of principle and determine whether Seldon v Davidson was correct in holding that payment of money to a stranger gives rise prima facie to an implied obligation to repay. Such a determination calls for an examination of the relevant authorities. 101. In Seldon v Davidson, the Court of Appeal regarded the two earlier authorities referred to as being of no assistance and distinguished them on the facts. To my mind, the two cases are not consistent with the approach taken by the Court of Appeal. In Cary v Gerrish[71]the plaintiff executors sought to establish the existence of a loan by proving that the defendant received cash for a draft drawn by the testator. Lord Kenyon held that there was no evidence to establish a debt, there being no evidence of the circumstances under which the draft was given. His Lordship made the point that no inference of loan could be drawn and said: “standing as a naked transaction, as this does, it is not evidence…[72]” 102. In the other case, Welch v Seaborn[73], Lord Ellenborough held that a loan of money by A to B is not to be inferred from the bare fact that A delivered a sum of money to B which A had borrowed from another. His Lordship pointed out: “that the presumption of law was, that money when paid is paid in liquidation of an antecedent debt”[74] 103. In a case not drawn to the attention of the Court of Appeal, Aubert v Walsh[75], the headnote records “Proof of the delivery and payment of a check to the Plaintiff is not sufficient evidence of a debt in order to support a set-off, unless it be shewn upon what consideration and under what circumstances, the check was given.” Mansfield CJ said: “I am sure I remember a case before Lord Mansfield CJ in which a check given was produced as evidence of a debt, and his Lordship held that alone was not sufficient.”[76] Chambre J said: “All our accounts would be in inextricable confusion if such evidence were allowed.”[77] 104. None of the three cases can be easily or persuasively reconciled with Seldon v Davidson. Moreover, in Heydon v The Perpetual Executors, Trustees & Agency Co (WA) Ltd[78], the High Court of Australia held that the onus of proving that a transaction involving the payment of money amounted to a loan or was received to the use of the testatrix lay upon the plaintiff executor and was not satisfied by proof merely of the payment by her to the defendant of the amount claimed. Dixon J in his judgment drew attention to Aubert v Walsh as a further authority in support of the view that the burden of proof was on the plaintiff[79]. In argument, in response to the submission that a “voluntary payment to a stranger raises a presumption of a resulting trust”, Dixon J interposed: “Godefroi, 3rd ed., at p. 195, 4th ed., at p. 145, says chattels which pass by delivery are not within the rule, and the presumption arising from a voluntary delivery of them is that a gift was intended, in the absence of circumstances; and in George v Howard (1819) 7 Price 646, at p. 651; 146, Richards, C.B. says: ‘If I deliver over money... to another, even although he should be a stranger, it would be prima facie a gift.’” 105. Subsequently in Joaquin v Hall, Jenkinson J, in the light of the decision in Heydon’s case, refused to follow Seldon v Davidson. His Honour considered that Heydon’s case rejected the presumption of an obligation to repay from the fact of payment to a stranger. His Honour went to say that “none of the cases cited … throws any doubt, in my opinion, on the proposition that no such presumption is raised by payment of money to a stranger”[80] 106. It follows that the reasoning in Seldon v Davidson is flawed, that it is inconsistent with earlier English authority and with Australian authority. That circumstance in itself is enough to justify the conclusion that the decision should not be followed in Hong Kong and that the presumption of an implied obligation to repay from the fact of payment to a stranger cannot be supported. Further, I am not persuaded that, as a matter of policy, recognition of such a presumption has anything to commend it. The making of a bare payment to another may in the nature of things be explicable by reference to a wide variety of possibilities. To my mind, the probability that there is an obligation to repay the amount is not so strong that it should become the subject of a presumption, even a presumption of fact as Stevens J held it to be in Keqing Zheng v Golden International Trading Ltd and another[81]. It is the making of the payment in the circumstances which surround it that will enable appropriate inferences to be drawn, in the light of any relevant traditional presumption. In such a situation it is preferable to avoid the making of a presumption and leave the character and effect of the payment to the drawing of inferences. Presumptions lend themselves to tactical ploys in litigation; it is better that parties be encouraged to present the totality of their case, particular in cases arising out of commercial transactions. 107. There are cases in Hong Kong where the courts have applied or referred to Seldon v Davidson without subjecting it to critical scrutiny – see Mak Ka Hing v Rang Ming Chung[82] (where judgment for the plaintiff was justified on the findings made by the Deputy Judge who disbelieved the defendant) and El Vince Ltd v Wu Wen Sheng[83] (where Seldon v Davidson was distinguished). Likewise, in England, Seldon v Davidson has been applied or referred to uncritically – see, for example, Clark v Mandoj[84], Patel v Mirza[85] and Chapman v Jaume[86]. 108. As there is nothing in these cases which provides additional support for the reasons that led to the conclusion reached in Seldon v Davidson, I consider that it should not be followed in Hong Kong. 109. It follows that the appellant’s case on the onus of proof fails. Mr Justice Ribeiro PJ: 110. The Court unanimously dismisses the appeal and makes an order nisi that the appellant pay the respondents’ costs, with liberty to the parties to lodge written submissions on costs within 14 days from the date of this judgment and that, in default of such submissions, the order nisi should stand as an order absolute without further direction. Mr Benjamin Yu SC and Ms Sara Tong, instructed by Rowdget W Young & Co, for the plaintiff (appellant) Mr Denis Chang SC, Mr Lawrence Ng and Mr Christopher Chain, instructed by Ford Kwan & Co, for the 1st and 2nd defendants (respondents) [1] [1968] 1 WLR 1083. [2] Para 11, Court of Appeal’s judgment. [3] The facts and pleadings are complicated, I will deal with them in some detail when I come to the plaintiff’s appeal against the trial judge’s rejection of its case. [4] Re-Amend Statement of Claim in HCA 1957/2005. [5] Para 7A. [6] Para 4 Re-Re-Re Amended Defence (“the defence”). [7] See para 6 defence. [8] Paras 9-11 defence. [9] Para 17A defence. [10] Para 13(a) defence. [11] Para 7, defence. [12] Para 15(a) and (g). In an earlier version, BIC denied that the signature on the Settlement Agreement was Mr Lee’s. [13] Paras 6 & 7 of the defence. [14] Para 8 reply. [15] Mr Ambrose Ho SC was counsel for the defendants. [16] I shall only mention grounds which are relevant to the appeal to us. [17] Para (1) at B61. [18] Notice of appeal, p 3, para (3); and repeated before us, para 28, Appellant’s case. [19] It is unlikely that the plaintiff would have been given leave to appeal on the facts. [20] In his oral submissions in this court, Mr Yu modified his position and asked for a new trial instead. He said, insufficient material had been placed before us to enable us to conclude the factual issues in the plaintiff’s favour. [21] See Chinachem Charitable Foundation Ltd v Chan Chun Chuen (2011) 14 HKCFAR 798, paras 40-58. [22] Para 95. [23] This was one of the reasons given by the learned judge in support of his rejection of the case based on FEAs. [24] The defence was that they were handed over to a Mr Li, an agent of the plaintiff. That was denied and the judge rejected the evidence because he was not satisfied with the oral evidence surrounding the agent. Unfortunately, the judge failed to consider the strong evidence of currency exchange. Nor the implication of BIC’s receipts acknowledging receipt of RMB dealt with in paras 28 &29 below. [25] On my calculation at the exchange rate of HK$1 to RMB 1.1201. [26] At exchange rates ranging from 1.0830 and 1.1201 according to my calculation. [27] Para 117. [28] Para 118. [29] Mr Yu relied on this in support of the loan agreement. But “temporary loan” is an unlikely description for a loan which was not repayable for a year. [30] Paras 118 and 119. [31] With the agreement of Kwan and Chu JJA. [32] No doubt, Mr Lee thought so too, hence his initial case that his signature was forged. [33] In this context, I also mention the first letter of demand relied on by BIC, which the judge rightly held to be a forgery. But in this letter, the plaintiff asked for the return of HK$ 104,554,260 with no mention of interest, saying “(BIC) now no longer requires the Renminbi Funds” supporting the defence that they were involved with an exchange of HK$ for RMB. [34] The Court of Appeal was of the view that this alternative defence was unmaintainable. Willmer LJ was of the view that the loans were repayable on demand and Edmund-Davies LJ said at 1090 because the defendant had repudiated the loans by claiming that they were gifts, that rendered the loans immediately repayable. Thus, the other discussions in Seldon arose in the context of a sole viable defence based on gift. [35] Willmer LJ at 1088B and Edmund-Davies LJ at 1089H. [36] I believe “voluntary” is used in the sense “as denoting the obtaining or giving of something without anything being obtained in return.” Overseers of The Savoy v Art Union of London (1896) AC 296 at 305. [37] Not even, as WLB v Islington shows, when the loan turned out to be ultra vires. See also the dictum of Browne-Wilkinson J in Re Sharpe [1980] 1 WLR 219 at 223B cited by Hobhouse LJ in Clark v Mandoj see below at para 51 in support of his statement at p 5 that the plaintiff’s evidence of “the loan directly went to negative the concept of a resulting trust.” [38] 708F and 709B, Lord Browne-Wilkinson. [39] In a matrimonial dispute over whether the husband had acquired any beneficial interest in the matrimonial home purchased by the wife in her sole name by way of a resulting trust as a result of work he had done on the house by way of redecoration and improvement. [40] With the agreement of Thorpe and Etherton LJJ. [41] At p 4, Hobhouse LJ emphasised the words “on the terms put forward by the plaintiff”. [42] Young v Queensland Trustees Ltd (1956) 99 CLR 560. [43] This is not the time to delve into the history or evolution of the presumption of a resulting trust, which in relation to immovable property appears to arise by analogy with “the rule of common law, that where a feoffment is made without consideration, the use results to the feoffor.” Lord Upjohn in Pettitt quoting Eyre CB in Dyer v Dyer (1788) 2 Cox, Eq. Cas. 92. WLB v Islington shows a resulting trust may arise on a voluntary payment, however, generally speaking, I believe the burden is on the payer to show that it was a voluntary payment. [44] Modern pleadings backed by statement of truth would support this view. [45] With the concurrence of the other members. [46] Kleinwort Benson Ltd v Lincoln City Council (1999) 2 AC 349 at 409B. [47] Mr Denis Chang SC for the defendants have referred us to, for example, Mitchell McInnes, “Unjust enrichment, Juristic Reasons, and Palm Tree Justice: Garland v Consumers’ Gas Co” (2004-2005) 41 Can Bus LJ 103. [48] Ribeiro PJ in Shanghai Tongji at para 76. [49] [1968] 1 WLR 1083. [50] Phipson on Evidence 18th ed para 6-06. [51] [1956] 99 CLR 560. [52] ibid at 564; see also at 568. [53] ibid at 569-570. [54] [1976] VR 788. [55] [1968] 1 WLR at 1088B. [56] ibid at 1088F. [57] ibid at 1089H. [58] ibid at 1090F-G. [59] [1976] VR 788 at 789. [60] 7th ed at 1210. [61] ibid. [62] (1956) 95 CLR 353 at 364. [63] (1980) 32 ALR 153 at 158. [64] (1984) 155 CLR 242 at 246-247, 266-267. [65] [1980] 1 WLR 219 at 223B. [66] [1996] AC 669 at 708A. [67] ibid at 708B. [68] ibid at 708C-D. [69] ibid at 708G. [70] See Hudson, Understanding Equity and Trusts, Routledge-Cavendish, 3rd ed 2008 at p 87. [71] 4 Esp 10. [72] ibid at 10-11. [73] 1 Stark 474. [74] ibid. [75] 4 Taunt 294. [76] ibid. [77] ibid. [78] [1930] 45 CLR 111. [79] ibid at 113. [80] [1976] VR 788 at 789. [81] High Court of the New Zealand Auckland Registry, CIV 2006-404-5720 [2007] NZHC 1827 (11 June 2007). [82] [2011] 1 HKLRD 347. [83] [2011] 4 HKLRD 541. [84] Court of Appeal unreported judgment delivered on 19 March 1998. [85] [2015] 1 ALL ER 326. [86] [2013] 1 FCR 619. 
Press Summary (English) Press Summary (Chinese) FACV No. 6 of 2012 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION FINAL APPEAL NO. 6 OF 2012 (CIVIL) (ON APPEAL FROM CACV NO. 251 OF 2010) Between : Chief Justice Ma: 1. I agree with the conclusion and orders set out in the judgment of Mr Justice Litton NPJ. So far as the reasons for the conclusions reached on the main issue in the appeal are concerned, I agree with the views set out in the judgment of Mr Justice Ribeiro PJ and Mr Justice Gleeson NPJ. I agree with the reasons contained in the judgment of Mr Justice Litton NPJ in relation to the cross appeal. Mr Justice Chan PJ: 2. I agree that this appeal should be allowed and that the Court of Appeal’s orders should be set aside and the orders made by the Judge restored. Brief facts 3. I would gratefully adopt the facts (which are not seriously in dispute) as set out in detail in the judgment of Mr Justice Litton NPJ. In brief, on 11 January 2008, the appellant and the respondent entered into a preliminary agreement whereby the appellant agreed to sell to the respondent the property at Rooms 1 to 5 of the Penthouse on the 37th floor of the Bank of America Tower, together with a car park space on the 4th floor (“the Property”) at the price of $138 million. The appellant was selling as confirmor, having entered into an earlier agreement (on 14 November 2007) to purchase the Property from Nesco Ltd. Both agreements were to be completed on 17 June 2008. The respondent had paid $13.8 million to the appellant as deposit. No formal sale and purchase agreement was signed pursuant to the preliminary agreement but no issue was taken by the parties. 4. In accordance with conveyancing practice, the appellant’s solicitors delivered to the respondent’s solicitors a total of 97 title deeds and documents as set out in a list, including the originals of the assignments executed in 1990 and 1991 which were accepted as the intermediate root title of the Property for the purpose of s.13 of the Conveyancing and Property Ordinance, Cap 219 (“the Ordinance”). The respondent’s solicitors raised various requisitions and the appellant’s solicitors purported to answer them. 5. On 17 June 2008, Nesco Ltd and the appellant completed the sale and purchase between them, but the respondent declined to complete, claiming that the appellant had failed to answer satisfactorily their requisitions, to show and prove good title to the Property and to produce the originals of certain documents. The appellant forfeited the deposit and the respondent demanded its return. Present appeal 6. It is now no longer necessary to deal with all the matters raised in the requisitions. The only grounds relied on by the respondent’s solicitors which are relevant to the present appeal are: (1) The appellant failed to produce on completion the originals of 20 pre-intermediate root title documents and 9 post-intermediate root title documents or to provide a satisfactory explanation for their absence. The respondents’ case is that the appellant had failed to comply with the common law obligation to deliver the originals of all title documents; and (2) The appellant failed to show and prove good title to the Property in that it had failed to produce the original or a certified copy of the original of 4 documents which were agreements entered into in 1979 and 1998 relating to the sea water cooling system of the Bank of America Tower. In this connection, the respondent contends that the absence of the 1979 Pump Pit Tenancy Agreement is fatal and that certified copies of the certified copies of the originals of the other documents is not acceptable under s.13(2) of the Ordinance. 7. The Court of Appeal (reversing the trial judge) held in favour of the respondent on the 1st ground mentioned above, holding that the appellant was under a common law obligation to deliver on completion the originals of all title deeds and documents, both pre-root and post-root, or to provide a satisfactory explanation for their absence but had failed to discharge such an obligation. The appellant appeals against this conclusion. 8. The Court of Appeal decided against the respondent on the 2nd ground above, holding that the 1979 agreement had ceased to have effect and that s.13(2) of the Ordinance permits the production of a certified copy of a certified copy of the original. The respondent seeks to overturn these findings in resisting the appeal. Showing and proving of title 9. There is no dispute that a vendor in a sale and purchase of real property transaction is under a duty to show and prove his title in the property before the date of completion and a duty (a separate and distinct duty) to give a good title to the purchaser on completion. The object of the transaction is to enable the purchaser to acquire all the rights and interests in the property which flow from the vendor’s title in the property according to the terms and conditions of the agreement on the agreed date of completion. The purchaser must be given a reasonable opportunity to examine the vendor’s title before completion and raise requisitions and ask questions relating to the title. 10. The practice in Hong Kong in the showing and proving of title is for the vendor’s solicitors to produce all the relevant documents of title to the purchaser’s solicitors for examination before the date for completion upon the undertaking that they would be returned to the vendor’s solicitors for the purpose of completion. The title documents required for showing and proving title are set out in s.13(1) of the Ordinance (namely, the Government lease and the post root documents of title). Under s.13(2), it is sufficient compliance of s.13(1) to produce attested or certified copies of the relevant title documents. However, although the production of attested and certified copies of title documents would suffice for the showing and proving of title, it is seldom that the vendor would choose to produce only the attested or certified copy of a relevant title document at this stage and keep the original for the completion stage if he has the original in his possession or power. Where the property is, for instance, still under a legal charge to a bank, the practice, I am given to understand, is for the vendor’s solicitors to “borrow” the originals of the title documents from the bank for the purpose of this stage of the transaction. Hence, if only the attested or certified copy but not the original of a title document is produced, it would be legitimate, where appropriate, for the purchaser’s solicitors to raise this with the vendor’s solicitors in advance so that the purchaser would know whether he can expect to receive the original of that document on completion. 11. This was apparently what happened in the present case. The respondent’s solicitors asked to see the originals of 29 title documents and were told that the vendor did not have them in its possession or power. The draft statutory declaration prepared by the vendor’s solicitors in connection with their absence was admittedly not satisfactory or acceptable. The respondent argued that the appellant had failed to comply with the obligation to deliver those original title documents and was thus entitled to treat the agreement as repudiated and to refuse to complete. Main issue in the appeal 12. The main issue in the appellant’s appeal concerns the obligation of the vendor to deliver the originals of the documents of title on completion. The parties accept that there is at common law an obligation on the part of the vendor to hand over the original documents of title in his possession or power but differ on the scope of such obligation and the consequence of a failure to comply with such obligation. This dispute involves a consideration of 3 separate but related aspects of this obligation: what documents are required to be delivered by the vendor to satisfy this obligation; whether the vendor is required to provide a satisfactory explanation for the absence of any original title document; and whether the purchaser is entitled to refuse completion if the vendor fails to comply with this obligation. Appellant’s position 13. The appellant’s position appears in paragraph 23 of its Written Case, that is: (1) A purchaser is entitled to insist on the vendor handing over to him the original documents of title which are in the vendor’s possession or power; and (2) A purchaser may be entitled to insist on the best explanation possible as to why the original of a title document cannot be produced if there is a realistic possibility that his failure to do so could lead to his being affected with notice of a prior equitable interest. 14. Mr John McDonnell QC (leading Ms Liza Cruden) for the appellant argues that a vendor is only obliged to deliver on completion those original documents of title which are in his possession or power; that if the original of any title document is not in the vendor’s possession or power, he is not required to provide any explanation for its absence and the purchaser is not entitled to refuse completion. However, he accepts that if there is a realistic possibility that the absence of the original of a title document could lead to the purchaser being affected with notice of a prior equitable interest and there is no satisfactory explanation for its absence, then the purchaser is entitled to refuse to complete. He submits that with the registration system under the Land Registration Ordinance in Hong Kong and on the facts of this case, there is no realistic possibility that the respondent would be taken to have notice of any prior equitable interest as a result of the appellant’s failure to provide the originals of the title documents in question. Respondent’s arguments 15. On the other hand, Mr Edward Chan SC (leading Mr Paul Lam SC) for the respondent quite fairly puts forth 4 alternative propositions: first, a vendor is under an obligation to deliver the originals of all title documents, both pre root as well as post root title documents; second, a vendor is obliged to deliver the originals of all post root title documents; third, a vendor is obliged to deliver the originals of all title documents which are material to the giving of his title to the purchaser; and fourth, a vendor is only obliged to hand over the originals of all title documents which are in his possession or power. He submits that he would succeed if the common law is represented by either the first or the second proposition. For the first proposition, he relies on Re Duthy and Jesson’s Contract [1898] 1 Ch 419 and Yiu Ping Fong & another v Lam Lai Hing [1999] 1 HKLRD 793 (and the cases which followed Yiu Ping Fong); and for the second proposition, he draws an analogy with s.13A of the Ordinance (although this provision which was enacted after the dispute in this case arose does not apply to the present case). He submits that the Court should accept either the first or the second proposition and reject the third and fourth as they are not supported by any authorities. Obligation to deliver originals 16. The vendor’s obligation to deliver the original title documents on completion was described by Romer J in Re Duthy and Jesson’s Contract [1898] 1 Ch 419 as “the ordinary obligation of handing over on completion all title deeds in their possession or power” (p.422). They are, according to him, “prima facie the deeds [which] the vendors would in the ordinary course be obliged to hand over on completion to the purchasers.” (See also the note to the judgment in Parr v Lovegrove (1857) 4 Drewry 170at 182-183.) Obviously, the title documents to be handed over are only those which “related solely” or “exclusively” to the land being sold (Emmet and Farrand on Title, #21.008 and Williams’ Title to Land (4th ed) p.547) and do not extend to title documents relating to any other land. 17. The vendor’s obligation to hand over on completion the original title documents in his possession or power was also considered in Yiu Ping Fong, where Yuen J (as she then was) said at p.798: “However, a vendor cannot make or give good title, by handing over only certified true copies at completion without an adequate explanation as to why the originals cannot be handed over. The handing over of original title deeds and documents (or at least those which relate exclusively to the property being sold) is an important part of the vendor’s obligation in a sale of land. That obligation is well established in the common law.” 18. The reasons for such an obligation are not difficult to understand: the purchaser has a proprietary right to these title documents after he has become the owner of the land to which these documents relate; the title documents are his; they run with the land; they are the evidence of his title. (See e.g. Re Williams and Newcastle’s Contract [1897] 2 Ch 144; Clayton v Clayton [1930] 2 Ch 12.) As Yuen J in Yiu Ping Fong put it at 798H: “A purchaser of land is entitled as a matter of proprietary right to possession of the original title deeds, Williams on Title (4th ed.) at p.547, which is the best evidence of ownership.” 19. There is another reason for imposing such an obligation: to ensure that the title is a good title which would be free from any encumbrance. The learned judge in Yiu Ping Fong added at 798I: “Further, it is important for him to get possession of the original title deeds so that he can be sure that the property is not subject to an equitable mortgage by deposit of title deeds. If the deeds are in the hands of some person other than the vendor, the purchaser is thereby placed on enquiry as to the reason for this, and he has constructive notice of the facts which such enquiry would disclose (see Emmet on Title, # 5. 147).” 20. It must be noted that it is far from clear from what the learned judge said in Yiu Ping Fong that she required that all title deeds, whether pre root or post root title deeds, must be handed over on completion. That was not a case concerning pre root title documents; the missing original title deed there was an assignment from the developer to a predecessor in title in 1986 which was a post root title document. This was pointed out by Cheung JA in Downpower Trading Ltd v Apexcom Ltd [2010] 1 HKLRD 915, #29. I would add that in view of what is required to be shown and proved under s.13(1), the absence of pre root title documents will now seldom, if ever, have any impact on the title to be acquired by a purchaser. I believe it is unlikely that Yuen J had in mind pre root title documents when discussing the vendor’s obligation and purported to lay down a requirement for the handing over of pre root title documents. 21. I should also mention that as a result of the decision of Yiu Ping Fong, s.13A was added to the Ordinance. The enactment of this provision would seem to acknowledge the existence of such a common law obligation on the part of the vendor to deliver the relevant title documents in his possession or power on completion. However, it would also appear that legal practitioners were worried that this decision would have the effect of requiring as a matter of law delivery of all the originals of even pre root title documents and that this would cast an overly onerous, if not impossible, burden on a vendor and his solicitors. The effect of s.13A (which is not applicable in this case) is to restrict the title documents to be handed over by the vendor to the purchaser on completion to those which the vendor is required to produce for showing and proving title under s. 13(1)(a) and (c). Satisfactory explanation for absence of original 22. If the vendor’s solicitors produces only a certified copy of a relevant title document at the showing of title stage, as discussed above, it is legitimate, where appropriate, for the purchaser’s solicitors to enquire whether the original is in the vendor’s possession or power and whether he, the purchaser, would receive it on completion. If the vendor has the original in his possession or power, he is obliged to hand it over to the purchaser on completion. But if he does not have the original in his possession or power, he is expected to give an explanation for its absence. There may be good reasons for its absence, such as loss or destruction over the years. This may already be obvious from the rest of the title documents. What if the vendor refuses to explain or cannot offer a satisfactory explanation? 23. This aspect of the vendor’s obligation did not arise for consideration in Re Duthy. In that case, the owner mortgaged the property in 1848 to certain settlement trustees to secure a loan which was subsequently paid off but no reconveyance was executed. Some of the mortgagee trustees had died and the mortgage deed and other title deeds were in the custody of the solicitors for the mortgagee trustees or their representatives. The purchaser required these deeds to be handed over on completion. The vendors applied to the solicitors for the return of these documents without success. The Court held that the vendor had a duty to obtain these documents at their own expense but had failed to take steps to do so. The documents in question were clearly in existence and within the powers of the vendors to obtain. There was no question in that case of requesting the vendors to provide a satisfactory explanation for not able to hand over those documents. 24. In Yiu Ping Fong, the learned judge considered that “it is legitimate for the purchaser to examine whether there is sufficient conveyancing evidence to explain [the loss of a title document]” (p.798J). However, she took the view that the vendor could not make or give good title “without an adequate explanation as to why the original cannot be handed over” (p.798F). What the judge said must be read with caution and in context. The title document whose original was missing in that case was the assignment in 1986 from the developer of the land to the first predecessor in title. There was a statutory declaration made in Taiwan explaining the loss, but the judge found that it could not be validly used in Hong Kong, suggesting that the vendor could supply the explanation through another means. When is a purchaser entitled to refuse to complete? 25. The question, which is the crucial question in the present appeal, then arises: whether the purchaser is entitled to refuse to complete if the vendor refuses to give any explanation or if he cannot offer a satisfactory explanation for the absence of an original title document. Mr Chan argues that the purchaser can refuse to complete, relying on what the judge said in Yiu Ping Fong. However, Mr McDonnell submits that the purchaser cannot rescind and is relegated to his right to sue the vendor in detinue. 26. It is important to bear in mind that we are not dealing with the situation where there is no evidence of the title document in question or it contents – the vendor is able to produce a certified copy of that document pursuant to s.13(1) of the Ordinance. We are dealing with the situation where the vendor is unable to deliver the original of the title document on completion and refuses to explain or cannot offer an explanation for its absence. 27. Obviously it would defeat the very object of a sale and purchase transaction if a purchaser is entitled to refuse completion even if the originals of the most insignificant title documents which have no realistic possibility of affecting good title are missing and the vendor cannot for some reason offer any satisfactory explanation for their absence. Several examples were discussed during the hearing of this appeal, Take the common example of an owner who had for some reason failed to pay management fees and as a result, the owners incorporation imposed a charge on his property and registered it at the Land Office but the arrears were later paid and the charge was removed. It would defy common sense to say that the purchaser is entitled to require delivery of these documents or a satisfactory explanation for their absence, failing which he can refuse completion. 28. This is to be contrasted with the example raised by the Chief Justice during argument of a vendor who had only acquired the property a few years prior to the sale and purchase in question but had without satisfactory explanation failed to hand over the original of the assignment to him and could only offer a certified copy of that title deed. The purchaser in such a case will legitimately be concerned as to whether there would be any problem with his title after the purchase is completed. If there is a problem with his title to the property, detinue is neither a sufficient relief nor consolation if the original is not in the vendor’s possession or power. 29. In my view, the fact that the purchaser has a right to the delivery of the title documents does not necessarily mean that he is automatically entitled to rescind and refuse completion if the vendor fails to do so or explain why. Whether the purchaser can do so must depend on the circumstances of each case. In considering whether he is entitled to refuse completion, one should bear in mind that the object of the transaction is for the purchaser to acquire a good title in the property. As discussed earlier, the purchaser as the owner of the property has a proprietary right in the relevant title documents which is evidence of his title in the property. As the Master of the Rolls said in Bryan v Busk (at p.4), the title deeds are the ordinary and primary means for asserting the owner’s title and defending his possession. In Yiu Ping Fong, Yuen J also said, the purchaser is naturally concerned that his title to the property is not subject to an equitable mortgage by deposit of title deeds by his predecessor. All this is the purpose of requiring the vendor to hand over the original of the title documents. 30. In Moulton v Edmonds, the Lord Chancellor said at p.249: “Perhaps the test may be, whether the recited deeds not produced cast any reasonable suspicion upon the title shown by the deeds produced.” 31. In my view, if there is a reasonable suspicion on the title where the vendor cannot deliver on completion the original of a relevant title document or to provide a satisfactory explanation for its absence, the purchaser should not be compelled to complete the transaction, running the risk that his title may in future be validly questioned. In such a case, but only in such a case, he is entitled to refuse completion. 32. In the present case, the originals of the title documents in question are clearly not in the appellant’s possession or power. But it is not contended that there is any realistic possibility that the absence of the originals of these title documents has any impact on the title to be acquired by the respondent. It is also accepted that the possibility of the creation of an equitable charge by the deposit of title deed is quite remote. In such circumstances, the respondent should not be allowed to refuse completion. For these reasons, the appellant’s appeal must succeed. Respondent’s other grounds 33. The respondent’s submissions can be disposed of very briefly. For the reasons given by Mr Justice Litton in his judgment, I agree that there is no substance in Mr Chan’s arguments. I would just add the following observations. First, in respect of the missing 1979 Pump Pit Tenancy Agreement, there are concurrent findings (beyond reasonable doubt) by the lower courts that it had ceased or expired. There is also not the slightest possibility that it would still have any legal effect. Secondly, the object of s.13(2) of the Ordinance is to facilitate the showing and proving of title. The very purpose of attestation and certification is to ensure that the copy which is produced for use after it has been duly attested or certified is the same as the original. Once it is accepted that attestation or certification involves a conscious verification process, there is no reason why an attested or certified copy of an attested or certified copy of the original should not be acceptable for the purpose of showing and proving title under s.13. Mr Justice Ribeiro PJ and Mr Justice Gleeson NPJ: 34. We have read the reasons for judgment of Mr Justice Litton NPJ, and agree with his conclusion and with the orders he proposes. He has dealt in detail with the facts and the issues that were argued, and we will confine ourselves to a brief statement of our views on what we regard as the central question in the appeal. 35. Although s 13A of the Conveyancing and Property Ordinance (“the CPO”) does not apply to this case because it did not come into effect until 11 July 2008, it is convenient to use the language of the CPO, including s 13A, to explain the issue that arises. 36. Sections 13 and 13A of the CPO refer to three different, but related, subjects: (a) the documents a purchaser is entitled to require a vendor to produce as proof of title at the stage of showing title (s 13); (b) the documents a purchaser is entitled to require the vendor to deliver upon completion for the purpose of giving title (s 13A(1) and (2)); and (c) the purchaser’s proprietary right in a document (s 13A(3)). 37. The language of s 13A(1), in referring to a purchaser’s entitlement to require the delivery of a document for the purpose of giving title, addresses the question whether a purchaser may refuse to complete if there is a failure to deliver a particular document, or, to put it in another way, whether a court will force title upon a purchaser without the delivery of the document. At common law, that may depend upon the availability of secondary evidence of the document. We note for future reference that s 13A(2) allows recourse to the common law if the specified originals cannot be delivered. 38. In Re the Halifax Commercial Banking Company Limited and Wood (1898) LT 536, Lord Lindley MR said at 539: “The purchaser was willing to complete, but the vendors could not produce the deeds, they having been unfortunately lost by their solicitors after the contract had been signed. There is no condition providing for the non-production of lost deeds. But the mere fact of their loss does not release the purchaser from the performance of his contract. He can be compelled to complete if he is furnished in proper time with satisfactory secondary evidence of the lost documents. This was decided in Bryant v Busk (4 Russ 1) and Moulton v Edmonds (1 De G.F & J 246).” 39. In the same case, Chitty LJ said, at 539-540: “The two authorities cited of Bryant v Busk and Moulton v Edmonds show that the court of equity will force a purchaser where the deeds are supposed to have been lost to take the title upon secondary evidence as to the contents of the deeds, and their having been duly executed and duly stamped. But in my opinion the secondary evidence produced ought to be clear and cogent … so that the purchaser may maintain his title against all those who may attack him when he is in possession and so that he may pass on the title in the ordinary way in the market to a purchaser or to a mortgagee.” 40. The existence of the secondary evidence qualification, and the rationale for it as explained by Chitty LJ, shows that the principle concerning the purchaser’s entitlement to refuse to complete is not co-extensive with the purchaser’s proprietary right to title deeds. As was submitted for the appellant in the present appeal, statements of the law sometimes tend to conflate the two. For some purposes that may not matter, but in the present case it does. The purchaser’s proprietary interest in documents of title is qualified in an important respect. The relevant principle was stated by Romer J in Re Duthy and Jesson’s Contract [1898] 1 Ch 419 at 422 as follows: “The purchaser is not asking to have these deeds produced for the verification of or for information as to the title, but is calling upon the vendors to fulfil the ordinary obligation they are under of handing over on completion all title-deeds in their possession or power.” The purchaser’s right to the title deeds as chattels, which may be vindicated, for example, by an action in detinue, is subject to the qualification in the concluding words of that passage. Secondary evidence does not come into it. The purchaser as the new owner will be entitled to all the title deeds that relate exclusively to the property, and, insofar as they are in the possession or power of the vendor, the vendor must hand them over. On the other hand, the purchaser’s interest in obtaining a good and marketable title, referred to by Chitty LJ, is protected by the requirement of clear and cogent secondary evidence, sometimes referred to as a “satisfactory explanation” of the absence of a document. 41. In this case, there is no argument about the matters referred to in para 36(a) and (c) above. Section 13 of the CPO dealt with showing or proving title, and the appellant-vendor was willing to hand over all the documents that were in its possession or power. The issue concerns para 36(b), and arises because, although the appellant had certified copies of certain documents of title, and although the register kept under the Land Registration Ordinance contained memorials of all such documents, the statutory declaration which the appellant intended to provide concerning the missing original documents did not satisfy the requirements for secondary evidence. It showed that the original documents were not in the possession or power of the appellant, but did not contain clear and cogent evidence of their loss or destruction (cf Phipson on Evidence, 17th ed., 41-23). It was, however, accepted in argument by the respondent that there were no grounds for suspecting that any of the original documents which were not in the appellant’s possession or power had been used in any unwritten and unregistered transaction giving rise to rights which could still be enforced in 2008 against the respondent if it had taken title. 42. The case of Moulton v Edmonds, referred to by Lord Lindley MR and Chitty LJ, left open a question which is of present importance. Lord Campbell LC ((1859) 1 De GF & J 246 at 249), after referring to missing title deeds that could not be produced, said: “Perhaps the test may be, whether the recited deeds not produced cast any reasonable suspicion upon the title shown by the deeds produced”. That question did not have to be decided because his Lordship went on to hold that it was shown or accepted in that case that due search had been made and that there was sufficient secondary evidence of the due execution and contents of the missing deeds. 43. The test that Lord Campbell raised as a possibility is consistent with what Chitty LJ later said was the rationale for the requirement of secondary evidence. It is to protect the purchaser against a risk of some future challenge to his title. If there is such a risk that, in turn, affects the marketability of the title. 44. The purchaser’s capacity to “maintain his title against all those who attack him while he is in possession” is related to questions of actual or constructive notice that may affect priorities in the case of an outstanding and competing interest. (See Agra Bank v Barry (1874) LR 7 HL 135 and Oliver v Hinton [1899] 2 Ch 264.) 45. In the present case, there was a combination of the following circumstances. First, the vendor satisfied all the requirements of s 13 of the CPO by way of proof of title. The vendor proved its title to the land by the means prescribed by the statute. Secondly, the vendor was willing to hand over on completion all documents of title in its possession or power. Thirdly, there were no grounds for suspecting that any of the original documents which were not in the appellant’s possession or power had been used in any unwritten and unregistered transaction giving rise to rights which could still be enforced in 2008 against the purchaser. Why, in those circumstances, should the purchaser not be required to complete? In our view, the vendor’s inability to produce an original document of title rather than a certified copy, and to account for the absence of the original to the standard required for secondary evidence, would only have justified refusal to complete where the absence of the original would indicate the realistic possibility of some transaction affecting the land which could affect the purchaser if it took title. 46. At the hearing, there was much discussion of the judgment of Yuen J (as Yuen JA then was) in Yiu Ping-fong v Lam Lai-hing [1999] 1 HKLRD 793 at 798. After pointing out that s 13(2) facilitates proving title by allowing certified copies, her Ladyship stated that that section “does not ... exonerate the vendor from producing at completion the originals of such title deeds and documents, at least those that relate exclusively to the property being sold.” And that “a vendor cannot make or give good title, by handing over only certified true copies at completion without an adequate explanation as to why the originals cannot be handed over.” Yuen J also noted the purchaser’s proprietary right to possession of the original title deeds and that the requirement for originals was to ensure that the property is not subject to an equitable mortgage by deposit of title deeds. 47. All judgments must be read bearing in mind the question which the court is actually having to decide. Yuen J was concerned with a case where the original of a key document in the chain of title – the first assignment from the developer in 1986 executed some 12 years before the relevant contract of sale – could not be delivered and where satisfactory secondary evidence was found to be unavailable. 48. Yuen J was not concerned with any questions regarding the giving of good title in relation to pre-intermediate root documents and certain first instance decisions which took her Ladyship to be suggesting that there was an obligation to deliver or explain the absence of the originals of documents relating exclusively to the property, however ancient, and whether or not they gave rise to any realistic risk that title may be affected, have erroneously taken that judgment out of context. Mr Justice Litton NPJ: Introduction 49. We are concerned here with a preliminary agreement for sale and purchase dated 11 January 2008. The Respondent is the purchaser. It was a confirmor sale: In other words, the vendor (Appellant) had earlier entered into an agreement to purchase the same property from a head vendor (Nescon Ltd), for completion of the sale earlier on the same day, 17 June 2008. The property comprises the Penthouse of Bank of America Tower, a 37-storey building standing on IL8294 Central Hong Kong: Described in the agreements as Rooms 1, 2, 3, 4 and 5 on the 37th Floor of the high-rise building together with a car parking space on the 4th Floor, being in total 76/10,000 share of IL8294. The vendor in Clause 3 of the agreement undertook to sell the property free from encumbrances. By Clause 4 the vendor undertook at completion to give vacant possession of the property to the purchaser except for Room 5[1]. 50. The purchase price for the property was $138 million. The deposit totalling $13.8 million was to be paid in two tranches: $4,267,800 upon signing the preliminary agreement and $9,532,200 upon signing a formal sale and purchase agreement on or before 18 February 2008. When the parties signed the preliminary agreement the vendor was represented by the solicitors’ firm Lo Wong & Tsui but the purchaser was not legally represented. The parties failed to agree upon the terms of the formal sale and purchase agreement but the sale still went ahead and the purchaser paid the second tranche of the deposit. $13.8 Million Forfeited 51. Nescon Ltd duly completed the sale on 17 June 2008 but the Respondent on its part repudiated the sale and demanded the return of the deposit paid of $13.8 million which the Appellant had forfeited. The Buyer’s Case 52. In essence, the buyer’s case was that because the vendor: (i) failed to produce “certified copies” of agreements dating back to 1979 and 1988 relevant to the sea-water cooling system of Bank of America Tower (and, in relation to a 1979 agreement any copy), and (ii) was able to produce only certified copies but not the originals of a whole host of other documents, the vendor had failed to answer the objections and requisitions satisfactorily and/or was not able to show, prove or give good title. The buyer repudiated the sale on these grounds and demanded the return of its deposit. 53. The vendor subsequent sold the property for $88.8 million. Legal Proceedings 54. The Respondent (buyer) commenced proceedings in the Court of First Instance for repayment of the $13.8 million (plus $86,910 for the costs of investigating title). The Appellant (vendor) counterclaimed for damages for wrongful repudiation. The matter went before Yam J who, by his judgment dated 20 October 2010, dismissed the buyer’s claim and gave judgment on the vendor’s counterclaim for damages to be assessed. 55. The buyer appealed. 56. The Court of Appeal (Tang VP, Fok JA and Lam J) agreed with the trial judge that the buyer’s repudiation based upon the ground referred to in para 52(i) above was baseless, but reversed Yam J on the objections and requisitions as mentioned in para 52(ii) above: More particularly, on what the Court of Appeal categorized as failure to produce before completion the originals of title documents (falling both inside and outside the period of 15 years referred to in s 13(1)(a)(ii) of the Conveyancing and Property Ordinance, Cap 219[2], CPO) or provide any satisfactory explanation for such failure. 57. The original documents which the vendor failed to produce were as follows[3]: (a) Two legal charges dated 30 May 1990 (evidenced by Memorials No. UB4460364 and UB4460370 registered in the Land Office) and a deed of release dated 12 October 1998 (Memorial No. UB7600635). The 30 May 1990 documents executed by Delvincourt Ltd, charged Rooms 4 and 5 in favour of Standard Chartered Bank in support of a loan. (b) The deed dated 12 October 1998 re-assigned Rooms 4 and 5 back to Delvincourt Ltd as absolute owner. (c) Four assignments: · 27 May 1985 Assignment of Room 4 and 15/10,000 share by National Bank of Canada to Hong Kong Carpet Manufacturers Ltd · 30 July 1985 Assignment of Room 5 and 15/10,000 share by Bank of Nova Scotia to Hong Kong Carpet Manufacturers Ltd · 26 August 1986 Assignment of Room 4 and 15/10,000 share by Hong Kong Carpets (Holdings) Ltd to Tai Ping Carpets Ltd · 26 August 1986 Assignment of Room 5 and 15/10,000 share by Hong Kong Carpets (Holdings) Ltd to Tai Ping Carpets Ltd 58. Originally, the purchaser also required the production of the originals of three tenancy agreements dated respectively 12 June 1993, 28 May 1994 and 1 November 1996. The terms under those agreements had long expired. Counsel before us did not pursue the point relating to these three instruments any further. 59. No question arises as regards Rooms 1, 2 and 3 of the Penthouse, and the carparking space on the 4th Floor. 60. The Court of Appeal found that the vendor’s failure to account for the originals of the documents (referred to in paras 57(a) (b) and (c) above) was fatal to the vendor’s case. Hence the vendor’s appeal to this court. Before us, the Respondent maintains its case on the original grounds before the trial judge and cross-appeals on the basis that the vendor’s failure to produce the “certified copies” as mentioned in para 52(i) above entitled it to succeed. The Chain of Title 61. The agreement for sale by Nescon Ltd (the head vendor) to the Appellant is dated 14 November 2007. 62. Nescon Ltd derived its title to the property from Delvincourt Ltd under an assignment about nine years earlier dated 12 October 1998. 63. Delvincourt Ltd in turn took its title from Tai Ping Carpets Ltd under two assignments: one dated 30 May 1990 relating to Rooms 4 and 5 with 30/10,000 share of IL8294, the other dated 20 August 1991 relating to Rooms 1, 2 and 3 and the carparking space on the 4th Floor, with 46/10,000 share of IL8294. 64. As mentioned earlier, the date for completion of both the head agreement and the agreement in question was 17 June 2008. Nearly four months before that date, on 21 February, the Appellant’s solicitors sent to the Respondent’s solicitors “all the title deeds and documents” set out in a list comprising 97 deeds and documents. This included originals of the assignments referred to in paras 62 and 63 above: The assignments by Tai Ping Carpets Ltd to Delvincourt Ltd (para 63 above) went back some eighteen years from the time of the agreement in question: Sufficient, prima facie, to satisfy the requirements of s 13(1)(a)(ii) of the Conveyancing and Property Ordinance for proof of the title of the head vendor Nescon Ltd and for the vendor to pass good title to the buyer. Objections to Conveyance 65. By letter dated 29 February, the buyer’s solicitors raised a whole series objections and requisitions. These will be dealt with in segments of this judgment below. Objections to Copies 66. These objections relate to matters in the 1970s and 1980s when arrangements were made between three parties: The Hong Kong Government, the owner of an adjoining high-rise building Hutchison House and the owner of Bank of America Tower (then known as Gammon House). These arrangements are referred to in para 52(i) above. They concern the sea-water cooling systems for the two buildings, serviced by pipes under government land. These arrangements were embodied in what were called Pump Pit Tenancy Agreements and Pipeline Wayleave Agreements. The Respondent’s solicitors wanted certified copies of those documents “for perusal”. In relation to a 1979 document the vendor had no copy. In relation to the 1988 copies, the solicitor’s certificate was appended on a copy previously attested by two solicitors’ clerks (what counsel calls “copies of copies”). 67. The buyer’s solicitor asserted the right to see those documents for this reason: The agreement of 11 January 2008 contained a schedule forming part of its terms. Clause 1 of the schedule says that the subject property was the same property as set out in an assignment registered by Memorial No. 7599026. When one turns to that assignment, dated 12 October 1998, between Delvincourt Ltd as vendor and Nescon Ltd as purchaser, one sees that the sale is said to be inter alia subject to and with the benefit of the agreements relating to the pump pit and pipe-lines (as referred to in para 52(i) above). 68. From a commonsense point of view, the purchaser’s request may seem surprising. The whole estate and interest in the land vested in Nescon Ltd, which the Appellant had contracted to sell on as confirmor, was no more than 76/10,000 share of the land and building (together with the exclusive right to use occupy and enjoy the Penthouse and the carparking space on the 4th Floor). How such an owner could be expected to have – or be able to procure – in 2008 copies of such documents, beyond those he already had (some of them going back 30-odd years) is difficult to understand. 69. The practical reality is that if, through some breach of the arrangements embodied in those tenancies and wayleave agreements, the cooling system failed, it would have affected the entire building and all its occupants. Likewise other arrangements for servicing a multi-storeyed office block, such as electricity supply and the servicing (and perhaps replacement) of the lifts if those failed. This is simply an incident of the management of high-rise buildings in multiple ownership, of which all buyers (and their solicitors) must be aware. A careful buyer, before entering into a contract to buy units in such a building, would have made inquiries as to the arrangements for such services, and whether a special fund[4] has been set up to provide for exceptional expenditures and the amount in such fund. These arrangements, good or bad, might well affect the price the buyer is willing to pay but do not affect the title of the owner to his units: unless of course some charge has been levied by the manager of the building, which has been registered against the owner’s units under the Land Registration Ordinance, Cap 128. 70. The fact that the pump pit tenancies and wayleave agreements appear as recitals to the assignment of 12 October 1998 (see para 67 above) does not take the matter any further. The fact that those arrangements exist is not in dispute. The question, at the end of the day, is whether they have any effect on the title sought to be conveyed. The answer is plainly No. 71. And yet, when it came to litigation between the parties, that was apparently the first point taken by the buyer. The trial judge Yam J devoted many pages of his judgment in disposing of this point against the Respondent, though not on the grounds outlined above. It was apparently assumed by all counsel that those documents were relevant to the title that the vendor had contracted to convey. So the argument at trial focussed on two matters: (i) whether contractual arrangements for the supply of sea-water for the cooling system in 1988 had replaced the earlier arrangements, making the absence of any copy of a 1979 Pump Pit Tenancy Agreement irrelevant, and (ii) whether the requirements of s 13(2)[5] of the CPO were satisfied by the production of certified copies of the 1988 agreements, where the solicitor’s signature is appended to a copy attested by two solicitors’ clerks. 72. The Judge was upheld in this regard by the Court of Appeal. Hence the Respondent’s cross-appeal. The Respondent’s cross-appeal 73. Mr Edward Chan, SC has focussed his submissions solely upon para 71(ii) above. 74. As the Court of Appeal said (para 23) the point turns on the construction of s 13(2)5. 75. This point can be tested by reference to one of the documents in issue, an agreement between the developers of Hutchison House and Gammon House (as the Bank of America Tower was then called). The document produced by the vendor bears two endorsements. (i) An endorsement made on 7 July 1981 whereby two clerks of the firm of solicitors Deacons asserted that they had examined the original document and “certified and attested” that the document was a true and correct copy of the original. (ii) A solicitor’s certificate, endorsed on either the same document or a copy thereof, dated 10 July 1985 (that is to say, after 1 November 1984, in terms of s 13(2)(b) of Cap 219) whereby he stated that having examined the document he was able to say that it was a true and complete copy “of the said Attested copy”. 76. As the Judge said, and the Court of Appeal agreed, what s 13(2) required was the production of a copy attested or certified to be a true copy. It did not in terms require an attested or certified copy of the original[6]: Although, in this case, the attestation was clear evidence that the two clerks had examined the original. What in effect the solicitor in turn was stating was this: On the evidence before him he was satisfied that it was a true copy of the original and so certified that it was “true”. Thus, even where a public officer or solicitor certifies as true a copy of a certified copy of a relevant document for the purposes of s 13(2)(b), what is being certified as a true copy is the copy of the original. The whole thrust of s 13(2) is to facilitate conveyancing. The section places reliance on the integrity and judgment of public officers and solicitors. When they have certified that a copy is true, the parties can rest content that it is so, unless extraneous evidence shows otherwise. 77. I would dismiss the Respondent’s cross appeal. A Common Law Rule: Yiu Ping Fong v Lam Lai Hing 78. I now turn to what appears to be the more substantial part of this appeal: A supposed common law rule which solicitors in Hong Kong thought had been laid down by Yuen J in Yiu Ping Fong v Lam Lai Hing [1999] 1 HKLRD 793 at 798 F – H to this effect: Unless excluded by express contractual provisions, a vendor at completion was duty bound to deliver to the purchaser all original title documents going back to the root of title, however remote, if those documents relate exclusively to the property sold. If, prior to completion, the vendor was unable to provide a satisfactory explanation as to why he would not be able to do this at completion, the purchaser was entitled to rescind. The way the learned judge puts it is this: “…A vendor’s solicitor can as part of proof of title under s.13(1) send certified true copies of title deeds and documents to the purchaser’s solicitor for perusal, without having to prove that the originals are lost. However, a vendor cannot make or give good title, by handing over only certified true copies at completion without an adequate explanation as to why the originals cannot be handed over. The handing over of original title deeds and documents (or at least those which relate exclusively to the property being sold) is an important part of the vendor’s obligation in a sale of land. That obligation is well established in the common law. A purchaser of land is entitled as a matter of proprietary right to possession of the original title deeds (Williams on Title, 4th ed. p.547), which is the best evidence of ownership. Further, it is important for him to get possession of the original title deeds so that he can be sure that the property is not subject to an equitable mortgage by deposit of title deeds. If the deeds are in the hands of some person other than the vendor, the purchaser is thereby placed on enquiry as to the reason for this, and he has constructive notice of the facts which such enquiry would disclose (SeeEmmet on Title, para 5.147). Where, therefore, as in this case, the Vendor has given notice that she would not be able to produce the original 1986 assignment on completion, it is legitimate for the purchaser to examine whether there is sufficient conveyancing evidence to explain its loss.” 79. As will be noted, Yuen J in Yiu Ping Fung stated (at 798 H) that the vendor’s obligation was “well established in the common law”. This is a very wide statement: And, as such statement, it was thought binding by solicitors and was followed in a number of subsequent cases, with unfortunate results for the vendor. Guang Zhou Real Estate v Summit Elegance 80. For example, Guang Zhou Real Estate v Summit Elegance [2000] 2 HKLRD 855 (Deputy Judge Chu): It concerned the sale of five lots of vacant land. The contract was made on 4 September 1997, for completion on 31 January 1998. The purchaser made requisitions, identified as No. 10 and 14. No. 10 related to the Crown lease of one of the lots. The vendor was only able to produce a certified copy of the original. The original was never in its possession. Requisition No. 14 concerned a deed of release relating to a right of way over one of the lots. The vendor could only produce a certified copy. It never had the original. 81. The sale agreement in Guang Zhou Real Estate (clause 7) required the vendor to “show and give good title…pursuant to s 13 of the Conveyancing and Property Ordinance….and furnish to the purchaser such attested or certified copies of any deeds or documents of title,..as may be necessary to prove such title.” [Emphasis added] The deputy judge construed Yiu Ping Fong as laying down a common law rule that to give good title, the vendor had to deliver the originals of title documents or provide a satisfactory explanation for their absence, however remote in time such documents might be. As will be noted later on in this judgment, the delivery of originals of documents, which deals with the right of property in those documents as chattels, is a different matter from the obligation to give good title. 82. Take another example where the application of the principle articulated in Yiu Ping Fong had unravelled the parties’ bargain: Loyal Hope Ltd v Leung Pui Ming and 2 others [HCA 136/2007, 20 March 2008 Master M Yuen, unreported] where the agreement for the sale of a ground floor unit at 267 Castle Peak Road Kowloon was made in October 2006. The ultimate root of title went back to the year 1957. The purchaser’s solicitor required the production, before completion, of the originals of 30 documents dated between 1957 and 1971, on the authority of Yiu Ping Fong. The vendor’s solicitor replied (p.6-H in the unreported judgment): “We opine that the Yiu Ping Fong case is distinguishable as the document in question in that case is not pre-intermediate root instrument”. In reply the purchaser’s solicitor said: “With respect, Yiu Ping Fong’s implication extends to all original title deeds relating exclusively to subject property reaffirms the common law position that a purchaser is entitle(d) to all title documents and there is no destination (distinction) between pre-root or post-root documents. We reiterate our requisition.” [p.6-L] 83. Eventually the vendor’s solicitor offered to produce certified copies of the requested documents with a statutory declaration by the vendors that they never had the originals ever since the date of their purchase of the unit back in 1989. To this offer, the purchase’s solicitors replied as follows: “We reiterate our stance for the original documents. The declaration of lost being conveyancing evidence to account for the non-production must be made by the person who has personal knowledge on how the document(s) was received and lost. The declaration proposed by you, according to Yiu Ping Fong’s ruling is not acceptable by our client.” 84. The outcome was that the buyers refused to complete. Of the 30 missing documents the vendors were able to produce photocopies of 17. As to the remaining 13 all they could do was to provide copies of the memoranda of registration, as appearing in the Land Office register. 85. Multiply parties were involved, going back to 1957. To require the vendor, who had acquired the property in 1989, to explain who, why and how those other parties had lost the originals would seem an impossible task. The Master nevertheless gave judgment in favour of the buyers. At para 55 of the judgment the Master said: “55. The [vendors] were unable to give good title to [the purchaser] unless [they] were able to provide a satisfactory declaration to account for the loss of the title documents” 86. In so concluding the Master was following Guang Zhou Real Estate (supra) and Deputy Judge To in Goldmex Ltd [2006] 2 HKLRD 795, whilst noting that Reyes J in C & W Watch Company [HCMP 920/2005, 10 October 2005, unreported] had doubted the correctness of Yiu Ping Fong. 87. Here, the Court of Appeal, in allowing the buyer’s appeal against Yam J’s judgment, affirmed the formulation of the common law duty in Yiu Ping Fong. At para 47 Tang VP said: “…I also agree with Yiu Ping Fong and Guang Zhou Real Estate Development (HK) Co Ltd. I am also of the view that the obligation to produce original title deeds and documents extend beyond the intermediate root. In other words, notwithstanding section 13, a vendor is required to supply all the original title deeds which relate exclusively to the property sold which should be in his possession or power. Otherwise, the vendor must provide clear and cogent proof of: the contents of the missing document; its due execution; and the fact of its loss or destruction. Wu Wing Kuen and Ors v Leung Kwai Lin and Another [1999] 3 HKLRD 738.” 88. This common law obligation is formulated in the leading Hong Kong text book on Hong Kong ConveyancingLaw and Practice by Sihombing and Wilkinson (Butterworths) Chapter VI para 79.5 in these terms; “The common law position [79.5] At common law, unless otherwise agreed, a vendor, who has agreed to give good title, has an obligation, by way of giving such title, to deliver to the purchaser the originals of all title deeds that relate exclusively to the property to be assigned right back to the ultimate root of title. If such original title documents are missing, the vendor must provide a statutory declaration as to how they have come to be missing….” 89. As the authors later on noted (para 79.6), solicitors (and their clients) faced serious problems in complying with the common law rule enunciated in Yiu Ping Fung, since, in many instances, vendors could not trace the original title deeds relating to the property extending back to the ultimate root of title. 90. This could mean going back as far as 1844, with a World War in between, during which Hong Kong was occupied by enemy forces. s.13A CPO 91. Seeing the unsettling effect on conveyancing practice caused by Yiu Ping Fong as applied in later cases the legislature amended s 13 of the CPO by adding s 13A in July 2008 which provides: “Section 13 A (1) Unless the contrary intention is expressed, a purchaser of land shall be entitled to require the vendor to deliver to him, for the purpose of giving title to that land, the original of both of the following only- (a) if there is a Government lease that relates exclusively to the land, the lease; and (b) any document that relates exclusively to the land and is required to be produced by the vendor as proof of title to that land under section 13(1)(a) and (c). (2) Subsection (1) does not affect any rule of common law under which the vendor may discharge his obligation to give title to that land otherwise than by delivering the Government lease or document to the purchaser. (3) If the vendor is not required to deliver to the purchaser a document in giving title to that land, the purchaser has no proprietary right or ownership in the document. (4) The fact that- (a) the vendor is not required to deliver to the purchaser a document in giving title to that land; and (b) the purchaser has no proprietary right or ownership in the document, does not affect the right or interest of any other person in that land.” The Common Law Rule 92. The effect of s 13A will be discussed later. Here, I confine myself to an analysis of the common law rule, which is the crux of this aspect of the appeal. The Two Extremities 93. The completion of a contract for the sale of land entails the execution of the conveyance by the seller and the payment of the purchase price by the buyer[7]. Or, perhaps more accurately, the delivery into the purchaser’s hands of a duly executed deed of assignment, against the payment of the purchase price. 94. As a note in the report of Parr v Lovegrove [1858] 4 Drewry 66 at 70 says: “In general a purchaser on completing is entitled to have delivered up to him the title-deeds, assuming that they are in the vendor’s power. His right to them rests on the fact that when he has become complete purchaser, the title-deeds are his; they are the evidences of his title, not of the vendor’s who has by the transaction ceased to have any estate or title.” In other words, on completion, the right of property in the title deeds passes to the purchaser who has immediate right of possession over them. 95. The question is this: To fulfill the vendor’s obligation to give title, is the delivery of the title deeds part of the process of giving good title, or is it merely an incidence of conveyancing? In a preliminary agreement such as we have in this case, where it makes no mention of the delivery of title deeds into the buyer’s hands, can the seller simply say to the buyer: “I have given you title by handing over to you a duly executed deed of assignment. As to the title documents in my custody, I acknowledge your proprietary rights over them. Sue me in detinue if you will, but pay me the balance of the purchase price in the meanwhile.” This treats the title documents as mere chattels, the ownership of which passes with the land; but the giving of title is achieved by delivery of the executed deed of assignment and not by delivery of the title documents. 96. This is one view, adumbrated by Mr McDonnell QC in the course of argument before us. 97. At the other extremity is the broad proposition in Yiu Ping Fong – at least as understood by the profession and followed in subsequent cases: To repeat the proposition in para 78 above: Unless excluded by express contractual provisions, a vendor, in order to give good title, must deliver to the purchaser at completion all original title documents going back to the root of title, however remote, if those documents relate exclusively to the property sold. Unless the vendor is able to provide certified copies of such documents and provide a satisfactory explanation as to why the originals cannot be produced, the purchaser is entitled to rescind. 98. The harshness of such a rule is self-evident. 99. The common law abhors extreme or unreasonable outcomes[8]. If, to give good title, the common law requires the delivery of title deeds (as an act that goes to the giving of title) some middle ground, resulting in balanced justice, must be found. What Is The Rule or Principle? 100. A contract for the sale of land depends upon the ordinary law of contract. As Lord Hoffmann said in Jumbo King Ltd v Faithful Properties [1999] 2 HKCFAR 279 at 299C: “Contracts for the sale of land are not exceptions to the principle that parties have freedom of contract and may agree to whatever terms they like.” Where the parties have, through their solicitors, entered into a formal contract of sale, it would generally be assumed that they have provided for all the essential terms. This leaves little room for implication of terms. 101. But here we have only a preliminary agreement which by clause 2 envisaged that the parties would, on or before 18 February 2008, enter into a formal agreement. This never took place. But it is a binding legal agreement nevertheless. It is then the court’s function to give effect to the parties’ bargain, according to law. 102. In Clause 3 the vendor “undertook” to “sell the premises to the purchaser…. free from encumbrances”. It says nothing about giving good title to the buyer at completion. But, to make the contract work, that is an obligation which is necessarily implied. 103. Given that the vendor has an obligation to convey good title to the purchaser on completion, how is this to be accomplished? To make the contract work there must be a further implied obligation to be undertaken by the vendor: That it must, some reasonable time before 17 June 2008 (the day fixed for completion), deliver to the purchaser or its solicitors all the title documents the vendor was able to get from Nescon Ltd the head vendor, to enable the purchaser (with the advice of its solicitors) to be satisfied that the purchaser would have good title on completion. This is the so-called Hong Kong conveyancing practice (see Yiu Ping Fong at 798A) whereby title deeds and documents are delivered to the other side for the purchaser’s solicitors’ inspection and approval, those solicitors impliedly undertaking to hold the deeds and documents to the vendor’s solicitors’ order and to return the same on demand. This is what Deputy Judge J. Chan said in Lau Mui Fun v K.C. Chan & Co. [HCMP No. 572 /1992, 6 July 1992, unreported] at p.10 as: “the special and unquestioned trust one solicitor reposes on another in respect of custody of title deeds in conveyancing transaction is a cornerstone in the conveyancing practice established in Hong Kong.” 104. The system works upon trust. At the stage of showing title, the delivery of the documents is for the purpose of perusal by the purchaser’s solicitors. There is an implied representation that at completion these are all the documents the buyer will get with the executed assignment. 105. In this case the vendor’s solicitors, following normal practice, delivered on 21 February the deeds and documents (97 in number) to the purchaser’s solicitors for perusal, they having previously obtained the deeds and documents from Nescon Ltd and given an identical undertaking themselves to Nescon Ltd’s solicitors. 106. The obligations imposed on the vendor in this case to show good title and give good title arise by necessary implication, in order to make the contract work, and give effect to the parties’ bargain[9]. 107. What then is meant by the vendor’s obligation to make good title? The principle as stated in the leading textbook in Hong Kong, Hong Kong Conveyancing Law and Practice Chapter VI para 2 is as follows: “In a case where the vendor has agreed to give or make good title, his duty is to ensure that, by the completion date….the nature of his title is neither defective nor defeasible. He must deal, in a way acceptable ultimately to the court, with all defects in that title so that, paraphrasing the words of Lord Russell of Killomen in MEPC Ltd v. Christian Edwards [1981] AC 205, the facts and circumstance are to compelling that, beyond reasonable doubt, the purchaser will not be at risk of a successful assertion against him of any encumbrance…” 108. This is the principle under which a buyer would have acquired the property; it is entirely reasonable that, when he came to be the seller, he should be under the same obligation. “Real Risk” 109. There have been many cases in the past where, on an objection as to title, the question boiled down to this: Is there a real risk that a third party might assert some right over the property, thereby encumbering it or defeating the title altogether? Take the instance where structures have been added to buildings after the occupation permit was issued. Might this give rise to the Building Authority asserting its rights under section 24 of the Building Ordinance, Cap 123 to order the demolition of the structure and having the order registered in the Land Registry against the premises[10]? As Godfrey JA observed in Jumbo King Ltd v Faithful Properties Ltd [1999] 3 HKLRD 231 at 244 G: “ ‘Unauthorised structures’ are all too common in Hong Kong and present a real problem for conveyancers. A vendor whose title is, or may be, open to objection because there is, or may have been, erected on the property of which he is granting exclusive use to his purchaser some unauthorised structure is well advised to protect himself by precluding his purchaser, by contract, from raising any requisition or objection to the title based on an ‘unauthorized structure’ point....” On a further appeal in that case, this Court left it open as to whether the presence of unauthorized structures within the premises constituted risk of encumbrance so as to affect the vendor’s obligation to pass a good title: see Jumbo King Ltd v. Faithful Properties Ltd [1999] 2 HKCFAR 279 at 294. This Court said that it was a matter of degree. In that case the contract itself precluded the objection being raised. 110. A clear example of a case where the court held that there was no risk is Jumbo Gold Investment Ltd v Yuen Cheong Leung [2000] 3 HKCFAR 52 where a block of flats had been erected many years ago in apparent breach of a height restriction in the Crown Lease. The question was whether there was a real risk of the government taking action to re-enter by registering a memorial of re-entry under s 4 of the Government Rights (Re-entry and Vesting Remedies) Ordinance, Cap 126, extinguishing the interest of the owner (and that of every other owner in the building). The unanimous answer of the Court was No. As Bokhary PJ said (at p.60-61): “In my view, the evidence….did not exclude a reasonable possibility of an unwaived breach of condition which gives the Government a right of re-entry. The question is therefore whether, assuming that the Government has that right, there is any real risk that it would actually take the drastic step of enforcing it to the detriment of innocent owners. I entirely agree with Mr Justice Litton PJ that the correct answer is in the negative. It is simply not in the nature of good government to harm innocent people unnecessarily like that. Accordingly it is safe to proceed on the basis that the Government would never do so. The way in which this Court has decided the present case will, I trust, prevent cases of this type arising in future.” 111. In the context of the present case, the question of risk focusses on the possibility that an original deed of assignment might have been deposited with a third party as security for money borrowed. This question cannot be examined in a vacuum. It must be seen against the background of the Land Registration Ordinance, Cap 128 and in particular s 3 and s 4 which provide: “Section 3 Priority of registered instruments; effect of non-registration (1) Subject to this Ordinance, all such deeds, conveyances, and other instruments in writing, and judgments, made, executed, or obtained, and registered in pursuance hereof, shall have priority one over the other according to the priority of their respective dates of registration, which dates shall be determined in accordance with regulations made under this Ordinance. (2) All such deeds, conveyances, and other instruments in writing, and judgments, as last aforesaid, which are not registered shall, as against any subsequent bona fide purchaser or mortgagee for valuable consideration of the same parcels of ground, tenements, or premises, be absolutely null and void to all intents and purposes. Section 4 Notice of unregistered instrument not to affect registered instrument No notice whatsoever, either actual or constructive, of any prior unregistered deed, conveyance, or other instrument in writing, or judgment, shall affect the priority of any such instrument as aforesaid as is duly registered.” 112. The Court of Appeal in this case (para 39) thought that there was a real risk of the purchaser’s title being blemished by a prior encumbrance created by deposit of title deeds. Tang VP referred to Oliver v Hinton [1899] 2 Ch. 264, the headnote of which accurately summarizes the position in English law: “In order that a purchaser for value, who has acquired the legal estate without notice of a prior equitable mortgage of the property, may be postponed to that mortgage, it is not necessary to shew that he has been guilty of fraud, or negligence amounting to fraud; it is sufficient that he has been guilty of negligence so gross as to render it unjust to deprive the prior mortgagee of his priority.” What the Court of Appeal appears to have overlooked (perhaps because the case was never cited by counsel) was this: A situation like that in Oliver v Hinton could not have arisen in Hong Kong. In that case the deposit of title deeds to secure the repayment of £400 was accompanied by a memorandum of the deposit, with an undertaking to execute a legal mortgage if asked to do so. In Hong Kong the memorandum would have been registrable. (The properties in Oliver v Hinton were not located in counties which had a system of registering deeds and documents). And when, two years after the deposit of the title deeds, the owner executed a conveyance of the property to a purchaser, whose agent (a former solicitor’s clerk) never asked to see the deeds, the equities in that case depended on whether the purchaser had acted with such with gross negligence that she had to be postponed to the equitable rights of the chargee. In Hong Kong, once the assignment to the purchaser is registered the purchaser takes priority over an unregistered chargee. 113. This position was entrenched a hundred years ago in the case of Kwok Siu Lau v Kan Yang Che [1913] HKLR 52. The purchaser of a property had his agreement for purchase registered with actual knowledge of the existence of a prior 5-year lease (unregistered) over part of the property. He completed the purchase and registered his assignment. By section 4 of the Land Registration Ordinance, 1844 – the equivalent of the present-day section 4 – “no notice … either actual or constructive, of any prior unregistered deed … shall affect the priority of any such instrument… as is duly registered in pursuance of this Ordinance”. Rees Davies CJ held, affirmed by the Full Court, that the purchaser took the property free of the 5-year lease. As Havilland de Sausmarez J, said in the Full Court (p 65) the Ordinance had, at a stroke, cut off the application of the equitable doctrine of notice to an registered deed. Assuming that a purchaser like the one in Oliver v Hinton had, prior to completion, actual knowledge of the existence of an unregistered memorandum of deposit covering the property and went ahead nevertheless to complete the sale and then had his assignment registered, he would take the property free of the equitable mortgage. 114. Many generations of solicitors have practised conveyancing under the registration system in Hong Kong, well aware of the importance of having instruments registered. The Land Registration Ordinance (Ord No 3 of 1844) is one of the oldest laws passed in Hong Kong. At trial no suggestion was raised in this case that the head vendor Nescon Ltd’s title could have been blemished by any prior encumbrance effected by deposit of title deeds. The passing reference to the possibility of equitable mortgages being created by the deposit of title deeds in Yiu Ping Fong (at para 798-I) was not made with regard to the circumstances of conveyancing in Hong Kong. The learned judge merely cited a passage in Emmet on Title, para 5.147 (now para 5.129 in the latest edition). This dealt with the position in England generally, without any system of registration of instruments affecting land[11]. The Hong Kong practice of conveyancing, like that in Ireland where there is a similar system of registration, gives “a premium to diligence in registration”, using the words of Lord Cairns LC in Agra Bank Ltd v Barry [1874] LR 7 HL135 at 149. Hong Kong conveyancers would be well aware of this. 115. Take a hypothetical case. Mr X buys a newly completed flat from a reputable developer. The flat (and an undivided share of the land and building) is assigned to him and the deed of assignment is forthwith registered in the Land Office. Two years later he sells the flat to Y. In the normal course of events, when X’s solicitors deliver the title documents to Y’s solicitors for perusal, prior to completion, the original deed of assignment would have been included. But it is missing. There is only a certified copy. The purchaser’s solicitors might well ask why. This springs from the common experience of solicitors, and from common sense. The important point to note is this: In this example, there has been no intermediate transaction: Hence the priority of registered instruments under the Land Registration Ordinance does not apply. If X, before completion, had deposited the original assignment with a lender, accompanied by a note saying: “I hereby deposit my assignment with you as lender by way of security for a loan of $2 million, repayable on demand” and the lender registered the note in the Land Office prior to the assignment to Y being registered, Y would take the flat subject to the equitable charge. The Obligation To Explain 116. Of course, in a case like that outlined above, the purchaser could choose to take a chance and not raise an objection. The nearer to completion, the less his chance of being over-reached by a third party registering a memorandum of deposit before he can have his assignment registered. But a court would not require a purchaser to take such a chance and, in the circumstance outlined above, might well hold that the requisition was properly raised. The burden would then fall on the vendor to explain why he was unable to produce the original deed of assignment. There is nothing oppressive in the request. Had the matter been formulated in terms of the “officious bystander”[12] test, both parties would have said “of course the missing deed must be found, or the loss explained”. The matter was within the personal knowledge of the vendor. The obligation to explain springs from the circumstances of the particular case. 117. The matter is one of degree. The risk is that of a lender emerging from the dark with an original title deed in his hands and registering a memorandum of deposit ahead of the purchaser. The degree of risk – whether it can be regarded as “fanciful” – depends on the circumstances. Hence the importance of candour in the course of requisitions as to title: Facts known to the vendor material to the requisition must be disclosed: A principle emphasized in Active Keen Industries Ltd v Fok Chi-keong [1994] HKLR 396 at 407. 118. So far, the risk discussed is that of a lender with an unregistered memorandum of deposit: A lender who does not wish his transaction made public, for one reason or another, by registering a memorandum of deposit, revealing its terms. What of the position of a loan by parol, without a scrap of paper evidencing the loan and its terms: A situation also covered by the passage in Emmet on Title cited in Yiu ping Fong (referred to in para 78 above)? 119. As Mr Mcdonnell QC, counsel for the Appellant said at the hearing (and not contradicted by Mr Edward Chan SC) there is simply not one instance in the reports going back to the earliest days in Hong Kong of equitable mortgagees by deposit of title deeds having defeated the title of purchasers. 120. It is difficult to imagine a situation where a person would lend a substantial sum of money with nothing more than an oral commitment and a deposit of title deeds, when he could have safeguarded his security by requiring a memorandum of deposit and having it registered. One is not dealing here simply with the transaction between the lender and the borrower, against whom there is a personal right of action to recover the loan. One is dealing with the lender’s rights against a third party, the owner of the land. If it is security for the loan which the lender wants, why does he not secure it by having a written instrument and having that registered against the property? The Rule in Yiu Ping Fong 121. Pulling the threads together and returning to the broad principle of common law stated in Yiu Ping Fong, the foundation of that wide proposition, according to counsel, is the case of Duthy and Jesson’s Contract [1898] 1 Ch. 419, Romer J, where the contract to sell a freehold house was made in 1897. Nearly 50 years before it had been mortgaged and the mortgage deed and other title documents were in the custody of the solicitors for the mortgagees, but by the time of the sale of the house in 1897 the mortgagees were all dead. The mortgage had been paid off. The solicitors were holding the documents as bare custodians, claiming no lien over the documents. The vendor had an undoubted right of possession over those title deeds, being the owner of the freehold. And at completion the purchaser would equally have had a right of possession over those deeds. So it was simply a matter of trouble and expense in taking physical possession of the deeds on the vendor’s part. The solicitors claimed no right over the deeds but said they could not hand them over without the consent of the mortgagees’ personal representatives. Had the vendor asserted his right of possession over those deeds by court proceedings, he would unquestionably have obtained physical possession. But he did nothing. 122. Duthy and Jesson’s Contract has nothing to do with lost deeds. The deeds were within the vendor’s power because he had the right of possession over those deeds. What he did not have was physical possession. 123. The question raised before Romer J (see p. 419) was simply whether the vendor was bound at his own expense to get hold of the deeds for the purpose of handing them over to the purchaser on completion. No question of title was involved in the case (see p.423). It is no foundation for the broad principle stated in Yiu Ping Fong at 798-F: “… a vendor cannot make or give good title, by handing over only certified true copies at completion without an adequate explanation as to why the originals cannot be handed over. ” 124. The judgment in Yiu Ping Fong dealt with two requisitions, only one of which is relevant to this case. As to that, the judge after saying that the issue before her was the effect and extent of s 13(2) of CPO, went on as follows: “There has been much debate in this Court as to the true meaning and effect of s.13(2), and a number of authorities have been quoted to me, in particular, Ip Fung Yee v Norwegian Missionary Society [1998] 1 HKLRD 94, which on one reading seems to suggest that so long as certified true copies are produced, it is not necessary for a vendor to explain the loss of original title deeds.” 125. It appears that counsel on both sides treated the matter as one of law rather than one turning on the facts of that case. That probably explains why the facts are so sparse on the point. The sale in issue took place in 1998. The missing deed of assignment was made well within the 15 years in s 13(1)(a)(ii) of the CPO: The vendor had taken an assignment of the flat from a person called Chiu two years before, in 1996; Chiu had acquired the flat from her mother in 1990, the mother having bought the flat from developers a few years before that – in 1986. 126. The risk to the purchaser in Yiu Ping Fong was what the vendor might have done with the deed: She had bought the flat only about two years before the sale in 1998, presumably through solicitors. Those solicitors would have on record what title documents passed on conveyance. They would have known what took place in 1996, when they acted for the vendor in acquiring the flat from Chiu. But they seem not to have been asked. There might well have been grounds in Yiu Ping Fong for the judge to conclude that the vendor’s bare assertion – that she never had the original 1986 assignment (see p.799-H) – was not good enough. But the requisition was not focussed in this way. This might yet be another case where the process of requisitions was not centered upon the essential point of the case. 127. In so far as Yiu Ping Fong has been taken as authority for the broad proposition in para 78 above that must be over-ruled. Section 13A CPO 128. Where section 13A(1) imposes an obligation on the vendor, for the purpose of giving title, to deliver to the purchaser an original title deed[13], that requirement is subject to the common law rule: see subsection (2). If an original in the chain of title is missing, and such loss gives rise to no risk that the title to be passed to the purchaser might be blemished, then the vendor would have fulfilled his obligation of giving good title under s.13A(1). The Objection to Title In This Case 129. The question boils down to this: is there a real risk that, in taking an assignment from Nescon Ltd, with the Appellant as confirmor, the purchaser was not receiving a good title to the property in this case? 130. The intermediate root of title went back to May 1990 (see para 63 above) and the vendor was able to prove title by producing the original title deeds. As regards the charges referred to in para 57(a) above, the evidence shows conclusively that they had been discharged. This leaves the deeds of assignment referred to para 57(c) above. Take the instance of the assignment of Room 5 to Tai Ping Carpets Ltd. Is it conceivable that someone might, over 20 years later, come forward to say that he holds the deed of assignment dated 26 August 1986 because he had lent money to Tai Ping Carpets Ltd: A loan by parol, without a memorandum evidencing the loan: a memorandum which would have been registrable under the Land Registration Ordinance and “absolutely null and void to all intents and purposes” (s.3(2) of that Ordinance) as against subsequent bona fide purchasers for value? 131. Take further the assignments of Rooms 4 and 5 to Tai Ping Carpets Ltd both dated 26 August 1986 (Memorials No. 3146075 and 3146074), of which the originals were missing. Assume that, prior to the completion of the subsequent sales of those Rooms by Tai Ping Carpets Ltd to Delvincourt Ltd on 30 May 1990, Tai Ping Carpets Ltd had deposited the originals of its deeds of assignment with a lender in the circumstances adumbrated in para 115 above: Once the deed of assignment to Delvincourt was registered (as it was by Memorial No. 4460361) any rights which the lender might have had over those two Rooms as equitable chargee would have been extinguished. The title deeds in the lenders’ hands would have been of no worth. And a loan by parol can for practical purposes be disregarded. 132. In this case, ignoring the three tenancy agreements referred to in para 58 above, the missing title documents went back some twenty years, with many registered instruments affecting title in between. 133. Objections and requisitions proceed on practical realities, not theoretical possibilities. Solicitors need on each occasion to ask themselves whether their client the purchaser is taking a real risk when the vendor, in showing title, is able only to produce a certified copy of a title document and not the original. Conclusion 134. In my judgment in so far as the common law rule as stated in Yiu Ping Fong always requires a satisfactory explanation for the absence of an original title document before the court could order a purchaser to take title on completion, it is far too wide. The cases following that statement of principle were wrongly decided. 135. I would allow this appeal and restore Yam J’s judgment, though not on his grounds. I would direct that the parties have 14 days from the date of handing down our judgments to lodge written submissions as to costs, with any written submissions in reply within 14 days thereafter. Chief Justice Ma: 136. This appeal is unanimously allowed. The Court also makes the orders set out in para.135 above. Mr John McDonnell QC & Ms Liza Jane Cruden, instructed by Lo Wong & Tsui, for the Appellant (Defendant) Mr Edward Chan SC and Mr Paul Lam, instructed by Woo Kwan Lee & Lo, for the Respondent (Plaintiff) [1] “Unit 3705” in the agreement was sold subject to a tenancy yielding HK$78,700 per month, exclusive of management and air-conditioning fees and other expenses. [2] Section13(1)(a)(ii): ….. a purchaser of land shall be entitled to require from the vendor, as proof of title to that land, only production of the Government lease relating to the land sold and – (a) proof of title to that land …. (ii) extending not less than 15 years before the contract of sale of that land commencing with an assignment….. dealing with the whole estate or interest in that land; [3] Paras. 36, 62, 68-70 Tang VP’s judgment. [4] See Schedule 7 to the Building Management Ordinance, Cap. 344 para 4(1): “The manager shall establish and maintain a special fund to provide for expenditure of a kind not expected by him to be incurred annually”. [5] Section 13 (1) Unless the contrary intention is expressed, a purchaser of land shall be entitled to require from the vendor, as proof of title to that land, only production of the Government lease relating to the land sold and – (a) proof of title to that land- (i) where the grant of the Government lease was less than 15 years before the contract of sale of that land, extending for the period since that grant; or (ii) in any other case, extending not less than 15 years before the contract of sale of that land commencing with an assignment, a mortgage by assignment or a legal charge, each dealing with the whole estate and interest in that land; (b) production of any document referred to in the assignment, mortgage or charge mentioned in paragraph (a) creating or disposing of an interest, power or obligation, which is not shown to have ceased or expired and subject to which any part of that land is disposed of; and (c) production of any power of attorney under which any document produced is executed where that document was executed less than 15 years before the contract of sale of that land. (2) Where this section requires the production of any document, it shall be sufficient to produce a copy - (a) attested, before 1 November 1984, by 2 solicitors’ clerks; or (b) certified by a public officer or a solicitor, to be a true copy. [6] As Tang VP noted in para 31, in practice solicitors have generally accepted that a certified copy of a certified copy would fall within s.13(2) and referred to the Law Society Circular No. 87-48 as follows: “Section 13(2) requires the production of a certified copy of the original document or a certified copy thereof: A certified copy of a plain photocopy will not suffice”. [7] See Emmet and Farrand on Title: 2012 ed. para. 8.007: Definition – “Completion” normally means the “complete conveyance of the estate and final settlement of the business” (Turmer V.C. in Lewis v South Wales Railway [1852] 10 Hare 113, adopted in Killmen v France [1946] 2 All ER 83….Final settlement of the business normally involves the handing over of documents of title and, if sale is with vacant possession, the giving possession, in return for payment of the purchase money or the balance. [8] A phrase taken from the judgment of Gaudron J in Ebner v. Official Trustee in Bankruptcy [2000] 176 ALR 644 para 164 [9] See Lee Gee Kee v Chong Kai Tai [1996] 1 HKC 105 as an example of the court by necessity implying a term in a provisional sale and purchase agreement, in order to give efficacy to the parties’ bargain. There was a chain of confirmors in relation to the sale of a flat under development; the sales agreement was silent as to the process of completion; the question was whether payment of the price by the last buyer in the chain was to be simultaneous with the delivery of an executed deed of assignment by the last vendor: see Ching JA at 119I – 120B “[the] points depend upon the implication of a term that completion was to be simultaneous with the payment of the balance. I accept that, in vacuo, it may seem to be common sense that a purchaser should not have to pay his money in advance of completion….In the circumstances, obtaining in Hong Kong and in this particular case I do not find it possible to imply that payment and completion were to be simultaneous”. Litton VP at 121-I: “I have therefore reached the conclusion that, to give business efficacy to the contract, it was not necessary to imply a condition that the obligations of the purchasers and vendors must be simultaneous.” [10] Section 24 (1) Where any building has been erected, or where any building works or street works have been or are being carried out in contravention of any of the provisions of this Ordinance the Building Authority may by order in writing require— (a) the demolition of the building, building works, or street works…” (2) An order made under subsection (1) shall be served on— (a) in the case of a building or building works but subject to subsection (2A), the owner of the land or premises on which the building has been erected or on which the building works have been or are being carried out; (2C) The Building Authority may, upon the service of an order under subsection (2)(a), (b) or (c)(iii) or (2A), cause the order to be registered in the Land Registry against— (a) if the order has been served on the owner of any land or premises in accordance with subsection (2)(a) or (c)(iii), the land or premises. [11] Middlesex and Yorkshire being exceptions. [12] The court would only imply a term when it is satisfied that both parties would, as reasonable persons, have agreed to it, had it been suggested to them: see Chitty on Contract: vol. 1 31st ed 2012 ed. p.991. The celebrated test for implying a term set out in Shirlaw v Southern Foundaries [1939] 2 KB 206 at 227 requires both parties to suppress the officious bystander with a common “oh, of course!”. [13] as required to be produced as proof of title under s.13(1)(a) and (c) 
1. This appeal originally concerned two cases in the Magistrate’s Court, namely ESCC 3350/2014 (HCMA 617/2015) and ESCC 3406/2014 (HCMA 438/2015). With consent from all parties, the two cases were heard together. 2. There were three defendants in the case of ESCC 3350/2014 (HCMA 617/2015), they were Yip Po‑lam (D1 at trial, hereinafter called “D1”), Cheung Hon‑yin (D2 at trial, hereinafter called “D2”) and Wong Kan‑yuen (D3 at trial, hereinafter called “D3”). 3. There were two defendants in the case of ESCC 3406/2014 (HCMA 438/2015), they were Cheung Kwai‑choi (D1 at trial, hereinafter called “D4”) and Chow Nok‑hang (D2 at trial, hereinafter called “D5”). 4. Each of these appellants was charged with the offence of contravening an Administrative Instruction (“section 11”) issued under section 8(3) of the Legislative Council (Powers and Privileges) Ordinance(“LCPPO”), namely failing to “behave in an orderly manner” within the precincts of the Chamber of the Legislative Council (“LegCo”)[1]. They pleaded not guilty and were all convicted after trial by the Magistrates.[2] They now appeal against the convictions. 5. D4 and D5 were fined. D1 was sentenced to an imprisonment term of two weeks, D2 a term of one week and D3 a term of three weeks. D1 ‑ D3 also appeal against the sentence. 6. At the hearing, D2 was absent allegedly due to health reason. As all parties agreed, and as a matter of expediency, it was ordered that the case of D2 be severed and heard separately. It has not yet been heard at the time of this judgment. Prosecution Case 7. The two cases stemmed from two different incidents of demonstration in the precincts of the Chamber of the LegCo. The facts may be summarized as follows.[3] ESCC 3350/2014 8. In the afternoon on 6 June 2014, the Finance Committee (“FC”) of LegCo held a meeting at which the funding application for the advance works of the “North-East New Territories Development Proposal” (“NENT Proposal”) was discussed. While the meeting was held inside the LegCo Complex, a number of protesters demonstrated in the designated public activity area (“DPAA”) outside the LegCo Complex. 9. There was a lobby on G/F of the LegCo Complex (“the Lobby”). The entrance of the Lobby (ie, Public Entrance 1 or “PE1”) comprised two sets of automated glass doors, namely the “Western Doors” and the “Eastern Doors”. Members of the public (including protesters) were required to go through a registration process in order to obtain visitor’s passes for entering the LegCo Complex and the Lobby. The Lobby and PE1, as well as the DPAA outside PE1, were within the “precincts of the Chamber” as defined in the LCPPO. 10. To enhance the security of the LegCo Complex while the FC meeting was held, at about 5pm on 6 June 2014, LegCo security staff closed the Western Doors of PE1. The Eastern Doors of PE1 were guarded by security officers, with the automated operation of the doors changed to manual operation. 11. At about 5:27pm on the same day, a visitor fell on the floor while leaving the Lobby through the Eastern Doors. As a result, security officers were unable to close the Eastern Doors. At this juncture, a large group of protesters outside the LegCo Complex rushed into the Lobby through the Eastern Doors. Some protesters blocked the Eastern Doors with their bodies and prevented the doors from being closed by the security staff. At about 5:42pm, the Western Doors, which had previously been closed, were tampered with by some protesters who succeeded in opening the doors. More protesters then entered the Lobby through the Western Doors. At about 5:46pm, security staff successfully closed the Eastern Doors. However, the Western Doors continued to be blocked by the protesters who prevented the doors from being closed by the security staff. 12. Tens of protesters who had entered the Lobby without permission continued to occupy the Lobby for demonstration. They chanted slogans, displayed banners and played drums in the Lobby. Security staff guarded the escalators at the Lobby to prevent the protesters from going upstairs to the venue of the FC meeting. 13. Meanwhile, the Western Doors of PE1 continued to be blocked by the protesters despite the LegCo security officers repeatedly asked them not to obstruct the doors. At about 8:55pm, the LegCo sought assistance of the police. 14. At about 9:17pm, police officers, accompanied by the LegCo security staff, went to the Western Doors and attempted to close the doors. Their action was met with fierce resistance of the protesters who prevented the doors from being closed. A number of the LegCo security staff (including PW7, PW8, PW9 and PW15) sustained various degrees of injuries during the struggle. Some protesters then sat on the ground blocking the Western Doors. 15. At about 9:45pm, the FC meeting was adjourned. The protesters who had been blocking the Western Doors ceased obstructing the doors at about 9:57pm. However, a large number of protesters remained in the Lobby for demonstration, and refused to leave the Lobby in spite of repeated warnings given by the LegCo security staff. At about 11:35pm, the majority of the protesters began to leave the Lobby. The last batch of protesters left the Lobby after midnight. ESCC 3406/2014 16. The FC meeting for the discussion of the funding application for the NENT Proposal resumed at 3pm on 13 June 2014. While the meeting was held inside the LegCo Complex, a number of protesters demonstrated in the DPAA at the car park of the LegCo Complex. At the entrance of the car park there was a walkway which was covered with a canopy (the “Canopy”, which was at least 8 meters above ground). The Canopy was within the “precincts of the Chamber” as defined in the LCPPO. 17. At about 5:05pm on the same day, D4 went onto the Canopy and hang some strings down from the Canopy. D5 and other protesters, who were on the ground, then tied some folded banners onto the strings so that D4 could lift up the banners onto the Canopy. A number of warnings were issued by the LegCo security staff through a loudhailer to D4 and D5, who ignored the warnings and continued with their acts. A security officer approached D5 and tried to stop his act, but he insisted to tie the banners onto the strings. 18. D4 lifted up altogether four banners onto the Canopy. Between about 5:22pm and 5:54pm, he hang the four banners bearing the slogans “撤回計劃”, “寸土必爭”, “寸步不讓” and “東南西北” (each of about 8m x 2m in size) down from the Canopy. 19. At about 6:09pm, D4 climbed down from the Canopy himself. Defence Case 20. At trial of both cases, all the appellants[4] exercised their rights and elected not to testify. No defence witness was called. Findings of the Magistrates 21. In both cases, there was challenge that section 11 and the related offence were unconstitutional. There was also dispute on admissibility of some evidence (which is not a concern in this appeal) as well as on factual allegations. 22. The two cases were dealt with by different Magistrates. Both Magistrates held that section 11 and the related offence were constitutional. Further, both of them found the prosecution proved beyond reasonable doubt all the elements of the offence and found all the appellants guilty of the offence. 23. In ESCC 3350/2014, the Magistrate found D1 and D3 had performed the following acts during the incident: (1) D1 had, without obtaining any permission or visitor’s pass, entered and remained in the Lobby for more than five hours (between 7:21pm on 6 June 2014 and 12:27am on 7 June 2014), during which she delivered two speeches (between 9:13pm and 9:25pm (the “First Speech”); and between 11:05pm and 11:09pm (the “Second Speech”)) to other protesters in the Lobby via a loudhailer. In her First Speech, D1 expressed that she was the organizer and asked the protesters to sit together and take a photo of “filling up the LegCo Lobby (in original words: 塞爆立法會大堂”. After the FC meeting was over, D1 remained in the Lobby. (2) Later when D1 was notified by LegCo security staff to leave the Lobby, she did not do so but remained until the end. D1 left the Lobby only at 12:27am on the next day, and was amongst the last batch of protesters leaving the Lobby. (3) At all material times, the protest in the Lobby was peaceful and there was no evidence that D1 (or other demonstrators with her in the Lobby) actually disturbed the LegCo sitting, attempted to go to the upper floor where the Chamber was, obstructed LegCo members from attending the meeting, or behaving in breach of peace or exercising any violence at all. (4) Between 9:07pm and 9:17pm, D3 and D2 blocked the Western Doors with their bodies, thereby preventing the doors from being closed. (5) At about 9:17pm, when the police and the LegCo security staff attempted to close the Western Doors, D3 together with others resisted (D2 left the doors at this time). In spite of repeated announcements made by the police through a loudhailer asking people to leave, D3 continued to obstruct the Western Doors until 9:57pm. The conduct of D3 and others had indirectly caused injuries to some LegCo security staff. Grounds of Appeal D1: Yip Po-lam 24. D1 was represented by Mr Martin Lee SC[5], Ms Linda Wong and Mr Geoffrey Yeung, the grounds put forward on behalf of D1 for the appeal against conviction are:[6] (1) The Magistrate erred in finding that the precincts of the Chamber is not a public place, and therefore erred in finding that the relevant rights are not engaged. (2) The Magistrate erred in failing to hold that section 11 of the Administrative Instruction is unconstitutional for lack of legal certainty and therefore failed the requirement of “prescribed by law”. (3) The Magistrate erred in law in not holding that section 11 imposes unnecessary and / or disproportionate restriction on the relevant rights and therefore is unconstitutional. (4) If the offence is constitutional, the Magistrate erred in finding D1 guilty when the prosecution failed to prove all the elements of the offence. (5) The Magistrate erred in holding that D1 committed the offence when the prosecution’s case based only on the “failure to behave in an orderly manner” limb of the offence. (6) The Magistrate erred in equating a “failure to behave in an orderly manner” under section 11 with “behaving in a disorderly manner” under section 17B of the Public Order Ordinance. (7) The Magistrate erred in convicting D1 of the offence without defining what constitutes a failure to “behave in an orderly manner” under section 11. (8) The Magistrate erred in holding that D1 failed to behave in an orderly manner when there was no order to speak of at the material times in the Lobby. (9) The Magistrate failed to give reasons or sufficient reasons for convicting D1. (10) The Magistrate erred in convicting D1 by assuming that section 11 applied to persons who were not admitted into the precincts of the LegCo Chamber. (11) The conviction of D1 was unsafe and unsatisfactory. 25. Before the hearing, Mr Lee SC stated in his written submission that in the light of the Court of Final Appeal decision in HKSAR v Fong Kwok Shan,[7] he would not further advance Grounds 1 and 2 and would only pursue the other grounds. D3: Wong Kan-yuen 26. D3 was represented by Mr Randy Shek,[8] the grounds he put forward for D3 are:[9] (1) Section 11 is unconstitutional, in that: (a) the Magistrate erred in holding that the precincts of the Chamber is not a public place, and therefore erred in finding that the relevant rights of D3 was not engaged; (b) the Magistrate erred in failing to hold that section 11 and the offence charged lacked legal certainty and therefore failed the requirement of being “prescribed by law”; and (c) the Magistrate erred in failing to hold that section 11 imposes unnecessary and / or disproportional restriction on the relevant rights and therefore is unconstitutional. (2) If section 11 is constitutional, the Magistrate erred in finding that D3 failed to “behave in an orderly manner”: (a) in the light that the prosecution case based solely on D3’s failure to “behave in an orderly manner”; (b) he erred in treating a failure to “behave in an orderly manner” as the same as “behaving in a disorderly manner” under section 17B of the Public Order Ordinance[10]; (c) he erred in law in convicting D3 of the section 11 offence without defining what is meant by “failing to behave in an orderly manner”; (d) he erred in law in convicting D3 of the offence since there was no order to speak of at the material time and place; and (e) he erred in applying section 11 to persons who have not been admitted to the precincts of the Chamber in convicting D3. (3) The conviction was unsafe and unsatisfactory. 27. Mr Shek informed the court that he would fully adopt the submission put forward by Mr Lee SC and Mr Tam, supplementing with oral submission. D4: Cheung Kwai-choi 28. D4 was represented by Mr Jeffrey Tam and Mr Andrew Lau. The following grounds were put up on behalf of D4:[11] (1) the charge is unconstitutional because it is not prescribed by law; (2) the Magistrate is wrong in law in applying ordre public as the legitimate purpose; (3) the charge is unconstitutional because it is a disproportionate restriction; (4) the Magistrate applied the wrong test and erred in finding that D4’s conduct amounted to not “behaving in an orderly manner”; (5) the Magistrate erred by applying section 11 to persons not admitted or persons who do not require a visitor’s pass to the precincts of the Chamber; and (6) the Magistrate erred in convicting D4 where the prosecution’s case is based solely on the limb of failure to “behave in an orderly manner”. 29. Before the hearing, Mr Tam also stated in his written submission that in the light of HKSAR v Fong Kwok Shan,[12] he would not further advance Grounds 1 and 2 and would only pursue the other grounds. D5: Chow Nok-hang 30. D5 was represented by Mr Randy Shek,[13] who put up similar grounds for D4, as follows:[14] (1) the Magistrate erred in holding that the charge is constitutional; (2) the Magistrate erred in law in applying ordre public as the legitimate purpose to justify the restriction of the right of D5 of peaceful assembly and freedom of expression; (3) the Magistrate applied the wrong test in holding that D5’s conduct amounted to not behaving in an orderly manner; (4) the Magistrate erred by applying section 11 to persons not admitted or persons who do not require a visitor’s pass to enter the precincts of the Chamber; and (5) the Magistrate erred in convicting D5 of the offence where the prosecution’s case based solely on failure to “behave in an orderly manner”. 31. Mr Shek also informed the court that he would adopt the submission of Mr Lee SC and Mr Tam, as appropriate, on behalf of D5, supplementing with oral submission. Discussion and Consideration The constitutional issue 32. The grounds put forward for each of the appellants include the constitutional issue: whether section 11 and the related offence are necessary and proportional. I will consider this issue first. 33. The relevant rights engaged in the present case are the rights to freedom of expression, demonstration and assembly. Such rights under the International Covenant on Civil and Political Rights (“ICCPR”) are implemented by the Hong Kong Bill of Rights Ordinance[15] (“BORO”) which enacts the Hong Kong Bill of Rights[16] (“BOR”). Articles 16 and 17 of the BOR are relevant, which respectively provide: “Article 16 Freedom of opinion and expression (1) Everyone shall have the right to hold opinions without interference. (2) Everyone shall have the right to freedom of expression; this right shall include freedom to seek, receive and impart information and ideas of all kinds, regardless of frontiers, either orally, in writing or in print, in the form of art, or through any other media of his choice. (3) The exercise of the rights provided for in paragraph (2) of this article carries with it special duties and responsibilities. It may therefore be subject to certain restrictions, but these shall only be such as are provided by law and are necessary— (a) for respect of the rights or reputations of others; or (b) for the protection of national security or of public order (ordre public), or of public health or morals. Article 17 Right of peaceful assembly The right of peaceful assembly shall be recognized. No restrictions may be placed on the exercise of this right other than those imposed in conformity with the law and which are necessary in a democratic society in the interests of national security or public safety, public order (ordre public), the protection of public health or morals or the protection of the rights and freedoms of others.” 34. The relevant rights are also provided in Article 27 of the Basic Law: “Hong Kong residents shall have freedom of speech, of the press and of publication; freedom of association, of assembly, of procession and of demonstration; … .” 35. The Court of Final Appeal held in Leung Kwok Hung v HKSAR[17] that there is no difference between the right of peaceful assembly guaranteed by Article 27 of the Basic Law and that provided in Article 17 of the BOR.[18] It was further held in HKSAR v Fong Kwok Shan[19] that the same applies to freedom of speech under Article 27 of the Basic Law and freedom of expression under Article 16 in the BOR. 36. Article 39 of the Basic Law provides: “The provisions of the International Covenant on Civil and Political Rights … as applied to Hong Kong shall remain in force and shall be implemented through the laws of Hong Kong Special Administrative Region. The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law. Such restrictions shall not contravene the provisions of the preceding paragraph of this Article.” 37. It is well established that the “BORO” implements the “ICCPR” and is given constitutional status by Article 39 of the Basic Law.[20] 38. Any restriction of the relevant rights therefore can only be made in accordance with paragraph 2 of Article 39 of the Basic Law.[21] 39. The Court of Final Appeal held in Leung Kwok Hung v HKSAR[22] that the relevant rights “may be subject to restrictions provided two requirements are satisfied: (1) The restriction must be prescribed by law (the ‘prescribed by law’ requirement). (2) The restriction must be necessary in a democratic society in the interests of national security or public safety, public order (ordre public), the protection of public health or morals or the protection of the rights and freedoms of others (the ‘necessity’ requirement). It will be convenient to refer to the specified purposes as ‘the legitimate purposes’.”[23] 40. As regards the necessity requirement, it was held that it involves the application of a proportionality test.[24] 41. It was submitted on behalf of the appellants that although section 11 had been considered by the Court of Final Appeal in the case of HKSAR v Fong Kwok Shan[25] and the provision was held to be constitutional, in that case the Court of Final Appeal had only been invited to consider whether the provision is properly prescribed by law and not the issue of proportionality. 42. It was therefore submitted that, whilst D1 would no longer pursue Grounds 1 and 2, so would the other Appellants in relation to the grounds on similar basis, I still have to consider the proportionality issue in particular as the Court of Final Appeal confirmed in HKSAR v Fong Kwok Shan[26] that the rights under the Basic Law and the Hong Kong Bill of Rights are applicable to demonstration within the precincts of the LegCo Chamber. 43. Counsel for the Respondent, Mr Derek Lai, SADPP(Ag)[27] rightly did not take issue on this point. 44. As Ribeiro PJ said in HKSAR v Fong Kwok Shan,[28] the burden is placed on the government to prove that the limitation of the guaranteed right is justified.[29] 45. There was criticism that the Magistrates based their decisions on the constitutional issue by reason of the nature of the premises of the LegCo or precincts of the Chamber. If either of the Magistrates had done so, this basis cannot sustain in the light of HKSAR v Fong Kwok Shan.[30] I do not need to go into the details in examining whether either of the Magistrates did err on the constitutional issue as magistracy appeal takes the form of re-hearing and I no doubt can make my decision on the point on proportionality which is a pure legal matter. 46. Mr Lee SC submitted that the LegCo is unique in that its proceedings and the work of its members are important, and pointed out that public protests are expected there as the LegCo is where important and controversial political decisions are often made. 47. He stressed that members of the public have the right to enter and stay in the precincts of the Chamber and such right is enshrined in section 8 of the LCPPO which provides: “sitting of the Council shall be open to the public” despite such right shall be subject to the Rules of Procedure or any resolution of the Council limiting or prohibiting the enjoyment of such right. He said no such Rules of Procedure or resolution was in force at the material time. 48. He also stressed that the offence for contravention of section 11 covers “a wide multitude of sins”, and does not as the two offences concerning disorderly conduct in section 17B of the Public Order Ordinance[31] require proof of either a further purpose or intent to provoke a breach of the peace on top of the disorderly conduct. 49. Relying on HKSAR v Fong Kwok Shan,[32] he submitted that section 11 does not mean to prohibit disorderly behaviour at large, and only with this interpretation that the provision can be constitutional. He quoted the observation of Ribeiro PJ, which he made in the course of considering the “prescribed by law” issue, that section 11 “has the benefit of a clear contextual focus and is not concerned with disorderly behavior at large”.[33] His Lordship also observed: “… [section 11] does not simply penalise a failure to ‘behave in an orderly manner’ without more. The orderly behaviour is demanded only of persons who enter or are within the precincts of the LegCo complex. The section also requires compliance with any directions given by an officer of the Council ‘for the purpose of keeping order’. The section is therefore self-evidently concerned with keeping order in those precincts. … … the clear purpose of [section 11] is to set a standard of orderly behaviour on the part of visitors congruent with LegCo’s institutional and social importance. Its context shows that it is part of a statutory framework aimed at creating a secure and dignified environment in the LegCo complex conducive to the legislature carrying out its constitutional functions at its sittings without disruption or disturbance, while permitting members of the public to observe the proceedings within the Chamber as an open legislative process. The section’s context is provided by its mother Ordinance, the LCPPO, and the other sections of the Administrative Instructions.”[34] 50. Apart from citing the abovesaid passage in the judgment of HKSAR v Fong Kwok Shan, Mr Lee SC also quoted certain provisions in the LCPPO as well as some relevant Hansard record and made the submission that the contextual focus of section 11 is twofold: (1) To ensure the right of the public to observe sittings of the LegCo. (2) To prevent LegCo proceedings from being disrupted or disturbed by members of the public present at the precincts of the Chamber. 51. He submitted that a failure to “behave in an orderly manner” in the context of section 11 must therefore be construed as behavior which is disorderly in terms of being likely to disrupt or disturb LegCo proceedings or other members of the public in observing such proceedings. To construe otherwise would amount to a prohibition on disorderly behavior at large, which would unnecessarily and impermissibly restrict the right of members of the public to enter and remain within the precincts of the Chamber. 52. It was the submission of Mr Lee SC that if the offence was properly confined to its contextual focus of preventing disruption or disturbance to LegCo proceedings and ensuring the right of other members of the public in observing such proceedings, it might not constitute a disproportional restriction of the relevant rights, however, if the offence covers conduct such as that of D1 in the present case, it would be disproportional as it is devoid of its contextual focus. 53. He pointed out that it was the finding of the Magistrate that there was no breach of the peace or disruption of LegCo proceedings involved and accepted that D1 was exercising her constitutional rights.[35] If the offence covers a situation like this, it is not proportional. 54. Mr Lee SC also argued that the offence may be proportional if the elements of the offence require proof of both failing to “behave in an orderly manner” and failure to “comply with any direction given by any officer of the Council for the purpose of keeping order”. For the reasons I give in considering Ground 5 of D1[36], I cannot accept that contravention of section 11 requires proof of both matters as asserted by Mr Lee SC. 55. Counsel for D4, Mr Tam, echoed Mr Lee SC and submitted that section 11 which criminalizes mere failure to “behave in an orderly manner” creates a “blanket prohibition” without taking into account the circumstances of the case. He stressed that what D4 did in the present case (hanging the banners in the car park which only became precincts of the Chamber on the day of meeting under the extended definition but was at the same time a DPAA) did not have the effect of actually disrupting and / or disturbing the LegCo meeting or the public’s right to observe such proceedings. If such acts are covered by section 11, the provision is disproportional and therefore unconstitutional. 56. Mr Tam cited the two offences under section 17B of the Public Order Ordinance[37] which either requires a specific purpose of “preventing the transaction of the business for which a public gathering area was called together”[38] or “with intent to provoke a breach of the peace, or whereby a breach of the peace is likely to be caused”[39] and submitted that the difference between these provisions and section 11 demonstrates a disproportionate restriction in the latter. 57. He also argued that the risk of an indiscriminate blanket prohibition is particularly pronounced in the present case because it applies to any protests and demonstrations that may take place in a DPAA just because the place happens to be part of the precincts of the Chamber as well. A loosely defined criminal offence can easily and readily be abused and used as an instrument to deter protests and demonstrations, as behavior which did not disrupt and / or disturb the meeting or the right of other members of the public in observing such proceedings will be caught. Holding the conduct of D4 amounted to the offence in such circumstances detracts from the contextual focus and purpose of section 11thus constitutes a disproportionate restriction on the relevant rights. 58. Counsel for the Respondent, Mr Lai, submitted that section 11, if properly construed in the light of its statutory context and purpose, does not lay down “blanket prohibition” and is proportional. He stressed that the restriction is of a limited scope both in terms of location and behavior: (1) it only regulates conduct of persons within a specific place: the precincts of the Chamber; and (2) its purpose is to set a standard of orderly behavior on the part of visitors congruent with LegCo’s institutional and social importance. 59. In support of his submission, Mr Lai quoted what Ribeiro PJ said on these points in HKSAR v Fong Kwok Shan[40], as set out in paragraph 49 above. 60. I accept the submission of Mr Lai that comparison between section 11 and other statutory provisions does not serve much use, in particular as the purposes of different provisions are not the same. 61. The question is whether the restriction imposed in section 11 is necessary and proportional. 62. In Mok Charles v Tam Wai Ho[41], Ma CJ held: “28. The proportionality test, which is a well known test in our courts, consists of the following analysis in respect of any restriction or limitation: (a) The restriction or limitation must pursue a legitimate aim. (b) The restriction or limitation must also be rationally connected to that legitimate aim. (c) The restriction or limitation must also be no more than is necessary to accomplish that legitimate aim.” 63. In Bank Mellat v Her Majesty’s Treasury (No 2),[42] Lord Sumption JSC described a similar approach but added one more step of consideration: “whether, having regard to these matters and to the severity of the consequences, a fair balance has been struck between the rights of the individual and the interests of the community”. 64. The Court of Final Appeal examined in Hysan Development Co Ltd v Town Planning Board[43] the approach to be adopted in assessing proportionality. In the judgment of Ribeiro PJ, to which all the other learned judges agreed, it was held that the four-step analysis should be adopted in Hong Kong and at the same time it was reiterated that there was no doubt on the correctness of any previous decision involving a three-step assessment. 65. Lord Sumption JSC noted that the elements of the four-step analysis inevitably overlap. Ribeiro PJ also observed that “in the great majority of cases, (the application of the fourth test) would not invalidate a restriction which has satisfied the requirements of the first three stages of the inquiry.”[44] 66. I believe it was not the suggestion of any appellants that the first two tests as set out in paragraph 62 above had not been met, in any case my judgment is that they are met. 67. Creating a secure and dignified environment which is conducive to the legislature carrying out its constitutional function is a legitimate aim. Setting a standard of orderly behavior and regulating the behavior of members of the public in the precincts of the Chamber are rationally connected to that aim. 68. I therefore focus on the examination of whether: (1) the restriction or limitation is no more than is necessary to accomplish the legitimate aim; and (2) a fair balance has been struck between the rights of the individual and the interests of the community. 69. The President of the LegCo is empowered under section 8(3) of the LCPPO to issue Administrative Instructions “for the purpose of maintaining the security of the precincts of the Chamber, ensuring the proper behavior and decorum of persons therein and for other administrative purposes”. It falls within the ambit of restriction approved in Leung Kwok Hung v HKSAR.[45] It is necessary in the interest of public order (ordre public) as well as protection of the rights of those who enter and stay in the precincts of the Chamber for a legitimate purpose. Whilst the concept of ordre public is imprecise and elusive, it includes “what is necessary for the protection of the general welfare or for the interests of the collectivity as a whole”. It must remain “a function of time, place and circumstances”.[46] 70. Mr Lai submitted that the restriction in section 11 passed the required tests. He stressed that the restriction is a limited one and does not prevent one from having a peaceful demonstration in an orderly manner inside the precincts of the Chamber, including any DPAA within it. It also does not prevent one from exercising the relevant rights lawfully at any other places outside the LegCo precincts. 71. Mr Lai cited Hysan Development Co Ltd v Town Planning Board[47] in which the Court of Final Appeal affirmed that the test is one of reasonable necessity, and a minimal impairment test does not mean that the restriction must be the very least intrusive method of securing the objective which might be imagined or devised. 72. He submitted that a reasonable balance has been struck by section 11 between the societal benefit of enabling the LegCo to properly carry out its constitutional functions on the one hand and the restriction on the relevant rights on the other. 73. Despite in HKSAR v Fong Kwok Shan[48], the Court of Final Appeal was not invited to examine the issue of proportionality in relation to section 11, I did examine it when I heard the case[49] and my conclusion was that the proportionality test had been passed. I agree that I have to consider the matter afresh in particular in the light of the observation of the Court of Final Appeal that the two strands in the Administrative Instruction are disjunctive. Having regard to submission in the present case, I maintain my view. 74. It requires section 11 to be construed to see if it does indeed lay down a blanket prohibition as argued or is in any case disproportionate. The context and purpose have to be considered[50], and they have been clearly identified in HKSAR v Fong Kwok Shan as stated in paragraph 49 above. 75. Mr Lee SC submitted that if the conduct of D1 in the present case amounts to an offence, it is devoid of the contextual context and constitutes a disproportionate restriction of the relevant rights. I do not agree. 76. The offence is clearly defined. The argument that section 11 lacks certainty and therefore fails the “prescribed by law” test was rejected by Court of Final Appeal in HKSAR v Fong Kwok Shan[51]. It has been held that it does not prohibit disorderly behaviour at large. It only prohibits failure to “behave in an orderly manner” within the precincts of the Chamber. Whether a conduct amounts to failure to “behave in an orderly manner” is to be judged on evidence by applying what was said in Chow Nok Hang[52]: the trial court is to apply the ordinary meaning of “behave in an orderly manner” to the time, place and circumstances in question, and having regard to the purpose of the Administrative Instruction, of which section 11 is a part, which is to create a secure and dignified environment in the LegCo Complex conducive to the legislature carrying out its constitutional functions at its sittings without disruption or disturbance, while permitting members of the public to observe proceedings within the Chamber as an open legislative process. Whether it is correct in convicting the appellant of the offence is one matter, whether the offence is constitutional is a totally different question. 77. If the regulation amounts to limitation to a constitutional right, whether the limitation complies with the tests is to be examined having regard to the context and purpose, taking into account the characteristic of the location where the limitation applies. It may be regarded as disproportionate at one place, but not so in another. The Administrative Instructions have application only within the precincts of the Chamber. 78. As parties argued, there is uniqueness of the LegCo. It has institutional uniqueness which stems from its constitutional role and functions as the sole legislature body in the HKSAR. Besides, the LegCo is where debates on public issues take place, some of which may attract much public concern. Decisions on controversial political matters are made there. It is also where members of the community will like to go to observe the proceedings, and express their view due to its prominence and importance symbolically. 79. The LegCo is important and unique from the perspective of its constitutional role and functions as well as from the perspective of the members of the community in exercising their rights. 80. Members of the community may go to the precincts of the Chamber to express their view, it is an important constitutional right. Nevertheless, exercise of the rights must be subject to regulation, otherwise the role and functions of the LegCo as well as the rights of members of the community will be jeopardised. 81. The regulation under section 11 confines to the precincts of the Chamber. Members of the community may exercise their rights so long as they “behave in an orderly manner”. Members of the community can also exercise their rights in other places, and there are many other channels for them to express their view. 82. A member of the community may go anywhere to exercise the right of expressing his view, wherever he goes it is out of the question that his conduct is not subject to any regulation. It may be that the place where he exercises his rights is a place designated for demonstration, for instance a DPAA within the boundary of the precincts of the Chamber. It is reasonable that when one exercises his rights inside such a DPAA, he is subject to the same regulation as in other places in the precincts of the Chamber. 83. The rights can be exercised at a place as close as just outside the precincts of the Chamber. If it happens, one may say there is just a fine distinction in the impact on the LegCo functions from what caused in ESCC 3406/2014 by the conduct inside the car park which became part of the precincts of the Chamber only by the extended definition. However, a line must be drawn. The precincts of the Chamber are or may be used by members and staff of the LegCo, officials who have to discharge duty at the LegCo, and members of the community who are exercising the legitimate rights to have access. Despite the nature and degree of the use of different places in the precincts of the Chamber vary, orderly behaviour is essential in keeping a secure and dignified environment in these places to ensure their proper use is maintained. What happens in places just beyond the precincts of the Chamber may also affect the environment but as I said, a line has to be drawn. That the conduct within or beyond the precincts of the Chamber is subject to different sets of rules does not necessarily mean that the restriction under section 11 is not proportional. Owing to the unique importance of the LegCo, it is important that people inside the precincts of the Chamber do not fail to “behave in an orderly manner”. 84. It is noted that the physical boundary of the precincts of the Chamber may vary. On the day when the Council or a committee is sitting, it extends beyond the LegCo Chambers building and covers places such as the enclosure or open space adjoining or appertaining to the building[53], including the car park where the incident in ESCC 3406/2014 took place. Notwithstanding the possible variation in boundary, section 11 only sanctions conduct amounting to failure to “behave in an orderly manner” or not complying with certain directions of the officers, within the precincts of the Chamber as defined. The definition is clearly stipulated in the legislation. The “extended precincts” only embraces a limited area surrounding the Chamber’s building and there is clear demarcation. If an incident which section 11 seeks to regulate occurs, there will be impact on the important duties and functions which the LegCo has to execute. 85. Having considered submission from Counsel, in my judgment, section 11 and the offence for contravening it met all the criteria set out in paragraphs 62 ‑ 65 above. They only regulate conduct of persons within the precincts of the Chamber. The restrictions may seem to be wide but it is clearly defined. It is important to set a standard of orderly behaviour congruent to the institutional and social importance of the LegCo. The importance of keeping the order in the precincts of the Chamber was stressed by Ribeiro PJ in HKSAR v Fong Kwok Shan[54]. Restricting the relevant right of the people inside the precincts of the Chamber by means of penalising a person for failing to “behave in an orderly manner” is rationally connected to a legitimate aim, the limitation is necessary and not disproportionate. It strikes the right balance between the rights of the individual and the interests of the community. I find that they are constitutional and valid. Other grounds of appeal 86. I now turn to the other grounds of appeal put forward by each of the appellants. D1 87. D1 did not pursue Grounds 1 and 2. I have already considered Ground 3 of D1. Grounds 4 and 7 88. Mr Lee SC argued these two grounds together. In gist, he submitted that the Magistrate failed to properly construe the words “behave in an orderly manner” and failed to apply the properly construed test to D1’s conduct. The crux of the submission is that to amount to failure to “behave in an orderly manner”, the conduct has to have the effect of disrupting or disturbing the LegCo proceedings or other members of the public in observing such proceedings, the law does not require a person to “behave in an orderly manner” at large.[55] To construe otherwise would amount to a prohibition on disorderly behavior at large, which would unnecessarily and impermissibly restrict the right of members of the public to enter and remain within the precincts of the Chamber. 89. It was submitted that as the conduct of D1 did not have such effect of disrupting or disturbing the LegCo proceedings or other members of the public in observing such proceedings, the offence was not committed. 90. Mr Lee SC stated, and correctly, that the present offence does not contain an element involving a breach of the peace. 91. The contextual focus which Mr Lee SC referred to is that set out in paragraph 49 above. 92. Apart from citing the passages in the judgment of Fong Kwok Shan[56], Mr Lee SC also quoted certain provisions in the LCPPO as well as some relevant Hansard records and made the submission that the contextual focus of section 11 is twofold, as stated in paragraph 50 above. 93. The submission of Mr Lee SC is that giving proper context to the provision, it was erred for the Magistrate to find D1 failed to “behave in an orderly manner”, in particular as there was no evidence of any actual disturbance of LegCo proceedings. The Magistrate took into account that D1 did not have a visitor’s pass. Mr Lee SC submitted that lack of the pass cannot in itself be equated with failure to “behave in an orderly manner”. The Magistrate also said D1 did not leave upon the requests of staff. Mr Lee SC submitted that D1 had not been charged with failing to comply with direction. He pointed out that the lobby was one floor below the Chamber where the FC was sitting. He stressed that evidence reveals that D1 took part in a peaceful assembly in the lobby of the LegCo complex, away from the Chamber where the FC was holding a meeting which indeed finished by 9:45pm, there was no evidence that she attempted to force her way into the upper floor, there was not any attempt or intention or likelihood on her part to disrupt or disturb the proceedings or those observing the proceedings. Despite there might be noise in the lobby, it is not a proper basis for a finding of not “behaving in an orderly manner”. There was no evidence that such noise would or was likely to reach the Chamber. The order of the lobby and the “vague notions of sanctity of LegCo” have nothing to do with the proceedings of the LegCo and / or the rights of other members of the public to observe such proceedings. The Magistrate was also criticized for failing to have due consideration of the speeches of D1 in particular the content. 94. In response, Mr Lai submitted that it had not been held in Fong Kwok Shan that disruption or disturbance to the LegCo proceedings is a necessary ingredient of the section 11 offence. He drew the attention of the court that it can be contrasted with section 17(c) of the LCPPO which specifically provides that “Any person who … creates or joins in any disturbance with interrupts or is likely to interrupt the proceedings of the Council … commits an offence.” Section 11 only requires one to “behave in an orderly manner” within the precincts of the Chamber. Section 2 of the LCPPO provides for an extended definition of precincts of the Chamber. It is clear legislative intent that by section 11 disorderly behavior is prohibited not only in the Chamber and conference rooms where LegCo proceedings are held, but also in the entire area designated as precincts of the Chamber. Even if disorderly conduct outside the Chamber and conference room does not necessarily lead to direct disruption or disturbance to the LegCo meeting, such conduct may still undermine the security and dignify functioning of the legislature. 95. I agree that requiring proof of direct and actual disruption or disturbance of the meeting does not accord with the judgment of Fong Kwok Shan. Whilst there will be little scope to argue that performing conduct which disrupts or disturbs the meeting is not a failure to “behave in an orderly manner”, it does not require such conduct to be so. 96. What the Court of Final Appeal said in Fong Kwok Shan should be considered in its entirety, the gist of which is set out in paragraph 49 above. Failure to “behave in an orderly manner” is not confined to conduct which directly and actually disrupted or disturbed any of the meetings of the LegCo. What people including D1 and D3 did would definitely hinder the legitimate use of the lobby by those who were entitled to do so. These people included members of the LegCo, officials, observers, and members of the press. They were at the least deprived of smooth leaving and returning to the building. As Mr Lee SC said, swift returning to the venue of meeting can be important in particular as the quorum is in question or when a vote is required. Besides, members of the public might be deterred to come and exercise their rights to observe the meeting. As the registration measures could not operate and staircase leading upstairs had to be guarded and effectively closed, the rights of these people were compromised. There will no longer be a secure and dignified environment in the LegCo complex conduces to the legislature carrying out its constitutional functions. 97. Firstly, the Magistrate clearly demonstrated that his analysis of the evidence was focused on whether D1 had failed to “behave in an orderly manner”. Secondly, in my judgment the Magistrate had demonstrated that he reached his decision having considered carefully the whole of the evidence. 98. The followings are the major finding of facts the Magistrate made in respect of D1: (1) No visitor’s pass had been issued to D1. (2) Despite so, she had remained in the lobby for more than five hours. (3) She was amongst the last batch of protestors leaving the lobby. (4) The system of issuing visitor’s passes could not be implemented as the protestors had rushed into the lobby and blocked the doors of the entrance. (5) D1 had given two speeches in the lobby. In the first speech,[57] she stated that she was the organizer and asked the protestors to stick together and took a photograph of “filling up the LegCo lobby”. Although what D1 said in the second speech was not revealed in evidence, the Magistrate drew the inference that she continued to participate in the protest at the lobby. (6) By giving the two speeches, D1 had encouraged other protesters to remain in the lobby. (7) There were at least around 100 demonstrators in the lobby, none of them wore a visitor’s pass. (8) Much noise was created in the lobby, including using public announcement equipment and hitting of drum, and had disrupted the order there. (9) There were so many people crowded in the lobby and a fire hazard had been created. 99. Whilst the Magistrate did not find D1 had disrupted or disturbed the meeting of the FC, or had caused a breach of the peace, he found that she had failed to “behave in an orderly manner” in the precincts of the Chamber. In reaching his decision, the Magistrate had also taken into account the circumstances such as that there were other facilities the access of which was from the lobby, the lobby was also the access to the upper floor where the LegCo Chamber is located and indeed at the material time a meeting was in process. 100. The Magistrate had taken into account the two speeches made by D1. Content of the first speech is known but not that of the second speech. However, taking into account the circumstances as shown in the video footage, I am of the view that the sentiment of the people in presence must have been nurtured by her speech is the only reasonable inference to be drawn from the evidence. It should also be noted that the time of the first speech coincided with the time demonstrators resisted the effort of officers to close the doors.[58] 101. The Magistrate was entitled to take into account that D1 had a prolonged stay in the Lobby, at a time when a meeting was in progress, a lot of noise was created, the speeches D1 gave, and the normal function of the Lobby had been compromised. Those who have legitimate use of the Lobby must have been hindered in the use. Functioning of the LegCo does not confine to the meeting room. What D1 did can hardly be described as orderly and conducive to the legislature carrying out its constitutional function, it went a long way to the opposite direction. 102. In all the circumstances, having perused the video footages already adduced as evidence at trial, I find the conclusion of the Magistrate reasonable, it accords with the legislative purpose and is well supported by evidence. Ground 5 103. Mr Lee SC submitted that the Magistrate erred in holding that the offence in section 11 is committed when the accused either “fails to behave in an orderly manner” or “fails to comply with direction given by any officer of the Council for the purpose of keeping order”. It was submitted that the offence would only be committed if both of the mentioned elements exist. It is said to be the right interpretation and meaningful and can avoid an absurd situation which may be inconsistent with the right of a person in the precincts of the Chamber. 104. In Fong Kwok Shan[59], Ribeiro PJ held that it is an error to treat the offence as involving proof of both disorderly behavior and a failure to comply with the direction of an officer of the Council.[60] Mr Lee SC submitted that it was only obiter dictum and urged me not to follow it. 105. Mr Lai submitted that the above statement of Ribeiro PJ was not obiter as it had direct bearing on the overall decision of the case. 106. In any case, having considered submission and the judgment in Fong Kwok Shan[61], I do not find myself able to adopt the course Mr Lee SC urged me to. Ground 6 107. Mr Lee SC quoted a passage in the Statement of Findings and submitted that the Magistrate had erred in law in treating a failure to “behave in an orderly manner” as the same as “behaving in a disorderly manner” under section 17B of the Public Order Ordinance. The passage may be roughly translated into this: “‘Behaving in an orderly manner’ or otherwise, in my view, just like the term ‘behaving in a disorderly manner’ in the Public Order Ordinance, can be applied in accordance with its ordinary meaning and determined in the light of the nature of the act and the time, place and circumstances in which it has taken place. It is a question of fact [rather than law], to be decided upon scrutiny of the facts of each case as a whole by the court.”[62] 108. What the Magistrate said should be read in context and in its entirety. I do not think he equated failure to “behave in an orderly manner” with “behaving in a disorderly manner”. In any case, this matter does not have material impact on the consideration of the appeal in particular as I am entitled to form my own view on the evidence and decide whether the conclusion of the Magistrate is correct. As held by the Court of Final Appeal in Chou Shih Bin v HKSAR:[63] magistracy appeal is “by way of rehearing on the evidence before the trial court supplemented by such further evidence as the intermediate appellate court may admit under its statutory power to do so.” This I did when I dealt with Grounds 4 and 7. Ground 8 109. Mr Lee SC submitted that the word “orderly” in section 11 should be construed and applied in the light of the time, place and circumstances of the incident. One such important circumstances must be whether there was “order” at the material time and place. Before a person can be expected to “behave in an orderly manner”, “order” must exist at that particular place. As the requirement of the provision is to “observe the order”, it presumes the existence of order at the relevant time and place. 110. He referred the court to the Chinese version of section 11 and argued that the words “遵守秩序” support his view as the expression presumes existence of “order” “秩序” before one can “observe the order”. 111. He argued that as there was no order in existence at the material time, there was no order to be observed at the precincts of the LegCo Chamber. Evidence shows that the system of applying for and granting visitors’ passes was not functioning as protestors were already inside the precincts of the Chamber. He said evidence shows protestors were implicitly allowed to enter the Lobby. The Magistrate found that the situation was a result of the conduct of demonstrators blocking the door. In this regard, he pointed out that there was no evidence that D1 was among those demonstrators who caused the disruption. He further submitted that in such a situation where a person was able to enter the precincts of the Chamber without intervention and found himself in an environment where there was no order to speak of, there is no reasonable or logical basis to convict him for failing to “behave in an orderly manner”. 112. In response, Mr Lai pointed out that this submission ignores the clear finding of the Magistrate that the security staff had never permitted the protestors to enter the Lobby for demonstration, it was that the system of issuing of visitor’s pass could not be implemented at the material time due to the protestors’ disorderly acts. He submitted that the fact that the order in the Lobby had already been jeopardized by the demonstration of the other protesters simply did not provide an excuse for D1 to fail to behave in an orderly manner within the precincts of the Chamber. If this argument is sound, it would mean that so long as the order in the precincts of the Chamber has been disrupted, people no longer need to behave orderly and may create further disturbance there without legal consequence. 113. I entirely agree to the view of Mr Lai. Whilst I accept that in deciding whether the conduct amounts to failure to “behave in an orderly manner”, the circumstances should be taken into account, it cannot be right to say as a matter of course that where there was no order in existence or that the proper order had been disturbed or disrupted, the duty of behaving in an orderly manner no longer exists. As to whether the conduct amounts to failure to “behave in an orderly manner”, it depends on whether what the prosecution has to prove, as stated in paragraph 76 above, is sufficiently proved by evidence. Ground 9 114. Mr Lee SC submitted that the Magistrate did not give sufficient reasons for the conviction and the reasons he gave were flaw. 115. Obviously the Magistrate did give reasons for the conviction. Whether the Magistrate had given sufficient reasons depends on the circumstances of each individual case. In the present case, I am of the view that the Magistrate cannot be said to have given insufficient reasons for the conviction. 116. The flaw which Mr Lee SC accused of were: (1) D1 did not have a visitor’s pass – comment of Mr Lee SC: lack of the pass cannot in itself be equated with a failure to “behave in an orderly manner”; (2) D1 did not leave the Lobby upon being asked by staff – comment of Mr Le SC: D1 was not charged with not complying with direction; and (3) D1 had used a loudhailer to make two speeches – comment of Mr Lee SC: there is no evidence that the meeting in progress had been disturbed or disrupted. 117. I cannot agree that the above matters are flaw. These are matters which the Magistrate was entitled to take into account in deciding whether the entirety of the evidence proved D1 had failed to “behave in an orderly manner”. Ground 10 118. It was submitted that the Administrative Instructions do not, and are not intended to, apply to such persons (like D1) who were not “admitted” to the LegCo precincts of the Chamber pursuant to the Administrative Instructions. 119. Mr Lee SC referred to section 8(3) of the LCPPO which enables the President to make administrative instructions: “The President may from time to time, for the purpose of maintaining the security of the precincts of the Chamber, ensuring the proper behaviour and decorum of persons therein and for other administrative purposes, issue such administrative instructions as he may deem necessary or expedient for regulating the admittance of persons (other than members or officers of the Council) to, and the conduct of such persons within, the Chamber and the precincts of the Chamber.” and argued that the regulations in the Administrative Instructions apply only to the people admitted into the precincts of the Chamber and not people like D1 who had not been officially admitted as visitors. 120. I do not agree to this view. In my judgment, the words “conduct of such persons” refer to “persons (other than members or officers of the Council)”. Section 8(3) enables the President for the specified purposes to make Administrative Instructions to: (1) regulate admittance of persons other than members or officers of the Council; and (2) regulate the conduct of such persons within the Chamber and the precincts of the Chamber. 121. Section 11 of the Administrative Instructions expressly applies to “persons entering or within the precincts of the Chamber” and section 20(b) of the LCPPO, the offence creating provision, expressly applies to “persons within the Chambers and the precincts of the Chamber”. Apart from the Chinese version of the title (see next paragraph), there is no reference to anything like persons being lawfully or officially admitted. 122. Mr Lee SC argued that the Chinese title of the Administrative Instruction supports his view. Whereas the English version is “Administrative Instructions for Regulating Admittance and Conduct of Persons”, the Chinese version is 《規限獲准進入立法會大樓的人士及其行為的行政指令》, he highlighted the expression “獲准進入” which literally does carry the meaning of having been permitted to enter. 123. He also referred to several provisions in the Administrative Instructions and submitted that the intention of the rules is to regulate admission into the Chambers or precincts of the Chamber. 124. He also pointed out that this suggested interpretation would not bring about the absurd consequence that the conduct of persons not officially admitted being unregulated as there are regulations to cover their conduct, for instance: (1) offences of unlawful assembly and forcible entry; (2) the offences under the Public Order Ordinance such as that contrary to section 17B(2); and (3) under section 8(2) of the LCPPO, “[t]he right of persons other than members or officers of the Council to enter or remain within the precincts of the Chamber shall be subject to the Rules of Procedure or any resolution of the Council limiting or prohibiting the enjoyment of such right.” 125. The title of a statutory provision is only one of many matters which the court may take into account in interpretation. Where the statutory provisions are set out in both the English and Chinese versions, Section 10B of the Interpretation and General Clauses Ordinance[64] is relevant: “(1) The English language text and the Chinese language text of an Ordinance shall be equally authentic, and the Ordinance shall be construed accordingly. (2) The provisions of an Ordinance are presumed to have the same meaning in each authentic text. (3) Where a comparison of the authentic texts of an Ordinance discloses a difference of meaning which the rules of statutory interpretation ordinarily applicable do not resolve, the meaning which best reconciles the texts, having regard to the object and purposes of the Ordinance, shall be adopted.” 126. In construing the provisions, the presumption that each authentic text has the same meaning should be borne in mind, and if there is a difference of meaning which cannot be resolved by the rules of statutory interpretation, the meaning which best reconciles the texts, having regard to the object and purposes of the ordinance should be adopted. 127. The Administrative Instructions obviously do not only target at persons officially permitted into the various places in the precincts of the Chamber as stipulated. Part II provides for regulation governing access and restriction on access. Part III, in which section 11 is a part, concerns conduct of people at specified place in the precincts of the Chamber. Part IV provides for facilities for the press as well as restriction of specified people of access and use of such facilities and taking of photographs. Part V provides for situation where evacuation is required. In many provisions, such as sections 4 and 5, the words “no person shall” are used. In section 11, the provision in question, the words used are “persons entering or within the precincts of the Chamber”. Having regard to the object and purposes of the Ordinance, I do not find the argument of Mr Lee SC sound. I do not find the Administrative Instructions, and in any case section 11, are meant to confine to people officially permitted into the precincts of the Chamber. 128. I accept that other statutory provisions or power may be applicable to regulate and control the conduct of people inside the LegCo, however it must be noted that different provisions cater for different situation and require different elements of offence to prove. 129. In the course of submission, Mr Lee SC pointed out that whilst there existed a mechanism say for example under section 8(2) of the LCPPO to take step to restrict or regulate behaviour of people, including trespassers, within the precincts of the Chamber, no such power had been invoked at the material time whereas the mechanism under section 8(3) and section 20 (b) of the LCPPO was erroneously invoked. 130. Section 8(2) of the LCPPO provides: “The right of persons other than members or officers of the Council to enter or remain within the precincts of the Chamber shall be subject to the Rules of Procedure or any resolution of the Council limiting or prohibiting the enjoyment of such right.” 131. Whilst separate power may exist to deal with the like situations, I do not agree that it is erroneous to invoke a power existing at the same time (issuing Administrative Instructions pursuant to section 8(3) of the LCPPO) when the circumstances and evidence justified taking such step. 132. This ground fails. Ground 11 133. This is not a separate ground as such. Having considered submission and the whole circumstances, I do not consider the conviction of D1 unsafe or unsatisfactory as alleged by Mr Lee SC. I therefore dismiss the appeal of D1 against her conviction. D3 134. As Counsel for D3, Mr Shek, adopted the submission of the other Counsel and in the light of my decision in relation to the argument on legal issues put forward by them, all of D3’s grounds of appeal so far as they relate to legal issues fail. 135. As far as the finding of facts by the Magistrate is concerned, Mr Shek was unable to convince me that it is unsafe or unsatisfactory. 136. As mentioned in paragraph 23 above, D3 blocked the door together with D2. When staff at around 9:17pm asked them to leave, D2 left but D3 continued to obstruct the doors until 9:57pm. To suggest, as Mr Shek did, the conduct of D3 as passive is not an apt description of what he did in the situation. Conduct of this crowd of people who stopped closing of the door caused injuries to a number of LegCo security staff. 137. The Magistrate was fully justified in finding D3 guilty of the offence. I therefore also dismiss the appeal of D3 against his conviction. D4 138. Grounds 1 and 2 were not pursued. Ground 3 concerns the constitutional issue and has been dealt with. Ground 4 139. Mr Tam alleged in Ground 4 that the Magistrate applied the wrong test and erred in finding that D4’s conduct amounted to not “behaving in an orderly manner”. He submitted that, as the conduct of D4 hanging the banners on the canopy could not have caused disruption and / or disturbance to the meeting then in progress, the conduct was consistent with “behaving in an orderly manner” or in any case not amounted to failure to “behave in an orderly manner”. 140. In the Amended Perfected Ground of Appeal, the Magistrate was criticized for refusing to regard the New Zealand cases of Brooker v Police[65] and Morse v Police[66] applicable in the current scenario and as a result erred in finding D4’s conduct amounted to not behave in an orderly manner. 141. In his written submission and during the appeal hearing, the focus of Mr Tam was that the Magistrate failed to take into account the impacts of D4’s actions and therefore erred in making his conclusion. 142. Mr Tam pointed out that the Court of Final Appeal held in HKSAR v Chow Nok Hang[67] that in deciding whether an accused was behaving in an orderly manner, the Court must look at the time, place and circumstances of the conduct in question. 143. He also pointed out that the Court of Final Appeal observed in HKSAR v Fong Kwok Shan[68]that “the orderly behavior is demanded only of persons who enter or are within precincts of the LegCo complex” and “the section is … self-evidently concerned with keeping order in those precincts”[69]. He also quoted the passage as set out in paragraph 49 above. 144. Mr Tam submitted that whilst anyone will appreciate that disrupting and / or disturbing meeting(s) of the LegCo or its committee(s) within the precincts of the Chamber, or disrupting and / or disturbing the right of other members of the public in observing such proceedings would be regarded as failing to “behave in an orderly manner”, it is not the same if the conduct took place in the car park, as it was in the present case, which only became part of the precincts of the Chamber by operation of section 2 of the LCPPO which provides: “… (2) includes, during the whole of any day the Council or a committee is sitting, the entire building in which the Chamber is situated and any forecourt, yard, garden, enclosure or open space adjoining or appertaining to such building and used or provided for the purposes of the Council; …” 145. Mr Tam said the court has to take into account the following matters: (1) D4 climbed up the canopy above the carpark. (2) At no time did he use violence. (3) He hang a total of four banners. (4) After hanging the banners, he came down voluntarily and subsequently surrendered. (5) The acts were done in an area outside the LegCo complex which was designated for public demonstration. 146. Mr Tam stressed that even though the car park was by definition part of the precincts of the Chamber, at the same time it was a DPAA, his submission was that the area was not intended to be, nor was it actually in use, by any LegCo members and / or staff for the purpose of any LegCo meeting at the material time. It was only an area for members of the public to express their opinions and grievances. Indeed, that was the reason why D4 went to the car park at that time. 147. Mr Tam submitted that a proper consideration of these matters should have led to the conclusion that the actions could not have caused disruption and / or disturbance to the meeting at the material time. As the Magistrate had failed to duly take into account the impact of D4’s action, he erred in finding that D4 failed to “behave in an orderly manner”. As a result, the conviction is unsafe and unsatisfactory. 148. Mr Lai submitted that the purpose of extending the definition of “precincts of the Chamber” on the day of meeting accords with the legislative objective of section 11, and indeed the LCPPO, to create a secure and dignified environment for LegCo members and officers to carry out their constitutional duties without fear, while permitting members of the public to observe safely the proceedings as an open legislative process. The whole of the precincts of the Chamber under the extended definition are places where all people having a legitimate purpose in attending the LegCo may have to pass through. I agree to this view. 149. The Magistrate held that the hanging of banners by D4 and D5 posed danger to other persons near the canopy, including the LegCo security staff performing duties there as well as other members of the public taking part in the demonstration in the designated protest area. 150. Having regard to the whole of the relevant circumstances, I am of the view that the Magistrate was entitled to find the conduct of D4 amounted to failing to “behave in an orderly manner”. Not only did his conduct pose danger to other persons near the canopy, he consciously placed himself in a dangerous situation by climbing up a high place where he ought to have known he should not go, and ought to have contemplated that staff would take step to stop him and sentiment of the people nearby would be aroused. He insisted to carry on when staff intervened and took action which was reasonable in the circumstances. Ground 5 151. This is indeed Ground 10 put forward by Mr Lee SC on behalf of D1. Mr Tam adopted the submission of Mr Lee SC and submitted that D4 was indeed in a better position as he was outside the LegCo complex at the material time and therefore was not even required to obtain a visitor’s pass for his presence. 152. For the reasons I have mentioned earlier, I do not accept the submission of Mr Lee SC in relation to this ground. 153. The fact that D4 was not inside the LegCo building is not relevant, he was inside the precincts of the LegCo Chamber as defined and has to “behave in an orderly manner”. 154. This ground fails. 155. All the grounds put forward for D4 fail. In my judgment, his conviction is neither unsafe nor unsatisfactory. His appeal against conviction is therefore dismissed. D5 156. Counsel for D5, Mr Shek, indicated that he adopted the submission of Mr Tam. All the grounds were framed as legal issues. I have earlier directly dealt with the legal issues in Grounds 1, 2, 4 and 5 and my conclusions are not favourable to Mr Shek. 157. For the grounds which alleged error in finding D5 guilty of the offence, I note that the Magistrate had the following findings of facts: (1) D5 was acting in concert with D4 and their common purpose was to have the banners hung up for display. (2) He continued his actions despite staff tried to stop him. He disregarded warning and direction of the staff. 158. In my judgment, the Magistrate was entitled to find D5 failed to “behave in an orderly manner” on the evidence. Clearly he was acting in concert with D4. The conviction of D5 is neither unsafe nor unsatisfactory. I dismiss his appeal against conviction. Appeals against sentence 159. D1 and D3 appeal against sentence. Grounds and Submission of the Appellants D1 160. D1 was sentenced to two weeks’ imprisonment. Mr Lee SC submitted that the sentence was wrong in principle and / or manifestly excessive, as the Magistrate: (1) failed to consider or give any or any proper weight to a number of mitigating factors favourable to D1; and (2) wrongly adopted a deterrent sentence when there was nothing to justify it. 161. He submitted that the Magistrate had failed to give any or any proper weight to the following matters: (1) D1 was merely exercising her constitutional rights to freedom of expression, assembly and demonstration, and the Magistrate said he understood the defendant has these rights. (2) As the Magistrate accepted, D1’s conduct was entirely peaceful, and there is no evidence that the demonstrators had used any violence or disrupted the LegCo proceedings. As held by the Court of Final Appeal in SJ v Wong Chi Fung[70]: “An offender’s motive for committing an offence is a relevant factor in a court’s decision as to the appropriate sentence to be imposed and can be a relevant mitigating factor”;[71] and “… the fact that an offence arises out of an occasion when constitutional rights to assemble and protest are being exercised is relevant to the background and context of the offending, particularly when those rights have been exercised peacefully and in accordance within the law up to the point when the offence was committed.”[72] (3) The maximum penalty for the offence is a fine of $2,000 and imprisonment for three months, this is much lower than the maximum penalty for the offence of causing disorder in a public place[73] which is a fine at level 2 and imprisonment for 12 months. (4) As the present offence did not involve any violence or breach of the peace, it must be less serious than any offence of unlawful assembly, and even for the latter offence it has been said in SJ v Wong Chi Fung[74] that a community service order (“CSO”) was a sentence frequently passed.[75] (5) The finding of the Magistrate that the two speeches had encouraged the other demonstrators to remain in the Lobby was not supported by evidence. (6) The Magistrate failed to have sufficient regard to the fact that this was the first time anyone was convicted and sentenced for this offence in Hong Kong. 162. Mr Lee SC submitted that a fine would have properly reflected the true nature of the offence, or alternatively a CSO should be considered. He pointed out that despite D1 had previously refused to give consent to perform unpaid work under a CSO, she now consents to do such work. D3 163. Counsel for D3, Mr Shek, stressed that D3 had a clear record before this conviction and submitted that the Magistrate erred in imposing a deterrent sentence on a first-time offender. He also submitted that a sentence of three weeks’ imprisonment is manifestly excessive and inappropriate. He pointed out that the demonstration had been peaceful, there is no evidence of any premeditation, there was no evidence that D3 intended such a number of persons to involve in the acts. He submitted that the role of D3 was minor and his conduct could be described as passive: just standing there. The court is asked to consider ordering CSO on D3. Submission of the Respondent 164. In response, Mr Lai quoted the case of SJ v Wong Chi Fung[76] in which the Court of Final Appeal repeated the remarks made by Ribeiro PJ in Chow Nok Hang: “Once a demonstrator becomes involved in violence or the threat of violence – somewhat archaically referred to as a ‘breach of the peace’ – that demonstrator crosses the line separating constitutionally protected peaceful demonstration from unlawful activity which is subject to legal sanctions and constraints. The same applies where the demonstrator crosses the line by unlawfully interfering with the rights and freedoms of others.” and added: “For this simple reason, a submission in mitigation of the offence of unlawful assembly (and certainly in the case of incitement) that the act was committed in the exercise of the constitutional rights to freedom of expression and freedom of assembly will be unlikely to carry any significant weight. The fact of a conviction of the offence will necessarily mean that the offender has crossed the line separating the lawful exercise of his constitutional rights from unlawful activity subject to sanctions and constraints. In such a case, there is little merit in a plea for leniency on the basis that the offender was merely exercising constitutional rights since, by definition, he was not doing so at the time when the offence was committed. This is all the more so when the facts of the offending involve violence, in particular on the part of the offender himself, since there is no constitutional justification for violent unlawful behaviour. In such a case involving violence, a deterrent sentence may be called for and will not be objectionable on the ground that it creates a ‘chilling effect’ on the exercise of a constitutional right, since there is no right to be violent. Quite simply the line of acceptability has been crossed.”[77] 165. Mr Lai submitted that the offence committed respectively by the two Appellants in the particular circumstances of the case were not minor. The offences hampered the proper functioning of the systems of security check on visitors and issuing of passes pursuant to the Administrative Instruction. As a result, people might enter the LegCo complex freely without any verification. Security of the complex was jeopardized, and dignity of LegCo was undermined. He submitted that the law never allows a person to exercise the constitutional rights through unlawful means. Reply of the Appellants 166. In reply, Mr Lee SC emphasized that the demonstration had been peaceful, at least as far as the circumstances in which D1 was involved, and she had not involved in any violence or threat of violence, let alone breach of the peace. 167. Mr Shek emphasized that there was no evidence that the act of D3 had directly caused anyone to suffer injuries. Discussion and Consideration 168. In sentencing, the Magistrate[78] recognized the rights of each citizen to have freedom of expression and that the community should accommodate different views and opinion. He took into account the following matters: (1) Nature and facts of the case, in particular as the FC was discussing on a matter which would have effects on many people, it had attracted widespread attention and there were different views on the matter. (2) Background of the appellants including the criminal conviction record of D1 and that D3 had a clear record. (3) CSO reports. (4) Mitigation. 169. He noted that the incident took place in the LegCo which has important constitutional function. It is therefore important to ensure that the members and staff can discharge their duty independently and in dignity. The regulation system is put in place to ensure this. 170. He took into account that the appellants were concerned about the matter discussed by the FC and had a strong view. 171. He also noted that the appellants chose an illegal means to achieve their aim. If everyone is to ignore the regulatory measures, there will be no order in the precincts of the Chamber. As a result, the proper and efficient functioning of the LegCo will be hampered, and this will have long term adverse effect on the development and running of the society. 172. He stressed that there were many channel for the appellants to lawfully and properly express their view. There was a demonstration area designated where people can lawfully and freely express their view. D1 chose to improperly stay in the Lobby for about five hours and did not leave despite repeated request by the staff. Her speech encouraged other demonstrators who had also entered without proper permission to stay, which seriously affected the function of the Lobby and blocked the access upstairs. Noise was created. Dignity of LegCo was affected. 173. As regards individual appellant, the Magistrate had the following observations and consideration. D1 174. He noted that D1 had over the years been enthusiastic in social affairs, she stressed that it was her intent to insert pressure on the authority and the members of the LegCo, and she also wanted to attract attention of the social media to have more concern about the matter as what she had done before did not seem to be effective. 175. He accepted that the demonstration inside the Lobby had been a peaceful one. He also accepted that there was no evidence to show that the demonstrators had obstructed the meeting of the FC or its members to attend the meeting, there was no evidence that any demonstrator had successfully run up to the upper floor, nor any one of them had used any violence. 176. However, he did not find the facts disclosed a not serious case and considered an imprisonment term was called for. He adopted two weeks as the starting point and considered there existed no grounds to justify any deduction. He therefore imposed an immediate imprisonment term of two weeks on D1. 177. Before he made the decision, the Magistrate had considered other sentencing options including CSO. CSO was not recommended not only because the reporting officer did not consider it appropriate, but also because D1 had indicated her unwillingness to perform unpaid work under a CSO. 178. In SJ v Wong Chi Fung[79], the Court of Final Appeal observed that “an offender’s motive for committing an offence is a relevant factor in a court’s decision as to the appropriate sentence to be imposed and can be a relevant mitigating factor.”[80] At the same time, the Court also delivered the passage as set out in paragraph 164 above. 179. The observation was made when the Court of Final Appeal was considering a case of unlawful assembly, a generally more serious offence. Nevertheless, I am of the view that such observation, in particular the underlying thought, has general application. 180. Even in cases of civil disobedience, which was not specifically raised by any of the Counsel as a concept applicable in the present case, where the court may take into account as motive for committing of the offence if the behavior of the offender is actuated by his conscientious objections and genuine belief of injustice, the Court of Final Appeal said: “the weight to be attached to that motive will necessarily vary depending on many other circumstances, including the facts of the offending and its consequences and the need for deterrence and punishment.”[81] The Court also said: “Even where an act of protest may properly be characterised as one of civil disobedience, … the court will not enter into an evaluation of the worthiness of the cause espoused. … It is not, however, the task of the courts to take sides on issues that are political or to prefer one set of social or other values over another.”[82] 181. The Magistrate demonstrated that he had given careful and thorough thought of the relevant matters before he decided on the sentence. 182. At the time of sentence, the option of CSO was obviously not available as D1 indicated that she was unwilling to perform unpaid work under a CSO and her consent is a statutory pre-requisite in imposing a CSO.[83] 183. Now D1 indicated that she would be willing if a CSO is imposed and Mr Lee SC asked the court to consider making such an order. 184. Mr Lee SC reminded the court that this was the first case in which a person has been charged for contravention of section 11. It is settled law that the sentence for an offence should be in accordance with the practice prevailing at the time of the commission of the offence,[84] and the court has yet at the material time indicated any view on the offence. 185. The Court of Appeal in SJ v Wong Chi Fung[85] observed that genuine remorse has always been considered a precondition for CSO and genuine remorse means the offender acknowledges that he has committed an offence and shows remorse for what he had done and caused, and not simply saying that he is willing to accept legal responsibility and punishment. Poon JA set out the matters which the court should take into account in deciding whether the convicted accused has genuine remorse: “Genuine remorse means that the offender acknowledges that he has committed an offence and shows remorse for what he has done and caused. Thus a genuinely remorseful person will normally plead guilty and explicitly accept his legal responsibility. It follows that, in deciding whether the offender is genuinely remorseful, one of the major factors that the court would look for is whether he has entered a timely plea. If an offender pleads not guilty, and expresses his remorse only after he is tried and convicted, the court would carefully examine such claim. Generally speaking: (1) Although the court will not reject his indication of remorse simply because he has contested his trial, neither will the court easily accept that he is genuinely remorseful for what he has done. (2) If the offender considers that it is his right under the presumption of innocence to ask the prosecution to prove its case, or he elects to plead not guilty because he disputes that the conduct he has admitted constitutes an offence, he is of course entitled to do so. If he is convicted after trial, the court will not enhance his sentence because of his conduct of defence. However, it does not mean that the court must necessarily accept that although he has conducted his defence in such a manner, he is genuinely remorseful. (3) If the offender considers that his prosecution is, in itself, not justified (as is the case of the 1st and 3rd respondents who consider that charging them with unlawful assembly was against their human rights) and maintains that particular view even upon conviction, that would be a clear indication that he is still refusing to accept what he has done is unlawful. In that case, the court will not accept his claim of genuine remorse, for such a claim contradicts his stance that the prosecution was unjustified. (4) If the offender insists that he is innocent after being convicted, or expresses firmly that he has not done anything wrong, it is a further indication that he has no genuine remorse; even if he says he is willing to accept legal responsibility and punishment, it cannot change the fact that he has no genuine remorse for the offence he has committed.” 186. Having considered the whole of the relevant circumstances, I do not consider D1 a suitable candidate for CSO, nor do I consider CSO appropriate, for the following reasons: (1) D1 did not play a minor role in this incident where the Lobby was occupied for a long time by people who had entered without complying with the rules causing substantial compromise in the order, what she did including giving of speeches, were likely to have the effect of nurturing the spirit of those present in continuing their actions. (2) D1 had been convicted of offences of like nature though not identical, she was convicted in 2012 of the following offences: (a) taking part in an unlawful assembly; (b) holding / convening / organizing / forming / collecting/ assisting in respect of an unauthorized public meeting / procession; and (c) taking part in an unauthorized assembly, and for these offences she was sentenced to four weeks’ imprisonment and fined a total of $3,000. (3) There is no indication that she is in genuine remorse. I form this view taking into account the followings: she pleaded not guilty, she appealed against the conviction, at the time of sentence she refused the option to perform unpaid work under a CSO, and what she told the reporting officer about how she thought of the incident and her conduct. 187. As regards the suggestion of a fine, it is my view that such a sentence is plainly insufficient to reflect the true culpability of D1 in the case. 188. I agree to the Magistrate that it is appropriate to impose an immediate imprisonment term. Having regard to SJ v Wade[86], I do not consider it appropriate to have the sentence suspended. The Magistrate adopted a starting point of two weeks, I do not consider it wrong in principle or manifestly excessive. I also agree that there is nothing which justifies a discount. 189. I therefore dismiss the appeal against sentence of D1. D3 190. In relation to D3, the Magistrate had the following observations: (1) He had a clear record. (2) He became concerned in social issues and had changed job so that he could have more time to participate in events concerning social issues. (3) D3 was not the persons who started to block the closing of the entrance door. (4) However, despite repeated requests and warnings, he did not leave the gate even after 9:17pm and continued to block the gate with other people, preventing officers from closing the door for about 50 minutes. (5) Situation was chaotic after 9:17pm and there was physical encounter between the staff and demonstrators, causing several security staff to suffer injury, one of them was still not able to resume duty at the time of the trial. (6) Despite injuries of the security staff were not directly caused by D3, his participation contributed to the situation which caused the injuries. (7) In such a chaotic situation and such degree of physical encounter, it was only luck that the injuries suffered by the staff were not even more serious. (8) He had been pushed away by the joint effort of several security staff. (9) Despite he was not one of the persons who started blocking of the door, his conduct damaged the system, affected the security and compromised effectiveness of the regulatory measures. (10) The culpability of D3 was most serious amongst the appellants. (11) CSO was not recommended. 191. The Magistrate considered an imprisonment term appropriate. He adopted three weeks as the starting point and did not consider there existed any reasons for discount. He therefore imposed an immediate imprisonment term of three weeks on D3. 192. In relation to D3, despite there was no evidence to prove that any injuries had been a direct consequence of his conduct, it was the chaotic situation in which he actively took part which caused the injuries of several staff, the remark of the Court of Final Appeal as quoted in paragraph 164 above is of relevance. 193. Mr Shek challenged the assessment that the culpability of D3 was most serious. I do not consider it meaningful and helpful to compare between the appellants for the purpose of this appeal as the conduct of D1 and D3 was of different nature. The question is whether the sentence now imposed on D3 is wrong in principle and / or manifestly excessive, or in any case inappropriate. 194. What the Magistrate stated, as set out in paragraph 190 above, are what he was entitled to take into account, and should have been taken into account, in consideration of sentence. He had carefully considered whether there was other appropriate sentencing option before he decided the sentence. Even though D3 did not have any prior criminal conviction, in the circumstances of the present case, I am also of the view that immediate custodial sentence is the only appropriate sentence to impose. 195. A fine is definitely insufficient to reflect the nature and seriousness of the conduct of D3 and the consequences (which must have been within the reasonable contemplation of D3). The fact that D3 was not the person who started to block the door did not alleviate his culpability as he joined realising the chaotic situation which had been caused and he carried on for 50 minutes during which there was physical encounter. 196. As regards CSO which has not been recommended mainly on the ground of the attitude of D3, having regard to the fact that this is the first conviction of D3, I similarly do not consider it appropriate due to the lack of genuine remorse on the part of D3. Not only that he did not plead guilty and has lodged an unsuccessful appeal against conviction, what he said to the reporting officer about his conduct at the material time did not indicate remorse. There is no sign that D3 accepted what he did was unlawful. 197. In all the circumstances, I am of the view that an imprisonment term is not wrong in principle. The length of the term cannot be said to be manifestly excessive. There was an argument that, if the term was so short the court should consider whether an imprisonment term should be imposed at all. Having considered the whole circumstances of the case, in my judgment the present sentence is appropriate and sufficiently reflects the nature and seriousness of the matter. 198. I am also of the view that there exists no justification to suspend the imprisonment term. 199. For these reasons, I also dismiss the appeal against sentence by D3. I do not consider it either wrong in principle or manifestly excessive. Mr Martin Lee SC, Ms Linda Wong and Mr Geoffrey Yeung, instructed by Ho, Tse, Wai & Partners, for the 1st appellant in HCMA 617/2015 (D1) Mr Jeffrey Tam and Mr Andrew Lau, instructed by Au Yeung, Chan & Ho, for the 1st appellant in HCMA 438/2015 (D4) Mr Randy Shek, instructed by Ho, Tse, Wai & Partners, assigned by the Director of Legal Aid, for the the 3rd appellant in HCMA 617/2015 (D3) and 2nd appellant in HCMA 438/2015 (D5) Mr Derek Lai, SADPP(Ag), Mr Derek Lau, SPP, and Ms Karen Ng, PP, of the Department of Justice, for the respondent in both cases The 2nd appellant in HCMA 617/2015 (D2) was not represented and was not present, his case was severed, see paragraph 6 in the judgment [1] Contrary to section 20(b) of LCPPO, Cap 382, Laws of Hong Kong. [2] Mr Lee Siu Ho in ESCC 3350/2014 and Mr Colin Wong in ESCC 3406/2014. [3] The summarized facts were drawn up by reference to the Statement of Findings, and were agreed by all parties. [4] At trial, D1 was represented by Ms Linda Wong, D2 appeared in person, D3 was represented by Mr Jeffrey Tam, D4 was also represented by Mr Jeffrey Tam, and D5 was represented by Mr Randy Shek. [5] Mr Lee SC did not represent D1 at trial, only Ms Wong did. [6] The Amended Perfected Grounds of Appeal filed on 6 August 2018. [7] [2017] 20 HKCFAR 425. [8] At trial, D3 was represented by Mr Jeffrey Tam. [9] Amended Perfect Grounds dated 6 August 2018. [10] Cap 245, Laws of Hong Kong. [11] Re-Amended Perfected Grounds dated 13 August 2018. [12] Supra. [13] Mr Shek also represented D5 at trial. [14] Re-Amended Perfected Grounds dated 13 August 2018. [15] Cap 383, Laws of Hong Kong. [16] Section 8 of the Ordinance. [17] (2005) 8 HKCFAR 229. [18] See paragraph 20 in the judgment. [19] Supra. [20] See Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 and Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229. [21] See paragraph 36 above. [22] Supra. [23] Paragraph 17 in the judgment. [24] See paragraph 33 in the judgment of Leung Kwok Hung. [25] Supra. [26] Supra. [27] Mr Lai appeared together with Mr Derek Lau, SPP, and Ms Karen Ng, PP. [28] Supra. [29] Paragraph 36 in the judgment. [30] See Part E in the judgment of the case. [31] Cap 245, Laws of Hong Kong. [32] Supra. [33] Paragraph 99 in the judgment. [34] Paragraphs 80 and 82 in the judgment. [35] Quoting paragraphs 125, 139 and 129 in the Statement of Findings. [36] See paragraphs 103 ‑ 106 in this judgment. [37] Cap 245, Laws of Hong Kong. [38] Section 17B(1) of the Public Order Ordinance. [39] Section 17B(2) of the Public Order Ordinance. [40] Supra. [41] (2010) 13 HKCFAR 762. [42] [2014] AC 700. [43] (2016) 19 HKCFAR 372. [44] Paragraph 73 in the judgment. [45] See paragraph 39 above. [46] See HKSAR v Ng Kung Siu & Another (1999) 2 HKCFAR 442, at 459 I. [47] Supra. [48] Supra. [49] HCMA 666/2015, [2017] 2 HKLRD 225, paragraphs 95 - 112. [50] See HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574 [63] and Fong Kwok Shan [78]. [51] Supra. [52] (2013) 16 HKCFAR 837. [53] Section 2 of the LCPPO and section 2 in the Administrative Instruction. [54] Supra. [55] Paragraph 99 of the judgment in Fong Kwok Shan was quoted. [56] Supra. [57] What D1 said was captured on a video footage, exhibit P21. [58] Paragraphs 14 and 23 in this judgment are relevant. [59] Supra. [60] Paragraph 100 in the judgment. [61] Supra. [62] The Statement of Findings was written in Chinese, its original text is: “「遵守秩序」與否，本席認為，無異於《公安條例》「擾亂秩序」四字，可按其通常意思加以運用，視行為的性質、發生的時間、地點和情況而決定，是一個事實（而非法律）問題，由法庭審視每宗案件整體案情而定。(HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837)” [63] (2005) 8 HKCFAR 70, at 78. [64] Cap 1, Laws of Hong Kong. [65] [2007] 3 NZLR 91. [66] [2012] 2 NZLR 1. [67] Supra, see paragraph 76 above. [68] Supra. [69] Paragraph 80 in the judgment. [70] [2018] 21 HKCFAR 35. [71] Paragraph 64 in the judgment. [72] Paragraph 67 in the judgment. [73] The offence contrary to section 17B(2) of the Public Order ordinance. [74] Supra. [75] See page 105 in the judgment. [76] Supra, paragraphs 68 ‑ 69 in the judgment. [77] Paragraph 69 in the judgment. [78] Mr Lee Siu Ho. [79] Supra. [80] Paragraph 64 in the judgment. [81] See paragraph 71 in the judgment of Wong Chi Fung. [82] See paragraph 75 in the judgment of Wong Chi Fung. [83] Section 4(3)(a) of the Community Service Orders Ordinance, Cap 378, Laws of Hong Kong. [84] See paragraph 77 in the Court of Final Appeal judgment of Wong Chi Fung. [85] [2018] 2 HKLRD 699, at paragraphs 146 - 147. [86] CAAR 1/2015. 
Mr Justice Ribeiro PJ: A. The main issues in this appeal 1. This case initially involved four parties: the plaintiff/respondent, Mr Fong Chak Kwan, a Hong Kong permanent resident (“Mr Fong”); the 1st defendant, Ascentic Limited, a Hong Kong company (“D1”); the 2nd defendant, Brentwood Industries, Inc, a company incorporated in Pennsylvania in the United States (“D2”); and the 3rd defendant/appellant, the Employees Compensation Assistance Fund Board (“the Board”), which administers the fund (“the Fund”) constituted under the Employees Compensation Assistance Ordinance (“Cap 365”). The present parties are the Board and Mr Fong. 2. In the circumstances described below, two issues have arisen between them. The first concerns the standing of the Board to take over D2’s defence and to challenge the Hong Kong Court’s assumption of jurisdiction against D2; and the second relates to the validity of such assumption of jurisdiction in respect of a tort allegedly committed outside of Hong Kong pursuant to Order 11 rule 1(1)(f) of the Rules of the High Court[1] which establishes a jurisdictional gateway (referred to here and below as “Gateway F”). 3. It will be apparent that the aforesaid jurisdictional issue arises in the first place between Mr Fong as the plaintiff and D2 as the foreign defendant. However, since judgment in default was entered against D2, the Board obtained leave to intervene in the proceedings to take over D2’s defence and to challenge the Court’s assumption of jurisdiction as aforesaid. It did so because it anticipated that the Fund might be liable to provide a statutory relief payment to Mr Fong if his claim against D2, when quantified, should be unsatisfied. The Board’s jurisdictional challenge below was unsuccessful, Marlene Ng J[2] and the Court of Appeal[3] holding that jurisdiction was properly founded via Gateway F. 4. The Board’s intervention gives rise to the first issue in this appeal, namely, as to the effect of section 33(d) of Cap 365 on the Fund’s potential liability. Its provisions are set out in Section E of this judgment. Because they exclude claims against the Fund in certain cases involving non-Hong Kong employment, the Appeal Committee[4] considered it apposite to examine the standing of the Board to intervene in this litigation (which had not previously been considered). On 22 February 2022, having sought submissions from the parties, it granted leave to appeal on the first issue, formulated as involving the following question: “Is section 33(d) of the Employees Compensation Assistance Ordinance (Cap 365) applicable and if so, what are its consequences for the appeal?” 5. The Appeal Committee also granted the Board leave to appeal on the following question concerning the second issue: “On a proper interpretation of the phrase ‘the damage was sustained ... within the jurisdiction’ in Gateway F, whether ‘the damage’ is limited to damage directly caused by the alleged tortious act, or whether it extends to indirect or consequential damage (e.g. the pecuniary expenditure or other loss resulting from the direct damage) sustained within the jurisdiction?” 6. I have had the benefit of reading in draft the judgment of Lord Collins of Mapesbury NPJ which deals with the second issue. I am respectfully wholly in agreement with his Lordship’s reasoning and conclusions. In this judgment, I deal with the first issue. B. The accident and Mr Fong’s bringing of proceedings in Hong Kong 7. On 10 October 2014, while Mr Fong was assigned to work as a site service specialist at a sewage treatment works in Ningbo on the Mainland, he accidentally fell more than 4 metres into the trench of a sedimentation tank when a metal plank on which he was walking gave way.[5] He consequently suffered serious personal injuries.[6] He returned to Hong Kong four days after the accident and received medical treatment in hospitals here and was granted sick leave for a total of 1,015 days. 8. On 7 March 2016, Mr Fong commenced the present action. D1 was served in Hong Kong and a defence was filed.[7] Then on 17 October 2016, Mr Fong obtained leave to issue and serve a concurrent writ[8] on D2 in Pennsylvania.[9] Such leave was obtained in reliance on five of the jurisdictional “gateways” in Order 11, namely, (1) that D2 was a necessary or proper party to the action brought against D1 (“Gateway C”);[10] (2) that the claim was brought for breach of a contract made within the jurisdiction (“Gateway D(i)”);[11] or (3) made by an agent within the jurisdiction on behalf of a principal outside (“Gateway D(ii)”) [12] or (4) governed by Hong Kong law (“Gateway D(iii)”);[13] and (5) that the claim was founded on a tort with the damage sustained within the jurisdiction, Gateway F referred to above.[14] 9. Mr Fong served the concurrent writ on D2 in the United States and, when D2 did not acknowledge service, on 4 July 2017 interlocutory judgment was entered against it for damages to be assessed.[15] D1 had disputed liability, contending inter alia that D2 and not D1, was Mr Fong’s employer.[16] D1 later settled Mr Fong’s claim without admission of liability, agreeing to pay him certain (relatively modest) sums under a consent order dated 18 May 2018.[17] C. The Board’s intervention and the decisions below 10. Initially, the Board did not consider the Fund at risk because it knew that D1 had an Employees Compensation insurance policy and that D1 was disputing liability. It did not have information as to whether D2 was insured but noted that D2 was denying that it was Mr Fong’s employer.[18] However, on 25 May 2018, the Board learned of Mr Fong’s settlement with D1. As the Board’s solicitor affirmed, this was seen to have major implications for the Board’s position: “... (a) the EC Policy had become irrelevant, (b) there was no other insurance policy known to be in force in relation to P/Accident insofar as section 25A of the ECO was concerned, and (c) there was no longer any ‘employer’ to properly contest P’s claim on both issues of liability and quantum, so the Board’s contingent interest in the outcome of the present action became more acute. Indeed, the following issues would affect the Board’s interests: (i) whether there was proper basis for P’s application to serve the [concurrent writ] on D2 out of jurisdiction, and (b) whether Hong Kong was the proper forum for bringing P’s claim against D2.”[19] 11. Taking the view that there were substantial issues to be contested, the Board submitted that in view of the Fund’s potential liability under Cap 365, it would be appropriate for the Board to intervene in the present action.[20] 12. Upon being joined as the 3rd defendant, the Board applied to set aside the Order granting Mr Fong leave to serve D2 outside the jurisdiction and also to set aside the default judgment on the grounds of material non-disclosure and to stay the proceedings on the basis of forum non conveniens. 13. Marlene Ng J found that Gateways C, D(iii) and F – but not Gateways D(i) and D(ii) – were available to Mr Fong. She rejected the non-disclosure and forum non conveniens applications.[21] 14. On appeal, the Court of Appeal disagreed with Marlene Ng J regarding the availability of Gateway C[22] and Gateway D(iii)[23] but upheld her judgment regarding Gateway F.[24] It is of present relevance that her Ladyship’s finding that Gateways D(i) and D(ii) were inapplicable was maintained. This is a point to which I shall return.[25] The Court of Appeal upheld Marlene Ng J’s rejection of the non-disclosure[26] and forum non conveniens[27] applications and they form no part of the present appeal. D. The issue of the Board’s standing 15. Cap 365 has as one of its main objects the making of relief payments to eligible persons in relation to damages awarded by a Hong Kong court of competent jurisdiction for employment-related injuries which they have been unable to recover despite reasonable enforcement efforts. Thus, section 20A of Cap 365 relevantly states: “(1) An eligible person who is unable to recover from an employer payment of an amount of damages for which the employer is liable may apply for a relief payment of that amount from the Fund.” (2) For the purposes of this section, an employer is not to be regarded as liable for the payment of an amount of damages unless the amount is payable pursuant to a judgment or order of a court of competent jurisdiction in Hong Kong.” 16. Similar provision is made regarding unsatisfied awards of employees’ compensation under the Employees’ Compensation Ordinance (“Cap 282”).[28] 17. An “eligible person” is, for present purposes, an “injured employee”; and a “relief payment” is a payment made from the Fund pursuant to an application under section 20A.[29] The amount of such a relief payment is the amount of common law damages which the employer is liable to pay in respect of personal injury suffered by the employee in an accident arising out of and in the course of his employment, subject to certain reductions.[30] 18. Cap 365 enables the Board, if faced with a contingent liability regarding a relief payment, to apply to intervene in existing proceedings, materially providing as follows: “If proceedings have been initiated to claim compensation or damages ... (a) where no policy of insurance is known to be in force at the time of the accident to which the proceedings relate, the Board may apply to the court to join in the proceedings as a party in accordance with Order 15, rule 6 of the Rules of the High Court ... to take over the defence as if it were the employer in the proceedings if— … (v) at any time when the employer fails to attend the hearing, leaving the claim uncontested ...”[31] 19. As we have seen, it was pursuant to those provisions that the Board, considering the Fund to be at risk of a claim by Mr Fong regarding unpaid damages awarded against D2, obtained leave to be joined as a party to take over the defence as if it were Mr Fong’s employer. 20. It is in this context that the issue of the Board’s standing arises. The questions are whether, on its true construction and on the facts established, section 33(d) applies to exclude potential claims by Mr Fong against the Fund; and if so, how such exclusion affects the Board’s intervention in the proceedings and what the consequences should be for the outcome of this appeal. E. The context and purpose of section 33(d) 21. Cap 365, section 33(d) provides: “No claim lies under this Part[32] in respect of— (d) any compensation or damages for injury to an employee who has been engaged outside Hong Kong by an employer who— (i) is outside Hong Kong; and (ii) has no place of business in Hong Kong (and whether or not the employer’s business is for gain).” 22. Section 33(d) thus excludes claims against the Fund involving injury to an employee (i) “who has been engaged outside Hong Kong” by an employer (ii) who “is outside Hong Kong” and (iii) who has “no place of business in Hong Kong”. They are features which locate the employment relationship outside of Hong Kong, denying access to the Fund where the three conditions apply. 23. The underlying policy is readily discernible when one considers how the Fund is financed and its purpose. The relief payments made under Cap 365 are funded by a portion of the levies collected under the Employees’ Compensation Insurance Levies Ordinance (“Cap 411”).[33] Cap 411 imposes a levy on each premium payable by employers under insurance policies issued for the purposes of Cap 282.[34] Such insurance, taken out by employers against liability for employees’ compensation[35] or common law damages[36] in respect of injuries to their employees in accidents arising out of and in the course of employment, is compulsory.[37] 24. Thus, Cap 282, section 40(1) provides that “no employer shall employ any employee in any employment unless there is in force in relation to such employee a policy of insurance issued by an insurer” for a specified amount “in respect of the liability of the employer”.[38] Cap 282, section 2 relevantly defines “employee” as “any person who has ... entered into or works under a contract of service or apprenticeship with an employer in any employment”, the focus being on the existence of a contract of employment. 25. The Fund accordingly represents a locally constituted monetary pool with the object of providing relief to injured employees who have been unable to recover compensation or damages awarded against their employers by Hong Kong courts of competent jurisdiction. 26. That is not to say that accidents occurring abroad are all excluded. Cap 282, section 30B(2) enables such a claim to be made (potentially leading to a relief payment claim under Cap 365), provided that the employment relationship is based in Hong Kong: “This Ordinance also applies where personal injury by accident arising out of and in the course of employment is caused to an employee outside Hong Kong where the employee’s contract of employment is entered into in Hong Kong with an employer who is a person carrying on business in Hong Kong.” Thus an employee who is injured when sent to do a job abroad is covered only if his or her contract of employment was entered into in Hong Kong with an employer carrying on business here. 27. In cases where an employer, carrying on business abroad, is prepared to submit to the jurisdiction of the Hong Kong court in relation to a claim based on an accident which occurred abroad, Cap 282, section 30B(5) lays down as a necessary condition of its application the requirement that the employee was “recruited or engaged” in Hong Kong: “If an employer who is a person carrying on business outside Hong Kong submits or has agreed to submit to the jurisdiction of the Court, then, notwithstanding that the accident causing the personal injury occurred outside Hong Kong, this Ordinance shall apply to employees within the meaning of this Ordinance who have been recruited or engaged in Hong Kong.” 28. It is noteworthy that Cap 282 makes it compulsory for employers who enter into employment contracts with employees in Hong Kong to acquire insurance against liability for accidents occurring abroad as catered for by the abovementioned provisions of section 30B. Thus, Cap 282, section 40(1A) states: “Subsection (1) does not require an employer to obtain insurance for any liability he may have in respect of damages awarded by a court outside Hong Kong to an employee referred to in section 30B.” By necessary implication, if the employee is employed in Hong Kong, section 40(1) in combination with section 2 require the employer to have insurance cover in respect of liability for damages awarded by a Hong Kong court. This clearly covers section 30B(2) cases and would also catch section 30B(5) cases where the employee was contractually “engaged” (as opposed to non-contractually “recruited”) in Hong Kong. The Fund’s liability to make compensation or relief payments is therefore premised on the relevant employer having been compulsorily required to insure against his own liability and thus required to contribute the levy financing the Fund. 29. The statutory intention of confining access to the Fund to Hong Kong-based employment relationships is also evident from the legislative history of section 33(d). Those provisions were introduced as one measure in a raft of legislative changes (described as a “rescue package”[39]) introduced by the Employees Compensation Assistance (Amendment) Bill 2002 in response to a financial crisis then faced by the Fund involving recurrent deficits due to various causes.[40] 30. It follows that the provisions of section 33(d) are intended to be construed restrictively, recognising that they were introduced to confine the Fund’s liability to claims arising out of Hong Kong-based employment relationships. F. Construction of section 33(d) 31. In the present case, it is uncontroversial that two of the three elements of section 33(d), namely, that the employer (i) “is outside Hong Kong” and (ii) “has no place of business in Hong Kong” are satisfied. There is no dispute as to the construction of those words nor as to the facts. D2 is a Pennsylvania company and therefore “outside Hong Kong”. It is also clear that D2 had no place of business in Hong Kong. The Judge found that “D2 had no business, presence or assets in Hong Kong”[41] and the Court of Appeal likewise noted that “[D2] was a US-based company with no presence in Hong Kong”.[42] 32. However, the parties differ as to the true construction and applicability of the element which excludes claims for injury to “an employee who has been engaged outside Hong Kong by an employer”. More particularly, they differ as to what “engaged” should be held to mean. 33. Mr Horace Wong SC, appearing for the Board,[43] rightly recognises that “engaged” has been given various meanings depending on the context. It is unnecessary to list the cases cited. He submits that for present purposes, the word should be interpreted as referring either (i) to the making of the contract of employment or (ii) the provision of services under that contract (or as he put it at the hearing, “providing the employee with work”). He contends that on either basis, upon the findings made by the Judge, this third element is also satisfied, making section 33(d) applicable and thus excluding claims against the Fund. 34. Ms Audrey Eu SC, appearing for Mr Fong,[44] relies principally on the Court of Appeal’s decision in Chan Sze Yuen v Tin Wo Engineering Co Ltd,[45] arguing that “engaged” in the present context bears a different meaning from “employed” and should be taken to be essentially the same as “recruited” and so treated as including “the preparatory steps leading to the conclusion of the contract of employment.”[46] She seeks to argue that Mr Fong was “engaged” in Hong Kong when he signed the agreement at D1’s premises, even though the contract with D2 was held not to have come into existence until D2’s subsequent acceptance in the United States. 35. I am unable to accept Ms Eu SC’s construction. Nor do I accept the second limb of Mr Wong SC’s submission suggesting that “engaged” should be interpreted to refer to the actual provision of services under the employment contract. In my view, in the present context, a person is “engaged” by an employer when that person enters into the contract of employment whereby he or she is hired as an employee. If that occurs outside of Hong Kong, the third element of section 33(d) is satisfied and, given satisfaction of the other two elements, the Fund’s liability excluded. 36. Chan Sze Yuen,[47] relied on by Ms Eu SC, was a case concerned with Cap 282, section 30B(5)[48] involving a non-Hong Kong employer who submitted to the local jurisdiction[49] in respect of a personal injury claim arising out of an accident occurring abroad. The claimant worked as a steel binder on a building construction site in Macau where he suffered a leg injury. The Court of Appeal had to consider whether he had been “recruited or engaged in Hong Kong” so as to make Cap 282 applicable. The main focus of the judgment is on the meaning of “recruited ... in Hong Kong”. 37. Thus Chu JA, writing for the Court of Appeal, noted that: “The trial judge found that at the time the applicant set off to Macau, his employment to work at the Site had already been confirmed and that he was not going to Macau merely to see whether he would be employed. Applying the ordinary English meaning of the word ‘recruitment’, he held that the applicant was recruited or engaged in Hong Kong and that section 30B(5) of the Ordinance applies.”[50] “... when he set off for Macau he was certain that he would be working at the Site. On this basis, the trial judge concluded that the applicant had been recruited in Hong Kong. In my view, on the evidence as accepted, there is ample room for the trial judge’s conclusion that the applicant was recruited or engaged in Hong Kong.”[51] 38. Her Ladyship held that “recruited” has a different meaning from “employed”.[52] I respectfully agree. The claimant was properly held to have been “recruited” in Hong Kong by a fellow worker who had suggested, on behalf of the foreman in Macau, that they should take up a job at the Macau building site, and he agreed. He evidently did not enter into any contract of employment until he arrived at the site (and so would not have been “engaged in Hong Kong” in the sense I have suggested). But the Court of Appeal found that the Judge had ample evidence to hold that he had been “recruited” here, satisfying the requirement that he was “recruited or engaged” in Hong Kong. A person can plainly be recruited here (eg, by a recruiting agent) and then be later engaged by the employer when entering into the contract abroad. Such pre-engagement “recruitment” in Chan Sze Yuen was sufficient to trigger section 30B(5) and the Court of Appeal’s construction is thus consistent with “engaged” being taken to mean, in contrast, “contractually hired as an employee”. 39. Chan Sze Yuen therefore does not support Ms Eu SC’s proposition that “engaged” should be treated as essentially the same as “recruited” so as to extend to “preparatory steps leading to the conclusion of the contract of employment”. Indeed, that decision runs counter to Counsel’s submission. Section 30B(5) applies to employees “within the meaning of this Ordinance who have been recruited or engaged in Hong Kong”. This refers to an employee who has entered into a contract of employment (and thus is an employee within the meaning of Cap 282) and had either been “recruited” or “engaged” in Hong Kong. As the Court of Appeal in Chan Sze Yuen held, the two words should be taken to mean two different things. As Bennion puts it: “… there is a presumption that where different words are used in an Act they have different meanings. So if a word is capable of bearing more than one meaning it is unlikely to be interpreted as having the same meaning as a different word in the same legislation.”[53] And as the Court held in HKSAR v Cheung Wai Kwong,[54] inclusion of two alternative activities (“driving” or “using” a vehicle) indicates that the Legislature “recognised and intended a distinction between the two activities”. 40. It is clear that Caps 282, 411 and 365 are in pari materia, for mingpart of the same statutory relief scheme and must be read together as a coherent whole, assuming uniformity in the use of language.[55] Section 33(d) does not mention “recruited” but refers to an employee “engaged outside Hong Kong”. Since 30B(5) uses “recruited” and “engaged” with distinct meanings, “engaged” in section 33(d) cannot coherently be equated with “recruited” within that scheme. 41. I am indebted to Lam PJ for, and gratefully adopt, his Lordship’s observation as to the difference between the two terms evident in the Chinese version of section 30B(5), as follows: In the Chinese version of Section 30B(5), the phrase “recruited or engaged” is rendered in Chinese as “招募或聘用”. As a matter of Chinese language, “招募” is a different concept from “聘用”. In the Chinese version, like the English version, the two expressions are put together by a conjunction “或” (meaning “or”). “招募” (recruited) refers to the initial process by which an employer searches for suitable candidates for employment and does not involve the making of any specific decision to employ a particular person. On the other hand, “聘用” (engaged) refers to a definite decision to employ a specific person and it is at the end stage in the employment process. Whilst as a matter of English usage the word “engaged”, in a context other than the way it is being used in the phrase “recruited or engaged” under section 30B(5), may sometimes be capable of being interpreted broadly to cover the whole process starting from recruitment to the decision to employ a person, it is not so for the Chinese expression “聘用”. For such broader concept of “engaged”, the appropriate Chinese rendering would be “招聘”. “聘用” carries a narrower connotation and involves the specific engagement of a particular candidate as an employee. When the English version is construed together with the Chinese version it is clear that “engaged” does not mean “recruited” in section 30B(5). 42. For all the aforesaid reasons, I cannot accept Ms Eu SC’s construction of section 33(d). 43. I do not agree with the second limb of Mr Wong SC’s construction because section 33(d) refers to an employee “who has been engaged outside Hong Kong by an employer” which naturally connotes being hired as an employee. The provision is not addressing a situation where an employee is “engaged in” or “engaged on” or “provided with” work of some kind. The distinction arises simply as a matter of language. Thus, for instance, in Benninga (Mitcham) Ltd v Bijstra,[56] in an entirely different context, MacKinnon LJ stated: “An employer ‘engages’ a servant when he makes an agreement with him for his services. A workman is ‘engaged’ on work when he is actually carrying it out.” 44. Moreover, the words “an employee who has been engaged outside Hong Kong” refer to a completed past event. They refer to the point in time when the parties entered into a contract of employment, commencing their employment relationship. 45. Construction of the relevant phrase to mean “an employee who has entered into a contract of employment outside Hong Kong” is reinforced when section 33(d) is read together and contrasted with section 30B(2) of Cap 282. That provision, as we have seen, states: “This Ordinance also applies where personal injury by accident arising out of and in the course of employment is caused to an employee outside Hong Kong where the employee’s contract of employment is entered into in Hong Kong with an employer who is a person carrying on business in Hong Kong.” (Italics supplied) 46. Section 30B(2) presents a mirror-image of section 33(d). While section 33(d) sets out three elements which operate to exclude claims from Cap 365, section 30B(2) sets out matching elements which, in reverse, function to include claims within Cap 282 (which therefore qualify for potential relief payment claims under Cap 365). 47. Thus, section 33(d) excludes Cap 365 claims which involve an injury to an employee (i) “who has been engaged outside Hong Kong” by an employer (ii) who “is outside Hong Kong” and (iii) who has “no place of business in Hong Kong”; while section 30B(2) permits claims where the injury is caused to an employee outside Hong Kong where (a) “the employee’s contract of employment is entered into in Hong Kong” (b) “with an employer who is a person carrying on business in Hong Kong”. 48. The inclusionary element (a) [an employee whose “contract of employment is entered into in Hong Kong”] is the obverse of the exclusionary element (i) [an employee “engaged outside Hong Kong”]; and element (b) [“an employer who is a person carrying on business in Hong Kong”] is the obverse of elements (ii) and (iii) [an employer who is “outside Hong Kong” and has “no place of business in Hong Kong”]. This strongly suggests that construed contextually, “engaged outside Hong Kong” is, like its mirror-image, also concerned with entering into a contract of employment. G. Application of section 33(d) 49. As previously mentioned the two elements of section 33(d) concerning the employer’s absence from Hong Kong are plainly satisfied. Since section 33(d) was not adverted to, the third element – whether Mr Fong had been “been engaged outside Hong Kong” – was not in terms addressed by the Courts below. However, in dealing with the jurisdictional avenues involving Gateway D, the Judge, upheld by the Court of Appeal, made findings directly bearing on whether Mr Fong had entered into the employment contract with D2 outside Hong Kong and hence on the applicability of the third element of section 33(d). 50. Before the Judge, Mr Fong sought to bring himself within Gateway D(i) under O 11 r 1(1)(d)(i) which materially provides: “... service of a writ out of the jurisdiction is permissible with the leave of the Court if in the action begun by the writ ... (d) the claim is brought to ... recover damages ... in respect of the breach of a contract, being ... a contract which (i) was made within the jurisdiction ...” 51. The Judge analysed in detail the evidence relating to the making of Mr Fong’s contract of employment[57] and concluded that Mr Fong “would have real difficulty to show a good arguable case that the Employment Contract was concluded in Hong Kong”.[58] Applying the postal rule, she held that the contract had been concluded in the United States and accordingly that Gateway D(i) was not open to Mr Fong.[59] That decision was upheld by the Court of Appeal.[60] 52. Mr Fong also relied on Gateway D(ii) under O 11 r 1(1)(d)(ii) which provides: “... service of a writ out of the jurisdiction is permissible with the leave of the Court if in the action begun by the writ ... (d) the claim is brought to ... recover damages ... in respect of the breach of a contract, being ... a contract which (ii) was made by or through an agent trading or residing within the jurisdiction on behalf of a principal trading or residing out of the jurisdiction.” 53. The Judge concluded that there was no basis for finding an agency and that this gateway was also not available.[61] The conclusion that the contract had been made in the United States – and not in Hong Kong – was therefore maintained. On my construction of section 33(d), this establishes that Mr Fong “had been engaged outside Hong Kong by” D2 as his employer. 54. The Judge did find that there was a good arguable case that the contract was governed by Hong Kong law and that Gateway D(iii) was thus available. But that was not a finding relevant to section 33(d). In any event, the Court of Appeal reversed her Ladyship on this point and held that on the evidence, Mr Fong had not made out a good arguable case for the availability of Gateway D(iii).[62] That decision is not under appeal. 55. It has therefore been decided concurrently by the Courts below that there is not even a good arguable case either that Mr Fong’s contract of employment had been made within the jurisdiction, or made by or through an agent trading or residing within the jurisdiction. The inescapable conclusion on those findings, albeit made in the context of Gateway D, is that the third element of section 33(d) is satisfied and that Mr Fong had been “engaged outside Hong Kong”. 56. Ms Eu SC submitted that since the section 33(d) point was not taken below, the well-established principles in Flywin Co Ltd v Strong & Associates Ltd[63] should be applied to give Mr Fong the opportunity of filing further evidence and making further arguments. However, when asked what evidence might be available, it was clear that Ms Eu SC was not suggesting that there was any further evidence to be had concerning the place where the contract of employment was made. She suggested that, assuming the correctness of her construction of “engaged” as encompassing pre-contractual conduct equated with “recruitment”, there might be evidence of a course of negotiations in Hong Kong which resulted in Mr Fong signing the written contract which was then concluded in the United States. The relevance, if any, of such evidence is thus wholly dependent on acceptance of Ms Eu SC’s construction of “engaged” as embracing such pre-contractual conduct. However, as previously indicated, I do not accept that construction. H. The consequences for this appeal 57. For the foregoing reasons, I conclude that section 33(d) would preclude Mr Fong from making a successful claim against the Fund for a relief payment. The Fund is and was therefore not at risk of any claim by him. Cap 365, section 25A caters for the Board’s intervention only if it can properly apply to be joined as a party in accordance with O 15 r 6.[64] Since no question or issue connected with any potential claim for relief against the Fund arises as between the Board and Mr Fong, the Board was not properly joined and it has no standing to challenge the Court’s assumption of jurisdiction over D2. 58. Ms Eu SC invited the Court to defer coming to the aforesaid decision. She submitted that, assuming that Mr Fong were to succeed on the Gateway F argument so that the default judgment against D2 remains in place, he should be allowed to proceed to an assessment of damages and then to seek to enforce the judgment in the United States (in order to satisfy the requirements of making reasonable enforcement efforts under Cap 365, section 20A(3)) and, if that proved unsuccessful, to claim a relief payment from the Fund; and if that was refused, to commence proceedings by writ against the Board under Cap 365, section 24. On the premise that Mr Fong succeeds on the Gateway F issue, there is of course nothing to stop him proceeding to an assessment of damages and seeking to enforce the judgment in the United States (or indeed elsewhere, such as on the Mainland), if he is so advised. However, if the ultimate object is to make a claim against the Fund, I do not think it would be beneficial to the parties or in the interests of justice to expend the additional time, effort and resources in pursuing the course of action outlined by Ms Eu SC where the outcome is now clear. 59. Accordingly, on the ground that the Board lacks standing to challenge the assumption of jurisdiction via Gateway F, I would dismiss the Board’s appeal. Mr Justice Fok PJ: 60. I agree with the judgment of Mr Justice Ribeiro PJ on the first issue and with that of Lord Collins of Mapesbury NPJ on the second issue. Mr Justice Lam PJ: 61. I agree that the appeal should be dismissed for the reasons set out in the judgments of Mr Justice Ribeiro PJ and Lord Collins of Mapesbury NPJ. Mr Justice Bokhary NPJ: 62. I agree with Mr Justice Ribeiro PJ and with Lord Collins of Mapesbury NPJ. Accordingly, I would dismiss this appeal. There is nothing I wish to add on the second issue. On the first issue, I would add simply this. If section 33(d) of the Employees Compensation Assistance Ordinance (Cap 365) stood alone, I might have construed the word “engaged” in it to mean “recruited”. But it does not stand alone. It has to be read together with section 30B(5) of the Employees’ Compensation Ordinance (Cap 282) in which the expression “recruited or engaged” is used. These two are statutes on closely related matters. That being so, the presumption is that they similarly use different words to mean different things. Such a presumption is rebuttable. But rebutting it is an uphill task. And the gradient is particularly steep where two different words are contrasted by having “or” inserted between them. There is nothing in the circumstances of the present case capable of propelling the “recruited” construction up so steep a gradient. 63. The second issue, which the Board got itself joined to argue, has been answered against the Board. In so far as the first issue has been answered in the Board’s favour, the answer means that the Board was not on risk, had no standing and should not have been joined. Lord Collins of Mapesbury NPJ: 64. I agree, in common with the other members of this Court, with Mr Justice Ribeiro PJ’s judgment on the first issue. In this judgment, I deal with the second issue, whether the default judgment against the second defendant should be set aside on the ground that the case did not come within the tort gateway in the Rules of the High Court, Order 11, r. 1(1)(f). 65. By Order 11, r. 1(1) “... service of a writ out of the jurisdiction is permissible with the leave of the Court if in the action begun by the writ – ... (f) the claim is founded on a tort and the damage was sustained, or resulted from an act committed, within the jurisdiction” (“Gateway F”). 66. The Statement of Claim alleged that the plaintiff’s fall at the sewage treatment plant in Ningbo on the Mainland was due to (inter alia) the negligence and/or breach of common law duty of care of the second defendant, and that as a result of the accident he sustained serious injuries to his left shoulder, left leg, and his head, thereby causing defect in his vision. He was granted leave ex parte to serve the writ out of the jurisdiction on the second defendant in the United States under (inter alia)[65] Gateway F on the basis that the negligence had been committed in Hong Kong. But after an interlocutory judgment had been entered against the second defendant and the Board had intervened, he made an affirmation that in any event Gateway F was satisfied because he had sustained and continued to sustain damage within Hong Kong.[66] His revised statement of damages claimed damages for pain, suffering and loss of amenities, loss of earnings, loss of earnings capacity, and special damages for medical and travel expenses. The judgments below 67. By the time of the hearing before Marlene Ng J, it was accepted on behalf of the plaintiff that the tort had not been committed in Hong Kong, and it was conceded on his behalf that he had suffered direct damage (i.e. immediate bodily injuries) as a result of the accident at the factory and therefore that the cause of action in negligence had been completed on the Mainland, and that he experienced only indirect damage in Hong Kong.[67] 68. After a very full analysis of the authorities in Hong Kong, England, Australia and Canada, including in particular Brownlie v Four Seasons Holdings Inc,[68] the judge, in line with the majority judgments of Lady Hale and Lord Wilson in that decision, and being unpersuaded by the minority view of Lord Sumption, decided that (a) the expression “damage” in Gateway F was not limited to damage which completed the cause of action; (b) the expression was not limited to direct damage as opposed to indirect/consequential damage; (c) where damage was felt in more than one jurisdiction, indirect/consequential damage qualified under Gateway F if it was of some significance; (d) the expression was to be given its ordinary and natural meaning, which embraced indirect/consequential damage; and (e) the consequences of a wide interpretation were sufficiently addressed by the discretion as to forum conveniens. 69. The Court of Appeal (Cheung and Yuen JJA), after an equally full analysis of the authorities (including the decision of the English Court of Appeal in Brownlie v FS Cairo (Nile Plaza) LLC[69]in the second phase of the claims by Lady Brownlie, which are elaborated below) dismissed the appeal. It followed its earlier decision in Dynasty Line Ltd v Sukamto Sia[70]that Gateway F applied if some significant damage had been sustained in Hong Kong. The Court of Appeal, like the judge, held that the reasoning of the majority in Brownlie v Four Seasons Holdings Inc was to be preferred to that of the minority.Damage included all of the heads of damage which might be suffered as a result of tortious conduct, including all the detriment, physical, financial and social which the plaintiff suffered as a result. The natural and ordinary meaning of Gateway F was clear, and there was no basis for drawing a distinction between direct and indirect damage. Nor was there any basis for applying the European jurisprudence on the Brussels Convention and Brussels I Regulations. Finally, the expression “the damage” in Gateway F did not mean that all the damage, or the damage which completed the cause of action, had to be sustained in Hong Kong. 70. Both Marlene Ng J and the Court of Appeal found that the discretion should be exercised in favour of the Hong Kong court, and leave to appeal was not sought from that finding. Arguments on appeal 71. For the Board it was argued that: (1) only direct damage satisfies Gateway F; (2) the significance of the damage is not determinative, since both direct damage and indirect damage can be significant or insignificant; (3) the critical consideration in interpreting Gateway F is not the ordinary or natural meaning, but its context and purpose; (4) the gateways represent the considered judgment of the legislature as to the classes of case in which a foreign defendant may be brought before the Hong Kong court, and their purpose is to identify a sufficient connection to justify the exercise of extra-territorial jurisdiction; (5) conferring jurisdiction on the place where the pecuniary consequences are felt, or where the pain and suffering are experienced, would in effect confer jurisdiction on the courts of the plaintiff’s domicile, which has never been a basis of jurisdiction in Hong Kong; (6) such a basis of jurisdiction would become portable, since a plaintiff could create a link with Hong Kong by moving there, and effectively create a form of universal jurisdiction; (7) it is no answer to say that excesses could be tempered by applying forum conveniens principles, since the existence of jurisdiction and its exercise are different matters, and serve different purposes; (8) nor is it an answer that the distinction between direct and indirect damage is elusive or meaningless, since the distinction is well-recognised and its application is straightforward. 72. For the plaintiff it was argued that: (1) the natural and ordinary meaning of the expression “the damage” is consistent with its context and purpose; (2) the wide meaning has worked well, whereas the narrow meaning is artificial and untested; (3) the jurisdiction would not be an exorbitant one; (4) it would be inconsistent with the purpose of the rule to interpret “damage” so as to in effect limit it to the place of commission of the tort; (5) the concept of indirect damage is not known to Hong Kong law, and a distinction between direct and indirect damage would be difficult to draw; (6) if the connection with the forum is not substantial, its weight can be balanced at the discretionary stage, which is an appropriate and effective mechanism to prevent acceptance of jurisdiction where the connection is casual or adventitious. 73. Since the judgments below, and this judgment also, to a large extent hinge on the applicability of the recent decisions in England, and in particular the two decisions of the UK Supreme Court in the Brownlie litigation, it is necessary to set out the history of the relevant provisions, to highlight the differences between the current rules in England and those in Hong Kong, and then, after a brief survey of decisions in Hong Kong and other jurisdictions, to consider whether the majority or the minority views in the Brownlie cases are to be preferred in Hong Kong. The evolution of Gateway F and cross-border torts 74. Prior to the introduction of the Rules of Court in the Supreme Court of Judicature Act 1873, jurisdiction over defendants abroad was governed by the Common Law Procedure Act 1852, sections 18 and 19, the effect of which was that service on a foreigner abroad was permissible where the cause of action “arose within the jurisdiction.” In Distillers Co Ltd v Thompson,[71] where the plaintiff had been born in New South Wales with severe disability as a result of his mother having taken the drug thalidomide, the Privy Council on an appeal from New South Wales (which retained the nineteenth century legislation) adopted the test of the place where in substance the act or omission occurred which gave the plaintiff his cause of action: since the essence of the complaint was failure to warn his mother in New South Wales of the danger of the product (rather than the faulty manufacture of the product in England), the cause of action was held to have arisen in New South Wales. 75. The Rules of Court scheduled to the Supreme Court of Judicature Act 1873 provided in Rule 6 that service out of the jurisdiction could be authorised “in a case in which the cause of action has arisen within the jurisdiction, or is properly cognizable against a defendant within the jurisdiction.” Order XI of the Rules of Court scheduled to the Supreme Court of Judicature Act 1875, permitted service out of the jurisdiction in relation to “any act or thing ... for which damages are sought to be recovered was or is to be done or is situate within the jurisdiction,” and in Bree v Marescaux[72]it was held that this did not extend to permit service abroad where only the damage had occurred in England: the plaintiff had alleged that he had been dismissed in England as a result of a false allegation of misconduct on a voyage to Jamaica. 76. This head of jurisdiction was not included in the Rules of the Supreme Court 1883, and it was held that service out of the jurisdiction was not generally available in tort actions unless they came under one of the specific heads of jurisdiction, such as necessary or proper party: Lenders v Anderson;[73] Croft v King;[74] Williams v Cartwright.[75] 77. But a specific tort head of jurisdiction was introduced in 1920 as RSC Order 11, r. 1(1)(ee), to cover cases where “the action is founded on a tort committed within the jurisdiction.” In George Monro, Ltd v American Cyanamid and Chemical Corp[76]the English distributor of a pest control product supplied by a US manufacturer sold the product to Boots Cash Chemists, which in turn sold it to a local council, which contracted to remove rats from a farm. The council was held liable to the farmer for negligence and the English distributor was ordered to contribute. The English distributor failed in its attempt to sue the US manufacturer in England. Goddard LJ seems to have thought that the tort jurisdiction did not extend to a case where the wrongful act was committed out of the jurisdiction but the damage resulting took place within the jurisdiction, and Du Parcq LJ asked: “Where was the wrongful act, from which the damage flows, in fact done? The question is not where was the damage suffered, even though damage may be of the gist of the action.”[77] 78. The principle in Distillers Co Ltd v Thompson was subsequently applied to the tort gateway in Order 11, r. 1(1), so that in the case of negligent and fraudulent misrepresentation, where the negligent or fraudulent statement was communicated from one country to another, the torts were held to have been committed where the statement was received and acted upon: Multinational Gas Co v Multinational Gas Services Ltd;[78] The Albaforth.[79] 79. With effect from 1987 the rule in England was changed to cover cases where “the claim is founded on a tort and the damage was sustained, or resulted from an act committed, within the jurisdiction.” The change was made to bring cases outside the sphere of the Brussels Convention into line, at least broadly, with the decision of the European Court in Bier v Mines de Potasse d’Alsace SA[80]that the expression “place where the harmful event occurred” in the Brussels Convention, Article 5(3), was intended to cover both the place where the damage occurred and the place of the event giving rise to it. Subsequently, in a series of cases on this provision (and its successors in the Brussels I Regulation and recast Brussels I Regulation), not all of which are easy to reconcile, the European Court decided that for this purpose damage did not include indirect or consequential financial loss: e.g. Dumez France SA v Hessische Landesbank;[81] Marinari v Lloyds Bank Plc;[82] Universal Music International Holding BV v Schilling;[83] contrast Kolassa v Barclays Bank Plc.[84] 80. In England, since the introduction of the Civil Procedure Rules 1998 the tort head of jurisdiction is to be found in what is now Practice Direction 6B, para 3.1(9), as: “A claim is made in tort where –(a) damage was sustained ... within the jurisdiction; or (b) damage which has been or will be sustained results from an act committed ... within the jurisdiction.”[85] A change had been made in 2000[86] from “the damage was sustained” to “damage was sustained” to bring the rule into line with the decision of the English Court of Appeal in Metall und Rohstoff AG v Donaldson Lufkin & Jenrette Inc,[87] that it was not necessary that all the damage should have been sustained within the jurisdiction, provided that damage has resulted from substantial and efficacious acts committed within the jurisdiction (whether or not other substantial and efficacious acts have been committed elsewhere). Differences between Hong Kong and English rules 81. The present Rules of the High Court were introduced in 1988, and included the current version of Gateway F as it had by then been amended in England. The relevant differences for present purposes between the English rule and the Hong Kong rule are as follows. 82. First, the English rule now refers to a claim “made in tort” whereas the Hong Kong rule continues to refer to a claim “founded on a tort.” No doubt the English formulation was intended to modernise the language as part of the reforms in the Civil Procedure Rules which replaced the Rules of the Supreme Court, although the wording in Hong Kong continues to be correct as a matter of legal terminology, since claims are not “made” in tort. 83. Second, the English rule no longer refers to “the damage,” as it did between 1987 and 2000, but now refers to “damage” simpliciter to take account of the decision in the Metall und Rohstoff AG case, above, whereas the Hong Kong rule remains unchanged. 84. A further difference between the Hong Kong rules and the English rules is that the Rules of High Court in Hong Kong provide in Order 11, r. 4(2) that leave to serve out of the jurisdiction shall not be granted “unless it shall be made sufficiently to appear to the Court that the case is a proper one for service out of the jurisdiction under this Order.” This is the same as the former RSC Order 11, r. 4(2), which was not reproduced in the Civil Procedure Rules. The English equivalent, introduced in 2000,[88] is now CPR r. 6.37(3): “The court will not give permission unless satisfied that England and Wales is the proper place in which to bring the claim.” The authorities in Hong Kong and other jurisdictions 85. The important issue raised on this part of the appeal is one which has been the subject of debate and decision in Hong Kong and elsewhere. Hong Kong: the decision in Dynasty Line Ltd v Sukamto Sia[89] 86. The effect of this decision, which supports the plaintiff’s case, is that no enquiry into the place where the cause of action is complete is necessary. The allegation was that a BVI company had been defrauded by its Singaporean shareholders, who had caused it to purchase shares in a Hong Kong listed company and procured it to hand over the shares to them before the vendors had been paid; the shareholders had then pledged the shares to financial institutions as security for loans, and the financial institutions sold the shares in diminution of the indebtedness. The original vendors of the shares sued the plaintiff BVI company for the price of shares which it should have, but no longer, owned, and the plaintiff sued the alleged fraudsters in Hong Kong. One of the defendants argued that the damage was sustained when the shares were first pledged, and there was a dispute about where that had occurred. It was held (applying Metall und Rohstoff AG v Donaldson Lufkin & Jenrette Inc, above) that, in considering whether damage was sustained in Hong Kong, it was sufficient if some significant damage had occurred in Hong Kong, and Gateway F applied because it had been pleaded that significant damage had been sustained in Hong Kong when the shares in the Hong Kong listed company were sold in Hong Kong by the pledgee financial institutions.[90] Australia 87. The Australian authorities favour the plaintiff’s approach. In New South Wales the rule (since repealed) that process may be served outside the State “where the proceedings are founded on, or are for the recovery of, damage suffered wholly or partly in the State caused by a tortious act or omission wherever occurring” was held to apply where the initial injury and damage occurred in Queensland, but the plaintiff suffered economic loss through not being able to carry out his occupation. The damage may be indirect, or consequential whether by way of bodily injury or financial or economic loss (Challenor v Douglas[91]) and is to be contrasted with the element necessary to complete a cause of action, and includes all the detriment, physical, financial and social which the plaintiff suffers as the result of the tortious conduct of the defendant (Flaherty v Girgis[92]). Canada 88. The early cases are helpful to the plaintiff’s case, but more recent authority disapproves the place of damage as a suitable basis for jurisdiction. As in Australia, the early Canadian authorities are largely concerned with damage occurring in the forum (Ontario) following accidents in another part of the federation. In cases involving a rule granting jurisdiction in the case of a claim “in respect of damage sustained in Ontario arising from a tort ... committed elsewhere” it was held that the Ontario court had jurisdiction where the accident occurred in another province and some damage had been sustained there, but consequential damage which flowed from the injury occurred in Ontario.[93] Those decisions are now obsolete, since it has been held by the Supreme Court of Canada that the existence and exercise of jurisdiction both with regard to defendants in other Canadian provinces and in foreign countries are subject to a constitutional imperative of a “real and substantial connection” with the forum. In determining whether a court can assume jurisdiction, the court relies on a set of specific factors which are given presumptive effect, and the place where the damage is sustained is not one of those factors. In Club Resorts Ltd v Van Breda[94]the Supreme Court of Canada said: “The use of damage sustained as a connecting factor may raise difficult issues. For torts like defamation, sustaining damage completes the commission of the tort and often tends to locate the tort in the jurisdiction where the damage is sustained. In other cases, the situation is less clear. The problem with accepting unreservedly that if damage is sustained at a particular place, the claim presumptively falls within the jurisdiction of the courts of the place, is that this risks sweeping into that jurisdiction claims that have only a limited relationship with the forum. An injury may happen in one place, but the pain and inconvenience resulting from it might be felt in another country and later in a third one. As a result, presumptive effect cannot be accorded to this connecting factor.”[95] England: the Brownlie decisions 89. It is not necessary for present purposes to do more than mention the authorities prior to the two decisions in the Brownlie litigation, because the judgments in those cases are exhaustive and supersede the previous decisions. In a series of decisions involving personal injuries suffered abroad, which were approved by the majority in both Brownlie decisions, it was held that the English court had jurisdiction where consequential loss was suffered in England.[96] There are also decisions involving economic torts,[97] including some which draw a distinction between direct and indirect damage, in the light of rulings of the European Court, which, as will be seen below, are not a secure basis for decision in the light of the Brownlie cases. 90. In Brownlie v Four Seasons Holdings Inc[98](“Brownlie I”)Lady Brownlie, the widow of Professor Sir Ian Brownlie QC, sued the defendant Canadian company in contract and tort for damages following the death of her husband and her own injury resulting from a car crash whilst on an excursion during a holiday in Egypt. The claim in tort was brought in her own capacity for damages for pain and suffering and loss of amenity and medical expenses; and in her capacity as the executrix of her husband’s will for damages suffered by her husband (under the Law Reform (Miscellaneous Provisions) Act 1934) and for damages for bereavement and loss of dependency under the Fatal Accidents Act 1976. In Brownlie I the Supreme Court held that there was no reasonable prospect that the claimant would be able to establish at trial that she had contracted with the defendant corporate entity or that it would be vicariously liable for the negligence of the driver of the car. A majority (Lady Hale, Lord Wilson, Lord Clarke; Lord Sumption, dissenting, with whom Lord Hughes agreed) concluded obiter that the meaning of the word “damage” in Civil Procedure Rules, Practice Direction 6B, para 3.1(9) refers to the actionable harm caused by the wrongful act alleged and includes all the detriment, physical, financial and social which the claimant suffers as a result. 91. In Brownlie v FS Cairo (Nile Plaza) LLC[99](“Brownlie II”)Lady Brownlie sued a separate corporate entity in the same group of companies. In a differently constituted panel, the majority (Lord Lloyd-Jones, with whom Lords Reed, Briggs and Burrows agreed, with Lord Leggatt dissenting) held that “damage” refers to actionable harm, and is not limited to the damage necessary to complete a cause of action. It was sufficient that some significant damage has been sustained in the jurisdiction, whether it be physical damage or economic loss. 92. The judgments in Brownlie I and Brownlie II are lengthy, but in essence the view of the majority in both decisions is founded upon what it regarded as the natural and ordinary meaning of the word “damage” in the context of the gateway, and its purpose. In Brownlie I, for Lady Hale the natural and ordinary meaning “when considered in the light of the purpose of the provision extends to the physical and financial damage caused by the wrongdoing, considerations which are apt to link a tort to the jurisdiction where such damage is suffered.”[100] Precisely the same formula was adopted by Lord Lloyd-Jones in Brownlie II,[101] and he also added later in his judgment that in the gateway “in a case of personal injury or wrongful death ‘damage’ ... extends, both in its natural and ordinary meaning and on a purposive reading, to the actionable harm caused by the tortious act, including all the bodily and consequential financial effects which the claimant suffers.”[102] The long line of first instance decisions in England adopting this approach was expressly approved by both Lady Hale and Lord Lloyd-Jones.[103] Support for the view of the natural and ordinary meaning of the gateway was drawn by both Lady Hale and Lord Lloyd-Jones from the Canadian and Australian authorities, and also by Lord Lloyd-Jones from the decisions in Hong Kong of the Court of Appeal in Dynasty Line Ltd v Sukamto Sia and of Marlene Ng J at first instance in this case.[104] 93. Alternative interpretations were rejected by Lady Hale and Lord Lloyd-Jones. First, even though the gateway had been changed to accommodate the decision of the European Court in the Bier case, there was no basis for adopting the distinction later drawn by the European Court between direct damage and indirect damage. There was no reason to think that those who framed the rule intended it precisely to mirror the interpretation later given to the tort jurisdiction head in the Brussels regime.[105] Nor was there any basis for an assumption that the domestic gateway was thereafter to be tied to European law as that was developed by later decisions of the European Court, the most significant of which were made some years after the amendment of the domestic rule.[106] In any event, the two schemes were different: the European scheme deliberately eschews any discretion in favour of clear and certain rules, in the context of a scheme which governs, not only jurisdiction, but also recognition and enforcement of the resulting judgments, by contrast with the discretionary element in the English scheme.[107] 94. Second, the word “damage” in the gateway could not be equated with the damage which completed the cause of action, since damage was not an essential part of every cause of action in tort, such as trespass to the person, trespass to goods, libel and some slanders; and there was no reason to think that completion of the cause of action is what the framers of the Rules had in mind when they used the word “damage.”[108] 95. To the extent that the exercise of such a potentially wide jurisdiction might encourage forum-shopping, or allow a claimant to choose the forum, or permit claims founded on only a tenuous amount of damage sustained in England, those concerns were met by requirement that England be an appropriate forum under the discretionary test, which was “robust enough”[109] or “sufficiently muscular.”[110] Lord Lloyd-Jones said:[111] “The discretionary test of forum non conveniens, well established in our law, is an appropriate and effective mechanism which can be trusted to prevent the acceptance of jurisdiction in situations where there is merely a casual or adventitious link between the claim and England. Where a claim passes through a qualifying gateway, there remains a burden on the claimant to persuade the court that England and Wales is the proper place in which to bring the claim. Unless that is established, permission to serve out of the jurisdiction will be refused ... In addition—and this is a point to which I attach particular importance—the forum non conveniens principle is not a mere general discretion, the application of which may vary according to the differing subjective views of different judges creating a danger of legal uncertainty. On the contrary, the principle applies a structured discretion, the details of which have been refined in the decided cases, in a readily predictable manner.” 96. The dissenting judgments in Brownlie I of Lord Sumption and in Brownlie II of Lord Leggatt concluded that “damage” meant direct damage, although their reasoning differed. 97. Lord Sumption’s reasoning can be summarised as follows: (1) The law of tort did not generally protect pecuniary interests as such, and was in general concerned with non-pecuniary interests, such as bodily integrity, physical property and reputation which were inherently entitled to its protection. Although “damage” as that word is used in the gateway was not necessarily limited to the damage which serves to complete a cause of action in tort, the two concepts were related, in the sense that the damage to the interest protected was sustained in country A where the claimant had been injured or killed, and the pecuniary measure of that damage might depend on things that happened elsewhere. It would have been possible for the draftsman of the gateway to provide that “damage” should extend to the financial or physical consequences of the damage, but there was nothing in the language to suggest that it was done. (2) The jurisdictional gateways were concerned to identify some substantial and not merely casual or adventitious link between the cause of action and the forum, and that purpose was better served by locating jurisdiction in the place where the relevant interest of the claimant was damaged than by asking where he or she experienced the effects of the damage. (3) In the context of personal injury, a principle which located damage in the place where the pecuniary consequences of the accident were felt or where any continuing pain, suffering or loss of amenity were experienced would in the great majority of cases confer jurisdiction on the country of the claimant's residence, which would amount to a universal jurisdiction to entertain claims by English residents for the more serious personal injuries suffered anywhere in the world. (4) The distinction between direct and indirect damage drawn by the European Court embodied an analysis of what constitutes “damage” which was unanswerable, and the jurisdictional gateway in tort was deliberately drafted so as to assimilate the tests for asserting jurisdiction over persons domiciled in an EU member state and persons domiciled elsewhere. (5) The gateways identified relevant connections with the jurisdiction, and their ambit was a question of law, by contrast with the discretion as to forum conveniens which authorised the court to decline a jurisdiction which it possesses as a matter of law. (6) The main determining factor in the exercise of the discretion on forum conveniens grounds was not the relationship between the cause of action with the jurisdiction but the practicalities of litigation, and English law did not accept jurisdiction simply on the basis that the English courts were a convenient or appropriate forum if the subject matter had no relevant jurisdictional connection with England. 98. Lord Leggatt did not agree with Lord Sumption that any inference could properly be drawn that the gateway was intended to have exactly the same scope as the tort head in the Brussels regime, although the Brussels regime did show that it was possible to distinguish between direct and indirect damage. Nor did he consider that there was any reason to equate the meaning of the “damage … sustained” referred to in the gateway with damage which completed a cause of action in tort. His reasons for dissent were these: (1) The territorial nature of jurisdiction demands that there should be a substantial connection between the territory of the State from which the court's authority derives and either the proposed defendant or something which that person has done before the assertion of personal jurisdiction over the proposed defendant is justified. (2) The provision in the Civil Procedure Rules (CPR r. 6.37(3)), that England is “the proper place in which to bring the claim” was not an adequate answer, since the purpose of having grounds for service out of the jurisdiction which must apply before the question whether England is the proper place in which to bring the claim arises is to make the existence of a sufficient connection with the jurisdiction a prerequisite to permitting a person outside the jurisdiction to be sued in England. (3) A test met by the occurrence in England of any harm resulting indirectly from a wrong committed elsewhere fails to achieve this purpose, and provides an irrational and ineffectual threshold requirement. (4) The question whether England is “the proper place in which to bring the claim” is intended to codify the forum conveniens doctrine, which applies where there is a sufficient basis for founding jurisdiction in England but there is also some other available forum, having competent jurisdiction, where the case could be tried. (5) The forum conveniens factors are not relevant to whether there is a sufficient connection between the defendant and England to make it legitimate for the English court to assume jurisdiction in a case where the defendant has not submitted to the jurisdiction of the English courts. (6) Different judges assessing whether England is the appropriate forum will inevitably attach different degrees of weight to different factors and may reach differing conclusions on similar facts without either conclusion being susceptible to legal challenge. Such inconsistency of outcome would be a source of injustice, and encourage satellite litigation causing defendants who have no real connection with England to have to incur the difficulty and expense of instructing English lawyers to apply to contest the jurisdiction of the English courts. Conclusions 99. This case is concerned, like Brownlie I and Brownlie II, with claims arising out of personal injuries. In such cases it is normally straightforward, as in this case, to identify the place where the tort was committed, and the place or places where the plaintiff suffered injuries of a physical or financial character. The process may be more difficult in other cases, such as professional negligence or conspiracy to defraud, and no doubt the principles will require to be refined in the future to encompass such cases. 100. In my judgment Marlene Ng J and the Court of Appeal were right to find that Gateway F applied to this case. My reasons are these. 101. There can be no doubt that the question is one of construction of Gateway F, and in this case the question is what is meant by “the damage” which must be “sustained ... within the jurisdiction.” 102. There are two preliminary matters in this exercise. The first is whether there is any distinction between “the damage” in Gateway F and “damage” in the equivalent English rule. As indicated above, the current English rule was introduced to give effect to the decision of the English Court of Appeal in the Metall und Rohstoff AG case that it was not necessary that all the damage should have been sustained within the jurisdiction, provided that damage has resulted from substantial and efficacious acts committed within the jurisdiction (whether or not other substantial and efficacious acts have been committed elsewhere). There is no reason why the decision in Metall und Rohstoff AG should not be followed in Hong Kong. 103. The second matter is whether the rulings of the European Court on the tort head of jurisdiction in the Brussels Convention and the Brussels I Regulations have any relevance. In Brownlie I and Brownlie II it was held that, although the place of damage was introduced to bring the English head of jurisdiction broadly into line with the Brussels Convention, the framers did not intend the gateway to be interpreted in accordance with European law and the decisions of the European Court. It would, of course, be absurd to suggest that the similar change in the Hong Kong rule in 1988 was intended to incorporate the European law principles and to treat Hong Kong as if the Brussels Convention (and the decisions of the European Court on the Convention and its successors) applied to it. The most which can be said about the European Court decisions is, as Lord Leggatt said in Brownlie II, which the Board adopts in its argument, that there is a well-recognised and workable distinction between direct and indirect damage in its jurisprudence. That would only arise in this case if that distinction fell to be made. 104. In my judgment the construction of the relevant part of Gateway F is clear, and the reasoning in the majority judgments in Brownlie I and Brownlie II is compelling. 105. The purpose of the gateways is to set out a list of the situations in which the legislator considers that there may be a sufficient link with Hong Kong to justify a defendant in another country or law district being sued in Hong Kong. In some cases the link may be tenuous. Four examples will suffice. First, the clearest example is Gateway C, which allows service abroad on a person who is a “necessary or proper party” to proceedings in Hong Kong, but this head of jurisdiction is not founded upon any territorial connection between the claim, the subject matter of the relevant action and the jurisdiction of the court: Tyne Improvement Commissioners v Armement Anversois SA (The Brabo);[112] Altimo Holdings and Investment Ltd v Kyrgyz Mobil Tel Ltd.[113] As Sir Francis Piggott, the Chief Justice of Hong Kong, said more than 100 years ago, the necessary or proper party head “throws the net of jurisdiction over a wider area; and the principle of considering the nature of the cause of action which pervades the whole subject, appears here to be ignored”: Piggott, Foreign Judgments and Jurisdiction, vol. 3, p. 238 (1910). Second, Gateway A permits service out of the jurisdiction on a defendant domiciled within the jurisdiction. Because of the common law notion of domicile of origin, such a person may have no real connection with Hong Kong. Third, Gateway D allows service out of the jurisdiction where a contract was made in Hong Kong, but the parties and the transaction may have nothing to do with Hong Kong. Fourth, Gateway M allows service out of the jurisdiction where the claim is brought to enforce any foreign judgment or arbitral award, and that jurisdiction is exercisable over foreigners even where the judgment debtor or award debtor has no assets within the jurisdiction: Tassarruf Mevduati Sigorta Fonu v Demirel.[114] 106. One of the results of the discretionary nature of the jurisdiction under Order 11, r. 1(1) is to mitigate the effect of such situations. It will be necessary to return to this aspect of the jurisdiction. 107. In my view, in the light of the legislative purpose, the natural and ordinary meaning of the word “damage” is just that, and the rule does not distinguish between the damage which completes a cause of action and that which does not, nor does it distinguish between direct or indirect damage, or between physical or financial damage. 108. The question is whether there is a legislative purpose, or a public policy, or an absurd or undesirable result, which justifies a narrower construction, to encompass only direct damage as opposed to indirect damage. 109. I have already indicated why Order 11, r. 1(1) does not reveal a legislative purpose that every gateway must be construed to require a real connection with Hong Kong. The Board has adopted the reasoning of Lord Leggatt on this appeal, and in my view there are three flaws in the reasoning. 110. The first flaw is that the heads of jurisdiction provide a necessary connection with the forum, and should be construed in accordance to conform with that connection. I have endeavoured to demonstrate that, although in some respects the heads of jurisdiction require a real connection with the forum, in others they do not. 111. The second flaw is the assumption (shared by Lord Sumption in Brownlie I)that the question of discretion is entirely separate from that of jurisdiction. The third flaw is the assumption that the exercise of discretion is exclusively or mainly related to forum conveniens. These points are inter-connected. 112. I agree with Lord Lloyd-Jones[115] that the jurisdictional gateways alone do not confer jurisdiction; they form only one element of the jurisdictional test, and before the court will give permission to serve proceedings out of the jurisdiction, not only must a claim pass through one of the gateways, but it must also be shown that England is the proper place in which to bring the claim. That is why from the earliest days of the operation of Order 11 it was emphasised that the “court ought to be exceedingly careful before it allows a Writ to be served out of the jurisdiction” (Société Générale de Paris v Dreyfus Bros[116]), and that the jurisdiction must be exercised “with care and forbearance” (Massey v Heynes[117]) or with “discrimination and with scrupulous fairness to the defendant” (Ocean Steamship Co Ltd v Queensland State Wheat Board[118]). This remains the guiding principle: Seaconsar Far East Ltd v Bank Markazi Jomhouri Islami Iran.[119] 113. It has been said that the jurisdiction in which a tort has been committed is prima facie the natural forum for the determination of a dispute: The Albaforth;[120] Berezovsky v Michaels.[121] But this is only a useful rule of thumb or a prima facie starting point: VTB Capital plc v Nutritek International Corp.[122] Although the burden is on the plaintiff to show that the action is brought in a forum conveniens, justice often requires a homeward trend. In the present case the forum conveniens discretion was deployed in favour of Hong Kong. Even though the accident happened on the Mainland and the witnesses were based there, Hong Kong was held to be an appropriate forum because damage had been sustained in Hong Kong; evidence could be given by video link or by taking evidence where the witnesses resided; expert evidence on Chinese law could be heard in Hong Kong; and the plaintiff had the benefit of legal aid in Hong Kong. If the Board had not intervened, these issues would not have arisen in practice because the second defendant allowed judgment to go by default, perhaps because it (or its insurers) had been advised that a Hong Kong default judgment arising out of an accident on the Mainland would not have been enforceable in the United States. 114. The third flaw is the assumption that the discretionary exercise is limited to considerations of the forum conveniens, which “requires the ... court to decide whether its jurisdiction or that of the suggested foreign court is the more suitable as a forum for the determination of the dispute between the parties. The traditional way in which this question has been framed speaks of the ‘forum in which the case can be suitably tried for the interests of all the parties and for the ends of justice’ (per Lord Collins JSC in Altimo Holdings and Investment Ltd v Kyrgyz Mobil Tel Ltd [2012] 1 WLR 1804, para 88, adopting the language of Lord Goff of Chieveley in Spiliada Maritime Corpn v Cansulex Ltd [1987] AC 460).”[123] 115. Order 11, r. 1(1) enumerates what are described as the principal cases in which service of a writ out of Hong Kong is permissible, and provides that service of the writ “is permissible with the leave of the Court” in those cases. Order 11, r. 4(1) provides that the application for grant of leave must be supported by an affidavit, and Order 11, r. 4(2) provides: “No such leave shall be granted unless it shall be made sufficiently to appear to the Court that the case is a proper one for service out of the jurisdiction under this Order.” 116. As indicated above, this is the same as the former RSC Order 11, r. 4(2), which has its origin in the 1883 Rules of Court. In England it has been replaced by: “The court will not give permission unless satisfied that England and Wales is the proper place in which to bring the claim” (CPR r. 6.37(3)). 117. It is well established that, in deciding whether a case falls within one of the heads of jurisdiction, the court must consider the substance of the matter and not merely whether the case technically falls within the letter of the head of jurisdiction in question: the case must be clearly within both the letter of the rule and the spirit of the head: Johnson v Taylor Bros & Co Ltd;[124] Rosler v Hilbery;[125]GAF Corp v Amchem Products Inc;[126]Metall und Rohstoff AG v Donaldson Lufkin & Jenrette Inc.[127] The authorities show, contrary to a recent dictum by the English Court of Appeal[128] in a case in which the authorities do not seem to have been cited,[129] that this principle applies to the question whether the claim falls within the spirit of a gateway, and not simply to the exercise of the discretion. Nor does it apply exclusively to the construction of the heads of jurisdiction, although it has sometimes been used in this sense, for example by Lord Mustill in Mercedes-Benz v Leiduck,[130] when discussing whether a Mareva injunction was within the injunction gateway. The wider, and principled, view is demonstrated by Metall und Rohstoff AG v Donaldson Lufkin & Jenrette Inc,[131]in which the English Court of Appeal said that “it would certainly contravene the spirit, and also we think the letter, of the rule if jurisdiction were assumed on the strength of some relatively minor or insignificant act having been committed here, perhaps fortuitously.” 118. The principle, not mentioned in Brownlie I or Brownlie II, that a claim must fall within the spirit of the rule provides the answer to the argument that an expansive interpretation of the damages aspect of the tort gateway might lead to a situation in which, as Lord Leggatt put it,[132] “a claimant can create a link with England which satisfies the gateway requirement for suing a foreign defendant in the English courts by travelling to England, for example for medical treatment, after the event giving rise to the damage has occurred.” If the plaintiff has no real connection with the forum, then the court may refuse leave, or set it aside, on the basis that the claim is not within the spirit of the rule, irrespective of whether the claim satisfies the discretionary forum conveniens factors (although in such a case they would normally not be satisfied). 119. No reliance was placed on this line of authorities in Brownlie I and Brownlie II because it may have been assumed that those authorities are inapplicable or obsolete following the replacement of Order 11, r. 4 by the requirement in CPR r. 6.37(3) to show that England is “the proper place in which to bring the claim” with the result that the discretion is concerned exclusively, or mainly, with forum conveniens. That assumption is implicit in Lungowe v Vedanta Resources plc,[133] in which “proper place” was treated simply as a translation of forum conveniens,and underpins the judgments in Brownlie I[134]and Brownlie II.[135] The contrary view (which I favour) would be that the replacement should be interpreted simply as part of the exercise in the Civil Procedure Rules to use less technical language than the Rules of the Supreme Court, and that it applies equally to the question whether a claim is within the spirit of the relevant head of jurisdiction, irrespective of forum conveniens factors. 120. But in any event the former provision remains part of Hong Kong law, and I have no doubt that it could be employed in Hong Kong, in addition to forum conveniens factors, to mitigate the excesses, such as the supposed portability of a claim to the plaintiff’s domicile or residence (or Lord Leggatt’s example of the Egyptian plaintiff who comes to England while still suffering pain or disability), to which an expansive interpretation of Gateway F is said to lead. 121. In summary, the natural and ordinary meaning of Gateway F leads to the conclusion that it includes indirect or consequential damage, and to the extent that construction leads in a particular case to a result where Hong Kong is not the forum conveniens or the case is outside the spirit of the rule, the Hong Kong court can, and should, exercise its discretion to refuse, or set aside, leave to serve out of the jurisdiction. For those reasons I would dismiss the appeal on the second issue. Mr Justice Ribeiro PJ: 122. The appeal is unanimously dismissed. We direct that the parties be at liberty to lodge written submissions, in the light of the aforesaid judgments, as to costs both here and in the Courts below. Such submissions should be lodged within 14 days of the publication of this judgment and any submissions in reply within 14 days thereafter. Mr Horace Wong SC leading Mr Clark Wang and Mr Adrian TY Wong, instructed by Gallant, for the 3rd Defendant (Appellant) Ms Audrey Eu SC leading Mr Anson Wong Yu Yat, instructed by Kenneth Lam, Solicitors, assigned by the Director of Legal Aid, and Mr Jason Kung (not assigned by the Director of Legal Aid), instructed by Kenneth Lam, Solicitors, for the Plaintiff (Respondent) [1] Cap 4A. [2] 		[2020] HKCFI 679. [3] Cheung and Yuen JJA [2021] HKCA 1138, Cheung JA writing for the Court. [4] Ribeiro, Fok and Lam PJJ [2022] HKCFA 4. [5] J§2. [6] Listed in CA§49. [7] J§3. [8] Pursuant to O 6 rr 6 and 7. [9] J§7. [10] Under O 11 r 1(1)(c). [11] Under O 11 r 1(1)(d)(i). [12] Under O 11 r 1(1)(d)(ii). [13] Under O 11 r 1(1)(d)(iii). [14] J§27. [15] J§8. [16] J§5. [17] J§12. D1 made “ex gratia” settlement payments totalling about $575,000. [18] J§32. [19] J§34. [20] J§36. [21] CA§§8-9. [22] CA§§55-59. [23] CA§§63-64. [24] CA§34. [25] In Section G of this judgment. [26] CA§§66-72. [27] CA§65. [28] Cap 365, section 16. In this judgment, the discussion focusses on the claim for a statutory relief payment. [29] Cap 365, section 2. [30] Cap 365, sections 3 and 20B; Cap 282, section 3. [31] Cap 365, section 25A. [32] Part IV of the Ordinance which includes section 20A and the other provisions for claiming relief for unpaid damages. [33] Cap 411, sections 4, 6, 7(1), Schedule 2. [34] Cap 411, sections 14 and 15. [35] Cap 282, section 5. [36] Cap 282, section 26. [37] Cap 282, section 40(1). [38] The liability in question is liability “under this Ordinance and independently of this Ordinance for any injury to his employee by accident arising out of and in the course of the employee’s employment”: Cap 282, section 40(1F). [39] Legislative Council Brief, EMB CR 4/4/3231/77, February 2002 §10. The package included government loans and a net increase in the levy attributable to the Fund; as well as a reduction in and restructuring of the Fund’s liability, replacing payment of common law damages claims with relief payments reducing the recoverable amounts and providing for instalment payments: §§11-16. The Board’s power to intervene under Cap 365, section 25A was also introduced as part of those measures. The financial difficulties and the amending legislation are discussed in detail in Kwan Kam Pui v Fung Man [2014] 6 HKC 361 at §20. [40] Such causes included escalating claims for common law damages, declining levy income and the insolvency of certain participating insurers resulting from the major insolvency of their Australian parent company: Legco Brief §§4, 7-9. [41] J§193. See also J§§133(a) and 143. [42] CA§63(4) and (6). [43] With Mr Clark Wang and Mr Adrian TY Wong. [44] With Mr Anson Wong Yu Yat and Mr Jason Kung. [45] Hartmann & Chu JJA and Lam J (as he then was), CACV 71/2011 (25 July 2012). [46] Respondent’s Written Case §§72-75. Counsel also relied on the Chinese text “聘用” which she submits “has the connotation of ‘hiring’”, but that does not advance her argument. [47] Supra. [48] Discussed in Section E above. It states: “If an employer who is a person carrying on business outside Hong Kong submits or has agreed to submit to the jurisdiction of the Court, then, notwithstanding that the accident causing the personal injury occurred outside Hong Kong, this Ordinance shall apply to employees within the meaning of this Ordinance who have been recruited or engaged in Hong Kong.” [49] Chan Sze Yuen (supra) at §13. [50] Ibid at §9. [51] Ibid at §19. [52] Ibid at §16. [53] Bennion, Bailey and Norbury on Statutory Interpretation (8th Ed) at p 651. [54] (2017) 20 HKCFAR 524, per Fok PJ at §32. [55] See HKSAR v Kwan Ka Hei (2020) 23 HKCFAR 229, per Cheung PJ (as he then was) at §§47-49. [56] [1946] 1 KB 58 at 62. [57] J§§21-22, 25-26, 27(b) and 122-129. [58] J§125. [59] J§§127-129. [60] CA§63(7) and (8). [61] J§131. [62] CA§§63-64. [63] (2002) 5 HKCFAR 356 at §§38-39. [64] O 15 r 6 (2)(b)(ii) materially states: Subject to the provision of this rule, at any stage of the proceedings in any cause or matter the Court may on such terms as it thinks just and either of its own motion or on application— (b) order any of the following persons to be added as a party, namely— (ii) any person between whom and any party to the cause or matter there may exist a question or issue arising out of or relating to or connected with any relief or remedy claimed in the cause or matter which in the opinion of the Court it would be just and convenient to determine as between him and that party as well as between the parties to the cause or matter. [65] Marlene Ng J held that Gateway C (necessary or proper party) and Gateway D (iii) (breach of contract governed by Hong Kong law) applied, but Gateways D (i) and (ii) (breach of contract within the jurisdiction or made by an agent within the jurisdiction) were inapplicable. The Court of Appeal held that neither Gateway C nor Gateway D (iii) was applicable. This Court was not asked to grant leave to appeal in relation to those gateways. [66] The judge decided that the plaintiff was not barred from relying on the damage limb of Gateway F notwithstanding that it had not been raised at the ex parte stage. [67] At [200]. [68] [2017] UKSC 80, [2018] 1 W.L.R. 192. [69] [2020] EWCA Civ 996, [2021] 2 All E.R. 605, affd. [2021] UKSC 45, [2021] 3 W.L.R. 1011. [70] [2009] 4 HKLRD 454. [71] [1971] A.C. 458. [72] (1881) 7 Q.B.D. 434 (CA). [73] (1889) 12 Q.B.D. 50 (Div Ct). [74] [1893] 1 Q.B. 419 (Div Ct). [75] [1895] 1 Q.B. 142 (CA). [76] [1944] K.B. 432 (CA). [77] At 441. [78] [1983] Ch. 258 (CA). [79] [1984] 2 Lloyd’s Rep. 91 (CA). [80] Case 21/76 [1976] E.C.R. 1735, [1978] Q.B. 708. [81] Case C-220/88 [1990] E.C.R. I-49. [82] Case C-364/93 [1995] E.C.R. I-2719, [1996] Q.B. 217. [83] Case C-12/15 [2016] Q.B. 967. [84] Case C-375/13 [2016] 1 All E.R. (Comm.) 733. [85] The words omitted deal with threatened torts, which are not expressly covered by Gateway F in Hong Kong. [86] SI 2000/221. [87] [1990] 1 Q.B. 391 (CA). [88] SI 2000/940. [89] [2009] 4 HKLRD 454. For Singapore, where the rule is for cases where “the claim is wholly or partly founded on, or is for the recovery of damages in respect of, damage suffered in Singapore caused by a tortious act or omission whenever occurring” see MAN Diesel & Turbo SE v IM Skaugen SE [2019] SGCA 80, [77]. [90] At [33]-[34]. [91] [1983] 2 NSWLR 405, 408. See also Brix-Neilsen v Oceaneering Australia Pty Ltd [1982] 2 NSWLR 173. [92] (1985) 63 A.L.R. 466, 482 (NSW Court of Appeal), affd. on other grounds: [1987] HCA 17, (1987) 162 C.L.R. 574. See also Voth v Manildra Flour Mills Pty Ltd [1990] HCA 55, (1990) 171 C.L.R. 538, [3]. [93] Vile v Von Wendt (1979) 103 D.L.R. (3d) 356; See also Poirier v Williston (1980) 113 D.L.R. (3d) 252, affd. (1980) 118 D.L.R. (3d) 576n; Skyrotors Ltd v Carriere Technical Industries Ltd (1979) 102 D.L.R. (3d) 323 (aircraft crash in Quebec; owner suffered financial loss in Ontario). [94] [2012] 1 S.C.R. 572. [95] At [89], applied most recently in Moore v. Kemgni, 2022 ONSC 2141; Thind v. Polycon Industries, 2022 ONSC 2322. [96] Booth v Phillips [2004] EWHC 1437 (Comm.), [2004] 1 W.L.R. 3292; Cooley v Ramsey [2008] EWHC 129 (Q.B.); Harty v Sabre International Security Ltd [2011] EWHC 852 (Q.B.); Wink v Croatia Osiguranje DD [2013] EWHC 1118 (Q.B.); Pike v Indian Hotels [2013] EWHC 4096 (Q.B.); Stylianou v Toyoshima [2013] EWHC 2188 (Q.B.). [97] Société Commerciale de Réassurance v Eras International Ltd (Eras EIL Actions) [1992] 1 Lloyd’s Rep. 570, 590-591 (CA); Bastone & Firminger Ltd v Nasima Enterprises (Nigeria) Ltd [1996] C.L.C. 1902, 1912; ABCI v Banque Franco-Tunisienne [2003] EWCA Civ 205, [2003] 2 Lloyd’s Rep. 146, [44]-[45]; Erste Group Bank AG v JSC ‘VMZ Red October’ [2015] EWCA Civ 379, [2015] 1 C.L.C. 706, [106]-[107]; Eurasia Sports Ltd v Tsai [2018] EWCA Civ 1742, [2018] 1 W.L.R. 6089, [21], [30]. [98] [2017] UKSC 80, [2018] 1 W.L.R. 192. [99] [2021] UKSC 45, [2021] 3 W.L.R. 1011, affirming [2020] EWCA Civ 996, [2021] 2 All E.R. 605. [100] Brownlie I, [55], [69]. [101] Brownlie II, [51]. [102] Brownlie II, [76]. [103] Brownlie I, [48]; Brownlie II, [64]. [104] Brownlie II, [66]-[67]. [105] 		 Brownlie I, [50]. [106] Brownlie II, [54]. [107] Brownlie II, [55]. [108] Brownlie I, [52]; Brownlie II, [81]. [109] Brownlie I, [54] (Lady Hale). [110] Brownlie I, [66] (Lord Wilson). [111] Brownlie II, [79]. [112] [1949] A.C. 326, 338. [113] [2011] UKPC 7, [2012] 1 W.L.R. 1804, [73]. [114] [2006] EWHC 3354 (Ch.), [2007] 2 All E.R. 815, affd. [2007] EWCA Civ 799, [2007] 1 W.L.R. 2508. [115] Brownlie II, [77]. [116] (1885) 29 Ch.D. 239, 243. [117] (1888) 21 Q.B.D. 330, 334 (CA). [118] [1941] 1 K.B. 402, 417 (CA). [119] [1994] 1 A.C. 438, 455. [120] [1984] 2 Lloyd’s Rep. 91 (CA). [121] [2000] 1 W.L.R. 1004, 1013 (HL). [122] [2013] UKSC 5, [2013] 2 A.C. 337, [18]. [123] Unwired Planet International Ltd v Huawei Technologies (UK) Ltd[2020] UKSC 37, [2020] Bus. L.R. 2422, [94]. [124] [1920] A.C. 144, 153. [125] [1925] Ch. 250, 259-260 (CA). [126] [1975] 1 Lloyd’s Rep 601 (Megarry J. and CA), 605, per Megarry J. [127] [1990] 1 Q.B. 391, 434 (CA). [128] Sharab v Al-Saud [2009] EWCA Civ 353, [2009] 2 Lloyd’s Rep. 160, [35]. [129] See Conductive Inkjet Technology Ltd v Uni-Pixel Displays Inc [2013] EWHC 2968 (Ch.), [2014] F.S.R. 475, [59]. [130] [1996] A.C. 284, 299. See also Convoy Collateral Ltd v Broad Idea International Ltd [2021] UKPC 24, [2022] 2 W.L.R. 703, [167]. [131] [1990] 1 Q.B. 391, 437 (CA). [132] Brownlie II, [194]. [133] [2019] UKSC 20, [2020] A.C. 1045, [38], [66], [69]. [134] At [3], [40]. [135] At [28], [55], [196]. 
Mr Justice Ribeiro PJ: 1. I agree that this appeal must be allowed and agree with the answers given to the questions in respect of which leave to appeal was granted. There is no doubt that the Court had jurisdiction over the respondent, a Hong Kong permanent resident who was duly served and acknowledged service within the jurisdiction. It was bound to assume jurisdiction over the respondent unless it was persuaded that the proceedings should be stayed, on forum non conveniens principles. Having assumed jurisdiction, it would be bound to consider whether the orders sought should be made, applying substantive discretionary principles relevant to the exercise of powers under the Ordinance. It was, with respect, unnecessary and confusing to stray into considerations relevant to custody issues and the inherent jurisdiction. There was no basis for interfering with the trial judge's exercise of discretion in refusing a stay. Mr Justice Tang PJ: Introduction 2. The applicant, a Mainland resident, is the mother of G, a girl, who was born on 8 August 2007. Her father is the respondent, a Hong Kong permanent resident, who was at the time and remains married with a wife and two children in Hong Kong. G was born in Hong Kong so she is a Hong Kong permanent resident by birth. Shortly after she was born, the applicant and the respondent separated[1] and G was brought back to the Mainland by her mother on 3 October 2007.[2] G has lived in Guangzhou with her since she was two years old.[3] The respondent is employed by a substantial real estate company listed in Hong Kong.[4] 3. This appeal is concerned with s 10 of the Guardianship of Minors Ordinance, Cap 13 (“GMO”). Section 10 of the GMO was modeled on s 9 of the Guardianship of Minors Act 1971 (“1971 Act”) as amended by the Guardianship Act 1973 (“1973 Act”).[5] There is however an important difference. Under s 9 of the 1971 Act as amended, a maintenance order may only be made when the court makes an order regarding the custody of the minor. Under s 10(1) of the GMO: “The court may, on the application of either of the parents of a minor … , make such order regarding – (a) the custody of the minor; and (b) the right of access to the minor of either of his or her parents, …” Under s 10(2): “The court may as regards a minor, on the application of a person with whom, whether by virtue of an order under subsection (1) or otherwise, custody of the minor lies at law, make in respect of the minor any one or more of the following orders – (a) an order requiring payment to the applicant by the parent … of such lump sum [payment] … ; (b) an order requiring payment to the applicant by such parent … of such periodical sum towards the maintenance of the minor as the court thinks reasonable … ” Thus, under s 10 of the GMO, an order for maintenance may be made whether or not a custody order was applied for or made. Family Court 4. By an Originating Summons dated 12 July 2012, the applicant applied in the Family Court for maintenance in favour of G under s 10 of the GMO against the respondent. The respondent acknowledged service on 6 August 2012. On or about 6 September 2012, the respondent commenced proceedings at the People’s Court of Siming District in Xiamen City seeking both custody and maintenance for G against the applicant. According to the respondent, the proceedings were commenced there because that was where the applicant had “registered residency status”. According to the respondent, he did not know that the applicant and G were living in Guangzhou.[6] 5. On 24 September 2012, the father applied for an order that: “(a) The Originating Summons be dismissed on the ground that the Court has no jurisdiction; (b) Alternatively, all further proceedings in this action be stayed on the ground of forum non [conveniens] and that the People’s Court of Siming District in Xiamen City, Fujian Province, the People’s Republic of China (‘the Siming Court’)(福建省厦門市思明區人民法院) is the more appropriate forum; (c) Alternatively, all further proceedings in this action be stayed pending the outcome of the proceedings in Mainland China.” 6. On 27 May 2013, Deputy District Judge Ivan Wong, dismissed the father’s application. His Honour was of the view that the court had jurisdiction because G was born in Hong Kong and a Chinese citizen with a right of abode in Hong Kong, and her father was a Hong Kong permanent resident and domiciled in Hong Kong.[7] On forum non conveniens, his Honour was of the view that Siming Court was not a more appropriate forum,[8] and concluded that in any event he would exercise his discretion against a stay.[9] Court of Appeal 7. The respondent appealed. On 18 June 2015, the Court of Appeal[10] allowed his appeal and stayed the mother’s application. Cheung JA, whose judgment was agreed to by the other members of the court, said that he was prepared to proceed on the basis that the court had jurisdiction to make the maintenance order sought[11] but concluded “the Court below should have declined jurisdiction.”[12] Leave to Appeal 8. On 29 September 2016, the Appeal Committee[13] granted leave to appeal on the following questions of law: “(a) Does the Court have jurisdiction to entertain applications under the … (‘GMO’), in respect of a child who is neither ordinarily resident nor present in Hong Kong? (b) If such jurisdiction exists, does the Court have a discretion (whether by reference to the Court’s inherent jurisdiction or otherwise) to decline jurisdiction other than on forum non conveniens principles? (c) If it has such a discretion, what are the criteria to be considered in deciding whether jurisdiction should be declined? 9. Leave was also granted on the “or otherwise” ground on the basis that if the applicant succeeds in establishing jurisdiction under the GMO or inherent jurisdiction, the question of whether the judge’s dismissal of the respondent’s forum non conveniens stay application stands is a question which ought to be dealt with by the court for the effective disposal of the appeal. Discussion 10. In the section of his judgment under “inherent jurisdiction”, Cheung JA examined in some detail English decisions on the inherent jurisdiction of the court over wardship and guardianship of children. In Re P (GE)(An Infant)[1965] Ch 568, a decision of the English Court of Appeal, a mother applied to make her son, who had been taken by his father to Israel, a ward of court under s 9 of the Law Reform (Miscellaneous Provisions) Act 1949, and for custody under the Guardianship of Infants Acts, 1886 and 1925. The father who was served out of the jurisdiction, applied inter alia, for an order that the mother’s originating summons be set aside for lack of jurisdiction. In this context, Lord Denning MR said: “Those Acts give no guidance as to the geographical jurisdiction of the court; and we have to determine it on principle by reference to the inherent jurisdiction.”[14] 11. GMO contains no explicit limit on the court’s jurisdiction. Cheung JA pointed out that: “Unlike the Matrimonial Causes Ordinance (Cap. 179) which restricts the right to petition for divorce in Hong Kong to those who are domiciled here or ordinarily residing here or has a substantial connection here, there is no such restriction in the GMO.” [15] 12. As for inherent jurisdiction, Cheung JA concluded that the Court of First Instance (“CFI”) enjoys similar inherent jurisdiction.[16] However, in relation to the Family Court, Cheung JA said the better view is that “the Family Court lacks such inherent jurisdiction.”[17] That does not matter because the Family Court was concerned with jurisdiction under s 10 of the GMO and not inherent jurisdiction. Inherent jurisdiction was in any event only relevant because presumably his Lordship was of the view that the exercise of the s 10 of the GMO jurisdiction should be informed by the ambit of such inherent jurisdiction. 13. It is sufficient for present purposes to note that in England, “The court has jurisdiction (a) to appoint a guardian for, or to make a custody order in respect of, a minor where at the commencement of the proceedings the minor is a British subject or owes local allegiance to the Crown by virtue of his ordinary residence or presence in England; ”[18] 14. That is because as Pearson LJ explained: “It is clear from the authorities that the English court has, by delegation from the Sovereign, jurisdiction to make a wardship order whenever the Sovereign as parens patriae has a quasi-parental relationship towards the infant. The infant owes a duty of allegiance and has a corresponding right to protection and therefore may be made a ward of court: Hope v Hope.”[19] 15. In this context, I note the observations of Diplock LJ in Garthwaite v Garthwaite [1964] P 356 at 387-8 regarding the two senses in which the expression “jurisdiction” of a court may be used. His Lordship said: “In its narrow and strict sense, the ‘jurisdiction’ of a validly constituted court connotes the limits which are imposed upon its power to hear and determine issues between persons seeking to avail themselves of its process by reference (1) to the subject-matter of the issue or (2) to the persons between whom the issue is joined or (3) to the kind of relief sought, or to any combination of these factors. In its wider sense it embraces also the settled practice of the court as to the way in which it will exercise its power to hear and determine issues which fall within its ‘jurisdiction’ (in the strict sense) or as to the circumstances in which it will grant a particular kind of relief which it has ‘jurisdiction’ (in the strict sense) to grant, including its settled practice to refuse to exercise such powers, or to grant such relief in particular circumstances.” I believe Dicey and Morris[20] had in mind jurisdiction in the wider sense. 16. Obviously, after 1997, British nationality cannot be the basis for the exercise of any inherent jurisdiction in Hong Kong. Nor, for obvious reasons, Chinese nationality. Cheung JA appeared to be of the view that inherent jurisdiction might be based on the minor’s status as a Hong Kong permanent resident. He said: “8.10 … in appropriate situations which involve a child who is a permanent resident of Hong Kong, the Court would step in and exercise its traditional protective role over the child even if he or she is not physically present in Hong Kong at the time of application. An example given at the appeal is a Hong Kong family of Hong Kong permanent residents living in the Mainland; the parents died, leaving the child uncared for and living alone in the Mainland; the Hong Kong grandparent of the child then applies for his custody in Hong Kong. In such a situation the Hong Kong Court would be most reluctant to decline jurisdiction simply because the child is not present in Hong Kong.” 17. In re Willoughby (An Infant) (1885) 30 Ch D 324 was concerned with an infant girl who lived with her father in Paris and had no property in England. She was a British subject by descent. Her mother had lived separately from the father since 1877 and the father had obtained an order from the French Court prohibiting the mother from visiting the infant. The father died in Paris intestate in 1884. At the time of his death, the infant was under the care of a schoolmistress in Paris. The mother applied in the French Court for access to the infant, and to remove her from the school, and to appoint guardians. That application was opposed by the infant’s grand aunt. The French Court postponed dealing with the mother’s application until it was ascertained what course the English Courts would adopt in the English proceedings where the infant applied by her next friend for guardians to be appointed of her personal estate although the infant had no property in England. The English proceedings were opposed on behalf of the mother. It was admitted on behalf of the mother for the purpose of the argument that “if the child were here the Court would appoint guardians other than the mother, and would not allow the child to remain in the mother’s custody”.[21] Kay J held, the infant being a British subject the court had undoubted jurisdiction to make the order but as to whether the jurisdiction ought to be exercised,[22] he said at 329-330: “Therefore although there is no property in this country which I can lay hold of to enable me to enforce the order I may make, I think it right that the Court should in a case in which it is admitted that the mother is not a proper person to be the guardian, appoint English guardians of this English child; and I have no doubt whatever that when this Court has done that, the Courts of France will pay the respect and attention to the orders of the English Court concerning this English child which, as I have said, the English Courts would pay in the converse case with respect to the order of the French Courts as to a French child residing in this country.” 18. Kay J’s order was affirmed on appeal. Cotton LJ at 331-332 said: “… Of course it is only under extraordinary circumstances that the Court would make an order when the infant is not here, and when there is no property here, and when the persons who have the custody of the infant are not subject to the jurisdiction, as they would be if resident in this country.” 19. His Lordship went on to consider whether they ought to abstain from exercising their jurisdiction, because they cannot enforce the order and decided not to abstain saying: “It may well be that this order may assist the French Court. It may well be that the French Court may give effect – though I do not say they ought to give effect – to our order, and make it operative by giving to the persons whom the Court here thinks ought to be appointed guardians power in France to deal with the infant and the property there, or it may, if it thinks fit, appoint those persons guardians. It will be entirely a question for the French Courts.”[23] 20. Lindley LJ said: “… there is a reason for departing from the usual rule, which is against making an order which cannot be enforced. The case is very special and very unusual.”[24] 21. In England as noted, nationality and the physical presence of the minor provide the basis for jurisdiction. That is based on what one might regard as archaic notions of allegiance to and the corollary of protection by the Crown. This is not the occasion to consider how the jurisdiction might be developed in Hong Kong in relation to a minor who is entitled to permanent residence and whether protection of the parens patriae must be balanced by allegiance. In the case postulated by Cheung JA, given that the jurisdiction of the court is not expressly limited by s 10 of the GMO and since in In re Willoughby the court made the order even though it could not be enforced, it may be, though I do not decide, that in a proper case an infant’s status as a permanent resident may persuade a Hong Kong court to make a guardianship order under s 10. 22. However, Cheung JA was of the view that in this case the judge should have declined jurisdiction because: “8.18 My view is that the starting point is that the jurisdiction should be sparingly exercised where the child is not an ordinary resident or present here. … I agree with the view that the Court should be extremely circumspect in assuming any jurisdiction in relation to a child who is currently physically present in some other jurisdiction simply because of her residence status here. This is certainly not a case where the welfare of the child is at risk if the Court declines to exercise jurisdiction. Although any order made [by] the Hong Kong Court may be enforced in Hong Kong where the father lives, the Court faces unfamiliar matters relevant to the child such as her living, education and medical expenses, especially in respect of her intended emigration to Canada in the future. In addition to this, there is the policy consideration of comity of nations which in the present context means the separate systems of law between the Mainland and Hong Kong. The Court below should have declined jurisdiction.” 23. It may be that Cheung JA thought that a maintenance order against a Hong Kong resident in favour of a minor who is not ordinarily resident or physically here is against the settled practice of the court to make, and that is why his Lordship said the judge ought to have declined jurisdiction. If so, with respect, I do not agree. The authorities examined by Cheung JA where similar observations had been made were concerned with guardianship of minors or their properties where the presence within the jurisdiction of the minor or property was an important consideration. As Lord Cranworth LC said in Hope v Hope in relation to an infant who at the time the jurisdiction is asked is not within the jurisdiction of the court: “… It may be that the child is placed under such circumstances that the jurisdiction of the Court cannot be exercised over it because no order I might issue could be enforced; but in that case there is not a want of jurisdiction, but a want of the power of enforcing it.”[25] 24. But here we are not concerned with an application for the custody of G. We are solely concerned with maintenance. The father is a Hong Kong resident, has assets here and enforceability can be taken for granted. Nor is comity of nations a real concern because the order is sought against a Hong Kong permanent resident. 25. If, however, Cheung JA’s remarks were directed at the judge’s exercise of his discretion, with respect, I do not believe he was entitled to interfere. The judge had fully considered the comparative merits of proceedings in Hong Kong and in Siming albeit in the context of forum non conveniens. 26. As the judge correctly pointed out: “In DGC v SLC (nee C) [2005] 3 HKC 293, the Court of Appeal noted that the Family Courts in Hong Kong have vast experience in dealing with cases of an international dimension.”[26] Thus, the fact that the Family Court would have to determine the order to make for maintenance for a child in the Mainland should present no difficulty. There is of course, as the judge said, the advantage of enforcement of the order in Hong Kong.[27] Such advantage is not to be underestimated in family cases where emotion can sometimes be overpowering. 27. G was born an ex-nuptial child. Had her parents been married, the court would have had jurisdiction to make an order of maintenance against her father on her mother’s application if either parent had a substantial connection with Hong Kong.[28] Since the father has a substantial connection with Hong Kong, an order for maintenance could be made under the Matrimonial Proceedings and Property Ordinance, Cap 192 if her parents had been married. Because G was ex-nuptial, the application had to be made under s 10 of the GMO. That being the case, subject to forum non conveniens which will be considered below, I see no reason why the Court should decline to exercise its statutory jurisdiction under s 10 in a case where it would make an order in favour of a child of a married couple.[29] The Judge rightly held that he had jurisdiction to make a maintenance order under s 10 GMO and this is a case where the court might make an order. Forum non conveniens 28. The applicable principles are settled and the learned judge[30] applied the principles stated by Cheung JA in SA v SPH [2013] 2 HKC 130 at 135F-136C. Briefly stated, the question is whether there is some other available forum having competent jurisdiction in which the action may be tried more suitably for the interests of all the parties and the ends of justice. The applicant for the stay has to establish that first, Hong Kong is not the natural or appropriate forum in the sense that it has not the most real and substantial connection with the action and second that, there is another available forum which is clearly or distinctly more appropriate than Hong Kong. Failure by the applicant to establish these matters is fatal. If the applicant is able to establish both of these matters, then the plaintiff in the Hong Kong proceedings has to show that he will be deprived of a legitimate personal or juridical advantage if the action is tried in a forum other than Hong Kong. If the plaintiff is able to establish this, the court will have to balance the advantages of the alternative forum with the disadvantages that the plaintiff may suffer. 29. Following such principles, the learned judge dismissed the father’s application to stay the proceedings on the ground of forum non conveniens in favour of a Mainland court. 30. The Court of Appeal, having held that the judge should have declined jurisdiction, did not go on to consider the correctness of the learned judge’s decision on forum non conveniens. Nothing has been placed before us to make us doubt the correctness of the judge’s decision. Disposition 31. For the above reasons and those given by Mr Justice Fok PJ and Mr Justice Gleeson NPJ in their respective judgments, I would allow the appeal and restore the order of the Family Court. The certified questions 32. It is clear from the foregoing that the present appeal falls within a narrow compass. 33. The first question: Does the Court have jurisdiction to entertain applications under the GMO, in respect of a child who is neither ordinarily resident nor present in Hong Kong? The answer to the first question is yes. 34. The second question: If such jurisdiction exists, does the Court have a discretion (whether by reference to the Court’s inherent jurisdiction or otherwise) to decline jurisdiction other than on forum non conveniens principles? Here the court’s power to decline jurisdiction when faced with an application for a stay based on the ground of forum non conveniens must be distinguished from the court refusing to exercise its discretion to make or refuse to make any order in guardianship proceedings before it. In respect of the former, obviously the court should be guided by the forum non conveniens principles, though always bearing in mind in the applications of these principles that the interest of the minor is of paramount importance. My answer is no. 35. The third question: If it has such a discretion, what are the criteria to be considered in deciding whether jurisdiction should be declined? The exercise of the court’s discretion must depend on the facts of the individual case. Even if it is possible to do so, which I doubt, I think it would be unwise to confine the exercise of discretion by strict criteria. Mr Justice Fok PJ: 36. I have had the benefit of reading in draft the judgment of 37. The first question for which leave to appeal was granted raises a question of construction of section 10(2) of the Guardianship of Minors Ordinance (Cap.13) (“GMO”), specifically whether the fact that a minor in respect of whom an order for maintenance is sought is not ordinarily resident or present in Hong Kong is a bar to a court in Hong Kong making an order for maintenance under section 10(2). This question of construction arises in the context of an originating summons which on its face was issued, pursuant to section 10(2), on the application of the appellant, the mother of G (a minor) and the person with whom custody of G lay, seeking an order for maintenance of G against the respondent, as G’s father. 38. In my judgment, section 10(2) should not be so construed. Although there is a presumption against the extra-territorial application of legislation,[31] the invocation of the court’s jurisdiction in the present case was founded on the basis that the respondent, G’s father, was physically present within the jurisdiction when he was duly served with, and acknowledged service of, the originating summons in this action seeking an order for the payment of maintenance in respect of G. 39. Such service of the originating process in this action within the jurisdiction on the respondent was sufficient to found jurisdiction as of right. Whether the Hong Kong court would ultimately exercise jurisdiction to make the maintenance order would then depend on (i) whether the court were to accede to an application on the part of the respondent to stay proceedings on the ground of forum non conveniens, or (ii) assuming it did not stay the proceedings, whether the court were to decline to make an order for maintenance on the merits. The question of jurisdiction of the court, in the (narrow or strict) sense[32] of whether it had power to hear and determine the intended application for an order for maintenance, would not arise in this case at stage (i) because, as already noted, the respondent was served within the court’s territorial jurisdiction. Nor would it arise at stage (ii) if the court did not stay the proceedings in favour of a more appropriate jurisdiction. At that latter stage, the court would consider whether to exercise its jurisdiction (in the wider sense)[33] to make an order for maintenance in respect of the minor. This does not exclude the possibility of the summary dismissal of the application on grounds related to the merits (e.g. on the basis of DNA evidence proving the respondent is not the minor’s parent). 40. As a matter of construction of section 10(2), the fact that the minor in respect of whom a maintenance order is sought is not ordinarily resident or physically present within Hong Kong, either at the time of service of the originating process or even at the date of trial, would not preclude the Hong Kong court from making an order for maintenance. 41. There is nothing in the wording of section 10(2) that requires the minor in respect of whom the application is made to be ordinarily resident or physically present in Hong Kong. This construction is supported, as a matter of context, by section 26 of the GMO, which expressly provides that jurisdiction under the ordinance is exercisable notwithstanding that any party to the proceedings is not domiciled in Hong Kong. It is also supported, contextually, by the fact that, unlike other family law statutes, there is no express requirement of residence or presence in Hong Kong as a condition for the exercise of jurisdiction.[34] Moreover, as a matter of purpose, the GMO requires the court, as a matter of general principle in relation to proceedings thereunder before it, to “regard the best interests of the minor as the first and paramount consideration”.[35] 42. Thus, faced with the respondent’s application in the present case, the sole question for the Family Court was whether the application of the ordinary principles of forum non conveniens led to the conclusion that the action, commenced as of right by service of originating process within the jurisdiction, should be stayed rather than allowed to proceed on its merits. That filtering procedure would be an adequate manner in which the Hong Kong court could ensure that a claim for maintenance under section 10(2) of the GMO was properly tried before it, rather than more appropriately tried in the courts of some other jurisdiction. This would have met the concern raised in argument by the respondent that the jurisdiction of the Hong Kong courts would be too easily and inappropriately invoked in respect of a minor resident or present elsewhere, for example by service of process on a parent fleetingly in Hong Kong whilst in transit at the airport. 43. The Family Court having exercised its discretion to decline to stay the appellant’s originating summons on the ground of forum non conveniens, and there being no proper reason to interfere with that exercise of discretion, the Court of Appeal should have dismissed the respondent’s appeal below. 44. There was therefore no need in the present case for the Court of Appeal to consider the court’s inherent jurisdiction and, with respect, it was in error in doing so. 45. Nor was this a case in which the appellant, G’s mother, needed to have resort to “long arm” jurisdiction in order to found jurisdiction against the respondent in Hong Kong. By service of the originating summons on the respondent in Hong Kong, jurisdiction was founded as of right. Had the respondent not been present in Hong Kong, it might have been necessary for the appellant to seek to serve the originating summons on the respondent out of the jurisdiction, for example in reliance on Order 11, rule 1(2)(b), of the Rules of the High Court (Cap.4A).[36] That would have raised a different question of construction of section 10(2) of the GMO but one that it is not necessary to address in this appeal. 46. The first question posed is therefore to be answered in the affirmative. 47. The second question for which leave to appeal was granted must be understood to arise in the context of the present case, namely one in which a party has been duly served within the jurisdiction with the originating process seeking an order under section 10(2) of the GMO. In that context, the question posed is to be answered in the negative. 48. Having so answered the first and second questions, the third question for which leave to appeal was granted does not arise and need not be addressed. Mr Justice Stock NPJ: 49. Jurisdiction to entertain the maintenance application in this case was established as of right by personal service on the respondent in Hong Kong. The Ordinance does not require a child in respect of whom maintenance is sought pursuant to s 10(2) to be ordinarily resident or present in Hong Kong. There was, with respect, no need for the Court of Appeal to import considerations of inherent jurisdiction and it was inappropriate to do so; and, further, there is, in my opinion, no valid basis upon which to disturb the decision of the first instance judge on the question of forum non conveniens. Accordingly, I agree that this appeal should be allowed and I agree as well with the answers which are proposed to the three questions in respect of which leave to appeal was granted. Mr Justice Gleeson NPJ: 50. I agree with Mr Justice Tang PJ and Mr Justice Fok PJ that, on the true construction of the Guardianship of Minors Ordinance (Cap 13), the grant of power in s 10(2) of the Ordinance is not limited in the manner contended for by the respondent; the power is capable of being exercised in respect of a minor who is not ordinarily resident or present in Hong Kong. 51. I also agree that, jurisdiction having been established by personal service in Hong Kong, considerations of forum non conveniens were relevant to a decision whether to exercise that jurisdiction. 52. Where, as in this case, such considerations are invoked, and the court has concluded that they do not warrant a stay of proceedings, then the court should proceed to deal with the application on its merits. 53. This was the approach of the primary judge, and in my view it was correct. No error was shown in the judge’s conclusion that, applying the principles relevant to an argument of forum non conveniens, Hong Kong was the appropriate forum. Nothing in the Ordinance warranted or required the application of some other and different principles as a potential basis for declining to exercise the power conferred by s 10. 54. Reference was made by the Court of Appeal to considerations of international comity. However, this was not a case involving the potential exercise of an exorbitant jurisdiction, and the reasoning on forum non conveniens involved a finding that Hong Kong was a natural and appropriate forum. 55. I agree that the first question should be answered in the affirmative, the second question should be answered in the negative and the third question does not arise. Mr Justice Ribeiro PJ: 56. The appeal is unanimously allowed and the order of the Court of Appeal dated 18 June 2015 is set aside and the matter is remitted to the Family Court for the substantive issues to be dealt with. We make an order nisi that the respondent pay the applicant's costs here and below, with liberty to the parties to lodge written submissions as to costs within 14 days from the date of this judgment, the costs order to stand as an order absolute without further direction in default of such submissions. Mr Robert Pang SC and Mr Eugene Yim, instructed by Christine M Koo & Ip, Solicitors & Notaries LLP for the applicant (appellant) Mr Kenny C P Lin and Mr Lincoln Cheung, instructed by Lo, Wong & Tsui, for the respondent (respondent) [1] CA judgment, para 3.1. [2] Presumably the mother was not entitled to stay in Hong Kong. [3] Social Welfare Investigation Report dated 9 January 2013, para 8. [4] The Social Welfare Investigation Report shows that although the father’s work (5 days a week) required him to spend most of his time in the Mainland, his home was in Hong Kong. Para 12. [5] Section 3 of the Guardianship of Infants Act, 1925, enabled an order for maintenance to be made in favour of the mother where the court makes an order giving the custody to the mother under s 5 of the Guardianship of Infants Act, 1886 (“1886 Act”). The 1886 Act was repealed by the 1971 Act, s 9 of which enabled an order for maintenance of the minor against the father in favour of the mother where the court makes an order under s 9(1) giving custody of the minor to the mother. The 1973 Act equalised the rights of fathers and mothers, such that thereafter an order for maintenance under s 9(2) of the 1971 Act may be made against a father or a mother where an order for custody was made under s 9(1). [6] Para 6, respondent’s affirmation dated 7 November 2014. [7] Ruling, para 56. [8] Ruling, para 94. [9] Ruling, para 111. [10] Lam VP, Cheung and Yuen JJA. [11] Ruling, para 7.11. [12] Ruling, para 8.18. [13] FAMV 24/2016, Ribeiro, Tang and Fok PJJ. [14] 582D. [15] Para 7.2. To the contrary, s 26 provides that “The jurisdiction … shall be exercisable notwithstanding that any party to the proceedings is not domiciled in Hong Kong”. [16] Para 8.10, s 25 of the GMO provides: “Nothing in this Ordinance shall restrict or affect the jurisdiction of the Court of First Instance to appoint or remove guardians or otherwise in respect of minors.” [17] CA judgment, para 8.14. [18] Dicey and Morris on the Conflict of Laws, 9th edition, at p 399. [19] In Re P (GE)(An Infant) at 587F. [20] See para 13 above. [21] 327, Kay J. [22] 328. [23] 334. [24] 335-336. [25] (1854) 4 De GM & G 328 at 345. [26] Ruling, para 86. [27] Ruling, para 110. [28] Section 8(2) of the Matrimonial Proceedings and Property Ordinance and s 5(d) of Matrimonial Causes Ordinance, Cap 179. [29] I ignore as irrelevant, legitimation by subsequent marriage. [30] Ruling, para 61. [31] Bennion on Statutory Interpretation (6th Ed.) at Section 130: “Unless the contrary intention appears, and subject to any relevant rules of private international law, an enactment is taken not to apply to foreigners and foreign matters outside the territory to which it extends.” [32] As explained by Diplock LJ (as he then was) in Garthwaite v Garthwaite [1964] P 356 at 387-8. [33] Ibid. [34] See, in this context, the Matrimonial Causes Ordinance (Cap.179), sections 3, 4 and 5; the Parent and Child Ordinance (Cap.429), section 6; and the Adoption Ordinance (Cap.290), sections 5(6) and 20C(5). [35] GMO, section 3(1). [36] Alternatively under Order 11, rule 1(2)(b), of the Rules of the District Court (Cap.336H). 
Introduction 1. The police have pursuant to a search warrant issued by a magistrate obtained from a hospital certain medical records relating to the applicant. This application for judicial review raises the narrow question of whether the fact that the search warrant has not been produced to the applicant has effectively obstructed her right of access to the courts. 2. It is appropriate to make clear at the outset that the present application is not about whether the police should have applied for the warrant, or whether the magistrate should have granted the warrant, or whether the warrant should be set aside. Factual background 3. On the evening of 11 August 2019, there were, according to the evidence filed, hundreds of violent masked protesters assembled outside the Tsim Sha Tsui Police Station, who were hurling bricks, stones, slingshots and other unknown objects at the police station and the police officers inside. It is unnecessary for the purposes of the present application to examine what actually happened that evening and, indeed, the court is not equipped with the evidence to do so in these proceedings. 4. In the midst of it all, at 7:24pm, the applicant, dressed in black and equipped with a helmet, a pair of goggles and a respirator — the typical outfit of some of the participants in similar assemblies, was found in Nathan Road near the junction with Austin Road having suffered injuries to her right eye. According to her, she was “hit by a suspected bean bag round shot by anti‑riot police”. She was conveyed to Queen Elizabeth Hospital (“Hospital”) for treatment and then admitted into the Hospital. 5. On 13 August, a Superintendent wrote on behalf of the Commissioner to the Hospital, stating that having regard to the serious nature of the injury, the police was obliged to look into the circumstances of the incident and would like to approach the applicant with a view to ascertaining the facts. (The police did not know the identity of the applicant at that time and simply referred to “a Chinese female” in relation to that incident.) The letter stated that with a view to minimising any disruption to her medical treatment, the police would not approach her at that moment, but sought the Hospital’s assistance in forwarding an enclosed letter to her. That letter, written in Chinese and addressed to whom it might concern, stated that the police wished to understand the personal injury incident in greater detail and asked the addressee to call the specified telephone numbers if she could assist. The applicant did not contact the police whether via those numbers or otherwise. 6. On 21 August, the police obtained a search warrant from a magistrate to get the applicant’s personal details from the Hospital (ie date of birth, HKID number, telephone number and address). This warrant was executed, and the police obtained the information on 27 August. On 29 August, the police obtained a second search warrant from a magistrate in respect of the medical records in the Hospital in relation to the applicant’s injury sustained on 11 August. Although there are two warrants thus issued, the present application for judicial review concerns the second one, and references to the warrant below should be understood as referring to that warrant. 7. Also on 29 August, at a police press conference, a Chief Superintendent mentioned that the police had applied or were applying for a search warrant to obtain the applicant’s medical records in question. 8. On 30 August, the applicant’s solicitors wrote to the Hospital stating that she would not consent to the disclosure of her medical records to any third party including the police, and that the Hospital owed her a duty of confidentiality. She asked to be provided with, among other things, particulars of any disclosed information and the recipients. 9. On 2 September, the applicant’s solicitors wrote to the Commissioner (copied to the Hospital), demanding: (1) if the police intended to apply for a warrant, that the applicant be notified of any such application; (2) if the police had obtained the warrant but not yet executed it, that the applicant be informed of the process by which it was obtained and be given a copy of the warrant and that the police refrain from executing the warrant until the applicant had instituted legal proceedings to review the granting of the warrant; and (3) if the police had already obtained and executed the warrant, that the applicant be informed of the manner in which the warrant was obtained and executed and the scope of the medical records obtained, and that the police seal up the records and refrain from further communicating the contents of the records until the resolution of further legal proceedings. 10. On 2 September, the Hospital replied to the applicant’s letter of 30 August, stating that it had provided to the police her personal details and was also required pursuant to a warrant served by the police to provide the medical records which were pending release as at that date. The Hospital stated that a search warrant empowered police officers to enter and search for and take possession of the documents covered and that it was obliged to cooperate and comply and did not wish to resist or obstruct police officers in the due execution of their duties. The Hospital asked the applicant to take up the matter directly with the police and expected to hear from her within the next 7 days. 11. On 3 September, the applicant’s solicitors wrote again to the Hospital asking it to provide her with a copy of the warrants and the records handed over to the police, and that the Hospital refrain from doing anything beyond lawful compulsion that would be in breach of the applicant’s privacy rights. The evidence does not show whether there was any response from the Hospital to this letter. 12. Also on 3 September, the Commissioner issued an interim reply to the applicant’s solicitors, acknowledging receipt of the letter of 2 September and stating that the police were considering the matters raised and seeking legal advice as appropriate and would reply as soon as possible. 13. The Commissioner’s letter was apparently missed by the applicant’s solicitors, who wrote again on 3 September to the Commissioner, asking for a copy of all the warrants obtained in relation to the personal data of the applicant, stating that the applicant intended to apply to set aside the warrants, and asking the police to refrain from further executing any warrants. 14. Apparently, as I was told by the Commissioner’s counsel at an earlier hearing on 12 September, the medical records covered by the second warrant were obtained by the police on 4 September, though this was not known to the applicant at the time. 15. On 6 September, the applicant’s solicitors wrote to the Commissioner, complaining that he had not responded to their previous letters (which is factually incorrect but is, I assume, an honest mistake by the solicitors) and stating that unless the information previously requested be provided, they were instructed to seek relief from the High Court. 16. In press conferences on 9 and 10 September, in response to questions from the press, a Senior Superintendent disclosed that the police had applied under s 50(7) of the Police Force Ordinance (Cap 232)[1] for search warrants and had obtained the applicant’s medical records from the Hospital. 17. On 9 September, the applicant lodged a Notice of application for leave to apply for judicial review (Form 86). An anonymity order was granted by this court.[2] Following an inter partes hearing on 12 September, the applicant was granted leave to apply for judicial review based on her Form 86.[3] Meanwhile, the Commissioner has voluntarily offered an undertaking to seal up the medical records obtained until the resolution of this judicial review in this court. The respondent’s evidence in response was filed on 4 October. Proposed amendment of the application for judicial review 18. The application for judicial review, for which leave has been given based on the Form 86, seeks to impugn the refusal of the Commissioner to produce to the applicant the warrant. It is said that this infringes the applicant’s right of access to the courts under Art 35 of the Basic Law. 19. By summons dated 21 October 2019 and returnable before this court on 4 November 2019 (ie the date of the substantive hearing of the judicial review), the applicant sought leave to amend her Form 86. One of the amendments proposed was to describe the decision attacked as the Commissioner’s “failure” rather than “refusal” to produce the warrant — to reflect the fact that there had been no actual refusal before proceedings were begun. This and a few other minor amendments were uncontentious and permitted at the hearing. 20. More controversially, however, the applicant also proposed to add a second decision to be reviewed, namely, the Commissioner’s “decision to fully execute the purported warrant to obtain the medical records of the Applicant”, on the ground that it was unfair for the Commissioner to proceed to fully execute the second warrant despite being on notice of the applicant’s intention to oppose the process such as through an application to set aside the warrant. It was said that the unfairness had been compounded by the Commissioner’s breach of the duty of candour in that the affirmation filed in these proceedings on his behalf failed to explain the circumstances of that decision. At the hearing I refused to allow these amendments. 21. This second “decision” appears to relate to the fact that the medical records sought under the second warrant were obtained by the police after 2 September 2019, since the Hospital’s letter of that date indicates the records were still “pending release” then. The complaint assumes there was such a second decision and alleges that it was wrong for the police to “further execute” the warrant after having received the applicant’s solicitors’ letter dated 2 September 2019. 22. Such a complaint is fact‑sensitive. But the details of the events leading to the police obtaining the medical records after 2 September 2019 (when the warrant had already been served on the Hospital) are not clear. The applicant attributed this to the respondent’s lack of candour, but I do not think this is fair. The original decision attacked was only the refusal to produce the warrant to the applicant. Pursuant to this court’s directions, the respondent’s affirmation was filed on 4 October 2019. The applicant only sought to add a new decision much later, and had never sought leave to apply for judicial review in relation to that decision. It is, to put the position at its lowest, not clear that the duty of candour required the respondent to deal with everything that the applicant now wishes to complain about. The respondent’s duty of candour is not an “open‑ended concept”: Chu Woan Chyi v Director of Immigration [2009] 6 HKC 77, §14(6). 23. In any event, the fact is that the respondent had not in his affirmation gone into all the matters that might be relevant to the new decision sought to be challenged, and this is not an application for discovery. In the circumstances, I did not think that it would be fair to permit the applicant to enlarge the scope of the judicial review at the hearing by a late amendment to which the respondent had not had a proper opportunity to respond. Accordingly, this application for judicial review must be adjudicated on the basis of the original Form 86 with the few minor amendments allowed at the substantive hearing. The applicant’s contentions 24. As stated at the outset, this application for judicial review concerns a narrow question. The originating summons filed pursuant to leave granted seeks only: “ A declaration that the Commissioner of Police’s refusal to produce the warrants obtained from the Magistrate for the entry and search of Queen Elizabeth Hospital in relation to the Applicant’s personal data and/or medical records is in breach of Article 35 of the Basic Law and Article 2(3) of the International Covenant on Civil and Political Rights.”[4] Some of the propositions initially advanced on behalf of the applicant in the Form 86 or in counsel’s skeleton submissions are however, on their face, very wide. Thus it was said that the Commissioner should have notified the applicant of his intention to apply for a warrant in relation to the medical records so that the applicant could be heard in an inter partes hearing.[5] At the hearing, however, Mr Robert Pang SC, appearing for the applicant, confirmed that he was not contending for a general rule that prior notice of an application for a search warrant on an institution such as a hospital or a bank should be given to the person whose information is contained in the documents targeted. Nor do I understand the applicant to maintain the proposition found in the Form 86 as a general rule that in such a case, the magistrate should be invited, in the ex parte application for the warrant, to impose conditions for the sealing of those records until an application is made by that person to set aside the warrant.[6] Likewise, Mr Pang disavowed any general contention that in such a case, having obtained a search warrant, the police are required to notify the person of it before taking steps to execute it so that the person has an opportunity of making an application to set it aside.[7] 25. What counsel do submit on behalf of the applicant is that when the lawfulness of the police’s seizure of medical records pursuant to a warrant is called into question by a person who has a sufficient interest in the subject matter, that person should be entitled to be shown the warrant. In that connection, it is contended that, by failing to produce the warrant to the applicant who had made clear her intention to apply to set it aside, the Commissioner has obstructed her from seeking access to the courts and from seeking relief to prevent interference with her privacy rights. 26. The position taken on behalf of the applicant is thus one of strict entitlement. There is no submission advanced for her on the basis that the decision impugned should be quashed because the Commissioner had a discretion which he exercised in a Wednesbury unreasonable manner[8] or without taking into account relevant considerations, and I express no opinion on the merits of such bases. The respondent’s case 27. In an affirmation made by a Woman Senior Inspector, the respondent stated that in the afternoon on 11 August 2019, a public meeting was held at Victoria Park which had been notified under the Public Order Ordinance (Cap 245) and not prohibited by the police. After its conclusion in the evening, however, a significant number of the protesters moved to different places in Tsim Sha Tsui and engaged in the violent activities referred to above and laid siege to the Tsim Sha Tsui Police Station. The protesters also occupied a sizeable section of Nathan Road, blocking its southbound and northbound carriageways with barriers and various objects. 28. Police investigation of the incident is still ongoing. Up to the date of the affirmation, charges had been laid against 17 protesters arrested at Park Lane Shopper’s Boulevard, one of whom for “Taking Part in a Riot” and the others for “Unlawful Assembly”. The scope of the investigation has been evolving and the major areas of investigation include ascertaining the facts of the unlawful activities, including the time, place and the events that took place; identifying the persons involved; obtaining relevant evidence and preserving evidence of damage and injury caused; and conducting relevant forensic examinations. 29. The respondent stresses that the revelation at the press conferences in relation to the warrants was made to address the public’s concerns over whether the applicant’s injury had been caused by the police, and did not reflect any change in the police’s position and practice that all information relating to ongoing investigation must be kept strictly confidential to preserve the integrity of the investigation. 30. The respondent explains that it is not the practice of the police to disclose to anyone any application for a search warrant or to inform anyone of the execution of any search warrant after its execution. It is stated that an important element of the integrity of police investigation is that the subject of the criminal investigation or his associates and representatives are not to be informed of the execution of search warrants or of the contents of such warrants, in order to avoid or reduce the risks of the subject of investigation, or his associates and representatives, taking steps to circumvent, obstruct, delay or interfere with the investigation and subsequent prosecution of criminal activities, including but not limited to removal or destruction of evidence or the making of collateral attacks on actions taken by the police whilst the investigation process is still ongoing. The relevant information of the investigation process (including search warrants) will normally be disclosed as part of the unused materials, should the subject be prosecuted in due course. If, on the other hand, no prosecution ensues, the confidentiality of the actions taken in the course of investigation will continue to be maintained as a matter of practice. 31. It is said that the provision of the warrant to the applicant in the present case will prejudice the public interest by compromising the principle and practice of confidentiality and the timeliness and effectiveness of the police’s ongoing investigation into, among other things, the unlawful activities on 11 August 2019. 32. On behalf of the respondent, Mr Johnny Mok SC submits that an application to set aside a search warrant will only be entertained on limited grounds as held in P v Commissioner of Independent Commission Against Corruption (2007) 10 HKCFAR 293 and that the applicant has not shown how there can be a sustainable set‑aside application on any of those grounds. He submits that the applicant is not prevented by the non‑disclosure of the warrant from taking legal action if there is a valid basis for doing so, including the making of an application for discovery, whether pre‑action or otherwise, against a defendant or a third party, if she has valid ground to do so. There is no free‑standing right, whether under Art 35 of the Basic Law or the common law, for anyone to ask the police to disclose documents of any kind relating to ongoing criminal investigation outside of the various existing regimes of document disclosure in legal proceedings. Nor does a person’s privacy mean that the police are at all times under an obligation to notify an affected person of their intention to apply for a warrant or to produce the warrant on demand (other than for the purpose and in the course of its execution). He submits that, if the applicant’s contention were correct, it would mean that any potential suspect who had got wind of an investigation would have the right to demand production by the police of all warrants obtained or executed and be able to find out the direction and progress of police investigation. Such a general rule would be detrimental to the effectiveness of criminal investigation which depends on a strong degree of confidentiality and covertness. Discussion 33. Having regard to the submissions made, I find it appropriate to divide my analysis into two aspects: first, whether there is a free‑standing right for the applicant to have the warrant produced to her on demand; and, secondly, whether the non‑production of the warrant to her has infringed her right of access to the courts. Whether there is a free‑standing right to production of warrant 34. A warrant is a document issued by a person in authority under power conferred in that behalf authorising the doing of an act which would otherwise be illegal: Inland Revenue Commissioners v Rossminster Ltd [1980] AC 952, 1000A per Lord Wilberforce. In the case of a search warrant, the act that would otherwise be unlawful without the warrant is the entry and search of the premises in question and the seizure of the targeted articles. In Hong Kong, there are numerous statutory provisions that confer power on law enforcement agencies to enter and search private premises with and without a search warrant. A detailed list may be found annexed to the judgment of the Court of Appeal in Keen Lloyd Holdings Ltd v Commissioner of Customs and Excise [2016] 2 HKLRD 1372. 35. The general provision used by the police, and the provision invoked in the present case, is s 50(7) of the Police Force Ordinance, which provides: “ Whenever it appears to a magistrate upon the oath of any person that there is reasonable cause to suspect that there is in any building, vessel (not being a ship of war or a ship having the status of a ship of war) or place any newspaper, book or other document, or any portion or extract therefrom, or any other article or chattel which is likely to be of value (whether by itself or together with anything else) to the investigation of any offence that has been committed, or that is reasonably suspected to have been committed or to be about to be committed or to be intended to be committed, such magistrate may by warrant directed to any police officer empower him with such assistants as may be necessary by day or by night– (a) to enter and if necessary to break into or forcibly enter such building, vessel or place and to search for and take possession of any such newspaper, book or other document or portion of or extract therefrom or any such other article or chattel which may be found therein …” 36. The question of production of the warrant in the course of the search is not dealt with in the Police Force Ordinance. This may be contrasted with the provisions in the Police and Criminal Evidence Act 1984 of the United Kingdom as well as other more modern legislation in Hong Kong. Sections 16(5)(b)‑(c) and 16(6) of the UK 1984 Act expressly provide that during execution of the search warrant, where the occupier or some other person who appears to be in charge of the premises which are to be entered and searched is present, the constable shall produce the warrant to him and supply him with a copy of it. Section 86(2) of the Interpretation and General Clauses Ordinance (Cap 1) likewise provides that a person executing a warrant to search for and seize journalistic material shall supply the occupier or person in charge of the premises with a copy of the warrant. 37. In Hong Kong, in the absence of specific legislation, the requirement for production of the warrant issued under s 50(7) of the Police Force Ordinance upon execution is governed by the common law. The common law recognises the requirement of an announcement to be made upon execution of a warrant, for the officer “first to signify the cause of his coming”: Semayne’s Case (1604) 5 Co Rep 91a. Further, the officer who executes the warrant must have it in his possession at the time: see Wah Kie v Cuddy (1914) 20 DLR 351, extending to search warrant the rule applicable to warrant for arrest. It seems right in principle that the search warrant should be produced upon the search, unless the exigencies of the circumstances require otherwise, such as where production before a search would frustrate the search itself. It has been said that a search warrant need not be produced unless the occupier asks to see it: Feldman, The Law Relating to Entry, Search & Seizure, §6.24,[9] but the authority cited, Hodges v Marks (1615) Cro Jac 485, concerned a warrant for arrest, and reliance on it in relation to a search warrant seems questionable. As is stated in the same work by Professor Feldman at §5.02, the production of the search warrant allows the occupier of the premises to be searched to satisfy himself that the officers who demanded to be admitted are acting lawfully, thus reducing the risk of misunderstandings, violence and the criminal charges which would follow, and makes the job of the police easier. 38. This being the position at common law, it is not surprising to find in paragraph 4 of Section 44‑04 of the Police General Orders (made by the Commissioner of Police under s 46 of the Police Force Ordinance) the stipulation that, when executing a search warrant, the senior police officer present shall, among other things, produce the warrant if in his possession, or if he is executing a warrant not actually in his possession, inform the person in charge of the premises that if he so wishes, the warrant will be available for inspection after the search has been completed, and inform him of where the warrant will be available for such inspection. 39. Neither the Police Force Ordinance nor the common law, however, confers on a person who is related to the documents seized or to be seized but who is not an occupier of the premises any right to have the warrant produced to him or her. The prime concern of the common law, in requiring entry and search of premises to be authorised by law, is “the inviolability of private premises from arbitrary intrusion” (per Stock VP in Philip K H Wong, Kennedy Y H Wong & Co v Commissioner of the Independent Commission Against Corruption [2009] 5 HKLRD 379, §47). This protection is now enshrined in Art 29 of the Basic Law.[10] “A man’s home is his castle”, so the saying goes: see Semayne’s Case, supra, and Mr Pang has also cited Entick v Carrington (1765) 19 State Tr 1029, which emphasised that “every invasion of private property, be it ever so minute, is a trespass”, unless empowered or excused by some positive law. The “citizens’ right to privacy” referred to by the Court of Appeal in Keen Lloyd, supra, at §49 alludes to the integrity and privacy of a man’s home and place of business, as referred to earlier in the same paragraph. 40. As explained above, the act that would otherwise be unlawful without a search warrant is the entry and search of the premises in question and the seizure of the targeted articles. The entry and seizure would attract tortious liabilities in trespass and conversion.[11] 41. Notwithstanding that the applicant is neither the occupier of the premises nor the owner of the documents obtained, it is submitted on her behalf that she has a “sufficient interest” in the subject matter of the warrant, giving her a strict right to its production. The applicant refers to R v Purdy [1975] QB 288, but that case concerned a warrant for arrest otherwise than for a criminal offence and simply applied the common law rule that a constable could not make an arrest based on a warrant unless he had the warrant with him at the time. The actual decision on the facts of that case was that the warrant was still to be regarded as being in the officers’ possession even though it was in their car some 50 or 60 yards away from the place of arrest. The rationale of the common law rule was explained by Roskill LJ at p 296H‑297B as follows: “ The mischief at which this common law rule is aimed is clear. Where a person is arrested on a warrant otherwise than for a criminal offence, it is essential that he should be able to know for what he is being arrested and whether he is being arrested for a matter of non‑payment of a fine, arrears of maintenance, arrears under an affiliation order or for something else; he can, if I may be forgiven the phrase, ‘buy’ his freedom from arrest by instant payment of the sum stated on the warrant. This therefore is the basic reason for the rule.” In my view, this case provides no support for the proposition that when the lawfulness of police action is called into question by persons who have a “sufficient interest” in the subject matter of a warrant, then the warrant must by law be produced to such persons. 42. Mr Pang invites the court to make a declaration that “upon the demand[12] of the person whose rights are affected by a search warrant, the relevant authority must produce a copy of the search warrant for inspection”.[13] Quite apart from the fact that this is not the declaration sought in the Form 86 or the originating summons, little imagination is needed for one to be able to see the far‑reaching implications of such a rule. As submitted by the respondents, it would mean that any potential suspect who had somehow learnt that he might be the subject of an investigation, could demand that the police produce to him all search warrants executed or to be executed in the ongoing investigation. In the case of a search conducted or to be conducted on a clinic, a bank or a commercial enterprise, the police would be obliged to produce upon demand the warrants for inspection by potentially very numerous persons whose information might be contained in the materials to be searched for and seized and whose privacy was therefore arguably affected. This rule is not found in any statute or the existing common law. Essentially the applicant contends that this court should create such a rule in her favour. 43. That a search warrant may have an impact on the privacy of persons who are not the occupier of the premises to be searched but whose information is contained in the items to be seized is, I think, beyond dispute. The question of the legal rights and obligations of such persons in relation to the warrant, and of the correlative powers and duties of the executive authority that executes the warrant, is a large and novel area that merits consideration by the legislature. It would not, however, be appropriate or conducive to the development of the law for this court to create a rule to cater for the unusual circumstances of the present case and in so doing make a crude and general declaration without proper delineations, distinctions and limitations which the common law is ill‑equipped to devise and which may need to rest on considerations of policies and practicalities rather than principle and logic. 44. A declaration of such free‑standing right to production of the warrant on demand is, in my view, not only inadvisable but also unnecessary in the context of the present case because in connection with legal proceedings to impugn a warrant, which is the purpose for which apparently the applicant had requested production, there are established legal mechanisms for the applicant to obtain access to the warrant if she has grounds to do so, such as by application for discovery within or before such proceedings — a subject that will be discussed further below. This brings me to the topic of whether the applicant has been effectively obstructed from obtaining redress from the courts, which is the crux of her complaint. Whether the applicant’s right of access to the courts infringed 45. In the system of entry, search and seizure by warrant issued under s 50(7) of the Police Force Ordinance, the law provides safeguards against excesses by the requirement of a warrant issued by a judicial officer. The significance of such prior scrutiny by an impartial authority has been explained by the Court of Appeal in Keen Lloyd at §§72‑74 and in Philip K H Wong at §§47‑49 in passages which provide a salutary reminder of the courts’ duty in this context which is one of high constitutional importance. 46. There is no prescribed form for a search warrant under s 50(7). As stated by the Court of Appeal in Philip K H Wong, supra, at §§85‑86, a warrant must of course contain the matters to which the statutory provision itself refers. In the case of s 50(7), this means that it must be an authorisation directed to the police; it must specify the premises to be searched; it must specify the offence that has been committed or reasonably suspected to have been committed or to be about to be committed or to be intended to be committed, to the investigation of which the articles sought are likely to be of value; it must indicate the articles which are sought; and it must show that the magistrate is satisfied there is reasonable cause to suspect that there is in the specified premises such articles likely to be of value to the investigation. 47. An application for a search warrant under s 50(7) is, as far as I know, generally made without notice to the occupier of the premises to be searched or to other parties. The grounds for the application would be set out in an information submitted to the magistrate. The information placed before the court for the purposes of an application for search warrant has been recognised to be covered, as a class, by public interest immunity from disclosure: Apple Daily Ltd v Commissioner of the Independent Commission Against Corruption (No 2) [2000] 1 HKLRD 647. While the applicant has queried why she was not given prior notice of the application, there is no challenge in these proceedings against the Commissioner’s decision to apply for the warrant ex parte and the magistrate’s decision to entertain the application and grant it on an ex parte basis. The magistrate has not been made the respondent or an interested party in these proceedings. At the hearing Mr Pang also disavowed any general proposition that a person in the position of the applicant should be given prior notice of an application for the search warrant or its execution. As such, the case of Re an application by Officer O for Judicial Review [2008] NIQB 52, which decided that the Police Ombudsman for Northern Ireland, before deciding to require the Chief Constable under a specific statutory provision to provide all medical and occupational health records relating to the officer’s medical condition, should have given advance notice to the officer, is not apposite. 48. Although the application for the issue of search warrant is made and dealt with on an ex parte basis, there is the further safeguard of a possible inter partes contest before the courts in respect of the legality of the search and seizure. That contest may take various forms. (1) In Keen Lloyd at §§32‑41, the Court of Appeal held that a party served with a search warrant can, at least up to the time before the warrant has been fully executed, apply to the judicial officer who issued the warrant to have it set aside. The grounds for such applications for discharge, however, seem to be limited to: (a) the legal limits for the grant of warrants; (b) the warrants were obtained by fraud or bad faith on the part of the law enforcement agency applying for the warrants; and (c) oppression. In particular, the court cannot entertain challenges as to whether a case of reasonable suspicion has been established or the likelihood of relevant materials or information being obtained: see Keen Lloyd, at §§38‑40. (2) Application to the Court of First Instance for judicial review is also available against decisions of magistrates to issue search warrants; see Re an application by Messrs Ip and Willis for leave to apply for judicial review [1990] 1 HKLR 154; Shun Tak Holdings Ltd & Others v Commissioner of Police (unrep, HCMP 1366 & 1367/1994, 27 July 1994); Philip K H Wong, supra, and Chan Kam Ching John Barry v Commissioner of Police [2014] 4 HKLRD 263 for examples of such application in Hong Kong and the grounds relied upon. In the UK, the House of Lords’ case of Rossminster Ltd, supra, also originated from an application for judicial review. A search warrant (other than those issued by the High Court itself) can be quashed by certiorari, and declarations and injunctions can be obtained: Feldman, op cit, §15.05. (3) Another possible avenue for relief is a civil action for injunction to restrain use of the documents seized and to procure their return or for damages: Stone, The Law of Entry, Search and Seizure (5th edition), §§3.106 to 3.110; Keen Lloyd, at §§42‑43. 49. For the purposes of these proceedings, as I understand his position, the respondent does not dispute that the medical records in question contain personal data of the applicant for which there is a reasonable expectation of privacy, and that the applicant may therefore have sufficient interest to institute the above actions, at any rate based on grounds relevant to her interest. 50. The applicant says that she seeks to re‑engage the judicial gatekeeping role in relation to the warrant and to challenge the lawful authority of the search and seizure of her medical records, but that the respondent has prevented her from doing so, thereby breaching Art 35 of the Basic Law. 51. Art 35 provides as follows: “ Hong Kong residents shall have the right to confidential legal advice, access to the courts, choice of lawyers for timely protection of their lawful rights and interests or for representation in the courts, and to judicial remedies. Hong Kong residents shall have the right to institute legal proceedings in the courts against the acts of the executive authorities and their personnel.” 52. In The Stock Exchange of Hong Kong Ltd v New World Development Co Ltd (2006) 9 HKCFAR 234, Ribeiro PJ said (at §50): “ … Art 35 ensures that the fundamental rights conferred by the Basic Law as well as the legal rights and obligations previously in force and carried through to apply in the HKSAR are enforceable by individuals and justiciable in the courts. It gives life and practical effect to the provisions which establish the courts as the institutions charged with exercising the independent judicial power in the Region. This dimension of Art 35 is therefore concerned with ensuring access to the courts for such purposes, buttressed by provisions aimed at making such access effective. …” 53. The parties agree that whether there has been an infringement of the right of access to justice is a question that depends on the facts of the individual case. In some of the previous cases, the issue was whether a particular statutory provision or rule restricted a person’s right of access to the courts. In Leung Chun Ying v Ho Chun Yan Albert (2013) 16 HKCFAR 735, it was contended that a provision in the Chief Executive Election Ordinance (Cap 569) that required any election petition to be lodged within a non‑extendable period of 7 working days infringed the constitutional right of access to the courts. Ma CJ stated that whether the restriction amounted to an infringement of the right depends on whether on analysis the essence of the right has been impaired. 54. In Lau Chun Ming v Deloitte Touche Tohmatsu (a firm) (2016) 19 HKCFAR 448, it was argued that s 94 of the Bankruptcy Ordinance (Cap 6), which provides for the release of trustees in bankruptcy and the consequent discharge of the trustee from all liability in respect of the administration of the bankruptcy, was inconsistent with Art 35 of the Basic Law in that it removed the creditors’ right to judicial remedies against the trustee. The Appeal Committee of the Court of Final Appeal stated that Art 35 ensured that fundamental rights as well as ordinary legal rights and obligations were enforceable by individuals and justiciable in the courts, and held that the applicant’s argument in that case was misconceived as there was no question of his having been denied access to the courts for such purposes. Section 94 allowed the applicant access to the court to ask for the trustee’s release to be withheld. The applicant did not avail himself of such access. Accordingly, Art 35 was not engaged. 55. It is of significance that the applicant has not pointed to any actual legal obstruction. On the face of it, there is in the present case no impediment in law preventing the applicant from instituting legal proceedings in the courts to ventilate her arguments about rights to privacy in respect of the medical records. 56. Is there any impediment in fact? The applicant refers to the case of Golder v United Kingdom (1979‑80) 1 EHRR 524. That was a case where a prisoner who wished to institute libel proceedings against a prison officer was refused permission to consult a solicitor for the purpose of commencing such proceedings. It was in that context that the court held (at §26) that by forbidding Golder to make contact with a solicitor, the Government of the United Kingdom “actually impeded” the launching of the action and “did in fact prevent him from commencing an action at that time”. Plainly there has been no actual hindrance of a comparable nature in the present case. The applicant has at all material times been legally advised and represented and capable of instituting such legal proceedings as she might be advised. 57. In Camenzind v Switzerland (1997) 28 EHRR 458, a search was conducted on a person’s home pursuant to a warrant issued by the telecommunications authority. No equipment of the type sought was found. The person applied to the Swiss Federal Court for a declaration that the search was a nullity, but the court dismissed the application on the basis that he no longer had an interest as he was no longer affected by the search. The European Court of Human Rights held that in those circumstances, the applicant did not have “an effective remedy before a national authority” within the meaning of Art 13 of the European Convention on Human Rights for pursuing his complaint of infringement of Art 8 of the Convention as a result of the search. Relying on this decision, it is submitted on behalf of the applicant that she has been deprived of the “effective remedy” of applying for the warrant to be set aside, leaving only the possibility of damages. 58. The respondent’s answer to this is that the applicant has not been prevented from taking legal action if there is any valid ground for doing so, and that Art 35 of the Basic Law does not confer on her any right to “fish” for material to see if she has such ground. The respondent submits that the existing litigation regime permits the applicant to apply for discovery, whether pre‑action or otherwise, from either a defendant or a third party, provided she meets the necessary conditions, including establishing a prima facie case, relevance of the documents and the necessity of disclosure for the fair disposal of the cause or matter or for saving costs: see sections 41 and 42 of the High Court Ordinance (Cap 4);[14] and Order 24 rules 7A and 8 of the Rules of the High Court (Cap 4A). Mr Mok refers to Taylor v Anderton (unrep, 16 October 1986) as an example of a pre‑action application for discovery of documents for the purpose of instituting proceedings against the police for procuring the issue of search warrants without reasonable cause and excessive execution of the search warrants. 59. In my opinion there is considerable force in the respondent’s submission. The Court of Appeal in Keen Lloyd was concerned with the position of the occupier of premises on whom the search warrant was served. It did not deal with the position of a person like the applicant herein. But assuming that the applicant (like a person on whom the warrant is served) also had the right to apply to the magistrate who issued the warrant to have it set aside, she was not “obstructed” in or prevented from taking that course. Mr Pang submits that without seeing the warrants, the applicant did not know which magistrate to approach. This in my view presents no real difficulty or actual impediment, for enquiry could be made with the magistrates’ courts or the application can be submitted which will no doubt be directed to the magistrate in question. At the leave hearing the respondent’s counsel also confirmed that if what was needed was simply the identity of the magistrate, then that could be readily supplied by the respondent. Without more, the threshold for “obstruction” has in my view not been met. 60. Equally there was nothing to prevent the applicant from applying to the Court of First Instance for leave to apply for judicial review of the warrant in question, or instituting other proceedings to impugn the warrant and to seek the return of the documents, if she has valid grounds for doing so. At the leave hearing the issue was mentioned whether the warrant could still be set aside if it had been “fully executed”. The Court of Appeal in Keen Lloyd at §§41‑42 suggested that it could not, but it seems that the court was there referring to an application to the judicial officer who granted the warrant to have it set aside, not to an application to the High Court for judicial review. There are many examples of judicial review of search warrants after they have been executed: see §48(2) above, and, where appropriate, interim measures can be put in place such as sealing up the documents to prevent them from further use by the authorities: see eg Apple Daily Ltd v Commissioner of the Independent Commission Against Corruption [2000] 1 HKLRD 595. In the present case, the applicant still seeks an injunction against the respondent to prohibit any further reading, use or distribution of her personal data obtained pursuant to the warrant. The respondent has given an undertaking to seal up the medical records obtained pursuant to the warrant pending the determination of these proceedings. Both parties seem therefore to have proceeded on the basis that there can still be meaningful relief associated with an application to set aside the warrant. It seems to me that, contrary to Mr Pang’s argument, assuming the applicant has valid grounds of complaint, she is not left with a claim for damages as her only remedy. 61. Despite that the applicant’s solicitors had indicated as early as on 3 September 2019 that they were preparing an application to set aside the warrant, it remains unclear what the ground is for such intended application. It seems to be suggested that the applicant intends to rely upon her privacy rights and to submit that the warrant is an unlawful intrusion into her privacy, but it is not explained how, if at all, such an application has been prevented or impeded by the non‑production at this stage of the warrant. 62. Mr Pang submits that without seeing the warrant, the applicant will not be able to raise grounds that attack the form and scope of the warrant. I am unable to accept this argument. It begs the question whether the applicant is entitled to production of the warrant in the first place — a question dealt with earlier. If she has no such free‑standing legal right, then it is for her to apply for discovery and production, if necessary before commencing substantive proceedings, if there are grounds for doing so. If in the event she is denied such discovery, that would not be an infringement of her right of access to the courts. 63. In these circumstances it seems to me the applicant has failed to establish that the essence of her right of access to the courts under Art 35 of the Basic Law has in any way been impaired. Conclusion 64. In summary, for the above reasons, I consider that the applicant does not have a free‑standing right to the production of the warrant on demand. There are existing mechanisms in the law for the applicant to seek, on proper grounds, production of the warrant in the context of actual or intended proceedings to impugn it. The fact that the respondent has thus far not produced the warrant to the applicant does not mean her right of access to the courts has been infringed. 65. Accordingly, the application for judicial review is dismissed. There will be an order nisi that the applicant do pay the respondent the costs of these proceedings with a certificate for two counsel, and that the applicant’s own costs be taxed in accordance with the Legal Aid Regulations. Mr Robert Pang SC, Ms Linda Wong and Mr Albert NB Wong, instructed by Ho Tse Wai & Partners, for the Applicant, assigned by the Director of Legal Aid Department Mr Johnny Mok SC, Mr Derek Chan SC and Mr Mike Lui, instructed by Department of Justice, for the Respondent The Interested Party, represented by Mayer Brown, did not appear [1] See §35 below for the provision of s 50(7). [2] See [2019] HKCFI 2307, §2. [3] See [2019] HKCFI 2307, §2. [4] Art 2(3) of the International Covenant on Civil and Political Rights provides: “ Each State Party to the present Covenant undertakes: (a) To ensure that any person whose rights or freedoms as herein recognized are violated shall have an effective remedy, notwithstanding that the violation has been committed by persons acting in an official capacity; (b) To ensure that any person claiming such a remedy shall have his right thereto determined by competent judicial, administrative or legislative authorities, or by any other competent authority provided for by the legal system of the State, and to develop the possibilities of judicial remedy; (c) To ensure that the competent authorities shall enforce such remedies when granted.” [5] Form 86, para 4(i); Skeleton Submissions, para 21. [6] Contrast Form 86, para 4(ii). [7] Contrast Form 86, para 4(iii). [8] See Associated Provincial Picture Houses Ltd v Wednesbury Corp [1948] 1 KB 223, 230, 234. [9] Cited by the Court of Appeal in Apple Daily Ltd v Commissioner of the Independent Commission Against Corruption (No 2) [2000] 1 HKLRD 647, 667G. [10] Which provides: “The homes and other premises of Hong Kong residents shall be inviolable. Arbitrary or unlawful search of, or intrusion into, a resident’s home or other premises shall be prohibited.” [11] See Keen Lloyd Holdings Ltd v Commissioner of Customs and Excise, supra, §45. [12] The applicant’s submission is that a verbal demand would suffice. [13] Skeleton Submissions of the Applicant, para 38. [14] Section 41 of the High Court Ordinance provides: “On the application, in accordance with rules of court, of a person who appears to the Court of First Instance to be likely to be a party to subsequent proceedings in that Court in which a claim is likely to be made, the Court of First Instance shall, in such circumstances as may be specified in the rules, have power to order a person who appears to the Court of First Instance to be likely to be a party to the proceedings and to be likely to have or to have had in his possession, custody or power any documents which are directly relevant to an issue arising or likely to arise out of that claim– … (b) to produce such of those documents as are in his possession, custody or power to the applicant …” 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Fok PJ and the orders contained therein. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Tang PJ: 3. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: 4. The present case raises questions as to what happens to a pending criminal appeal to the Court of Final Appeal when the appellant dies before the appeal can be heard. Does the Court have jurisdiction to permit the appeal to continue? If so, what is the basis of that jurisdiction and how and in what circumstances should it be exercised? These questions arise on a summons issued by the appellant’s widow and it would not normally be necessary for such an application to be addressed by a full five-member Court. The full Court has, however, heard this application because in answering the questions raised it is necessary to revisit dicta in a previous decision of the Court[1]. A. Background facts 5. On 4 October 2012, the appellant, Cheng Chee-tock Theodore, was convicted after trial[2] before HH Stanley Chan DJ of one count of conspiracy to defraud and sentenced, on 30 October 2012, to 5 months’ imprisonment and to a disqualification order[3] for 3 years. He was released from prison on 7 February 2013 after serving his sentence. On 14 March 2014, the Court of Appeal[4] dismissed his application for leave to appeal against conviction and, on 12 June 2014, refused to certify that a point of law of great and general importance was involved in their decision. 6. On 11 April 2014, the appellant then applied to this Court for leave to appeal. On 4 August 2014, the Appeal Committee[5] granted leave to the appellant to appeal to the Court of Final Appeal on the grounds that (a) a point of law of great and general importance is involved in the decision of the Court of Appeal[6], and (b) it was reasonably arguable that substantial and grave injustice has been done. 7. The appeal was originally fixed to be heard on 26 February 2015. On 22 December 2014, however, the appellant passed away. The appellant’s widow, Madam Leonora Yung, wishes to continue the appellant’s appeal. On 14 January 2015, she took out a summons for leave to be made a party to the appeal and for the appeal to be carried on as if she had been substituted for the appellant. Directions were given[7] for the hearing of the summons and, pending its disposal, for the substantive appeal to be adjourned until further order[8]. Madam Yung has now been granted letters of administration in respect of the appellant’s estate. B. The parties’ arguments and the issues arising 8. It was submitted on behalf of Madam Yung that there is jurisdiction to continue the appeal, either in the name of the appellant or in her name. It was her case that the Court has a discretion to permit an appeal to continue notwithstanding the death of the appellant and that, in the present case, there are special circumstances which justify the exercise of that discretion in favour of permitting the present appeal to continue. 9. For its part, the prosecution (respondent to the appeal) submitted that the appeal was personal to the appellant and that, upon his death, the appeal abated (i.e. terminated). It was further submitted that, having abated, there is no jurisdiction to continue with the appeal. Even if the appeal has not abated, the prosecution contended that an appeal requires a living appellant and that there is neither procedure nor power to make Madam Yung a party to the appeal, or to substitute Madam Yung for the appellant, or to continue the appeal on such terms as the Court thinks fit. Finally, it was submitted that, even if such procedure or power exists, this is not a case for the exercise of discretion in favour of Madam Yung’s application. 10. In the circumstances, the following issues arise for determination on this application: (1) Is there jurisdiction to continue a criminal appeal to this Court after an appellant’s death and, if so, on what basis, or does the appeal simply abate? (The Jurisdiction Issue) (2) If there is jurisdiction to continue, is it necessary to substitute a living party for a deceased appellant and, if so, what is the power and procedure by which the Court can do so? (The Substitution Issue) (3) Assuming there is jurisdiction to continue the appeal, how is the Court’s discretion to be exercised in the present case? (The Exercise of Discretion Issue) 11. It should be stressed at the outset that these issues are raised in the context of a criminal appeal to this Court. Different considerations may apply in respect of criminal appeals to the Court of Appeal or to the Court of First Instance and it should not be assumed that the answers in this judgment to the questions posed will apply in respect of criminal appeals to those courts[9]. C. The Jurisdiction Issue C.1 The appellate jurisdiction of the Court of Final Appeal 12. The Court of Final Appeal was established under Article 81 of the Basic Law of the Hong Kong Special Administrative Region. It sits at the apex of the court system in Hong Kong and exercises the power of final adjudication in accordance with Article 82 of the Basic Law. The Court’s power of final adjudication is a power exercisable only on final appeal: Solicitor v Law Society of Hong Kong & Secretary for Justice[10]. 13. There is no inherent appellate jurisdiction. Instead, an appeal, whether to this Court or any intermediate court of appeal (be it the Court of Appeal or Court of First Instance), is a creature of statute[11]. In the case of the Court, the governing statute is the Hong Kong Court of Final Appeal Ordinance[12] and this regulates and limits appeals to the Court[13]. 14. Part III of the Ordinance (comprising sections 28 to 37) applies to appeals in any criminal cause or matter. Section 31 of the Ordinance limits the decisions which may be appealed to the Court and also regulates at whose instance an appeal may lie. The section provides: “An appeal shall, at the discretion of the Court, lie to the Court in any criminal cause or matter, at the instance of any party to the proceedings, from – (a) any final decision of the Court of Appeal; (b) any final decision of the Court of First Instance (not being a verdict or finding of a jury) from which no appeal lies to the Court of Appeal.” 15. Section 32 of the Ordinance provides for the need for leave, which is the means by which the Court exercises its discretion referred to in section 31 to entertain an appeal. Thus: no appeal shall be admitted unless leave to appeal has been granted by the Court (s.32(1)); leave shall not be granted unless the court from which the appeal lies has certified that a point of law of great and general importance is involved in the decision, or it is shown that substantial and grave injustice has been done (s.32(2)); and, if the court below declines to certify, the Court itself may do so and grant leave to appeal (s.32(3)). 16. Section 33(1) stipulates the requirement to make an application for leave to appeal within the time specified from the date of the final decision to be appealed from (or such time as the Court may extend (s.33(2)) and rules relating to such application are contained in Part II of the Hong Kong Court of Final Appeal Rules[14]. Section 33(3) is important to the context of the relevant provisions under consideration and provides: “An appeal to the Court shall be treated as pending until any application for leave to appeal is disposed of and, if leave to appeal is granted, until the appeal is disposed of.” 17. Thus, a criminal appeal to the Court is an appeal from a final decision of an intermediate court of appeal, in respect of which the Court has granted leave to appeal upon the certification (either of the court below or the Court itself) that a point of law of great and general importance is involved or on the ground that it is reasonably arguable that substantial and grave injustice has been done. These thresholds are high thresholds for the grant of leave in keeping with the Court’s primary role in the administration of criminal justice in this jurisdiction, namely to resolve real controversy on points of law of great and general importance[15]. This role, of hearing appeals only on points of law of the requisite degree of importance, is reflected also in the context of civil appeals by the recent abolition of “as of right” appeals to the Court[16]. However, neither the Ordinance nor the Rules deal expressly with the death of an appellant whilst a criminal appeal is pending. C.2 Abatement of an appeal upon death in other jurisdictions 18. The respondent relies on a line of English authorities to support the proposition that, in the absence of statutory language to the contrary, a criminal appeal is personal to a convicted person and abates upon that person’s death. 19. In R v Jefferies[17], the English Court of Appeal considered an application by the widow of a man who had given notice of application for leave to appeal to the Court of Appeal against his conviction and sentence to continue the appeal notwithstanding his death. Widgery LJ, giving the judgment of the court, said[18]: “ Whatever may be the powers of courts exercising a jurisdiction that does not derive from statute, the powers of this court are derived from, and confined to, those given by the Criminal Appeal Act of 1907. We take it to be a general principle that whenever a party to proceedings dies, the proceedings must abate, unless his personal representatives both have an interest in the subject-matter and can by virtue of the express terms of a statute (or from rules of court made by virtue of jurisdiction given by a statute) take the appropriate steps to have themselves substituted for the deceased as a party to the proceedings.” 20. This statement of general principle in Jefferies, for which no authority was cited in support, was approved by the House of Lords in R v Kearley (No.2)[19]. In Kearley, a man was convicted and sentenced and had a confiscation order made against him. An appeal to the Court of Appeal against conviction was dismissed but, on further appeal, the House of Lords quashed the convictions on most counts and remitted to the Court of Appeal the issue of whether to set aside or vary the confiscation order. Before that remitter was heard, the man died and the question arose as to whether the Court of Appeal had jurisdiction to entertain the remitted proceedings. It was held that the right of appeal under Part I of the Criminal Appeal Act 1968 was personal to the convicted person and that proceedings abated on that person’s death. In his speech, Lord Jauncey of Tullichettle said[20]: “ My Lords, as a pure matter of construction untrammelled by authority I should have had little hesitation in concluding that a right of appeal to the Court of Appeal under Part I of the Act of 1968 was personal to the convicted person. Widgery L.J. with his far greater experience in these matters reached the same conclusion in Reg. v. Jefferies which fortifies me in the conclusion which I have reached. Reg. v. Jefferies has stood unchallenged for 26 years, has been accepted as an accurate statement of the law in Reg. v. Maguire [1992] Q.B. 936 , 945, accords entirely with the natural construction of the relevant sections and, in my view, accurately states the law. I therefore reject the argument that it should be overruled and would for the foregoing reasons dismiss the appeal and answer the certified question in the negative.” 21. It is to be noted, however, that the statutory appeal provisions engaged in both Jefferies and Kearley were expressed in terms of “a person convicted” having a right to appeal against his conviction[21]. These provisions were held, as a matter of construction, to give rise to a personal right on the part of the convicted person to appeal to the Court of Appeal.[22] 22. It was recognised that these decisions could result in hardship in some circumstances and, accordingly, by the Criminal Appeal Act 1995, section 44A has been introduced into the Criminal Appeal Act 1968 to expressly empower the Court of Appeal[23] to approve the bringing or continuation of an appeal on behalf of a deceased appellant. 23. The Canadian courts, in contrast, have taken a different approach to that in the English authorities referred to above. In R v Smith[24], the Canadian Supreme Court had to consider whether the Newfoundland and Labrador Court of Appeal was right to hold that an appeal could continue after the appellant’s death. Binnie J held (at §20) that, when an appellant died, the court retained jurisdiction to proceed in the interests of justice, but that it was a jurisdiction that should be sparingly exercised. Although the statutory appeal provision in that case[25] was expressed in similar terms to the English provisions considered in Jefferies and Kearley, Binnie J (delivering the judgment of the court) construed it simply to mean that the court’s jurisdiction had to be invoked by the convicted person when still alive but that, once invoked, the court retained jurisdiction notwithstanding the person’s death. Of the relevant statutory appeal provision, he said (at §§21 and 23): “This language presupposes that at the time of the filing of the notice of appeal, the person convicted is alive and thus competent to initiate the appeal. … The continuing jurisdiction of the Court of Appeal in this case rested on the notice of appeal that was properly filed during Smith’s lifetime. … An appellant thus exercises his or her ‘personal right’ to appeal when the notice of appeal is filed. The filing is the root of the appellate court’s jurisdiction. Whether the court chooses to proceed with the appeal or not is a matter of discretion to be exercised according to the principles … hereinafter discussed.” 24. The abatement principle is applied very differently in the United States, where, in cases of appeals as of right, the principle is applied to render all proceedings from the inception of the prosecution, including the conviction, void ab initio[26]. It is, however, not necessary to consider the relevant United States jurisprudence as it was not contended on behalf of Madam Yung that this approach should be adopted in Hong Kong. C.3 Hin Lin Yee v HKSAR 25. On behalf of Madam Yung, Ms Wing Kay Po[27] submitted that in Hin Lin Yee v HKSAR[28] this Court has already held that there is jurisdiction to continue an appeal after the death of an appellant. In that case, a medical practitioner (D1) and his clinic assistant (D2) were appealing to this Court against their convictions, upheld on intermediate appeal, for certain statutory offences relating to the sale of drugs. However, after the appeal to this Court had been brought but before it could be heard, D2 died. Since D1 and D2 were jointly represented by counsel and solicitors, were advancing the same grounds of appeal and D1 was still alive at the time of the appeal, the Court permitted D2’s appeal to proceed notwithstanding her death and without any directions substituting any party for D2. In the result, D1’s appeal was dismissed unanimously and D2’s appeal was dismissed by a majority. 26. There was no argument in Hin Lin Yee as to the effect of D2’s death on her appeal. This is not surprising, given the identical interests of D1 and D2 in the appeal and the fact that D1’s appeal would proceed in any event. Ribeiro PJ touched briefly on the position of D2 at the beginning of his judgment, saying: “18. The first appellant is a medical practitioner. The second appellant, who unfortunately died recently, was his assistant at his clinic in Tung Chung. The Court will, at the request of her counsel, proceed in any event to deal with the second appellant’s appeal with which it is seized.” 27. For his part, in his partly dissenting judgment, Bokhary PJ said this: “2. What about the assistant? Tragically, she has passed away. That was after her appeal to this Court had been lodged. If a conviction is unjust, it is not rendered otherwise by an appellant’s death. On a proper understanding of access to the courts under our constitutional arrangements, this Court has a discretion to entertain an appeal even though the appellant has died since it was launched. We can entertain the appeal if justice so demands. It can be justice to the reputation of the dead, to the feelings of the living, to the finances of the estate, to the purity of the law or to all or any of those interests. The finances of the estate are not involved in the present appeal. But all those other interests are. Combined they are involved in such a way and to such a degree as to make it just that the assistant’s appeal be entertained despite her death. Of those other interests, the deceased’s reputation is by no means the least important. Reputations last longer than life. Actual conviction is graver than mere defamation. And the judiciary’s responsibility is greater, for convictions are by the judiciary.” 28. Since none of the other members of the Court dealt with the position of D2, the above paragraph in Bokhary PJ’s judgment would appear to be the primary source of the first paragraph of the headnote in the Hong Kong Court of Final Appeal Reports, which reads: “(1) An appellate court retained jurisdiction, to be exercised on a discretionary basis, to hear and determine an appeal against conviction and against pecuniary sentence. The right of appellate courts to correct potential injustice was immanent in the Basic Law and provided the necessary jurisdiction. (See paras.2, 18.)” 29. For the respondent, Mr Nicholas Cooney SC[29], submitted that this paragraph in the headnote of Hin Lin Yee is wrong and that the Court ought not to follow its decision to hear an appeal notwithstanding the death of an appellant. For her part, Ms Po submitted that Hin Lin Yee, being a decision of this Court, should not be departed from except in exceptional circumstances. She invited us to follow Hin Lin Yee in respect of the issue of jurisdiction to continue an appeal. 30. It is quite clear that the issue now under consideration was not argued or considered in detail in Hin Lin Yee and so it falls to be considered for the first time in this application. Since, for the reasons stated in Section C.1 above, the Court’s appellate jurisdiction is statutory, the correct answer to the Jurisdiction Issue is a matter of construction of the relevant provisions of the Ordinance, principally section 31. C.4 Jurisdiction to continue a criminal appeal to this Court after death 31. Section 31 of the Ordinance gives a right to “any party to the proceedings” to appeal. In context, section 31 provides the first stage in engaging the jurisdiction of the Court to hear a final appeal in a criminal cause or matter. An appeal does not lie as of right but instead lies “at the discretion of the Court”. The right, therefore, given to any party to the proceedings, is to make an application for leave to appeal (pursuant to section 33 and subject to the rules governing such application in Part II of the Rules) and the Court will then determine whether to grant leave to bring a final appeal on the grounds specified in section 32. But, once that section 31 right is exercised, the appeal is treated as pending (s.33(3)). 32. In contrast to other statutory appeal provisions, e.g. section 82 of the Criminal Procedure Ordinance[30], section 31 is not expressed as a right of appeal limited to a convicted person since the right extends also to any party to the proceedings, including the Secretary for Justice. For this reason, Jefferies, Kearley and also Smith are distinguishable in this regard, as those cases concern appeal provisions personal to a convicted person. 33. There are further distinct features of a criminal appeal to this Court. First, such an appeal is a final appeal in a court hierarchy in which there will have already been one appeal to an intermediate court of appeal. Secondly, the appeal is one subject to the grant of leave by the Appeal Committee of the Court. Leave is only granted where there is a point of law of great and general importance or it is reasonably arguable that substantial and grave injustice has occurred. Where leave to appeal is granted, there is a strong public interest in such an appeal being heard, which interest goes beyond that of an individual appellant. These features were not present in Jefferies, Kearley or Smith, which were concerned with appeals to intermediate courts of appeal. 34. Accordingly, it is necessary to construe section 31 in its context within the scheme of the Ordinance and purposively having regard to the role of the Court as the final appellate court hearing appeals on important points of law. Doing so, I conclude that, once a party to the proceedings has invoked the jurisdiction of the Court under section 31 by making an application for leave to appeal pursuant to section 33, the Court retains jurisdiction to hear a final criminal appeal notwithstanding the subsequent death of a party (be he appellant or respondent to the appeal) and has a discretion whether to do so or not. Where a convicted person, whose appeal against conviction has been dismissed by the Court of Appeal, has invoked the right whilst he is alive and competent to institute an appeal by filing an application for leave to appeal, the Court retains jurisdiction to entertain the appeal even if the appellant thereafter dies. For this reason, it was clearly within the Court’s jurisdiction to exercise its discretion to proceed to hear D2’s appeal in Hin Lin Yee. 35. This construction of section 31, permitting an appeal to continue notwithstanding an appellant’s death, is consistent with the Court’s preparedness to hear a civil appeal having regard to the public importance of the matter despite the ultimate issue having become academic as between the parties[31]. 36. I reach this conclusion on the Jurisdiction Issue as a matter of construction of the relevant statutory provisions in the Ordinance and not by reference to the constitutional arrangements regarding access to the courts. I do not, with respect, agree with the headnote in Hin Lin Yee insofar as it suggests that such jurisdiction is immanent[32] in the Basic Law or that this jurisdiction in respect of criminal appeals in this Court extends to any other appeal court. The constitutional arrangements are, of course, relevant to the context and purpose of the Ordinance and therefore to the construction of its provisions. However, jurisdiction to continue an appeal in this Court after the death of an appellant is to be found in the Ordinance alone, properly construed according to its context and purpose, and, as indicated above, this judgment only addresses criminal appeals to this Court. How this issue is to be resolved in respect of criminal appeals to intermediate courts of appeal will turn on the construction of different statutory provisions and remains to be dealt with on some other occasion. Similarly, whether there is jurisdiction (a) for the Secretary for Justice to commence a criminal appeal in this Court where an accused who has been acquitted on appeal dies before the application for leave to appeal is filed, or (b) for a criminal appeal to be commenced in this Court where the convicted person dies before invoking his right under section 31 by filing an application for leave to appeal, are questions not raised by the Jurisdiction Issue in the present case and I would leave them open to be dealt with when necessary. 37. Mr Cooney submitted that it would be “inconsistent and odd” if a right of appeal were to survive where an appellant died while appealing to this Court but to abate in the same situation in a pending appeal to the Court of Appeal. This, of course, assumes that an appeal to the Court of Appeal will abate on the appellant’s death, which may or may not be a correct assumption[33]. But, even on that assumption, I do not think it is either inconsistent or odd. The statutory appeal provisions applicable to appeals to the Court of Appeal and to this Court are different and they fall to be construed in the light of their own particular context and purpose which are quite different. This Court’s final appellate role is well-recognised[34] and its primary role to resolve real controversy on points of law of great and general importance is different to the function of an ordinary court of criminal appeal[35]. The retention of such jurisdiction even after an appellant has died is therefore far from odd. On the contrary, it would be odd if the Court did not retain that jurisdiction. 38. By whom and how the discretion is to be exercised will be addressed in Section E.1 below but it is first convenient to address the Substitution Issue. D. The Substitution Issue D.1 Is it necessary to substitute a living party for a deceased appellant? 39. In the case of a criminal appeal to this Court, neither the Ordinance nor the Rules contain express provisions requiring the substitution of a living party for a deceased party in order for a deceased appellant’s criminal appeal to continue. 40. In Smith, Binnie J considered that the deceased’s appeal became irregular upon his death because, as from that time, the appeal was in the name of a non-existent person and this irregularity ought to have been addressed by an application by the executor or personal representative of the deceased appellant to pursue the appeal in substitution for the deceased[36]. He considered that a live appellant was necessary for the appeal to be carried on and that the means for substitution of a live party for a deceased party was provided for in the relevant Newfoundland civil procedural rules which were applicable.[37] 41. In Hin Lin Yee, on the other hand, the need for substitution of a living party for D2 was not argued and, since there was an identity of D1 and D2’s interests in the appeal to this Court and they were jointly represented by counsel and solicitors, it is understandable that the Court did not think it necessary to consider whether D2’s appeal could or should properly proceed in the absence of another party in substitution for her. But there is no suggestion that the absence of such a party caused any prejudice to the case for the appellants or that for the respondent or led to any difficulties for the Court. On the particular facts of that case, the decision to allow D2’s appeal to continue without a living appellant in substitution for her was a plainly correct decision. 42. As Hin Lin Yee shows, it is not always necessary that a living party be substituted for a deceased appellant in order for an appeal to be properly argued and determined by the Court. Instead, it will be a matter of discretion in any particular case whether the Court considers that substitution of a living party for the deceased appellant is appropriate when it is exercising its jurisdiction to permit the appeal to continue after the appellant’s death. D.2 The discretion whether to order substitution and, if so, by whom 43. It may well be that, save for the situation of a joint appeal with appellants having the same interests as in Hin Lin Yee, the Court will (subject to the issue of whether there is power to do so, addressed in Section D.3 below) usually exercise its discretion to order the substitution of a living party for the deceased appellant. In Smith, Binnie J identified[38] two reasons of general application demonstrating why there is a need for a live appellant for the purposes of a criminal appeal, namely because the dead cannot give instructions and are no longer amenable to the direction of the court. The weight of these two factors will vary from case to case but they are certainly relevant to a consideration of how the discretion to permit an appeal to continue after an appellant’s death might be exercised. 44. Similarly, the question of the identity of the person to be substituted is also a matter of discretion for the Court depending on the circumstances of the case. As the reasoning of Binnie J in Smith illustrates, the most natural choice of a party to substitute a deceased appellant may be the appellant’s executor or personal representative. This is also supported by Widgery LJ’s judgment in Jefferies in which the deceased appellant’s personal representatives were identified as the party who might, depending on the existence of jurisdiction, be substituted for the deceased appellant. But it may not always be the case that, if there is to be substitution, it must be by the executor or personal representative of the deceased appellant. 45. In England, where the Criminal Appeal Act 1995 has added section 44A to the Criminal Appeal Act 1968, the Court of Appeal may approve the bringing of an appeal begun by a deceased person by that person’s widow or widower, their personal representative or “any other person appearing to the Court of Appeal to have, by reason of a family or similar relationship with the dead person, a substantial financial or other interest in the determination of a relevant appeal relating to him.”[39] The specific statutory provision therefore contemplates the pursuit of the appeal by a person not having a particular personal relationship to the deceased but rather having a substantial financial or other interest in the appeal. 46. Whilst there is no equivalent provision in the Ordinance or Rules, there may be circumstances in which a party other than the deceased appellant’s executor or personal representative is the appropriate party to be substituted for the deceased appellant where the appeal is to be continued after the appellant’s death. D.3 The Court’s power to order substitution 47. In the written submissions for Madam Yung, it was contended there are two sources of power to order substitution, namely (i) Rule 78 of the Rules[40] and (ii) an appellate court’s implied power, as explained in Taylor v Lawrence[41], to regulate its own practice and procedure. In her oral submissions, Ms Po placed primary reliance on the Court’s implied power rather than Rule 78. 48. Where a court derives its jurisdiction from a statute, it acquires by implication from the statute all powers necessary for the exercise of that jurisdiction: HKSAR v Lam Kwong Wai & Another[42]. The English Court of Appeal cases of Taylor v Lawrence and G v Home Secretary[43] similarly establish that, as a creature of statute, an appellate court has implied jurisdiction to regulate its practice and procedure which is ancillary to its statutory jurisdiction. Having decided that the Court has jurisdiction to hear an appeal notwithstanding an appellant’s death after the lodging of an application for leave to appeal (see Section C.4 above), the Court’s exercise of discretion to order substitution of a living party for the deceased appellant in order to achieve that purpose is plainly an exercise of implied power ancillary to the Court’s statutory jurisdiction. Accordingly, if the Court considers that it should exercise its discretion to order the substitution of a living party for the deceased appellant, it undoubtedly has implied power to order the substitution of Madam Yung for her former husband. 49. It is therefore unnecessary to consider the other source of power relied upon in Madam Yung’s written submissions, namely Rule 78 of the Rules. There may be difficulties in reliance on Rule 78 since that rule is expressed in general terms relating to the practice and procedure “in the Court” to be determined by the Chief Justice rather than that to be adopted in an individual case by the Court (whether the full Court or the Appeal Committee). However, as stated, it is not necessary to consider this rule in the present context. E. The Exercise of Discretion Issue E.1 The exercise of discretion in general 50. Like any other discretion of the Court, the discretion whether to continue an appeal after an appellant’s death must be exercised on a principled basis. In Smith, Binnie J identified a number of factors that should be applied in the context of an appeal to a criminal court of appeal in Canada[44]. His identification of those factors was necessarily influenced by the particular statutory provisions he was considering. In Hin Lin Yee, Bokhary PJ identified various factors that he considered might determine that justice demanded the continuation of an appeal. However, those factors were neither the subject of argument, nor were they endorsed by the other members of the Court, nor were they based on the proper jurisdictional basis of the discretion as identified above. It is therefore necessary to determine afresh the appropriate factors relevant to the exercise of the discretion. 51. Having regard, therefore, to the context and purpose of section 31 of the Ordinance (addressed in Section C above), the factors relevant to the exercise of the discretion to continue an appeal notwithstanding the death of an appellant will include the following: (1) Whether leave to appeal has been or will be granted: The Court will be guided primarily by whether leave to appeal has been granted or, if leave has not yet been granted, by the basis on which leave to appeal is sought and the likelihood of such leave being granted. Where leave to appeal is sought on a point of law, this will probably be a critical factor in any given case since the grant of leave to appeal on that ground will serve to demonstrate the Appeal Committee’s view that the appeal is of sufficient importance to be heard by the Court as the final appellate court in Hong Kong. Similarly, although not as critical as the point of law ground, the grant of leave to appeal on the basis of substantial and grave injustice will be a relevant factor, the weight of which will depend on the circumstances of the particular case. (2) The existence of an applicant: The existence of a party who wishes the appellant’s appeal to be continued and who makes an application for such continuation will usually be an essential requirement. It would be most unlikely that the Court would consider exercising the discretion to continue if there were no such party. Indeed, in the absence of such a party, the application of the provisions for dismissal of an appeal for non-prosecution[45] (where leave has been granted) will probably result in the dismissal of the appeal. (3) The continuation of the appeal on a proper adversarial basis: The Court will be concerned that the issues to be determined by it in any continued appeal are to be properly argued. The willingness of the party applying for the appeal to be continued to be substituted as appellant will be relevant in the event the Court considers such substitution to be necessary. Similarly, the Court will wish to be satisfied that the appeal will continue on a proper adversarial basis with appropriate legal representation to argue the appeal. It may wish to be satisfied that legal aid is available for the appellant’s case or that the applicant is otherwise able to instruct legal representatives for the appeal. In cases of exceptional importance, the Court may see fit to appoint an amicus curiae. (4) The interest of the applicant in the continuation of the appeal: The reason why the party applying wishes the appeal to be continued will be relevant. The financial interest of an appellant’s estate in the pending appeal may be relevant where the executors or personal representatives of the appellant apply for the appeal to be continued. If financial interest is relied upon, the nature of that interest and its amount will be relevant to the weight to be attached to this factor. Similarly, the applicant’s interest in restoring the reputation of the appellant may be relevant but this will be a factor to which varying weight will attach: the graver the offence of which the appellant has been convicted, the more weight this factor will carry. 52. The above list of factors is not intended to be an exhaustive list of the relevant factors, nor is it intended to be applied as a checklist of factors that must be present in every case. 53. In the context of an appeal to this Court, the question whether to exercise the discretion to permit an appeal to continue notwithstanding the appellant’s death may arise at different stages. This is because the jurisdiction will arise as soon as a living applicant makes an application for leave to appeal pursuant to section 33 of the Ordinance and if, unfortunately he should die, this may occur at any time thereafter (including before the Registrar has decided whether to issue a Rule 7 summons or before the Appeal Committee has considered any submissions served in response to such a summons or before the Appeal Committee has determined the application for leave to appeal). As provided in section 33(3), an appeal to the Court is treated as pending until any application for leave to appeal is disposed of. 54. For the reasons stated above, the grant of leave to appeal will likely be a critical factor. Given the importance of the grant of leave to the exercise of this discretion, where a putative appellant dies after filing his notice of application for leave to appeal but before that application is heard, the question of whether the appeal should be continued is one which should be decided by the Appeal Committee. Thus, if the Registrar is minded to issue a Rule 7 summons, he should direct the party applying for the appeal to be continued to show cause both as to why the application should not be dismissed and why, also, if the leave application is to be heard by the Appeal Committee, the appeal should be continued notwithstanding the putative appellant’s death. The Appeal Committee will then decide both whether the application for leave to appeal should be heard and, if so, whether the applicant should be heard on behalf of the putative appellant. If the Registrar is not minded to issue a Rule 7 summons, the Appeal Committee will have to decide whether, if it is minded to grant leave to appeal, to exercise its discretion to allow the appeal to continue. 55. Similarly, if the appellant dies after leave to appeal has been granted, an application to continue the appeal, although one which a single permanent judge of the Court would have jurisdiction to determine[46], should be heard by the Appeal Committee which, having granted leave, should have the opportunity to take the change of circumstances into account. For this reason, the summons seeking the continuation of the appeal should be made returnable before the Appeal Committee. The present application has been heard by the full Court for the particular reason explained above and, in future cases, it will not normally be necessary for a similar application to be heard by the full Court. E.2 The exercise of discretion in this case 56. In the present case, the Appeal Committee certified that a point of law of great and general importance was involved in the decision of the Court of Appeal and granted leave to appeal on that ground and also on the ground that it was reasonably arguable that substantial and grave injustice has been done in respect of the appellant. The appeal is therefore demonstrably of the requisite importance to be heard by the Court and, as indicated above, this is a critical factor supporting the exercise of discretion to permit the appeal to continue. 57. Madam Yung is a willing applicant and there is no reason to think her wish for the appeal to continue is anything but genuine. As the widow of the appellant, she has a substantive relational interest in the continuation of the appeal in order to overturn her former husband’s conviction. The appellant was convicted of a serious criminal offence involving fraud, the gravity of which is reflected in the sentence of imprisonment imposed. Although the appellant fully served his sentence and his disqualification order became moot upon his death, the reputational harm due to his conviction remains. 58. In her application, Madam Yung asserted, as a ground for the exercise of discretion to permit her to continue the appeal, that the appellant’s estate would most likely suffer “significant financial loss”[47]. She referred to three civil claims or potential claims against the appellant’s estate in which his conviction is said to be relevant. The value of one of those claims is said to be HK$6,580,000. The existence and viability of these claims against the estate was disputed by the respondent but it is unnecessary to resolve this dispute since this factor, of significant financial loss to the appellant’s estate, must be discounted. This is because, in the Schedule of Assets and Liabilities of the appellant annexed to the Letters of Administration granted to Madam Yung, the appellant’s estate is shown only to have assets of approximately HK$60,000 so that, assuming the claims against it to be valid, the estate is technically insolvent. In any event, as against those assets of HK$60,000, the schedule refers to liabilities in the form of legal costs due to the appellant’s solicitors in the criminal proceedings giving rise to this appeal. Although no details are given, it is a safe assumption that such costs will have already exceeded the stated assets of the estate so that, on that basis also, the estate is insolvent. 59. Nevertheless, taking all the relevant factors into account, in particular the critical factor of the grant of leave to appeal, this is a case in which it would be appropriate to exercise the discretion to permit the appeal to continue notwithstanding the appellant’s death. 60. In the present case, unless Madam Yung or some other natural person is substituted for the appellant, there would be no other party capable of giving instructions or of being amenable to the direction of the Court (in relation, for example, to an order for costs of the appeal). This suggests that it would be appropriate to order the substitution of some other party for the deceased appellant. As indicated above, the most natural person to substitute for a deceased appellant will usually be the appellant’s executor or personal representative. Here, since (i) Madam Yung has now been granted letters of administration in respect of the appellant’s estate, (ii) she wishes the appeal to be prosecuted and (iii) no other party has sought to be substituted or wishes to pursue the appeal on behalf of the deceased appellant, Madam Yung is the most obvious person to be substituted for the appellant. Moreover, she was represented in this application by the same counsel and solicitors who have been representing the appellant on his appeal to this Court. Preparation for the hearing of the appeal had already reached an advanced stage at the time of the appellant’s death; the parties’ printed cases for the appeal had already been filed and the appeal was ready to be heard. If Madam Yung were substituted for the appellant, there is no doubt that the appeal would proceed on a proper adversarial basis. 61. For these reasons, it would therefore be appropriate to make an order that she be substituted for the appellant and that the appeal be continued in her name in her capacity as the personal representative of the appellant’s estate. The joinder of Madam Yung to the appeal in that capacity will enable the Court to consider, depending on the outcome of the appeal, its discretion in relation to costs as against the estate[48]. Although the argument that the estate has a financial interest in the continuation of the appeal has been discounted on the basis of the assets declared, this is not to be taken to preclude the prosecution proving that assets exist in the event this becomes relevant in the context of costs or otherwise. F. Disposition 62. For the above reasons, I would exercise the Court’s discretion in favour of Madam Yung and order, notwithstanding the death of the appellant, that: (1) Madam Yung be made a party to the appeal in her capacity as the personal representative of the estate of Cheng Chee-tock Theodore in substitution for the appellant; and (2) the appeal in FACC 7/2014 be continued in the name of Madam Yung in such capacity aforesaid. 63. Finally, I would direct that any arguments as regards the costs of this application be reserved to the hearing of the appeal. Sir Anthony Mason NPJ: 64. I agree with the judgment of Mr Justice Fok PJ. Ms Wing Kay Po, Ms Doris Li and Mr Newton Mak,instructed by Peter K.S. Chan & Co., for the Appellant and the Applicant Mr Nicholas Cooney, SC, on fiat and Miss Jessie Sham, SPP (Ag) of the Department of Justice, for the Respondent [1] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 [2] In DCCC 476/2011. [3] Pursuant to s.168E of the Companies Ordinance (Cap.32). [4] Stock VP, Lunn JA and Barnes J (CACC 460/2012) [5] Ma CJ, Ribeiro & Tang PJJ (FAMC 23/2014) [6] Viz., “Where an asset is sold by a company (X Co) to a listed issuer (Y Co), do the ‘panoply of the rights of shareholders, including that in respect of the receipt of dividends’, either individually or cumulatively, constitute an interest in the assets of X Co which is sufficient to render any acquisition of any of those assets by Y Co a ‘transaction’ between A and Y Co such that the transaction is a “connected transaction” within the meaning of Rule 14A.13(1)(a) of the Rules Governing the Listing of Securities on The Stock Exchange of Hong Kong Limited (“the Listing Rules”), if A is a ‘connected person’ to Y Co within the meaning of Rule 14A.11(1) of the Listing Rules?” [7] By Ribeiro PJ, on 23 January 2015. [8] On 23 February 2015, Madam Yung applied to amend her summons to seek an order that the appeal be continued on such terms as the Court thinks fit. Although initially opposed by the respondent, Mr Nicholas Cooney SC, leading counsel for the respondent, indicated at the outset of the hearing that the respondent did not object to the amendment to the summons. [9] For this reason, the letter to the Registrar from Mr Cooney, received after the conclusion of the hearing, drawing the Court’s attention to instances in which the Court of Appeal has treated an appeal as abated upon death was not relevant. [10] (2003) 6 HKCFAR 570 at §29 [11] Ibid. at §31 [12] (Cap.484) (“the Ordinance”) [13] Solicitor v Law Society of Hong Kong & Secretary for Justice (supra) at §35 [14] (Cap.484A) (“the Rules”) [15] So Yiu Fung v HKSAR (1999) 2 HKCFAR 539 at pp.541I-542B & p.543E-H. [16] Administration of Justice (Miscellaneous Provisions) Ordinance 2014, s.8 [17] [1969] 1 QB 120 [18] At p.124B-D [19] [1994] 2 AC 414 [20] At p.422F-G [21] In Jefferies, s.3 of the Criminal Appeal Act 1907 (“A person convicted on indictment may appeal under this Act to the Court of Appeal … (a) against his conviction…”); in Kearley, s.1(1) of the Criminal Appeal Act 1968 (“… a person convicted of an offence on indictment may appeal to the Court of Appeal against his conviction …”). [22] A different conclusion was reached in R v. Maguire [1992] QB 936, where the English Court of Appeal construed section 17 of the Criminal Appeal Act 1968 as permitting the case of a deceased person to be referred to the court by the Home Secretary and that, after referral, it was to be treated as an appeal by that person (see p.947D). [23] By section 44A(2)(b), this power also applies in respect of appeals to the UK Supreme Court. [24] [2004] 1 SCR 385 [25] Criminal Code, RSC 1985, c.C-46, s.675(1): “[a] person who is convicted … may appeal to the court of appeal … against his conviction”. [26] Durham v United States, 401 US 481 (1971); the position appears to be different for appeals heard on a discretionary basis (where the appeal abates but the conviction is left intact), see Dove v United States, 423 US 325 (1976). [27] Leading Ms Doris Li and Mr Newton Mak [28] (2010) 13 HKCFAR 142 [29] Leading Miss Jessie Sham, SPP (Ag) [30] (Cap.221), governing criminal appeals to the Court of Appeal, which provides: “A person convicted of an offence on indictment may appeal to the Court of Appeal against his conviction.” [31] See e.g. Secretary for Security v Sakthevel Prabakar (2003) 6 HKCFAR 397 at §42. [32] Existing or operating within; inherent. [33] Any inconsistency or oddity would be removed if the relevant statutory provisions governing intermediate appeals were to be construed in a manner similar to section 31 of the Ordinance. As indicated above, this is an open question not addressed on this appeal but this result is one which the reasoning in Smith supports. [34] Solicitor v Law Society of Hong Kong & Secretary for Justice (supra) [35] So Yiu Fung v HKSAR (supra) [36] At §26 [37] At §§27-29 [38] At §26 [39] Criminal Appeal Act 1968, section 44A(1) and (3). Section 44A(1) provides: “(1) Where a person has died – (a) any relevant appeal which might have been begun by him had he remained alive may be begun by a person approved by the Court of Appeal; …”; and section 44A(3) provides: “(3)Approval for the purposes of this section may only be given to – (a) the widow or widower of the dead person; (b) a person who is the personal representative (within the meaning of section 55(1)(xi) of the Administration of Estates Act 1925) of the dead person; or (c) any other person appearing to the Court of Appeal to have, by reason of a family or similar relationship with the dead person, a substantial financial or other interest in the determination of a relevant appeal relating to him.” [40] This provides: “In any matter not provided for in these Rules, the practice and procedure in the Court shall be such as may be decided by the Chief Justice who may, if he thinks fit, be guided by the practice and procedure of the High Court.” [41] [2003] QB 528, per Lord Woolf CJ at §§17, 26 & 50 [42] (2006) 9 HKCFAR 574 at §§67 to 71, esp. at §69 [43] [2004] 1 WLR 1349, per Lord Phillips of Worth Matravers at §13 [44] At §50 [45] In Rule 18 of the Rules. [46] See section 46(2); such decision of a single permanent judge being subject to review by the Appeal Committee under section 46(3). [47] Applicant’s Affirmation, 20.1.15, §9 [48] Recently discussed in HKSAR v Wong Tak Keung (FACC 8/2014), Judgment on Costs, 20 April 2015. 
Chief Justice Ma: 1. At the conclusion of the hearing, we dismissed the appeal and indicated that we would give our reasons later. They are now set out in the judgment of Mr Justice Chan PJ. Mr Justice Chan PJ: 2. The appellant was convicted after trial in the District Court (H H Judge Geiser) of 5 charges: aiding, abetting, counseling and procuring the making of child pornography (charge 1), criminal intimidation (charges 2 and 4) and indecent assault (charges 3 and 5). His appeal to the Court of Appeal (Cheung and Hartmann JJA and Barnes J) was dismissed. We are now concerned only with the 5th charge, leave to appeal having been refused in respect of the first 4 charges. The issue in this appeal 3. The appellant relies on both the point of law (what, if any, relevance the lateness of a notice of alibi has on the weight of the alibi evidence) and the substantial and grave injustice limbs. However, in essence, the question to be determined in the present appeal is whether the trial judge had properly considered the alibi evidence given on behalf of the appellant in relation to the 5th charge. The Prosecution case 4. Although we are only concerned with the 5th charge, it is pertinent to refer, as a matter of background, to the events leading to the other 4 charges. In the summer of 2003, the complainant (Miss X) was then only 12 and in primary 6. Through the ICQ network, she came to know the appellant who was then 24. Initially they discussed trivial matters but in their later contacts, the appellant asked her whether she had any boyfriend and whether she had any sexual experience. He then succeeded in persuading her to take two naked photographs of herself (with one showing her face) and have them sent to him (the 1st charge). 5. Several months later, in March or April 2004, the appellant contacted Miss X and asked to meet her but was refused. He then threatened to expose her naked photographs to the media and on the internet and to send them to her father if she refused (the 2nd charge). She reluctantly agreed and they met. He took her to what he said was his home where he indecently assaulted her and forced her to masturbate him and perform oral sex on him. He also took photographs of her in the nude against her will (the 3rd charge). 6. In July 2005, more than a year later, the appellant called Miss X at home asking her why she did not give him her new mobile phone number. He asked to see her but she said she did not want to see him again as he had made her unhappy. Upon his threat to reveal her naked photographs to the media and on the internet and send them to her father, she agreed to meet him (the 4th charge). 7. They met on 5 August 2005. She said it was between 12 noon and 3 p.m. He took her to a nearby hourly hotel where he indecently assaulted her by asking her to undress and forcing her to masturbate him and perform oral sex on him. He further demanded to have sexual intercourse which she refused. She was then told to kneel on the bed with her back to the appellant who moved his penis in between her thighs and she was again forced to masturbate him and perform oral sex on him. He also took naked photographs of her. This constituted the 5th charge. She said she could remember that date because it was her sister’s birthday. The defence case 8. The appellant did not give evidence at the trial but it was not denied that he had contacts with Miss X through the ICQ network. His defence through questions put to her during cross examination was that he did not discuss sexual topics with her; nor did he ask her to take naked photographs or threaten her or indecently assault her. It was put to her that she fabricated her evidence against him and that if what she said had happened, it happened with someone else. It was also suggested that the relationship between the appellant and Miss X over the internet developed into an intimate relationship in December 2008 but turned sour when she found out subsequently that the appellant had become engaged and that it was out of spite or jealousy that she reported to the police and fabricated these allegations against him. These suggestions were strongly denied by Miss X. There was no evidence in support of these suggestions and the judge rejected them. The alibi evidence 9. In respect of the 5th charge, apart from the defence that the allegations against him were fabricated, the appellant also relied on alibi. It must be noted that there was no dispute that the appellant must have known for almost 6 months the alleged date and time of the indecent assault in relation to the 5th charge: the date was mentioned in Miss X’s 1st witness statement (22 March 2010) and the time was given in her 2nd witness statement (23 March 2010). Further, no specific date was given in the particulars of the first 4 charges and it was only in relation to the 5th charge that the particular date of the alleged offence was specified. However, no alibi notice was served in accordance with s.65D of the Criminal Procedure Ordinance, Cap221 before the commencement of the trial. 10. In the afternoon on the 2nd day of the trial, during the cross examination of Miss X, counsel for the appellant applied to the trial judge for leave to serve an alibi notice and informed the judge that he (counsel) had been told in that morning “that the defendant actually has an alibi for that day (5 August 2005) which I (counsel) was unaware of until today.” The application was supported by an affirmation made by the appellant’s wife. As the prosecution did not resist the application, the judge granted him leave to lead alibi evidence. 11. Since the appellant did not testify, there was no evidence from him as to where he was on 5 August 2005, especially in the afternoon. His wife’s elder brother (DW1) was called pursuant to the alibi notice. DW1 said that in 2005, he came to know the appellant who was then his sister’s boyfriend. On 5 August 2005, he was told by his mother to go to the Buddhist Hospital to visit his maternal grandfather who was dying. When he arrived at the hospital, he saw his sister and the appellant there. They stayed from 12.30 p.m. until 2 p.m. and left to have lunch. Shortly after that, DW1 also left. He returned to the hospital later that evening after he was informed that his grandfather had passed away. The suggestion was that the appellant was not with Miss X from 12 noon to 3 p.m. on 5 August 2005 as alleged by her. If this alibi was or might be true, this would cast doubt on Miss X’s evidence. The complainant’s credibility 12. As the judge noted, the crucial issue in the case was the credibility of Miss X. With regard to her evidence, the judge, having considered all the evidence, found her to be an impressive witness and was quite satisfied that she was entirely truthful. He gave his reasons for doing so in the Reasons for Verdict: “16. For myself, I say that I found her to be a most impressive witness who gave compelling evidence of these distressing events which form the subject matter of these five charges. Despite the lapse of time since the incidents occurred, she was able to give a detailed description as to what by any account must have been extremely unpleasant and embarrassing for her. I found her to be entirely truthful and reject out of hand the suggestion that she has fabricated the events she testified to and deliberately lied to this court. Of course, it is right to say that there are some inconsistencies in her evidence and the statements that she made to the police but I find that those that do exist to be perfectly understandable given the nature of the case and the lapse of time since the offences occurred and I further find that they in no way undermine her general credibility.” The way the judge dealt with the alibi evidence 13. The judge did not accept the alibi evidence of DW1. He set out his conclusion as follows: “17. As to the evidence of DW1, Mr Ma Kai-yin, the brother-in-law of the defendant, I attach little or no weight to it. The application to produce alibi evidence of this nature was made after Miss X had given evidence under cross-examination that the assault which is the subject matter of charge 5 took place on 5 August 2005 between 12 noon and 3 pm. No proper explanation has been put forward as to the reasons for the lateness of this application. In any event, I have already accepted the evidence of Miss X that in relation to charge 5 she was the subject of a gross indecent assault at the hands of the defendant between 12 noon and 3 pm and to the extent that DW1 testified that between 12.30 pm and 2 pm the defendant was at the Buddhist Hospital in Wong Tai Sin, I reject it. As to precisely what time the defendant may have been at the hospital that day, if he ever was, I cannot speculate.” 14. What the judge exactly meant in this paragraph and whether his approach to the alibi evidence was erroneous is very much the bone of contention in this case. The appellant made a number of criticisms on this paragraph as one of the grounds in his appeal to the Court of Appeal (which is now the main ground in the present appeal). The Court of Appeal rejected all his criticisms. They were satisfied that what the judge meant to convey in that paragraph was that he found Miss X’s evidence credible and that the alibi evidence did not throw doubt on her evidence. The Court of Appeal also said that the fact the prosecution did not resist the late application and did not put to DW1 that he was either lying or mistaken did not mean that the judge was bound to accept the alibi evidence. Criticisms against the judge’s decision 15. In submitting that the conviction on the 5th charge cannot stand, Mr McGowan on behalf of the appellant makes in effect two main submissions which can be summarized as follows: (1) The judge was wrong to attach little or no weight to the alibi evidence solely on the ground that the alibi notice was filed late. Having granted leave to the appellant to serve an alibi notice, the judge should not have rejected such evidence on this basis. If he had taken the view that it was necessary, whenever there was a late alibi notice, for the defence to provide an explanation for the delay, this was wrong, being inconsistent with the requirement that it is for the prosecution to disprove an alibi. If the judge had regarded the absence of explanation for the delay in serving an alibi notice as a relevant consideration, then he should have raised this with the parties. By not doing so, he had deprived the defence of an opportunity to adduce evidence (such as to re-examine DW1 or call the wife) and offer an explanation on this matter. (“The lateness argument”.) (2) The judge was wrong to have rejected DW1’s evidence. First, it was never put to DW1 in his cross-examination that his alibi evidence was being challenged. This was a breach of the rule in Browne v Dunn (1894) 6 R 67. Secondly, the judge had failed to properly consider DW1’s evidence and the reason given by him seemed to indicate that he had already accepted Miss X’s evidence even before considering the alibi evidence. This was an entirely wrong approach. (“The improper rejection argument”.) 16. It is submitted that the judge, in making these errors, had departed from accepted norms as a result of which the appellant has suffered substantial and grave injustice. Counsel further submits that if the conviction on the 5th charge is quashed on the ground that the alibi evidence is true or may be true, this would affect Miss X’s overall credibility and since the whole case depends on her credibility, the convictions on the other 4 charges must also be set aside, notwithstanding that no leave has been granted in respect of these charges. The lateness argument 17. This argument must be rejected. In the third and fourth sentences of paragraph 17 of the Reasons for Verdict, the judge referred to the fact that the alibi notice was filed late and there was no explanation for the lateness of the application. However, when these sentences are read in context, especially with the use of the phrase “in any event” in the sentence which immediately followed, it is quite clear that the lateness of the alibi notice and the absence of explanation was, contrary to counsel’s contention, not the main, let alone the sole, reason for the judge’s rejection of the alibi evidence adduced on behalf of the appellant. It is a fact that the appellant, unlike his wife and DW1, had known of the alleged date and time of the offence under the 5th charge for nearly 6 months. There might be an acceptable reason for the appellant not being able to recall the importance of this date and time. But the fact remains that he did not give evidence and his wife and DW1 were obviously not in a position to speak on his behalf. This is a matter which, as accepted by counsel, the judge was entitled to take into account. But whether there might or might not be a reason for the lateness of the application, it is quite obvious that this was not a decisive factor for the judge’s rejection of the alibi evidence. 18. Secondly, whether an alibi notice should be allowed to be served late is a matter of discretion for the judge in deciding whether the alibi evidence should be admitted. This is quite separate from the question whether the alibi evidence if allowed to be adduced should be given any weight which is a matter for the tribunal of fact. But the fact that an alibi notice is served late and there is no satisfactory explanation for the delay is a relevant factor in deciding the second question, i.e. what weight, if any, can be attached to the alibi evidence. The answer to this question clearly depends on the facts and circumstances of each case. However, what is of greater importance must ultimately be the cogency of the prosecution evidence and the quality of the alibi evidence. There is no question of shifting the burden to the accused to prove his alibi or relieving the prosecution of its burden of disproving it. The improper rejection argument 19. In respect of this argument, it is first said that there was a breach of the rule in Browne v Dunn in that neither the prosecution nor the judge had put to DW1 that his evidence was being impeached. There is nothing in this point. The Browne v Dunn rule is to ensure fairness to a witness whose evidence or any of the points to which he has testified is being questioned: he should be told that he is not to be believed on his evidence or on the point in question so that he can have an opportunity to offer an explanation unless it is obvious to him that his evidence is being challenged. (See Lord Herschell LC at p.71 and Lord Halsbury, p.76.) In the present case, it cannot be said that DW1 could have been under any misapprehension that his evidence was not being challenged. It must have been quite plain to him from the way questions were asked when he was cross-examined that his evidence was being tested and that the prosecution was impeaching his reliability if not also his credibility. There is nothing unfair to him by not putting directly to him that his evidence was not to be accepted. 20. The main basis of counsel’s complaint arises from paragraph 17 of the Reasons for Verdict. With respect to the trial judge, it can be argued that this paragraph is not clearly expressed and may understandably be open to criticism. But this paragraph must not be read in isolation. When it is read in the light of the rest of the judgment, particularly the paragraphs preceding it, what the judge meant is, in our view, sufficiently clear. 21. It can readily be seen from the Reasons for Verdict that the judge had carefully considered all the evidence adduced before him and fairly assessed Miss X’s credibility, bearing in mind the alibi evidence. He had described her evidence in detail (paragraphs 3 to 8) and considered what was put to her in cross-examination and the inconsistencies which had been drawn to his attention (paragraphs 9 to 11). He then referred to the alibi evidence given by DW1 in paragraph 14. It was only after he had done that that he came to the conclusion in paragraph 16 that Miss X was impressive and compelling and that her evidence was entirely truthful. Thus it cannot be said that the judge had adopted a wrong approach to the evidence and had come to accept Miss X’s evidence without having regard to the defence case and DW1’s evidence. The judge must have been satisfied that the alibi evidence did not cast any doubt on Miss X’s evidence before he accepted her testimony. When he said in paragraph 17 that he had “already accepted the evidence of Miss X”, it would seem that he was merely trying to make reference to his earlier conclusion with regard to Miss X’s evidence in the immediately preceding paragraph. 22. The judge had given good reasons for believing that Miss X was telling the truth notwithstanding her cross-examination and the alibi evidence. With regard to DW1’s evidence, as Ms Agnes Chan for the prosecution has pointed out in paragraph 45 of her written case, which is not really disputed, the quality of such evidence was far from satisfactory: apart from the singularly material fact which was in dispute (i.e. he saw the appellant in the hospital from 12.30 p.m. to 2 p.m. on 5 August 2005), he could not recall any other detail such as what the appellant had done in the hospital during that period and whether he saw the appellant again in the later part of that day when he went to the hospital after he had been informed of the death of his grandfather. It is not surprising that the judge rejected his evidence. As noted by the Court of Appeal, the judge was justified on the evidence in accepting Miss X’s evidence and rejecting DW1’s alibi evidence. 23. The judge had heard Miss X (who had given evidence for 3 days) and DW1 and had carefully evaluated her evidence having regard to the latter’s alibi evidence. Unless it can be shown that he had adopted an erroneous approach to the evidence, his assessment of the credibility and reliability of the witnesses should not be disturbed. We are not persuaded that the judge had taken a wrong approach towards the alibi evidence. There was no departure from accepted norms to the prejudice of the appellant. The conviction on the 5th charge must stand. Conclusion 24. For the reasons given above, there is no merit in this appeal which must be dismissed. Mr James H M McGowan, instructed by Jimmie K S Wong & Partners, for the appellant Miss Agnes Chan, ADPP of the Department of Justice, for the respondent 
Mr Justice Fok PJ: 1. This uncontested appeal turns on the meaning of the word “tampers” in section 49 of the Road Traffic Ordinance[1]. It is an example, unfortunately, of how in an adversarial system, particularly where an individual is initially unrepresented at trial, there are cases in which a material point of law is not raised in the courts below. Fortunately, however, with the assistance of pro bono legal representation[2], the appellant’s case has been pursued to this court and, for the reasons that follow, a wrongly convicted appellant has been able to overturn that conviction on a new point of law raised for the first time in this court. That he had, in the meantime, to serve a sentence of six weeks’ imprisonment for the offence of which he was convicted is naturally most regrettable but this could have been mitigated by a more timely appeal coupled with an application for bail. The Facts 2. The incident giving rise to the single charge preferred against the appellant occurred in the afternoon of 21 February 2014. On that day, at about 4pm, a van driver (who gave evidence for the prosecution as PW1) was delivering goods from his light goods vehicle (bearing registration number SL3521) whilst he was parked in Tsuen Wan. From a distance of about 10 feet behind the vehicle, PW1 saw through the open rear door of the vehicle and its left rear window a man, later identified as the appellant, closing the left front passenger door of the vehicle. Being suspicious, PW1 immediately went to the vehicle and checked the front seat where he had left his mobile phone. This, he discovered, was missing. Suspecting the appellant of having stolen his mobile phone, he chased the appellant and intercepted him about 30 feet away from the vehicle. The matter was reported to the police and the appellant was arrested for theft. There was, however, no evidence to support a charge of theft. The phone was not found on the appellant and there was no evidence that he had taken the phone. The course of the proceedings 3. The appellant was charged with one count of tampering with a motor vehicle contrary to section 49 of the Ordinance (which is set out below). The particulars of the offence charged that the appellant “did, without lawful authority or reasonable excuse, tamper with the nearside front passenger-side door of a vehicle, namely, a light goods vehicle displaying registration mark SL3521”. 4. The appellant was tried in Tsuen Wan Magistrates’ Court before Deputy Magistrate Jim Chun-ki[3] on 22 May 2014. The point now forming the basis of this appeal was not advanced by the appellant, who was unrepresented at trial. Instead, the Deputy Magistrate addressed the issues of whether PW1’s evidence was honest and reliable, whether PW1 really saw the appellant close the nearside front passenger door of the vehicle and whether the appellant, being the person intercepted by PW1, was the person who had closed the nearside front passenger door of the vehicle[4]. He found in favour of the prosecution and against the appellant on those issues and, satisfied of his guilt beyond reasonable doubt, convicted the appellant of the offence of tampering with the vehicle. On 5 June 2014, the Deputy Magistrate sentenced the appellant to six weeks’ imprisonment but granted bail pending appeal. 5. The appellant appealed to the Court of First Instance[5]. He was represented by counsel assigned by the Director of Legal Aid for the appeal but none of the issues raised included the issue of law as to the meaning of “tampers” which is now raised on this appeal. The Judge[6] dismissed the appellant’s appeal on 17 December 2014 and he then proceeded to serve his sentence of imprisonment. 6. On 22 April 2015, by this time assisted by the pro bono legal representation already referred to, the appellant applied to this Court for leave to appeal out of time. The application was initially subject to a Rule 7 summons[7] issued by the Registrar of this Court for the appellant to show cause why the application should not be dismissed. Prior to the Appeal Committee determining the application subject to the Rule 7 procedure, the prosecution indicated it was prepared to consent to the application for leave with a view to consenting to the appeal itself on the ground of appeal now advanced. For the reasons set out more fully in its Determination dated 3 July 2015[8], the Appeal Committee granted leave to appeal on the ground that the mere act of the appellant’s closing the door of the vehicle could not support a charge of tampering with a motor vehicle but dismissed the balance of the application for leave to appeal. Directions were given, pursuant to the procedure set out in HKSAR v Shum Wan Foon[9] for the filing by the parties of a Joint Case. 7. Having considered the Joint Case, the Court is prepared to dispose of the appeal on the papers and without an oral hearing. This is the judgment of the Court on the appeal. The meaning of “tampers” in section 49 of the Ordinance 8. Section 49 of the Ordinance provides: “If a person otherwise than with lawful authority or reasonable excuse gets on to a vehicle or tampers with any part of the vehicle, he commits an offence and is liable to a fine of $5,000 and to imprisonment for 12 months.” 9. The offence is based on section 29 of the UK Road Traffic Act 1972[10] and was originally introduced by the Road Traffic Bill 1982[11] which replaced the Road Traffic Ordinance 1957 (Cap.220) with the Ordinance. The offence was then new to the law of Hong Kong[12]. 10. The word “tampers” in section 49 is not defined in the Ordinance and, as will be seen, no court in Hong Kong has sought to provide a definition of it[13]. The word is not a technical term, nor is there any indication that it used as such in the Ordinance. The dictionary definition of “tamper” includes the following meanings of the word: “2.… Foll. by with: meddle or interfere with so as to cause alteration or harm; make unauthorized changes in. … 3. … Bias, influence, corrupt; meddle with, alter improperly.”[14] 11. Section 49 is in Part V of the Ordinance, which is headed “Traffic Offences” and includes a number of separate offences. Section 49 is a discrete offence committed by the act of: (a) getting on to a vehicle (i.e. including entering into or climbing onto a vehicle), or (b) tampering with any part of it, otherwise than with (c) lawful authority (i.e. permission of the vehicle’s owner), or (d) reasonable excuse (e.g. some emergency or other necessity justifying the particular act). The purpose of the section is self-evidently to deter persons from doing certain acts in relation to vehicles unless those acts are done with lawful authority or reasonable excuse. The nature of the acts prohibited, namely getting on to a vehicle or tampering “with any part of the vehicle”, indicate that the statutory purpose is designed to afford broad protection to owners and users of vehicles or those who might be affected by their use (such as other road users or passengers). 12. The section does not use the words “interferes”[15], “meddles” or “touches” but instead uses “tampers”, which as indicated by its dictionary meaning, connotes something more than the mere interference or meddling with, or touching of, that part and implies that there is something improper in the act done. The quality of that impropriety, again following the dictionary definition of the word “tamper”, should be such as to cause alteration or harm to, or to bring about unauthorised change in, the thing tampered with. Therefore, tampering within section 49 means an act constituting either interference or meddling with part of a vehicle so as to cause alteration or harm to it, or the making of an unauthorised change to it. Beyond this broad definition, it is not necessary for the purposes of this appeal to seek to give a more specific definition of what acts may amount to tampering under the section. Since, (i) the word “tampers” is a word used in a general sense which may embrace a large range of different acts, (ii) those acts may be done in respect of any vehicle of which there are many different types, and (iii) the offence may be committed by tampering “with any part of the vehicle”, there will necessarily be many acts which could amount to improper meddling or interference with a particular part of a vehicle, or the making of an unauthorised change in it. 13. As already noted, no Hong Kong court has attempted to define what amounts to tampering with a vehicle. In The Attorney-General v Choi Wah Hang & Anor.[16], the Court of Appeal addressed the question of whether the section 49 offence was one of specific intent such that the defence of self-induced intoxication was available as a defence. The court held that the offence was not one of specific intent so that that defence was not available. Cons V-P[17], delivering the judgment of the court, said (at p.106B-D): “We do not propose to attempt an exhaustive definition of the word ‘tamper’, which is a common and well understood word of the English language. There may be occasions where it will not necessarily be easy to say whether particular conduct amounts to tampering or not. But we are satisfied that there is no reason whatsoever to add to its basic meaning the further requirement that [counsel for the 2nd respondent] suggests [i.e. that the word ‘tamper’ connotes an intention to do something which is disadvantageous to the owner of the vehicle or alternatively has that effect].” 14. Unfortunately, the judgment in The Attorney-General v Choi Wah Hang & Anor. does not indicate the facts on which the magistrate had found that the two defendants had tampered with the vehicle in question and so the case does not materially assist in providing a definition of the word “tampers”. 15. The parties’ Joint Case has identified other Hong Kong cases involving convictions under section 49 for tampering with a vehicle. These, however, do not offer any definition of the word “tampers” in that section, nor do they provide any analysis of what that word means. They may be referred to in order to illustrate what other courts have previously considered constituted an act of tampering with a vehicle. Thus, in HKSAR v Wong Fuk & Ors[18], the defendants were convicted of the offence where the evidence disclosed they had been seen to use a screwdriver to prise the lock off a light goods vehicle. They had done this in order to replace the lock with another one, unbeknownst to the owner of the vehicle, and then to make a copy of the key of the stolen lock so that they could later return and, using that copied key, steal the vehicle. In HKSAR v Ansir Mahmood[19], the defendant was convicted of the offence where he had acted as lookout for his employee who used a screwdriver to prise the door of a goods vehicle. The Queen v Wong Wing Ching[20] was an appeal against sentence in which the appellant had been convicted of the offence where the evidence disclosed he had been seen squatting near the front of a goods vehicle, whilst another man tampered with the dash board of the vehicle. The vehicle was found, on examination, to have had one of its windows smashed, the lock prised open and the electrical wiring under the dash board pulled out. 16. The fact that the acts in those other cases have previously been held to be within section 49 of the Ordinance does not assist in providing a definition of what is meant by tampering but the acts in question would clearly fall within the broad definition set out in paragraph 12 above. 17. The broad definition of “tampers” set out in paragraph 12 above is supported by a decision of the Supreme Court of Victoria in Harris v Sumner[21], a case concerning a charge of tampering with a motor vehicle under section 82A of the Motor Car Act 1958[22]. The evidence in that case was that the defendant walked around the car in question, paying close attention to the panels; he opened the front near side door, looked inside the car and then closed the door; he then rocked the car by holding it by the body above one of the rear wheels. On appeal after conviction by the magistrates, Menhennitt J held[23]: “ As a matter of ordinary language it appears to me that the word ‘tampers’ involves the concepts [sic] that a person interferes in some way with an object so as to change it. This appears to me to accord with the fourth definition in the Shorter Oxford English Dictionary which is: ‘To meddle or interfere with (a thing) so as to misuse, alter, corrupt or pervert it.’ Underlying those latter words is, I think, a concept of change or alteration. The word ‘interferes’, in my view, does not involve necessarily the same concepts, and it is unnecessary and undesirable for me to attempt to define the word ‘interferes’ in the context. But I reiterate, in my view the word ‘tampers’ means interference which produces some change.” Since the evidence did not indicate that the defendant did anything to change the vehicle, the judge quashed the conviction. 18. The parties’ Joint Case referred to two other Australian decisions[24] in which Harris v Sumner was considered but, since those decisions concern the meaning of the word “interferes” in the particular statutory provisions, they do not assist in the determination of the issue raised in this appeal. 19. The parties’ Joint Case also referred to the English Crown Court decision in Reynolds and Warren v Metropolitan Police[25] concerning section 9 of the Criminal Attempts Act 1981 under which a person commits an offence if he interferes with a motor vehicle or trailer or with anything carried in or on a motor vehicle or trailer with the intention to commit certain specified offences (namely theft of the vehicle or trailer, or anything carried in or on either, or the offence of taking and driving away without consent). The case (and the academic commentary on it[26]) is therefore of some relevance in relation to the words “interferes with” but, as we have seen, the offence in section 49 of the Ordinance is constituted by tampering, which connotes something more than mere interference. The submissions in support of the appeal 20. In support of this appeal, the parties jointly submit that, on the evidence before him, it was not open to the Deputy Magistrate to convict the appellant of the offence under section 49 of the Ordinance and, by consent, invite the Court to allow the appeal and quash the conviction. It is submitted that it would have been stretching the meaning of the word “tampers” and the mischief of section 49 of the Ordinance beyond permissible limits to cover the mere act of closing the door of a vehicle parked on a road. The appellant’s conviction on this basis, it is submitted, constituted a departure from an accepted norm. 21. This must be correct. Applying the broad definition of “tampers” set out above, the evidence does not disclose any act by the appellant that would constitute interference or meddling with part of a vehicle so as to cause alteration or harm to it, or the making of an unauthorised change in it. All that the evidence discloses is that the appellant closed the door of the vehicle. That was a mere act of touching the vehicle and did not result in any alteration or harm to it or change in it. His act of closing the door could simply have been an act of helpfulness towards the vehicle owner. In any event, there is no evidence of any alteration or harm to the door or any other part of the vehicle, or any change in it, as a result of the appellant’s interaction with the vehicle. In the circumstances, therefore, the evidence could not support the offence charged and the appellant’s conviction was wrongful and should be quashed. 22. This submission that the facts disclosed by the evidence did not, in law, amount to an act of tampering within section 49 of the Ordinance is, as already mentioned, not one which was advanced in either court below. It is a new point pursued for the first time in this Court. However, it is a pure point of law and no facts need be investigated in order to resolve the point on appeal. Accordingly, albeit that the Court will only do so rarely, this is an appropriate case in which to entertain the fresh point of law on appeal[27]. Disposition 23. For the above reasons, the Court: (1) Allows the appeal and quashes the appellant’s conviction; (2) Orders, in accordance with the parties’ agreement, that the respondent pay the appellant the sum of HK$10,000 for the costs incurred in the courts below and the sum of HK$4,235 for disbursements incurred by the appellant for the leave application and appeal before this Court. Joint Written Submissions by: Mr Eric T M Cheung (Solicitor Advocate, with HRA in Criminal Cases) and Ms Stephanie T Y Lam, instructed by ONC Lawyers, for the Appellant Mr Andrew Li SPP, of the Department of Justice, for the Respondent [1] (Cap.374) (“the Ordinance”). [2] Afforded by the Clinical Legal Education Programme of the Faculty of Law of the University of Hong Kong. [3] In TWCC 823/2014. [4] Statement of Findings at [8]. [5] In HCMA 359/2014. [6] P. Li J. [7] Issued pursuant to Rule 7 of the Hong Kong Court of Final Appeal Rules (Cap.484A). [8] In FAMC 24/2015. [9] (2014) 17 HKCFAR 303. [10] This provides: “If, while a motor vehicle is on a road or on a parking place provided by a local authority, a person otherwise than with lawful authority or reasonable cause gets on to the vehicle or tampers with the brake or other part of its mechanism, he shall be guilty of an offence.” [11] Clause 48. [12] See Road Traffic Bill 1982, Explanatory Memorandum at [9]. [13] Nor has any UK court sought to define the word “tampers” in section 29 of the Road Traffic Act 1972. [14] The Shorter Oxford English Dictionary, 6th Edition (2007), Vol.II, p.3174. The parties’ Joint Case cites the definition of “tamper” in the 3rd Edition of the SOED (published in 1944), at Vol.II, p.2240, viz.: “3. To have to do or interfere with improperly, to meddle with (a thing) [1601]. 4. To meddle or interfere with (a thing) so as to misuse, alter, corrupt or pervert it [1593].” The differences between the definitions in the two editions of the SOED are not material to the issue in this appeal. [15] Compare section 50 of the Ordinance under which a person who, without relevant consent “moves, damages or interferes in any way with any traffic sign or prescribed traffic sign”, commits an offence. [16] [1987] 1 HKC 104 (Cons V-P, Clough JA and Power J). [17] Note that the reported judgment in [1987] 1 HKC indicates that the judgment of the court was delivered by Power J, whereas the unreported judgment of the appeal (HCMA 990/1986, 11 November 1986) indicates that Cons V-P did so. [18] CACV 253/2010, unrep., 24 February 2012. [19] HCMA 349/2011, unrep., 15 June 2011 (Line J). [20] CACC 521/1988, unrep., 19 January 1989. [21] [1979] VR 343. [22] This provided that “Any person who without just cause or excuse (the proof whereof shall lie upon him) tampers or interferes with a motor car owned by any other person shall be guilty of an offence against this Act.” [23] At pp.343-344, 346. [24] In Collidge v Russo [1984] WAR 1, concerning section 90 of the Road Traffic Act 1974, and The Queen v Wayne Michael Galvin [1998] ACTSC 104, concerning section 27(3)(g) of the Crimes Act 1900 (ACT). [25] [1982] Crim LR 831. [26] See the commentary in [1982] Crim LR at 832 and in Wilkinson’s Road Traffic Offences (26th Ed.) Vol. 1 at §15.45. [27] Archer v Hong Kong Channel Ltd (1997-98) 1 HKCFAR 298 at p.305D. 
Press Summary (English) Press Summary (Chinese) [2018] HKMagC 4 KCCC 788/2018 IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO KCCC 788 OF 2018 Before : Joseph To, Magistrate Dates of Hearing : 5, 6, 7, 13, 18, 19 & 24 September 2018 Date of Reasons for Verdict : 15 October 2018 Reasons for Verdict INTRODUCTION 1. A doctor in a family is invariably treated as the doctor for the family. It is not unusual – it is almost expected, in fact – that members of the family would consult him when they fall sick. And it is normal, being a member of the family, for the doctor to provide his professional opinions, to give his recommendation on the medication required, and to advise on what bodily samples should be tested and analysed, for this is how family bonds are forged, how family ties are deepened and enlivened. 2. In this case, however, the Independent Commission Against Corruption charges to court an ex-government doctor for having gone beyond the norm, alleging that she has ordered 75 medical tests to be performed free of charge for her family members at two government laboratories for 12 years, and that the misconduct she has committed is so serious that she should be convicted of the offence of misconduct in public office.[1] 3. It is alleged in the Particulars of Offence[2], in substance, that between July 2004 and April 2016, as a public official, namely, the Medical and Health Officer of the Department of Health in charge of the San Po Kong Elderly Health Centre, the defendant has, in the course of or in relation to her public office, without reasonable excuse or justification, wilfully and intentionally, misconducted herself by – (a) making or causing members of her staff to process 75 requests for laboratory tests on clinical specimens collected from four of her family members, who were not eligible for such services; (b) falsely representing to two laboratories under the management of the Department of Health that the 75 specimens were collected from persons eligible for such services; and (c) requiring and causing the two laboratories to carry out the 75 tests and to issue reports to the defendant as the officer-in-charge of the centre. 4. If the Commission’s criminal complaint is sustained, the defendant might have (but it is unnecessary and inappropriate, indeed, for this court to decide) breached the relevant Civil Service Bureau Circular governing conflict of interest and abuse of official position.[3] But if all that she has saved for her family members over 12 years was merely $2,970,[4] if her acts have caused no more than minimal clerical work which members of her staff would not otherwise have had to do, if there was only trivial or negligible effect – if at all – on other users and the overall workload of the health centre, and if there was only trivial or negligible effect – again, if at all – on the overall workload of the two laboratories concerned, is the defendant’s misconduct so serious as would reasonably justify a criminal conviction against her? 5. There is in this prosecution an important consideration. Regardless of age, all HKID-card holders are entitled to medical services as general out-patients on payment of $45 per attendance.[5] The services include “medical attention and treatment, prescriptions…, pathology investigations, radiology and other examination.”[6] Holders of HKID cards, the defendant’s family members (except her daughter, who was entitled to similar services at a families clinic free of charge)[7] could have obtained medical services as out-patients at $45 per visit. And the out-patient services would have included the 75 tests referred to in the charge – free of charge. 6. All along, the defendant must have been known of her family members’ entitlement to such services. So, did she – a person of good character, a professional of high status, and a public official of responsibility and long-standing – have the criminal intention to commit so serious a crime as misconduct in public office? Or, did her misconduct, given its nature, frequency, and consequence, sound only in a disciplinary action,[8] had she not resigned? 7. Material relied on by the prosecution shows that there was a case in which a government officer borrowed money on multiple occasions from two immediate subordinates; rather than a bribery charge of an agent soliciting and accepting advantage, a disciplinary tribunal was convened to deal with his case.[9] In yet another case, an officer made at his workplace on-line enquiries on a company for private purposes; he was never charged with the offence of unauthorised access to computer with dishonest gains but was dealt with by way of disciplinary proceedings.[10] 8. In neither case was the officer taken to court; but the defendant here is. What justifications are there for the apparent disparity in treatment between the two cases and hers? 9. To deal with these issues, a discussion on theraison d'être and on the development at common law of the subject offence is necessary. PROSECUTION CASE Defendant and the health centre 10. The defendant began working as a Medical and Health Officer of the Department of Health in March 1994.[11] In her third posting which took effect in September 1999, she became officer-in-charge of the health centre in San Po Kong.[12] The centre caters for persons of 65 or above. They have to pay a membership fee to register themselves as eligible or non-eligible members of the centre.[13] Eligible members have to pay $45 for each medical consultation;[14] non-eligible ones are charged a higher fee for membership and medical consultation.[15] 11. For the 12 years in question, the defendant has been the only doctor at the centre,[16] the only person possessing the power to make requests for clinical specimens collected at the centre for tests to be conducted at two laboratories in government service. Conflict of interest and abuse of official position 12. The Department sends to its employees on a regular basis relevant circulars and guidelines on the required standard of integrity;[17] the prosecution relies on five of them here.[18] As a matter of common sense, nothing in the required standard of behaviour is unreasonable, unfathomable, or unimaginable; in fact, everything in them is just common sense. 13. According to the circulars and guidelines, “the essential feature of the offence [of misconduct in public office] is an abuse by the public official of the powers, discretions or duties exercisable by virtue of his official position conferred on him for the public benefit.” [19] A government officer is expected to serve the public with integrity, honesty, and impartiality.[20] He must avoid conflicts, actual or potential, between his official duties and his private interests;[21] he must not use his official position to further his private interests;[22] and he must not coerce or induce his subordinates and others to do anything to his own benefit or that of his family and friends.[23] Loyalties to family and friends are of course to be encouraged, but the officer must not, in his official position, act or be seen to be acting preferentially in favour of them.[24] Defendant’s family members and the laboratory tests 14. None of the defendant’s four family members concerned were, at the time, 65 or above; so none of them were members, eligible or non-eligible, of the centre.[25] But between July 2004 and April 2016, the defendant had, on 67 times,[26] made a total of 75 requests from the centre to the two laboratories for tests to be conducted on clinical specimens[27] belonging to her four family members (including her daughter).[28] The requests were administratively processed at the centre and the laboratories, the tests and analyses were carried out in the laboratories, and 75 laboratory reports were sent to the centre.[29] Complaint against the defendant 15. In early September 2016, a member of the centre made a complaint against the defendant, the only complaint against her during the 17 years of her service there.[30] The complaint was that she had provided consultation to, and had arranged for blood tests and health assessments for, two of her family members.[31] 16. In response to a Questionnaire given to her regarding the complaint, [32] the defendant replied in writing, the significant part of which is reproduced in the admitted facts in these proceedings.[33] In part[34] of her answer to Question 5, she said that her sister and daughter had consulted her when they felt unwell, she had arranged for the necessary laboratory tests for them out of concern for their health, it had never occurred to her that doing so at the centre was improper and she would not have done so had she known that it was improper, and she had never intended to deceive or take advantage of anyone; and she apologised for her ignorance and undertook not to do so again.[35] DEFENCE CASE 17. The defendant is a doctor of positive good character. She elects not to give evidence in her defence, and she calls no witnesses to her defence. By way of cross-examination (which has obviously been well-planned and which has been executed with exemplary professionalism) of the prosecution witnesses, Mr Cheng Huan SC puts the Commission’s case in context, factual and legal, and submits powerfully that the paramount consideration in this case is the “context.” In addition, reliance is placed on appraisal reports,[36] character references, donation records, appreciation letters, and an award for volunteer service.[37] ANALYSIS General considerations 18. The burden of proof is on the prosecution throughout to satisfy the court of the doctor’s guilt; she has no burden to prove anything, for she has the benefit of the presumption of innocence. She exercises her legal right not to testify in court; the court is not permitted to draw any adverse inferences arising from the way in which she exercises her legal right. The doctor has a clear record, and a significant amount of evidence and materials has been placed before this court in support of her positive good character. This shows that she is not disposed to committing crime, let alone so serious a crime as misconduct in public office. In addition, her positive good character enhances her credibility in what she has said out-of-court in response to allegations of misconduct, in particular, with respect to the issues of dishonesty and improper motive. 19. In this analysis, the court will deal with the following issues – (i) the common law offence of misconduct in public office, (ii) the instant case, (iii) findings of facts, and (iv) application of the law on the facts found. Prosecuted as the case has been by experienced prosecutors, many things have gone wrong in the present prosecution. They are serious matters, and they will be canvassed below. The first issue: Misconduct in public office (1) Raison d'être 20. The existence for a long time of any common law offences does not necessarily justify their existence in the first place or their continuing existence in changed times. Common law offences were mostly called into existence when cases were brought before the court in situations showing that there existed a social need or danger which, if unheeded by the court, would have resulted in serious damage to the public interests and to the overall safeguard of justice and fairness of the land. 21. In respect of misconduct in public office, what then was the social need or danger that common law judges had in mind, and why was that social need or danger so important, so pressing, and so grave that the conduct concerned must be criminalised? And why should public officers be prosecuted in situations where private employees would just be dismissed? 22. As part of the terms of their employment with the government, civil servants are subject to the Civil Service Rules, and departmental guidelines and circulars. In discharge of their contractual duty of loyalty to their employer, they must comply with the relevant civil service rules and departmental guidelines. The common law offence of misconduct in public office is concerned, however, with a different duty toward a different value, namely, the public trust. The fundamental issue here is this: what is entrusted to the public officer by reason of his office as such? 23. To begin with, it must be noted that the subject offence has a special feature: it challenges the notion of equality. Equality before the law means that the law must be applied equally to everyone, but it does not mean that individual legal rights and liabilities must, no matter what, be the same for all. 24. In the private setting, an employee may be employed to perform functions that affect others, but the focus is on fulfilling contractual obligations. Unlike private employees, public officers are invariably entrusted with public powers, public duties, and public discretions, along with all necessary plants, tools, and equipment, over which they are entrusted with custody and control – all to be used for public purpose. In short, public officers hold these powers, etc., on public trust, in the sense that they are not to be used or exercised as if they were private, as if they were personal, as if they were their own. 25. The nature and extent of the public powers, duties, and discretions entrusted to public officers vary widely. So are the benefits sought to be conferred upon the public by the valid exercise of these powers and the valid performance of these duties. As to the ways in which public powers can be abused, one would wonder whether the sky can provide the limit. As to the consequences of such abuse, it can be – on the face of it – none. For example, a senior prison officer in the UK was sentenced, on appeal, to 27 months’ imprisonment for failing to disclose her intimate relationship with an ex-prisoner.[38] But the consequences can also be tragic deaths to some.[39] 26. Given its wide variety, misconduct in public office could not all be fitted into criminal offences known to the law, and abuse of public powers could result in serious damage to the public interests and could cause serious consequences. Common law judges saw great dangers to society in leaving serious misconduct committed by public officers to departmental or disciplinary determination. The common law intervened, therefore, and it did so in England as early as the 13th century.[40] In R v Bembridge,[41] Lord Mansfield, speaking in 1738, held: “[A] man accepting an office of trust concerning the public, especially if attended with profit, is answerable criminally to the King for misbehaviour in his office; this is true, by whomever and in whatever way the officer is appointed… [W]here there is a breach of trust, fraud, or imposition, in a matter concerning the public, though as between individuals it would only be actionable, yet as between the King and the subject it is indictable. That such should be the rule is essential to the existence of the country.” (emphasis supplied) 27. A public officer, in wilfully misconducting himself, breaches the public trust reposed in him – in failing the public purpose of that public office. When dealing with this offence, the court’s focus is not on the relationship between the public officer and the government, but on the status, power, and discretion conferred upon the officer on public trust for public purpose. In a prosecution for misconduct in public office, the standard of behaviour prescribed by the civil service rules has limited relevance, for the following reasons. 28. The offence of misconduct in public office was not promulgated to enforce the contractual relationship between the public officer and the government. The inability on the part of the prosecution to prove any breach of the civil service rules cannot by itself afford a defence to the offence. And breaches of the rules, no matter how serious and how frequent, do not mean that a prosecution for this offence is a must. Sir Anthony Mason said: “[T]here is … a borderland in which the common law offence and the disciplinary offences overlap.”[42] But ultimately the court’s duty remains: a public officer can be convicted only when the elements of this offence are established to the required criminal standard. 29. In conclusion, Lord Mansfield’s line of reasoning was as valid in England at His Lordship’s time as it is in Hong Kong today, and it provides the raison d'être of this common law offence. The essence of the offence is in a breach of public trust, committed in a matter concerning the public. (2) Development of the offence at common law 30. While HKSAR v Sin Kam Wah[43] contains the authoritative description of the elements of the offence of misconduct in public office, the founding jurisprudence on this offence lies elsewhere. In the earlier case of HKSAR v Shum Kwok Sher,[44] the Court of Final Appeal discussed this common law offence at length and provided the authoritative definition of this offence, not just for Hong Kong, but elsewhere.[45] 31. The defendant in Shum Kwok Sher’s case[46]was a Chief Property Manager in government employ. He caused lucrative government contracts to be awarded to companies in which his family members had interests. It was never part of the prosecution case that any of the contracts concerned had been poorly or improperly performed, or that the government had suffered any quantifiable loss. 32. Acknowledging that the offence was “not easy to define,” [47] Sir Anthony Mason sitting at the Court of Final Appeal provided examples to illustrate the ambit of the offence: “82. The critical question is: what is the mental element required to constitute commission of the offence? In the case of nonfeasance, non-performance of a duty arising by virtue of the office or the employment, all that is required is wilful intent, accompanied by absence of reasonable excuse or justification. Mere inadvertence is not enough. … 83. In other cases, the question is more complex. That is because outside the area of non-performance of a duty, an additional element is generally, if not always required, to establish misconduct which is culpable for the purposes of the offence. In such cases, in the absence of breach of duty, the element of wilful intent will not be enough in itself to stamp the conduct a culpable misconduct. A dishonest or corrupt motive will be necessary as in situations where the officer is exercising a power or discretion with a view to conferring a benefit or advantage on himself, a relative or friend. A malicious motive will be necessary where the officer exercises a power or discretion with a view to harming another. And a corrupt, dishonest or malicious motive will be required where an officer acts in excess of power. The point about these cases is that, absent the relevant improper motive, be it dishonest, corrupt or malicious, the exercise of the power or discretion would not, or might not, amount to culpable misconduct. Although the examples constitute some only of the range of situations which fall within the reach of misconduct in public office, they are enough to illustrate the proposition that the existence of an improper motive, beyond the existence of a basic wilful intent, is necessary to stamp various categories of conduct by a public officer as culpable misconduct for the purposes of the offence.” (emphasis supplied) 33. Sir Anthony Mason captured the essence of the offence clearly and elegantly in these two paragraphs: “84. … A public official culpably misconducts himself if he wilfully and intentionally neglects or fails to perform a duty to which he is subject by virtue of his office or employment without reasonable excuse or justification. A public official also culpably misconducts himself if, with an improper motive, he wilfully and intentionally exercises a power or discretion which he has by virtue of his office or employment without reasonable excuse or justification… 86. … [T]he misconduct complained of must be serious misconduct. Whether it is serious misconduct in this context is to be determined having regard to the responsibilities of the office and the officeholder, the importance of the public objects which they serve and the nature and extent of the departure from those responsibilities.” 34. How serious must the misconduct be? On this topic, the Supreme Court of Canada in R v Boulanger,[48] a case referred to in HKSAR v Tsang Yam-kuen, Donald[49] and relied on by Mr Cheng Huan SC here, said: ‘52. … The public is entitled to expect that public officials entrusted with these powers and responsibilities exercise them for the public benefits. Public officials are therefore made answerable to the public in a way that private actors may not be. This said, perfection has never been the standard for criminal culpability in this domain; “mistake” and “errors in judgment” have always been excluded. To establish the criminal offence of breach of trust by a public officer, more is required. The conduct at issue, in addition to being carried out with the requisite mens rea, must be sufficiently serious to move it from the realm of administrative fault to that of criminal behaviour. … What is required is “conduct so far below acceptable standards as to amount to an abuse of public’s trust in the office holder” (Attorney General’s Reference, at para. 56). As stated in R. v. Creighton, [1993] 3 S.C.R. 3, “[t]he law does not lightly brand a person as a criminal.” 53. The question posed by Sir Anthony Mason of the Court of Final Appeal in Shum Kwok Sher provide a sound definition of the parameters of the inquiry into whether the conduct constitutes a marked departure from accepted standards. The inquiry must take place against the background of the responsibilities of the office and the importance of the public objects they serve: Whether it is serious misconduct in this context is to be determined having regard to the responsibilities of the office and the officeholder, the importance of the public objects which they serve and the nature and extent of the departure from those responsibilities. [Emphasis added; para. 86]’ 35. Many cases of misconduct in public office have been decided in the common law world since Sir Anthony Mason’s formulation of the offence in Shum Kwok Sher’s case.[50] Several have been relied on by the parties here. Significant in terms of development of the law are the case of HKSAR v Wong Lin Kay,[51] cited by the prosecution, and the English Court of Appeal cases of R v W(M)[52]and R v Chapman and ors,[53] not cited by the parties in the present proceedings. 36. In Wong Lin Kay’s case,[54] a government driver was disqualified from driving but, without disclosing his disqualification, he continued driving for work purpose. The Court of Final Appeal found for him and rejected the prosecution’s attempt to reverse his acquittal. The unanimous ruling of the court is this: ‘22. The correct approach is therefore not to attempt somehow to decide in the abstract or in isolation whether a person is or is not a “public officer”. One must examine what, if any, powers, discretions or duties have been entrusted to the defendant in his official position for the public benefit, asking how, if at all, the misconduct alleged involves an abuse of those powers in any of the ways identified in Shum Kwok Sher v HKSAR. If the defendant occupies a position which confers no such powers on him, he is not a candidate for prosecution for the offence, even if he is employed by a government department or by an analogous public body.’ (emphasis supplied) 37. In R v W(M),[55]a decision which has triggered mixed responses from legal scholars,[56] a police officer was supplied with a credit card for work purposes, but he used it and incurred substantial personal expenses. At trial, he alleged it was common for his colleagues to use office credit cards in the same way and that his intention was to make reimbursement when he was able to do so, thus putting in issue whether an improper motive was proved against him. 38. The English Court of Appeal held that in circumstances involving an act of theft or fraud, dishonesty must be proved as an element of the offence of misconduct in public office. The court said: ‘11. … Thus, nearly 200 years ago, in R. v Borron (1820) 3 B and Ald 432, in the context of misconduct by a magistrate, Abbott CJ said: “… the question has always been, not whether the act done might, upon full and mature investigation, be found strictly right, but from what motive it had proceeded; whether from a dishonest, oppressive, or corrupt motive, under which description fear and favour may generally be included, or from mistake or error. In the former case, alone, they have become the objects of punishment.” (Our emphasis) 13. We have been unable to discover any authority which suggests that the observations in Borron are no longer apposite to a case like the present… Consistently with principle, and addressing the realities, it is in any event difficult to see how the defendant can have fallen so far below the standards required of him … by his superiors for “expenses”, his conduct is properly to be stigmatised as criminal, unless he was dishonest.’ 39. In R v Chapman and ors,[57] the English Court of Appeal dealt with proper directions to be given to the jury with respect to the offence of misconduct in public offence. The court remarked: “As was observed in R. v Borron (1820) 3 B. & Ald. 432; 106 E.R. 721, to condemn anyone who had fallen into error or made a mistake, belonged only to the law of a despotic state.”[58] It is clear, therefore, that a high threshold must be met before a public officer can be convicted. 40. The court held that the jury must be specifically directed that a breach of duty or of trust is not enough.[59] They must be told that the threshold of seriousness is that the misconduct must amount to “an abuse of the public’s trust in the office holder, … to [the] level … where it is calculated to injure, that is to say has the effect of injuring, the public interest so as to call for condemnation and punishment.”[60] 41. In summary, this is the law with respect to the offence of misconduct in public office as it currently stands. Whether or for how long it should remain a common law offence is an issue for another forum, another time. The second issue: The instant case (1) Particulars of offence framed contrary to Sin Kam Wah’s case[61] 42. The prosecution charges the defendant for having “wilfully and intentionally misconducted herself.” Before adjourning the case for written submission, the court observes that the law on misconduct in public office is settled. The prosecution agrees and puts forward Sin Kam Wah’s case[62] for the court’s consideration. The court then invites submission, also, on the Shum Kwok Sher’s case.[63] 43. In closing submission, the defence contends that the prosecution frames the charge, prosecutes the defendant to court, and must therefore strictly prove its pleaded case. Reliance is placed on HKSAR v Chan Tak Ming,[64] in which the Court of Final Appeal said: “[I]n criminal cases …, pleadings are not meant merely to inform the party to whom they are delivered. They are also meant to limit the party who delivers them. … In R v Landy [65]… the English Court of Appeal said (at p.362) that particulars of offence are meant: … to stop the prosecution shifting their ground during the course of the case without the leave of the trial judge and the making of an amendment.” 44. If the prosecution has informed itself – adequately or at all – of the rulings in Shum Kwok Sher’s case,[66] AG’s Reference (No 3 of 2003),[67] and Sin Kam Wah’s case,[68] the prosecution would never have failed to notice that the particular of offence here is plainly wrong, in that the phrase “wilfully and intentionally” should never have been pleaded. The reasons are as follows. 45. On the issue of mens rea, Sir Anthony Mason, who has in Shum Kwok Sher’s case[69] provided a first formulation of the offence, said: “84. … A public official culpably misconducts himself if he wilfully and intentionally neglects or fails to perform a duty to which he is subject by virtue of his office or employment without reasonable excuse or justification. A public official also culpably misconducts himself if, with an improper motive, he wilfully and intentionally exercises a power or discretion which he has by virtue of his office or employment without reasonable excuse or justification. Subject to two qualifications … 85. The first qualification is that … I consider that the misconduct must be “wilful” as well as “intentional”. … “[W]ilfully” signifies knowledge or advertence to the consequence, as well as intent to do an act or refrain from doing an act. Wilfulness in this sense is the requisite mental element in the offence of misconduct in public office, most notably in cases of non-feasance. There is no reason why the same mental element should not be requisite in cases of malfeasance and other forms of misconduct in public office. For this reason “wilfully and “intentionally” are not employed disjunctively …” 46. In AG’s Reference (No 3 of 2003),[70] the English Court of Appeal considered in depth Sir Anthony Mason’s formulation and then put forward its own, with emphasis on “abuse of the public’s trust in the office holder.”[71] On the issue of mens rea, the court said: ‘45. We share, with respect, counsel’s difficulty in understanding, upon the first qualification, the need for conduct to be both wilful and intentional. The need for both was considered important by Sir Anthony Mason and also by Bohkary PJ. … The judge [Sir Anthony Mason] was also underlining what he had said earlier, at para. 82, that “mere inadvertence is not enough”. Having cited the two limbs of the test in Sheppard[72] as disjunctive, it is unlikely that Sir Anthony Mason would require intent as distinct from recklessness in relation to advertence to the consequences. If there is a difference, we adopt the approach in G,[73] which in any event binds us. The decision of the Final Court of Appeal in Shum Kwok Sher[74] of course precedes that decision.’ 47. It was a result of the above comments of the English Court of Appeal that the Hong Kong formulation was modified in Sin Kam Wah’s case.[75] There Sir Anthony Mason said: “45. Since this Court delivered judgment in Shum Kwok Sher[76] …, the English Court of Appeal, in Re A-G’s Reference (No 3 of 2003)[77]… has considered the elements of the offence of misconduct in public office. In that case, the Court of Appeal took account of the decision in R v G & Another[78] … in considering the relationship between recklessness and wilful misconduct, as well as the judgments in Shum Kwok Sher[79]… In the light of the detailed consideration of the problem in Re A-G’s Reference (No 3 of 2003),[80] I would re-formulate the elements of offence … as follows. The offence is committed where: (1) a public official; (2) in the course of or in relation to his public office; (3) wilfully misconducts himself; by act or omission, for example, by willfully neglecting or failing to perform his duty; (4) without reasonable excuse or justification; and (5) where such misconduct is serious, not trivial, having regard to the responsibilities of the office and the officeholder, the importance of the public objects which they serve and the nature and extent of the departure from those responsibilities. 46. The misconduct must be deliberate rather than accidental in the sense that the official either knew that his conduct was unlawful or wilfully disregarded the risk that his conduct was unlawful. Wilful misconduct which is without reasonable excuse or justification is culpable.” (emphasis supplied) 48. The correct particular of offence must be “wilfully misconducted herself”, rather than “wilfully and intentionally misconducted herself,” as the particular is pleaded here. A number of cases have been cited by the parties before this court. Where it can be ascertained, the cases in which the charge has been correctly pleaded in accordance with Sin Kam Wah’s case[81] are (i) Chan Tak Ming’s case[82], (ii) HKSAR v Tsang Yam-kuen, Donald[83], (iii) HKSAR v Hui Rafael Junior,[84] and (iv) HKSAR v So Ping Chi.[85] And the cases in which the particulars have been wrongly pleaded are (i) HKSAR v Lee Wing-tai,[86] and (ii) the present one. 49. The difference between the two phrases is not without significance. It has attracted serious attention and in-depth consideration of judicial authorities of high stature – in fact, of the highest stature. Sin Kam Wah’s case[87] was decided in 2005, but the present charge is still incorrectly particularised. It should be noted that the court has expressly invited submission on Shum Kwok Sher’s case,[88] but the prosecution has never sought to rectify the particular. It is not for this court to decide whether this has been a case of prosecutorial inattention, or prosecutorial incompetence, or worse, prosecutorial indifference; this issue will have to be looked at elsewhere. 50. In respect of elements of offence, this court is charged with the duty to administer the law as per Sin Kam Wah’s case.[89] The court will treat as redundant, therefore, the two words “and intentionally” as they appear in the particulars of offence. (2) Prejudicial matter introduced by the prosecution 51. When dealing with the Questionnaire,[90] the prosecution reveals to court, in examination-in-chief of Dr Lee Siu-yin,[91] that the defendant has faced, not just one, but several complaints. The court takes exception to the prosecution approach, whereupon the prosecution explains that those other complaints have been dismissed. 52. This rather begs the question why the court has to know about those complaints in the first place. While the defendant has in the end elected not to give evidence, her response to the Questionnaire (see paragraph 16 above) [92] constitutes a mixed statement, the exonerating part of which is just as admissible as evidence of facts as the incriminating part of it; and it is a matter of weight, in respect of the two parts, to be assessed and attached. 53. In view of the admitted facts, it must have been clear to the prosecution – right from the start – that the defendant is putting in issue her motive and knowledge as to the propriety and appropriateness of her acts, the resolution of which issue necessarily depends on the court’s assessment on her credibility with respect to the exculpatory portion of her response to the Questionnaire. On the issue of credibility, the defendant has, at law, the benefit of her positive good character. Revealing in court complaints against her of a nature that cannot, in prosecutorial propriety and at law, be made clear to the court undermines her good character. 54. Prosecutors are presumed to be conversant with the duty of fairness, a duty they owe not only to the community,[93] but also to court, to due administration of justice. This is basic. The court will say nothing, with regard to this instance, whether the prosecutor has discharged the duty to the required professional standard. But it is most important, it must be noted here, for all prosecutors to avoid all appearance of being unfair, or worse, of being insidious. 55. The court ignores suggestions of other complaints against the doctor and notes the prosecutor’s apology in closing speech.[94] (3) Credibility and reliability of prosecution witnesses, and Prosecution’s failure to disclose material information 56. The issue of credibility and reliability of prosecution witnesses here should have been a straightforward one: the defence does not seriously challenge the witnesses on this front. But complications have arisen due to the prosecution’s defective discovery of unused material and defective understanding and performance of its duty of disclosure. What happened at the hearing was this. 57. The defence informs the court at the commencement of the trial that it has made a request, a week before the hearing, for disclosure of unused material, and was supplied with a bundle of documents the day before trial. It transpires that the material contains handwritten notes prepared by the Commission’s investigators on meetings or telephone conversation with the prosecution witnesses for the purpose of preparing draft statements for the witnesses’ endorsement. 58. The notes are plainly discoverable as unused material. They contain the best and only records of the witnesses’ prior oral statements on material issues. The court enquires with the prosecution why those notes have not been disclosed in a timely fashion; to that enquiry, the prosecution replies that the notes are documents of persons (namely, the investigators) whom the prosecution has no intention to call as witnesses in the trial. This, as a reason for non-disclosure, is plainly wrong. The court further enquires with the prosecution on the legal test for the scope of the prosecution’s duty of disclosure. Here comes the response: “Relevance to an issue in dispute.” This, as a matter of law, is again plainly wrong. 59. The notes should have been disclosed to the defence – unasked and well before trial. The prosecution’s reason for withholding the written notes is plainly wrong in light of the authoritative pronouncement on the prosecution’s duty of disclosure in the Court of Final Appeal case of HKSAR v Lee Ming-tee.[95] The editors of Archbold Hong Kong (2018) describe the duty in these two paragraphs: ‘ The prosecution is under a positive duty in law to make disclosure to the accused all material information in their possession which is or may be relevant to an issue in the case and which may help the accused in the conduct of his defence: HKSAR v Lee Ming-tee (No. 2) (2003) 6 HK[C]FAR 336 … The duty to disclose is pro-active and continuing and does not depend on the request of the accused. In R v Hennessey (Timothy) (1978) 68 Cr App R 419, Lawson LJ observed that: “ … those who prepare and conduct prosecutions owe a duty to the courts to ensure that all relevant evidence of help to an accused is either led by them or made available to the defence.” … ’[96] (emphasis supplied) ‘ It is only information that is material that is required to be disclosed. … The responsibility lies on the prosecution in determining whether information is material and needs to be disclosed. … The court will interpret the test liberally and thus material information is not restricted to that in the prosecution which is itself in the form of admissible evidence but to information recorded in however form which was or might have been material to the issue at the trial and “helpful” to the defence case: HKSAR v Lee Ming-tee (No. 2) (2003) 6 HK[C]FAR 336 … Thus “material” information is not confined to admissible evidence but may include that…which may directly affect a witness’s credibility…’[97] (emphasis supplied) 60. If the trial is not affected in fairness, failure by the prosecution to make disclosure rarely affords a ground of complaint; and Mr Cheng Huan SC is not complaining. But when the material in question is relevant to the court’s assessment of the credibility and reliability of witnesses, relevant to whether immunity from prosecution should be given to the witnesses, and relevant to whether warnings against self-incrimination should be administered on the witnesses, the defective disclosure, arising as it does from a defective understanding of a fundamental duty, takes on a different dimension. The problems in this case have arisen in the following circumstances. 61. Dr Lee Siu-yin[98]gives evidence that she does not consider the defendant’s conduct serious, provided that the services to other users of the centre were not affected. Mr Cheng Huan SC shows her the handwritten note of the investigator who has spoken with her. The note says categorically: “no effect.” 62. Dr Lee no longer recalls now what exactly she has told the investigator. She is prepared, however, to adopt the note as part of her evidence, subject to the proviso that services to others at the centre be unaffected. Mr Cheng Huan SC then produces to Dr Lee a draft statement, prepared by the investigator subsequent to the interview. It is at once clear that the proviso appeared for the very first time in the whole of the investigation stage only in that draft statement. Now what does this all mean? 63. The investigator’s handwritten note shows that Dr Lee has made a negative statement orally: “no effect.” But, when reduced into writing, this definite statement of hers somehow became a proposition, the validity of which depended upon services provided to others not being affected. And there is no evidence that the proviso has been an after-thought, there being no written records to that effect. 64. The issue then becomes: where has this proviso originated from? Origin of information adduced in court under oath is important to the court’s evaluation of evidence; in this case Dr Lee’s proviso has been shown to be of suspect originality. No weight can therefore be attached to the proviso. Insofar as the court is concerned, Dr Lee is and was of the view that the defendant’s misconduct has had no effect on the operation of the centre. 65. That is not the only problem. A staff nurse Madam Ku Siu-hing[99] gives evidence for the prosecution, in the course of which the defence raises the issue of self-incrimination concerning a potential offence of aiding and abetting. The prosecution assures the court that consideration has been given – on the basis of Madam Ku’s witness statement – to the issues of immunity and of self-incrimination, and that it has been the prosecution’s conclusion that no immunity would be given and no warnings on self-incrimination need be given. There are no dangers therefore that the witness might be prosecuted – in view of the prosecution’s stance. The court rules therefore that a warning against self-incrimination is not required. But complications arise when another staff nurse Madam Chan Lai-kuen[100] takes the stand. 66. As to Madam Chan, the court assumes – because the issue is not raised again – that the prosecution has likewise considered the issue of immunity and of self-incrimination. The defence cross-examines the nurse on the basis of her statement taker’s handwritten note. It transpires that she has told the investigator orally that she knew the defendant’s family members were below 65. But in the draft statement given to her for her endorsement, her knowledge in this respect has been grossly glossed over, so much so that if one were to look at her statement alone, little if any attention would have been drawn to the issue of Madam Chan’s knowledge of the age of the defendant’s family members. If the prosecution has relied on Madam Chan’s statement alone, issues of immunising her and of warning her not to incriminate herself could not possibly have been fully considered. 67. In the court’s view, Madam Chan must have known the age of the defendant’s family members. It is a matter of regret that this court has been denied the opportunity to have the issue of self-incrimination ventilated at trial. The court enquires with the prosecution when it was that the handwritten notes were first supplied to the Department of Justice as a result of the defence request for specific discovery. 68. The prosecution reveals to court at the end of its closing speech that it was two days before the hearing and, for the first time in the trial, the prosecution says it has – on top of the witnesses’ statements – considered the handwritten notes before coming to the conclusion that no immunities from prosecution need be granted to the nurses, and that neither of them need be warned about self-incrimination. 69. To conclude this part of the discussion, the court needs only to ask this question: what if the defence had never sought specific discovery of those handwritten notes? It seems most probable that important lines of cross-examination touching on credibility and reliability of prosecution witnesses would have been denied the defence; and the prosecution would have failed fundamentally to discharge its duty concerning whether the witnesses themselves should be prosecuted, whether they should be immunised instead, and whether warnings against self-incrimination should be administered. If the defence had not sought specific discovery, one wonders what would the prosecution’s failure to make disclosure of the handwritten notes – unasked – have done to due process? 70. Apart from the problems identified above, the court finds the prosecution witnesses truthful and reliable, except in regard to the costs caused by the defendant’s 75 requests for laboratory tests (see paragraphs 76 to 84 below). The third issue: Findings of facts 71. The court finds proved and established beyond all reasonable doubts the facts, acts, and circumstances contained in paragraphs 10 to 16 above. In the court’s view, the defendant’s cautioned statement does not take the matter any further. But the effects of the defendant’s misconduct have to be dealt with in detail. Effects of the defendant’s misconduct (i) Role of the staff nurses 72. The staff nurses have to cause a medical file[101] and a booklet[102] to be opened and maintained at the centre for the members, to perform medical assessments on the members, and to collect clinical specimens from the members. There is no evidence before the court that any of the nurses have done these tasks for the defendant’s family members.[103] 73. In the 67 offending occasions, the nurses received from the defendant labelled clinical specimens belonging to her family members, put the specimens (together with the request forms) inside sealed packets for delivery,[104] made entries in the daily registers stipulating that the specimens were sent to the laboratories “care of” the defendant,[105] and, when the reports were received at the centre, had them handed over to her.[106] In addition, upon receipt of the reports, a clerical officer of the centre would chop them with the receipt stamp.[107] 74. In support of the charge, the prosecution relies on the above menial tasks done at the defendant’s behest by her staff, tasks that they would not otherwise have had to do.[108] But no evidence has been adduced to show that the staff members have done more than these menial tasks.[109] (ii) Workload of the centre and of the two laboratories 75. No preponderance of evidence has been presented to court to prove beyond all reasonable doubt the following. First, as a result of the 75 requests over the 12-year period, members of the centre have had to wait longer (for their first medical assessments[110] and for their sessions at the consultation room) than they would otherwise have to. Second, as a result of the 75 requests in the 12-year period, the overall workload of the centre has been noticeably affected. Third, as a result of the 75 requests in the 12-year period, the overall workload at the two laboratories – which must have conducted thousands of thousands of tests and analyses – has been noticeably affected.[111] In its written submission[112] and in argument, the prosecution has not sought to argue otherwise. (iii) Prosecution stance on the laboratory costs 76. An important – if not the most important – issue at trial concerns the costs to the two laboratories that the defendant’s acts are said to have caused. On this issue, the evidence of Dr Janice Lo[113] is this: (i) she has told the Commission’s investigator that the costs for some of the tests were gazetted in 2003 or 2004, and the gazetted costs were ascertainable via her computer at work, (ii) outdated as they may be, the gazetted costs have remained legally applicable for the gazetted tests conducted between 2004 and 2016, (iii) for non-gazetted items, there was a formula by which the costs could be quantified; the formula was updated once a year; and it was last updated in 2017, (iv) the investigator asked her for a “more recent” figure, so the 2017 formula was applied for all but one of the 75 tests – in order to achieve uniformity, (v) with respect to the one test for which the 2017 formula was not used in this case, the 2015 formula was adopted instead because the test has become obsolete, and (vi) the laboratory costs, by the above calculation, totalled $24,116. 77. Gazetted and non-gazetted, there are only 75 tests with which the Commission, the parties, and the court are concerned. The costs published in the gazette of 2003 or 2004 will have failed to take account of the changes in labour costs, etc., between 2004 and 2016 and, to that extent, can be described as outdated. There are however no valid and, more importantly, legal alternatives by which the costs of gazetted tests conducted in those 12 years can be assessed. For non-gazetted items, on the other hand, it defies logic and common sense that the formulae – which have been revised annually – have not been used at all. Using the 2017 formula to calculate the laboratory costs of any of the 75 tests must be glaringly wrong. 78. Evidently, no evidence has been put forward to prove the exact costs that the two laboratories have incurred as a result of the 75 requests over the 12-year period. But the prosecution in its written submission maintains: “[T]he Defendant’s misconduct was serious: … [Dr Janice Lo]’s evidence[114] [is] that the total costs for the 75 tests is HK$24,116.”[115] The prosecution does not, in argument, resile from this position. 79. The prosecution seeks to justify the figure of $24,116 on the grounds that it was the only figure with which the prosecution has been supplied, and that the 2017 formula was used to achieve uniformity.[116] The court highlights Dr Lo’s evidence to the effect that the laboratory costs have been calculated by reference to the 2017 pricing policy,[117] and indicates that it was a wrong benchmark for tests and analyses done between 2004 and 2016. 80. In the end, the prosecution informs the court that the figure has been put forward “for reference” only, emphasising that it has never been the prosecution’s intention to provide the court with an inflated figure.[118] The prosecution approach adopted in respect of this issue is as objectionable as it is unprofessional: objectionable as it lacks focus; unprofessional as it involves placing before the court material that is prejudicial to a fair trial. The bases for this conclusion are these. 81. Apart from evidence, reference materials are, as a reality of trial nowadays, frequently placed before the court; but they are produced in court in a professional manner for purposes that are clearly defined. Reference materials are tendered, invariably by agreement of the parties, when they can provide background information that is in nature neutral, neutral in the sense that it does not make the commission of the crime more or less likely, or when they can assist the court to come quickly to grips with non-controversial matters. Introduction at trial of these materials can save court time. For issues in dispute, however, the law is clear. Only relevant and admissible evidence can be adduced, and the probativeness of the evidence sought to be produced must outweigh its prejudicial value. And this is again basic, in the court’s view. 82. Pricing policy of 2017 is blatantly irrelevant to costs incurred in the 12 years between 2004 and 2016. As a matter of common sense, the 2017 formula must have taken into account the inflationary tendencies of the laboratory costs, labour and material, over the years; so, using the 2017 formula in this case must be wrong because it is plainly prejudicial. 83. Prejudicial effect arising from evidence sought to be adduced cannot be determined solely by reference to the interests of the defence but must be assessed in light of the overall fairness of the trial process. In view of the importance of the issue concerned, the calculation premised upon the 2017 formula is plainly irrelevant and grossly prejudicial, prejudicial not to the defence but to a fair trial. No weight can be attached to the figure of $24,116; and this leaves a lacuna in the prosecution case. 84. The lacuna could have been avoided. Dr Janice Lo’s evidence shows that she has had e-mail correspondence with the investigator when calculating the costs. It is clear from those e-mails, which have been served on the defence as unused material, that the use of the 2017 benchmark was plainly wrong. If the prosecution had considered the unused material or had done so with care, the problem would have been spotted easily, and Dr Janice Lo could have been invited to provide a correct figure for court purposes. The fourth issue: Application of the law to the facts found 85. The defence contests the charge putting all elements of the offence in dispute, in particular, with respect to the following three issues: (i) whether the defendant has wilfully misconducted herself by making the 75 requests, (ii) whether her conduct was without reasonable excuse or justification, and (iii) whether her conduct was “serious, not trivial” as would amount to a criminal offence. (1) Public officer acting in the course of public office 86. Was the defendant a public officer, first of all, for the purposes of the subject offence? To answer this question, “[one] must examine what, if any, powers, discretions or duties have been entrusted to the defendant in his official position for the public benefit, asking how, if at all, the misconduct alleged involves an abuse of those powers in any of the ways identified in Shum Kwok Sher v HKSAR.”[119] The defendant’s official powers, duties, and discretions entrusted to her as the medical officer cum officer-in-charge of the centre must be taken into account, against the Commission’s allegations of misconduct. 87. The evidence shows, and this is not disputed, that the health centre was established to provide medical and health services to the elderly. The defendant, as the only doctor there, represented the only source of professional services to all members of the centre; and her official duties were to provide such services there. True it is that she has merely worked for the centre, but the centre would not have existed but for her work. She was the heart and soul of the centre, in fact, the raison d'être of it, for a health centre without a doctor is not a health centre. 88. No direct evidence has been adduced at the trial to detail the defendant’s official powers, duties, and discretions. From the entirety of the evidence, the court draws the only reasonable and irresistible inference as follows: (i) the defendant was well familiar with the criteria of the members’ eligibility for services provided at the centre, (ii) she was the only person who could make clinical decisions and who had the discretion to make or not to make requests for analyses on clinical specimens collected from the members, (iii) she was aware that her clinical decisions would be implemented or carried out by the nurses at the centre, (iv) she had the supervisory role on the nurses’ work and performance in respect of clinical procedures, eg., the taking of bodily samples for testing, and of the medical services, eg., what drugs were to be supplied to the members, all as per her instructions, and (iv) she was well aware that requests for laboratory tests on clinical specimens sent from the centre would be processed by the two laboratories. 89. The Commission’s allegations are simply that the defendant has in her official capacity abused her powers and discretions by making the 75 requests “in the course of her public office”[120] for the benefit, not of any members of the centre, but of her family members. In light of all the evidence, in light of the discussion above, the court finds proved beyond all reasonable doubt that the defendant was a public officer and that the alleged misconduct has taken place in the course of her office. (2) Wilful misconduct (i) The defence argument 90. The defence raises two points here. Firstly, it is argued that the defendant has never made any misrepresentations to the laboratories. There is no evidence to prove that any of the personal particulars – including the age of the family members – entered on the request forms were false. The defence contends that making or causing the staff nurses to process the request forms at the centre amounts to no misconduct as none of the request forms were false and none of the nurses have been forced or coerced into processing them. 91. Secondly, the defence submits that in situation where the prosecution alleges, as it does here, that the misconduct concerned involved a false misrepresentation, the prosecution must prove that the defendant has acted dishonestly in accordance with R v Ghosh.[121] The defence argues that the defendant has never acted dishonestly: none of the request forms contained any false representations. (ii) Misrepresentation 92. A responsible officer of the laboratories[122] gives evidence that the laboratories have no role in verifying the details provided in the request forms, and that they would rely on the good judgments of the requesting doctors and would conduct the tests accordingly. The court finds beyond all reasonable doubt that the defendant must be aware of this long-standing practice of the laboratories. 93. False representations can be made by words, or by conduct, or both. The court finds beyond all reasonable doubt that the defendant has misrepresented to the laboratories that her family members were members of the centre and were entitled to the laboratory services as such, and that she has done so by conduct, namely, by using the request forms belonging to the centre, by causing the staff nurses to process the forms, resulting in the forms and specimens being sent to the laboratories and in the laboratories conducting the tests as per the defendant’s requests. (iii) Dishonesty 94. The court would for the time being leave aside the issue whether dishonesty must be proved as an element of offence as the defence contends. Assuming that it must be, the court finds, on the basis of the entirety of evidence, that the defendant’s acts were clearly dishonest. The reasons are as follows. 95. The staff nurses at the centre were under the defendant’s supervision in terms of clinical procedures and services and must comply with her instructions on those matters: the nurses say as much, though not in so many words in evidence. In the court’s view, if the nurses fail to carry out the defendant’s instructions on clinical procedures and on medical matters, there could be serious consequences to the members. 96. The defendant must have known that the nurses would follow her instructions on clinical and medical matters without demur, even when the clinical specimens had originated not from the members but the defendant’s family members. The court finds that it was a dishonest act on the defendant’s part to cause the nurses to deal with requests for clinical tests for the benefit of the defendant’s family members as if the tests had been requested for the benefit of members of the centre. 97. By falsely misrepresenting by conduct that her family members were entitled to the laboratory services, the defendant has, for her family members’ benefit, obtained professional analyses on their clinical specimens. The court finds beyond all reasonable doubt that the defendant’s wrongful exercises of her powers and discretions in the manner stipulated in paragraphs 92 to 96 above were dishonest according to ordinary standards of a reasonable and honest person (the first limb in Ghosh’s test). But did the defendant herself realise that her acts were dishonest by those standards (the second limb in Ghosh’s test)? 98. Here the defence seeks to rely on the defendant’s response to the Questionnaire in which she said it had never occurred to her that it was improper for her to do as she did. 99. The defendant is a well-educated professional in public service since the mid-1990s. Her work experience in the civil service in this period must have brought home to her that there are rules and regulations concerning conflict of interests and abuse of official position, that there are practices and procedures to deal with cases involving family and friends. On the issue of dishonesty in situations of misfeasance in public office, the High Court of Australia said in Northern Territory of Australia v Mengel,[123] a case that was cited with approval in AG’s Reference (No 3 of 2003) ,[124] said: “It is the absence of an honest attempt to perform the functions of the office that constitutes the abuse of the office. Misfeasance in public office consists of a purported exercise of some power or authority by a public officer otherwise than in an honest attempt to perform the functions of his or her office … Malice, knowledge and reckless indifference are states of mind that stamp on a purported but invalid exercise of power the character of abuse of or misfeasance in public office …” 100. As a result, despite her clear record and positive good character, in light of all the evidence and of the factors considered in paragraph 99 above, the court is not prepared to attach any weight at all on the exculpation part of the defendant’s answer to the Questionnaire, and finds beyond all reasonable doubt that the defendant herself must have known that her acts were dishonest by the ordinary standards of a reasonable and honest person (the second limb in Ghosh’s test). (iv) Dishonesty as a separate element of offence? 101. But is it necessary to prove dishonesty on top of the intention to commit the impugned acts? R v W(M)[125]suggests that it is. Sir Anthony Mason in Shum Kwok Sher[126] has provided examples showing that in cases of malfeasance, an additional element of improper motive is generally, if not always, required (see paragraph 31 above), saying: “[A] corrupt, dishonest or malicious motive will be required where an officer acts in excess of power … absent the relevant improper motive, … the exercise of the power or discretion would not, or might not, amount to culpable misconduct.” (emphasis supplied) 102. By using the phrases “generally, if not always” and “would not, or might not,” Sir Anthony Mason has left open the issue whether an additional element of improper motive is required in malfeasance cases. In the court’s view, a corrupt or dishonest motive in malfeasance cases must be determined against the exercise or purported exercise of the public powers or discretions concerned (see Northern Territory of Australia v Mengel[127]). 103. The general proposition remains that not only must the prosecution prove that the defendant has committed an intentional act, which includes an act recklessly done,[128] but the prosecution must also prove that the defendant has acted knowing that his act was unlawful or that he has acted being reckless as to its lawfulness.[129] In the view of this court, lawfulness of the acts concerned can involve a consideration whether the very acts themselves are or are not of a corrupt or dishonest nature. On this issue, given the long history of the offence, some old wisdom perhaps can help. 104. In R v Wooler,[130] an 1817 case, Lord Ellenborough CJ said: “[I]f he acted partially … it would be corrupt exercise of his functions.” In Kennedy v Bent,[131] an 1830 decision, Pedder CJ exclaimed: “I cannot conceive how it is possible for a public officer to shew partially in the exercise of his office, without being corrupt.” And in R v Jeyes,[132] an 1835 judgment, another Chief Justice Lord Denman CJ declared: “[P]artiality would be a corrupt motive.” (emphasis supplied) In the view of this court, the word “corrupt” when used to castigate certain misconduct can mean dishonest. Although these cases dealt with situations of partiality, there is no reason why other forms of malfeasance should be treated differently. (iv) Primary and secondary improper motives 105. To put the matter beyond doubt, the court is of the view that in a malfeasance case, if the prosecution is able to prove a secondary motive, eg., of corrupt or dishonest gains, the requirement of dishonesty is necessarily satisfied. But where the prosecution is unable to prove such a secondary motive, it does not necessarily mean that the dishonest requirement is not fulfilled. It depends, firstly, on the nature of the public powers in question, and secondly, on the manner of their abuse. 106. On the basis of the evidence, the court finds beyond all reasonable doubt that the defendant has had a secondary improper motive, namely, to dishonestly obtain laboratory reports of diagnostic significance for the benefit of her family members. (3) Has the defendant any reasonable excuse or justification 107. Without reasonable excuse or justification is an element of the offence of misconduct in public office.[133] The burden is upon the prosecution to prove it beyond all reasonable doubt (see HKSAR v Lam Yuk Fai[134]). 108. In most cases, reasonable excuse or justification falls within the exclusive knowledge of the defendant. As a reality of trial, if the prosecution can prove a secondary motive of corrupt or dishonest gains, there will not be any scope for argument of reasonable excuse or justification. Where the evidence produced, be it from the prosecution or the defence, tends to suggest that a reasonable excuse or justification might exist, the burden would be on the prosecution to negative it beyond all reasonable doubt. 109. The defence here repeats its submission on dishonesty and improper motive and argues that the prosecution has failed to discharge its onus of proof to establish the element of lack of reasonable excuse or justification. The court repeats its rulings under “Wilful misconduct”, in particular, the findings in paragraph 106 above, and rejects the defence submission on reasonable excuse or justification. In addition, the court finds that there is no room for arguments of “mistake” or “error of judgment”; there is no evidential basis for them. (4) “Serious, not trivial” (i) Defence submission 110. The Court of Final Appeal has, in Sin Kam Wah’s case,[135] provided the legal test for the required degree of seriousness for the purposes of this offence (see paragraph 47 above). While the law is clear, its application is not. On this issue, the defence submits – rightly – that “context” is everything. The defence relies, in particular, on the following cases: (i) Boulanger’s case,[136] (ii) Wong Lin Kay’s case,[137] and (iii) HKSAR v Ho Hung Kwan Michael.[138] (ii) The court’s consideration 111. In Boulanger’s case,[139] the defendant was Director of Public Security of the locality in question whose daughter was involved in a car accident. His subordinate prepared a report on the accident, but the defendant sought (as he was entitled to) a second, more detailed report, which absolved his daughter from blame, resulting in his saving CAN$250 insurance deductible. The Supreme Court of Canada held that neither the actus reus nor the mens rea for the offence was established: ‘50. … [I]t cannot be that every breach of the appropriate standard of conduct, no matter how minor, will engender a breach of the public’s trust. … 52. … [P]erfection has never been the standard for criminal culpability… To establish the criminal offence of breach of trust by a public officer, more is required. … 54. … [T]he public official’s conduct must represent a “marked” departure from the standards expected of an individual in the accused’s position of public trust. … 57. As with any offence, the mens rea is inferred from the circumstances. … [T]hat a public officer obtains a benefit is not conclusive of a culpable mens rea. … Conversely, the offence may be made out where no personal benefit is involved. 62. … [T]his may be enough to bring Mr. Boulanger within the ambit of disciplinary action …, the criminal offence of breach of trust by a public officer requires more. Specifically, it requires Mr. Boulanger to have acted with the intention to use his public office for a purpose other than the public good, for example for a dishonest, partial, corrupt or oppressive purpose which, along with his actions, represents a serious and marked departure from the standards expected of an individual in his positon.’ 112. In the court’s view, Boulanger’s case[140] is clearly distinguishable. The official duties of Director of Public Security were not detailed in the judgment. Instructing a subordinate to provide a traffic accident report did not appear to go anywhere near the Director’s core public powers, duties, and discretions apposite to public security. But the core duty of the defendant here was to provide medical and health services to the elderly. She abused her powers and discretions for personal purpose, which was clearly contrary to the public object pursued by the Department of Health. 113. It is unnecessary to discuss at length Wong Lin Kay’s case[141] and Ho Hung Kwan Michael[142], as they too are distinguishable on the same basis. But in the present case, has the defendant’s culpability met the requirement for the offence? On this issue, the court’s view is as follows. (iii) Misappropriation of valuable public resources as a species of misconduct in public office 114. The defendant’s misconduct has caused little or no demonstrable loss to the government. The fact that the defendant’s family members were entitled to alternative public medical services at out-patient and family clinics is just another way of saying that there has been no demonstrable loss to the government. But as is pointed out in Boulanger’s case,[143] the fact the government has suffered no loss does not necessarily mean that the offence is not made out. The defendant here exercised her professional judgment and decided for the benefit of her family members that their clinical specimens should be tested. But she has gone further and used the resources at the centre and the laboratories to conduct the tests. 115. When the defendant used the request forms belonging to the centre, misrepresenting that her family members were members of the centre, she had engaged herself in a dishonest exercise of her powers and discretions. Her actions resulted in the nurses having to do her private bidding in processing the requests at the centre. Worse, the two laboratories – which must have been staffed by professional chemists, analysts, and trained technicians – were caused to do the defendant’s private bidding in performing the tests and in compiling the 75 reports. 116. The court is of the view that this is a case of dishonest appropriation by a public officer of valuable human and technological resources belonging to the public and the government, public resources that are of obvious importance, because the laboratories reports contained technical information of diagnostic significance. 117. Misappropriation of valuable public human and technological resources, as a species of misconduct in public office, does not appear to have been considered in previous cases. As observed in paragraph 24 above, there are no limits to the ways in which public powers can be corruptly abused. In the court’s view, there is, in principle, no reason why misappropriation of valuable resources belonging to the public cannot support a charge of misconduct in public office. (iv) Prosecution’s concession misconceived 118. In closing submission, the prosecution makes a high-sounding concession: “If the defendant has engaged herself in the alleged misconduct only once, we would not be here.” When asked to make clear the basis of the concession, the prosecution becomes lost and incoherent. 119. The prosecution is unable to make clear whether it is accepted that the defendant’s misconduct is not so serious that unless it is repeated on multiple occasions, it would not have supported a conviction of this charge, or her misconduct is serious enough to ground a prosecution even if it is committed once, but the prosecution would not have authorised the prosecution on the basis of insufficiency of public interests to pursue the defendant to court. 120. In response, the defence replies sharply that if one instance of the defendant’s misconduct amounts to mere suspicion of a crime, she should never have been prosecuted at all because suspicion upon suspicion always remains suspicion. 121. The high-sounding concession of the prosecution is unsound. In the court’s view, in cases of dishonest appropriation of valuable public resources, one transgression is one too many. Provided that all elements of offence are present, a singular incident of misconduct of such misappropriation, subject to the consideration of reasonable excuse or justification, is capable of supporting a conviction of misconduct in public office. 122. Whether in any given case the prosecution would – on consideration of public interests – exercise its prosecutorial prerogative not to send the case to court is a matter for the prosecution. The court will place no weight on the prosecution’s concession. Conclusion 123. In light of the evidence, the parties’ submission, and the court’s consideration and findings above, the court finds all elements of the offence proved beyond all reasonable doubt, and that the defendant is guilty as charged. Representation: Mr. Derek Lai, Ag. SADPP, and Mr. Andrew Cheng, SPP, for the Prosecution Mr. Cheng Huan SC, leading Mr. Paul C. L. Leung, Mr. Quinson Leung and Mr. Dixon Tse, instructed by Yiu & Associates, for the Defendant [1] An offence preferred against the defendant in common law, punishable under section 101I(1) of the Criminal Procedure Ordinance, Cap. 221. [2] Amended Charge Sheet dated 25 July 2018 [3] Para. 9 of Civil Service Bureau Circular No.2/2004 (Exh. P99) [4] Dr Lee Siu-yin (PW1) confirmed under cross-examination that there had only been 67 occasions on which the defendant’s requests were made, in some of which more clinical tests than one to be done, making a total of 75 tests. The fees that the defendant’s family members would have had to pay if they had sought out-patient treatments were therefore $45 x 67, amounting to $3,015. In argument, the prosecution concedes that the defendant’s daughter would have been entitled to free medical services at a families clinic. The total amount of fees the defendant’s family members would have had to pay is adjusted to $2,970. See also para. 2 of Mr Leung Lai-ping’s statement dated 16 July 2018 (Exh. P111). [5] Para. 2 of Mr Chan King-hong’s statement dated 1 August 2018 (Exh. P110) [6] Para. 2 of Mr Chan King-hong’s statement dated 1 August 2018 (Exh. P110) [7] Para. 2 of Mr Leung Lai-ping’s statement dated 16 July 2018 (Exh. P111), and in argument, the prosecution concedes that the defendant’s daughter would have been entitled to free medical services at a families clinic. [8] Para. 88b of the Defendant’s Closing Submissions dated 24 September 2018 [9] Case 4 under Use of Official Position on page 5 of Department of Health Standing Circular No.12/2005 (Civil Servants’ Guide to Good Practices) (Exh. P100) [10] Case 5 under Use of Official Position on page 5 of Department of Health Standing Circular No.12/2005 (Civil Servants’ Guide to Good Practices) (Exh. P100) [11] Para. 1 of the Admitted Facts (Exh. P105) [12] Paras. 2 and 3 of the Admitted Facts (Exh. P105) [13] Paras. 6 and 7 of the Admitted Facts (Exh. P105) [14] Para. 7 of the Admitted Facts (Exh. P105) [15] Para. 7 of the Admitted Facts (Exh. P105) [16] Para. 3 and 9 of the Admitted Facts (Exh. P105) [17] Para. 28 of the Civil Service Bureau Circular no. 2/2004 (Exh. P99) [18] Para. 22 of the Admitted Facts (Exh. P105), and Exhs. P99 – P103 [19] Para. 3.6 of Guide on Conduct and Discipline (Department of Health/April 2014) (Exh. P103) [20] Para. 3 of Department of Health Standing Circular No. 12/2005 (Exh. P100), para. 1.1 of Guide on Conduct and Discipline (Department of Health/April 2014) (Exh. P103) [21] Para. 3 of the Civil Service Bureau Circular no. 2/2004 (Exh. P99), para. 2(a) of Department of Health Standing Circular No. 12/2005 (Exh. P100), para. 5.2 of Guide on Conduct and Discipline (Department of Health/April 2014) (Exh. P103) [22] Para. 9 of the Civil Service Bureau Circular no. 2/2004 (Exh. P99), para. 2(b) of Department of Health Standing Circular No. 12/2005 (Exh. P100), para. 5.5 of Guide on Conduct and Discipline (Department of Health/April 2014) (Exh. P103) [23] Para. 9 of the Civil Service Bureau Circular no. 2/2004 (Exh. P99), para. 2(b) of Department of Health Standing Circular No. 12/2005 (Exh. P100), para. 5.5 of Guide on Conduct and Discipline (Department of Health/April 2014) (Exh. P103) [24] Paras. 3 and 4 of the Civil Service Bureau Circular no. 2/2004 (Exh. P99), para. 2(a) of Department of Health Standing Circular No. 12/2005 (Exh. P100), para. 5.3 of Guide on Conduct and Discipline (Department of Health/April 2014) (Exh. P103) [25] Paras. 11 to 14 of the Admitted Facts (Exh. P105) [26] Dr Lee Siu-yin (PW1) confirmed under cross-examination that there had only been 67 requests, in some of which more clinical tests than one were asked for, making a total of 75 tests. [27] Namely, blood, urine, nail, and vaginal swaps [28] Para. 17 of the Admitted Facts, and Annex entitled “Summary of details of Test No.1 to 75” (Exh. P105) [29] Paras. 17 to 19 of the Admitted Facts (Exh. P105), Exhs. P1 – P75, and Exhs. P84 and P85 [30] Dr Lee Siu-yin (PW1)’s evidence under cross-examination [31] Para. 23 of the Admitted Facts (Exh. P105) [32] Exh. P108 and P108A [33] Para. 23 of the Admitted Facts (Exh. P105) [34] Dr Lee Siu-yin (PW1)’s evidence in re-examination [35] Para. 23 of the Admitted Facts (Exh. P105) [36] Exhs. D1 and D2 [37] Admitted Facts (Part III) dated 19 September 2018 (Exh. D11) [38] R v Rebecca King [2014] 1 Cr App R (S) 73 [39] R v Dytham [1979] 1 QB 722 and AG’s Reference (No 3 of 2003) [2004] 2 Cr App R 23 [40] R v Chapman [2015] 2 Cr App R 10, at para. 17 [41] R v Bembridge (1783) 3 Dougl 327; 99 ER 679 [42] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381, at 410C [43] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [44] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [45] See, for example, the South Australian case of R v Quach (2010) 210 A Crim R 522 [46] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [47] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381, at 404I-J [48] R v Boulanger [2006] 2 RCS 49 [49] HKSAR v Tsang Yam-kuen, Donald [2018] HKCA 425, at paras. 58 and 129 - 131 [50] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [51] HKSAR v Wong Lin Kay (2012) 15 HKCFAR 185 [52] R v W(M) [2010] 1 Cr App R 28 [53] R v Chapman and ors [2015] 2 Cr App R 10 [54] HKSAR v Wong Lin Kay (2012) 15 HKCFAR 185 [55] R v W(M) [2010] 1 Cr App R 28 [56] J R Spencer, “Police behaving badly – the abuse of misconduct in office” (2010) 69(3) Cambridge Law Journal 423; Cronin A, “Misconduct in public office: dishonesty is an element if the misconduct amounts to theft or fraud” (2010) 74 Journal of Criminal Law 290; and Davids and McMahon, “Police misconduct as a breach of public trust: the offence of misconduct in public office” 19 (vol.1) Deakin Law Review 89 [57] R v Chapman and ors [2015] 2 Cr App R 10 [58] R v Chapman and ors [2015] 2 Cr App R 10, at para. 31 [59] R v Chapman and ors [2015] 2 Cr App R 10, at para. 31 [60] R v Chapman and ors [2015] 2 Cr App R 10, at para. 32 [61] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [62] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [63] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [64] HKSAR v Chan Tak Ming (2010) 13 HKCFAR 745 [65] R v Landy [1981] 1 WLR 355, approved in R v Chan Kang To [1997] HKLRD 412, 416A-B [66] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [67] AG’s Reference (No 3 of 2003) [2004] 2 Cr App R 23 [68] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [69] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [70] AG’s Reference (No 3 of 2003) [2004] 2 Cr App R 23 [71] AG’s Reference (No 3 of 2003) [2004] 2 Cr App R 23, point (3) of para. 61 [72] R v Sheppard (James Martin) [1981] AC 394 [73] R v G [2004] 1 Cr App R 237 [74] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [75] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [76] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [77] AG’s Reference (No 3 of 2003) [2004] 2 Cr App R 23 [78] R v G [2004] 1 Cr App R 237 [79] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [80] AG’s Reference (No 3 of 2003) [2004] 2 Cr App R 23 [81] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [82] HKSAR v Chan Tak Ming (2010) 13 HKCFAR 745, at para. 5 [83] HKSAR v Tsang Yam-kuen, Donald [2018] HKCA 425, at para.14 [84] HKSAR v Hui Rafael Junior (2017) 20 HKCFAR 264, at para.20 [85] HKSAR v So Ping Chi CACC186/2016 (Judgment of McWalters JA, dated 21/11/2016, at para. 5, unreported) [86] HKSAR v Lee Wing-tai ESCC1017/2010 [87] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [88] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [89] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [90] Exhs. P108 and P108A [91] Dr Lee Siu-yin (PW1)’s evidence given on 5 September 2018 [92] Para. 23 of the Admitted Facts (Exh. P105) [93] Para. 3.11 of the Statement of Prosecution Policy and Practice [94] Prosecution’s oral submission made on 24 September 2018 [95] HKSAR v Lee Ming-tee (No. 2)(2003) 6 HKCFAR 336 [96] Para. 4-125 of Archbold Hong Kong (2018) [97] Para. 4-126 of Archbold Hong Kong (2018) [98] PW1 [99] PW4 [100] PW5 [101] Exh. D4, and the evidence of Mdm Ku Siu-hing (PW4) [102] Exh. D5, and the evidence of Mdm Ku Siu-hing (PW4) [103] Evidence of Mdm Ku Siu-hing (PW4) and of Mdm Chan Lai-kuen (PW5) [104] Eg., Exhs P76 – P83, and the evidence of Mdm Ku Siu-hing (PW4) [105] Eg., Exhs P86 – P98, and the evidence of Mdm Ku Siu-hing (PW4) and of Mdm Chan Lai-kuen (PW5) [106] Evidence of Mdm Ku Siu-hing (PW4) and of Mdm Chan Lai-kuen (PW5) [107] Evidence of Mdm Chan Lai-kuen (PW5) [108] Prosecution’s oral submission made on 24 September 2018 [109] Prosecution’s oral submission made on 24 September 2018 [110] Dr Lee Siu-yin (PW1)’s evidence in cross-examination [111] Prosecution’s oral submission made on 24 September 2018 [112] Para. 19 of the Prosecution’s closing submission dated 21 September 2018 [113] Dr Janice Lo is the third prosecution witness (PW3) who gave evidence on 6 and 7 September 2018. [114] Evidence of Dr Janice Lo given on 6 and 7 September 2018 [115] Para. 19 of the Prosecution’s closing submission dated 21 September 2018 [116] Prosecution’s oral submission made on 24 September 2018 [117] Evidence of Dr Janice Lo given on 6 and 7 September 2018 [118] Prosecution’s oral submission made on 24 September 2018 [119] HKSAR v Wong Lin Kay (2012) 15 HKCFAR 185 [120] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [121] R v Ghosh [1982] QB 1053 [122] PW3 [123] Northern Territory of Australia v Mengel (1995) 69 ALJR 527, at p. 547 [124] AG’s Reference (No 3 of 2003) [2004] 2 Cr App R 23 [125] R v W(M) [2010] 1 Cr App R 28 [126] HKSAR v Shum Kwok Sher (2002) 5 HKCFAR 381 [127] Northern Territory of Australia v Mengel (1995) 69 ALJR 527, at p. 547 [128] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [129] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [130] R v Wooler [1817] 106 ER 71 [131] Kennedy v Bent [1830] TASupC 9 [132] R v Jeyes [1835] 111 ER 471 [133] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [134] cf. HKSAR v Lam Yuk Fai (2006) 9 HKCFAR 281 [135] HKSAR v Sin Kam Wah and Anor (2005) 8 HKCFAR 192 [136] R v Boulanger [2006] 2 RCS 49 [137] HKSAR v Wong Lin Kay (2012) 15 HKCFAR 185 [138] HKSAR v Ho Hung Kwan Michael (2013) 16 HKCFAR 525 [139] R v Boulanger [2006] 2 RCS 49 [140] R v Boulanger [2006] 2 RCS 49 [141] HKSAR v Wong Lin Kay (2012) 15 HKCFAR 185 [142] HKSAR v Ho Hung Kwan Michael (2013) 16 HKCFAR 525 [143] R v Boulanger [2006] 2 RCS 49 
Introduction 1. This is a criminal contempt application, pursuant to leave granted on 30 July 2015, by the Secretary for Justice for the committal of the nine respondents in relation to events which took place on 26 November 2014 in the portion of Nathan Road between Argyle Street and Dundas Street in Mong Kok, Kowloon (“the Area”). This case is not about the right or wrong of the “Occupy Movement”. 2. The relief sought in the originating summons against the nine respondents includes that the respondent be committed to prison and / or fined for criminal contempt for interfering with and / or impeding the due execution on 26 November 2014 of the injunction order made by Au J on 10 November 2014 as amended on 21 November 2014 in HCA 2104 of 2014 by refusing to leave the Area covered by the injunction order despite repeated warnings. 3. The grounds for the application are set out in the Amended Statement of Facts filed in August 2015 in respect of each respondent. The Amended Statement of Facts sets out the background of the case. Most of the background facts are not disputed. They can be used conveniently as the background in this case. Background The Amended Injunction Order 4. As a result of what was generally known as the “Occupy Movement”, a significant part of the public highway in Mong Kok, namely, the portions of Nathan Road between Argyle Street and Dundas Street had been occupied by different individuals since late September 2014. The general public was thereby prevented to a varying extent from using the relevant roads/streets. 5. On 20 October 2014, Lai Hoi Ping (suing on his own behalf and on the behalf of all other members of Hong Kong Taxi Association) and Tam Chun Hung (suing on his own behalf and on the behalf of all other members of Taxi Drivers and Operators Association) (collectively “the Plaintiffs” in HCA 2104 of 2014 (“the underlying civil action”)) made an ex parte application on notice for an injunction against “persons occupying portions of Nathan Road near to and between Argyle Street and Dundas Street to prevent or obstruct normal vehicular traffic from passing and repassing the occupied areas” (“the 1st Defendants in the underlying civil action”). 6. On the same day, before Poon J (as he then was) in chambers (open to the public), an injunction order was granted (“the ex parte injunction order”) until the return date on 24 October 2014. 7. The granting of the ex parte injunction order was widely reported in the mass media including, inter alia, English and Chinese newspapers with wide circulation in Hong Kong, major radio and television service providers such as Radio Television Hong Kong, Commercial Radio Hong Kong, and Television Broadcasts Limited, etc, and various sources on the internet (“the Local Media”). 8. At the hearing of the inter partes application on 24 and 27 October 2014, before Au J in chambers (open to the public), judgment was reserved after the inter partes hearing and the ex parte injunction order was continued in the meantime. 9. The inter partes hearing and the continuation of the injunction were also widely reported by the Local Media. 10. By judgment dated 10 November 2014, Au J ordered that the ex parte injunction order be continued (“the Injunction Order”). The terms of the order made were finalised after a directions hearing on 13 November 2014 and amended by the judgment of 21 November 2014 (“the Amended Injunction Order”). As amended, the terms were as follows: - (1) The Defendants, and each of them, whether by themselves or agent or servant or howsoever, be restrained until trial or further order of the Court, from doing, any of the following acts, namely: - (a) Occupying portions of Nathan Road between Argyle Street and Dundas Street (“the Area”) to prevent or obstruct vehicular traffic from passing and repassing the Area; (b) Erecting, building or otherwise set up tents, canopies, barriers, barricades or other structures obstacles or obstruction, or doing any other act, to prevent or obstruct vehicular traffic from passing and repassing the Area; or (c) Obstructing or interfering with, or doing any other act which deters, the Plaintiffs through their agents properly authorised in writing from or in dismantling or removing barriers and other obstacles and obstruction in or from the Area. (2) The bailiff does take all reasonable and necessary steps to assist the Plaintiffs and its agents to effect the clearance and removal of the obstructions as provided in the Injunction Order. (3) The bailiff be authorized and directed to request the assistance of the Police where necessary. (4) Any police officer be authorized to arrest and remove any person who the police officer reasonably believes or suspects to be obstructing or interfering any bailiff in carrying out his or her duties in enforcing the terms of the Injunction Order, provided that the person to be arrested has been informed of the gist of the terms of the Injunction Order and this Order and that his action is likely to constitute a breach of the Injunction Order and this Order and obstruction of the administration of justice, and that he may be arrested if he does not desist. (5) Without prejudice to the provisions of the Police Force Ordinance (Cap 232) (including S 51 and S 52 thereof), any person so arrested by the police shall be brought before the Court, under lawful or legal processes, as soon as possible for further directions. The handing down of the judgment and the sealing of the Amended Injunction Order were also widely reported by the Local Media. 11. After the handing down of the judgment on 10 November 2104, various applications for leave to appeal were dismissed by Au J and by the Court of Appeal (Cheung CJHC, Lam VP) respectively. The dismissal of the leave applications was likewise widely reported by the Local Media. Service of the Amended Injunction Order 12. Pursuant to paragraph 6 of the Amended Injunction Order, Au J granted leave to serve the Amended Injunction Order on the 1st Defendants in the underlying civil action, by placing an advertisement in one Chinese and one English newspaper circulating in Hong Kong; and placing the Order (with a Chinese translation of the same) in clear plastic envelopes and attached to a board or post, or other prominent location in the Area (with the permission of the Police or another government authority), together with a notice (in Chinese and English) notifying the 1st Defendants that they could obtain copies of the Order, the Inter Partes Summons, Writ of Summons and supporting Affirmations by applying to the Plaintiffs’ solicitors at an identified address during normal office hours upon payment of the usual photocopying charges. 13. An assistant solicitor of the solicitors for the Plaintiffs in the underlying civil action made an affirmation that on 24 November 2014, an advertisement of the Amended Injunction Order was made in one Chinese and one English newspapers, namely Sing Tao Daily and The Standard respectively, circulating in Hong Kong. 14. On 9 March 2015, Lee Chi Leung, a bailiff officer made an affirmation for the following actions: - (1) on 24 November 2014, he served the Amended Injunction Order (together with a Chinese translation of the same) by placing them in a clear plastic envelope and attached to eleven different prominent locations at the portions of Nathan Road between Argyle Street and Dundas Street; and (2) on 24 November 2014 he served the Notice of Injunction Order dated 24 November 2014 (together with a Chinese translation of the same) by placing them in a clear plastic envelope and attached to eleven different prominent locations at the portions of Nathan Road between Argyle Street and Dundas Street. 15. The service of the Amended Injunction Order was widely reported by the Local Media. Execution of the Amended Injunction Order on 26 November 2014 16. Prior to 26 November 2014, it was already widely reported in the Local Media that the Plaintiffs and Bailiff Officers were going to take action on 26 November 2014 to execute / carry out the Amended Injunction Order. 17. Indeed, the Bailiff Officers did on 25 November 2014, execute / carry out the injunction order in similar terms granted by Au J in another case (ie HCA 2086 of 2014) over portions of Argyle Street between the junction of Tung Choi Street and that of Portland Street. The enforcement action of the said injunction order was widely reported by the Local Media. 18. On 26 November 2014, the Bailiff Officers executed the Amended Injunction Order in Mong Kok at the Plaintiffs’ request. Unless otherwise stated, all references as to time herein below refer to the time of 26 November 2014. 19. At about 8:00 am, Sergeant Li Kwong (“PW1”) (together with six other police constables) set up six loudspeakers and two metal stages at Nathan Road junction with Argyle Street (northbound and southbound carriageways), and two loudspeakers at Nathan Road junction with Dundas Street (northbound and southbound carriageways) (“the Police Public Announcement (“PA”) System”). The PA System was working properly throughout the operation on 26 November 2014. 20. Around 8:05 am, the Plaintiffs’ legal representatives and lawful agents, certain Bailiff Officers and Police Officers (“the Execution Team”) gathered at Argyle Street for the preparation of the enforcement of the Amended Injunction Order. The First Announcements (Argyle Street junction with Nathan Road) 21. At about 8:40 am, Assistant Chief Bailiff Chiu Shuk Man (“PW2”) made an announcement at the junction of Argyle Street and Nathan Road (near to HSBC at No 675, Nathan Road). Through the loudhailer, PW2 explained to the crowd at the scene that the Bailiff Officers were there to enforce the Amended Injunction Order. The following announcement was made: “各位在場人士注意，我哋係司法機構執達事務組嘅執達主任。 根據高等法院案件編號2014 年第2104 號於2014 年11 月10 日發出並於2014 年11 月21 日修訂嘅禁制令，執達主任會協助原告人嘅合法代理人，清除及移走阻塞車輛在介乎亞皆老街與登打士街之間的彌敦道各部份正常交通運行嘅障礙物。請在上述範圍內嘅人士，立即收拾你哋嘅物品離開。 任何人士，如果作出任何阻礙執達主任執行職務嘅行為，將有機會觸犯藐視法庭罪，執達主任會要求警方協助作出拘捕行動。 以下時間會由原告人代表律師，宣讀禁制令嘅內容。” (“the Bailiff’s announcement”) Following that, the Plaintiffs’ legal representatives made an announcement in Chinese of the contents of the Amended Injunction Order through the loudhailer in the following terms: “根據區慶祥法官於2014 年11 月10 日頒下並於2014 年11 月21 日修訂嘅法庭命令，要點為任何人均不得阻礙原告人嘅合法代理人清除及移走阻塞在介乎亞皆老街與登打士街之間的彌敦道各部份的正常交通運行嘅障礙物。 該法庭命令亦禁止任何人在上述範圍豎立或放置任何障礙物以堵塞該些入口及通道。 任何人不服從該法庭命令，有可能構成藐視法庭。 同時，法庭亦指示執達主任採取所有合理和所需嘅步驟協助原告人嘅合法代理人清除及移走該些障礙物。 在有需要時，執達主任可要求警方協助。 而當警方合理地認為或相信任何人正妨礙或阻礙執達主任履行本法庭命令，只要該等人士已獲悉 （1） 本法庭命令嘅要點； （2） 佢咁嘅行動可能構成違反法庭命令及妨礙司法工作； （3） 而如果佢仍然不停止咁嘅行動有可能會俾警方拘捕； 係咁嘅情況下警方有權拘捕或移走任何有關人士，並將被拘捕人士在切實可行範圍內儘快送到法庭上，作進一步嘅指示。” (“the Plaintiffs’ announcement”) 22. Following that, shortly before 8:44 am, PW2 made another announcement to the effect that the lawful agents of the Plaintiffs would execute the Amended Injunction Order and clear the obstacles at the Area within 30 minutes. The crowd was also advised to obey the Amended Injunction Order and to leave the Area immediately. The Second Announcements (Nelson Street junction with Nathan Road) 23. At about 8:45 am, Acting Assistant Chief Bailiff Yu Tak Shun (“PW4”) repeated the Bailiff’s announcement through a loudhailer near the junction of Nelson Street and Nathan Road outside Watson’s at 637 Nathan Road. Following that, the Plaintiffs’ legal representatives also repeated the Plaintiffs’ announcement at about the same location. 24. Shortly before 8:50 am, PW4 made another announcement to the effect that the lawful agents of the Plaintiffs would execute the Amended Injunction Order and clear the obstacles at the Area within 30 minutes. The crowd was also advised to obey the Amended Injunction Order and to leave the Area immediately. 25. After the aforesaid announcements, the Execution Team marched down Nathan Road towards Dundas Street. During the course, a crowd of people (including Mr Wong Chi Fung and Mr Wong Ho Ming) kept asking about the terms of the Amended Injunction Order and the details of the enforcement action. In response, Chief Bailiff Kwan Sek Nam Sunny (“PW3”) informed them that the Bailiff Officers would soon announce the details and invited them to listen carefully to such announcements. The Third Announcements (Shantung Street junction with Nathan Road) 26. At about 8:51 am, PW2 repeated the Bailiff’s announcement through a loudhailer at the junction of Shantung Street and Nathan Road. Following that, the Plaintiffs’ legal representatives also repeated the Plaintiffs’ announcement at about the same location. 27. Shortly before 8:56 am, PW2 made another announcement to the effect that the lawful agents of the Plaintiffs would execute the Amended Injunction Order and clear the obstacles at the Area within 30 minutes. The crowd was also advised to obey the Amended Injunction Order and to leave the Area immediately. The Fourth Announcements (Dundas Street junction with Nathan Road) 28. At about 9:05 am, PW4 repeated the Bailiff’s announcement through a loudhailer at the junction of Dundas Street and Nathan Road. Following that, the Plaintiffs’ legal representatives also repeated the Plaintiffs’ announcement at about the same location. 29. Following that, shortly before 9:10 am, PW4 made another announcement to the effect that the lawful agents of the Plaintiffs would execute the Amended Injunction Order and clear the obstacles at the Area within 30 minutes. The crowd was also advised to obey the Amended Injunction Order and to leave the Area immediately. The Final Announcements (Argyle Street junction with Nathan Road) 30. At about 9:49 am, the Execution Team was back at the junction of Argyle Street and Nathan Road near the barricades set up there. At the time, PW4 and PW2 respectively announced, through the loudhailer, in Chinese and English, that the lawful agents of the Plaintiffs would immediately execute the Amended Injunction Order and clear the obstacles in the Area. Following the announcements, at about 9:53 am, PW2 requested the Plaintiffs’ legal representatives and the lawful agents to clear the barricades (which consisted of Mills barriers, planks and other objects) in front of them. Clearance of the Barricades 31. During the clearance of the barricades near HSBC (at the junction of Argyle Street and Nathan Road), Mr Wong Chi Fung and Mr Wong Ho Ming kept asking through loudhailer about the identity of the Plaintiffs’ lawful agents, and quarreled with the Plaintiffs’ lawful agents who wore red‑and‑white T‑shirts and red caps for easy identification on that date. The Bailiff Officers repeatedly requested both parties to keep calm. 32. At about 9:59 am, part of the barricades at the junction of Argyle Street and Nathan Road had already been cleared by the Plaintiffs’ lawful agents. At this juncture, there was a commotion in the crowd which cried out loudly and pushed against the Execution Team. Assistant Chief Bailiff Chim Fung Ling (“PW5”) immediately advised the crowd (through the loudhailer) to keep calm and not to obstruct or impede the Bailiff Officers from carrying out their duties. During the confusion, PW5 was pushed; PW2’s sunglasses fell off to the ground; and PW4’s right shank was hit by a Mills barrier. Request for Police Assistance 33. PW5 immediately made the following warning: “根據區慶祥法官於2014 年11 月10 日頒下並於2014 年11 月21 日修訂嘅法庭命令，要點為任何人均不得阻礙原告人嘅合法代理人清除及移走阻塞車輛在介乎亞皆老街與登打士街之間的彌敦道各部份正常交通運行嘅障礙物。 你嘅行為係會阻礙執達主任執行職務。如再繼續此行為，有可能觸犯藐視法庭罪，請你立刻停止。否則，我哋會要求警方協助去拘捕你。” 34. Despite such warning, the crowd did not desist. Therefore, Chief Bailiff Choi Tak Ming (“PW6”) formally requested the assistance from the police at about 10:00 am pursuant to paragraph 3 of the Amended Injunction Order. Acting on such request, the police started to take charge of controlling the crowd and clearing the barricades. Police Actions 35. At about 10:05 am, Chief Inspector of Police Lam Chi Yuen (“PW7”) issued a verbal warning (via the PA System) at Nathan Road near Argyle Street to the crowd in Chinese and English in the following terms: “ 前面人群注意。這是警方發出的警告。高等法院的執達主任，正在此根據高等法院的命令，清除及移走路上的障礙物。高等法院命令的副本，已在附近張貼，並已在報章刊登。執達主任/代表原告人的律師亦已在這裡現場解釋法院命令的要點。任何人阻礙或干擾執達主任及協助執達主任的人員執行高等法院的命令，可能違反法院命令和阻礙司法工作。這些行為可能構成刑事藐視法庭和干犯香港法例第228章《簡易程序治罪條例》第23條的抗拒或阻礙公職人員罪，干犯此兩項的任何一項，都可被判監禁及罰款，此外，這些行為亦可能構成其他的刑事罪行。你們必須立即停止阻礙或干擾執達主任及協助執達主任的人員，否則你們可能會被拘捕及檢控。 This is a Police warning to the crowd in front of me. The bailiffs of the High Court are now clearing and removing the obstructions here in accordance with an order / orders of the High Court. Copies of the order(s) is / are displayed nearby and has/have been published in the newspapers. Bailiffs / solicitors acting for the plaintiff(s) have explained the gist of the court order(s) here. Any person who obstructs or interferes with any bailiff or person assisting him in enforcing the High Court order(s) may be in breach of court order(s) and obstructing administration of justice. Such conduct may constitute criminal contempt of court and an offence of Resisting or Obstructing a public officer under section 23 of the Summary Offences Ordinance (Cap. 228). Either of these may result in imprisonment and a fine. Furthermore, such conduct may also constitute other criminal offences. You must immediately stop your obstruction and interference with the bailiffs and those assisting them, otherwise you may be arrested and prosecuted.” 36. At about 10:12 am, PW7 made an announcement to the press at the scene requesting them to follow the instruction of the Police Media Liaison Team to retreat from the area of Police operation. 37. At about 10:15 am, as the crowd ignored the verbal warning, PW7 gave a final verbal warning (via the PA System) to the crowd at the same location in Chinese and English in the following terms: “ 前面人群注意。這是警方發出的最後警告。你們必須立即停止阻礙或干擾高等法院執達主任及協助他們的人員清除及移走路上的障礙物。如你們不立即停止，警方會按涉嫌刑事藐視法庭，抗拒或阻礙公職人員罪，和你們已涉嫌干犯的其他罪行拒捕你們，如有必要會使用最低的武力執行拘捕，而不會再作任何警告。 This is the final warning made by the Police to the crowd in front of me. You must immediately stop obstructing or interfering with bailiffs of the High Court or those assisting them in clearing and removing the obstructions on the road. If you do not stop immediately, police officers will arrest you for suspected criminal contempt of court, resisting or obstructing a public officer, and other offences which you are suspected to have committed, and police officers will use minimum force for making arrests if necessary. No further warning of arrest will be issued before arrest action.” (“the final warnings”) 38. Notwithstanding the repeated warnings from the Bailiff Officers and the Police as aforementioned, the crowd still refused to leave the Area. As a result, PW7 instructed the police officers at the scene to arrest those people who caused obstruction in the Area and refused to leave. 39. With such instruction of PW7, the police check line (which was formed by police officers) progressed from the junction of Argyle Street and Nathan Road along Nathan Road towards Tsim Sha Tsui direction. 40. Between 10:25 am and 12:00 pm, Chief Inspector of Police Sean Lin (“PW8”) and Senior Inspector of Police Cheung Kin Pan (“PW9”) took turn to make public announcements through the PA System and a microphone along Nathan Road from Argyle Street towards Dundas Street. The content of such public announcements included advice to the crowd, onlookers and press at scene to take care of their own safety, that the police might raise the level of appropriate use of force, including the use of water jet pack of pepper‑spray‑based solution, if deemed necessary. PW8 and PW9 also advised the crowd to leave the Area via Tsim Sha Tsui direction in an orderly manner and not to obstruct police officers in the execution of their duties. 41. At about 11:07 am, the police check line reached the junction of Shantung Street and Nathan Road. At the time, the crowd gathered again at the junction of Shantung Street and Nathan Road. Whilst the police officers at the scene continued to advise the crowd to leave the Area via Tsim Sha Tsui direction, PW7 issued the final warnings once again. Afterwards, the police officers started arresting those people causing obstruction and refusing to leave. 42. The enforcement action of the police officers continued up to 12:40 pm when all the barricades on the portions of Nathan Road from Argyle Street to Dundas Street were cleared. 43. During the operation, several police video teams were deployed to capture the incident from different locations. 44. Before the consideration of each of the respondent’s case, a number of issues, common to some if not all, had been raised. Criminal Contempt 45. One of the issues argued by the respondents, in particular, respondents represented by Mr Lok, was the elements of the offence of criminal contempt. It was submitted that, in order to succeed, the applicant had to establish the necessary mens rea, namely that each respondent had the intention to interfere with the administration of justice. 46. Mr Lok essentially relied on two judgments. The first was Chiu Luen Public Light Bus Co Ltd v Persons Unlawfully Occupying or Remaining on the Public Highway and others (2014) 6 HKC 298. When Au J dealt with the police authorisation direction, he mentioned in his judgment the following passages: “123. Any person who obstructs the bailiff's performance of his duties would be in criminal contempt of the court: (1) An intention and act to interfere with or impede the due administration of justice is a criminal contempt: Halsbury's Laws of England, Vol 22, paragraph 5. (2) Thus, a person who, with knowledge of a court order, and deliberately impedes the bailiff in the due execution of his duties has both the requisite actus reus and mens rea of a criminal contempt as his conduct intentionally and deliberately interferes with or impedes the due administration of justice: AG v Times Newspapers [1992] 1 AC 191 at 208E‑F, 208H‑209B, 216A‑D; Dobson v Hastings [1992] Ch 393 at 402D‑403D. (3) It is therefore a criminal contempt to obstruct or impede a bailiff in the execution of his duties: Halsbury's Laws of England, Vol 22, paragraphs 17 and 49. 124. Criminal contempt is a common law offence punishable by imprisonment: Arlidge Eady & Smith on Contempt (4th ed), paragraph 3‑67. 125. At the same time, Police are empowered under the Police Force Ordinance (Cap 232) to arrest a person suspected of being guilty of criminal contempt as: (1) Under s 50(1)(a), it shall be lawful for any police officer to apprehend any person who he reasonably believes will be charged with or whom he reasonably suspects of being guilty of any offence for which a person, among others, may (on a first conviction for that offence) be sentenced for imprisonment; (2) Under s 10(j), the duties of the police force shall include taking lawful measures for executing summonses, subpoenas, warrants, commitments and other process issued by the courts. 126. In the premises, a police officer is empowered in law to arrest any person who he reasonably believes or suspects of being guilty of criminal contempt (which is [punishable] by imprisonment on a first conviction) in obstructing or impeding the bailiff in executing his duties in enforcing a court order, such as the injunctions granted herein. 127. Thus, the police authorisation direction is only to re-state what the law has empowered the police to do.” 47. Mr Lok placed particular emphasis on paragraph 123(2). After a careful read of the judgment, it seems clear to me that Au J was principally examining the origin of the police power to arrest, hence the offence of criminal contempt was mentioned. What Au J did not do was to analyze and assess the mens rea requirement in any depth. Cases cited to him were cases from England post 1981. For reasons stated below, we now know that those cases should be read with caution. 48. The same can be said of the second judgment relied on by Mr Lok. In the Secretary for Justice v Yuen Oi Yee Lisa (unreported, HCMP 2390/2008, 25 October 2010), Chu J made the following observation: “20. As for criminal contempt (such as conduct which interferes with or threatens the due administration of justice), what has to be shown is the intentional and conscious doing of an act or the bringing of a state of affairs which objectively is likely to bring about such interference. The mens rea required is an intent to interfere with the course of justice, but it needs not be the sole intent and it may be inferred: Attorney General v. Punch Ltd [2003] 1 AC 1046, at paras.4, 66 and 87 & Attorney General v. Newspaper Publishing Plc [1988] Ch 333 at 383B‑C.” 49. Again, Mr Lok placed emphasis on “the intent to interfere with the course of justice”. In fact, four days before on 21 October 2010, Wright J delivered the judgment in the Secretary for Justice v Ng Wai Bing [2011] 5 HKLRD 620 in which, after consideration of the history of criminal contempt and relevant authorities from England as well as Australia and New Zealand in details, he came to the conclusion that the proof of specific intent to interfere with the administration of justice was not required. 50. Wright J mentioned in his judgment: “57. In the circumstances I am satisfied that the applicant need only show basic intent, that is that the respondents intended to perform the acts which constituted the actus reus of the contempt.” 51. As to cases cited from England prior to 1981, Wright J sounded the warning: “45. .... However, in 1981 the Contempt of Court Act was passed in England which had the effect of limiting the application of the strict liability rule in contempt proceedings. No such equivalent legislation exists in Hong Kong. Consequently, the decisions of the English Court of Appeal as well as views expressed by commentators subsequent to 1981 discussing the question of intent in criminal contempt proceedings need to be approached with caution.” 52. Nowhere in Chiu Luen Public Light Bus or Yuen Oi Yee Lisa was there any detailed argument or discussion about the necessary requirements of mens rea in criminal contempt. 53. In addition to discussion regarding mens rea, the actus reus of the offence was also discussed in Ng Wai Bing. Paragraphs 58 and 60 of Wright J’s judgment continued as follows: “The nature of the act 58. The act which is said to constitute contempt of court is conduct calculated to prejudice or interfere with the due administration of justice, is inherently likely to do so. … 60. There is no merit in the initial contention by the first and second respondents, albeit not later pursued, that the Court is confined to already established examples of contempt of court. ‘Contempt of court has indeed been called “the Proteus of the legal world, assuming an almost infinite diversity of forms.”’ (per Kriegler J in S v Mamabolo 2001 (3) SA 409, para.13). It will be a matter of fact and degree in every instance whether or not the alleged contempt is made out. ‘Whether such a risk has been established will depend on the circumstances of each case including the nature of the act done or the language of the publication used.’ – Secretary for Justice v Oriental Press Group Ltd.” 54. The case went to the Court of Appeal (Tang and Stock V‑PP and Fok JA as they then were). Tang VP expressed his agreement: “23. I have no doubt that the conduct of the first and second respondents (see, for example, the agreed facts quoted in para.9 above) amounted to a criminal contempt of court, and I agree with Wright J that: [53] … proof of a specific intent to interfere with the administration of justice would not be a prerequisite for liability for criminal contempt … … [57] In the circumstances I am satisfied that the applicant need only show basic intent, that is that the respondents intended to perform the acts which constituted the actus reus of the contempt. The nature of the act [58] The act which is said to constitute contempt [of] court is conduct calculated to prejudice or interfere with the due administration of justice, is inherently likely to do so. … [69] It follows that I am satisfied that the conduct of unauthorized recording of conversations of witnesses or potential witnesses at court will constitute a contempt of court.” (emphasis added) As far as this court is concerned, the elements of the offence of criminal contempt have been settled by Ng Wai Bing[1]. That decision is binding. 55. Translating that into the present case, the applicant in order to succeed, needed only to prove that each of the respondents with the intention to remain, did remain in the Area. Further, their conduct in remaining was inherently likely to prejudice or interfere with the due administration of justice given the circumstances that the Bailiffs and the Plaintiffs’ agents were about to execute or executing the Amended Injunction Order in effecting the clearance and removal of obstacles. Whether each of the respondents did, in fact, obstruct or interfere with those who were responsible for clearing up the obstacles was not part of any legal requirement. Of course, if the evidence shows that the respondent did in fact obstruct or interfere with the clearing up of the obstacles, a fortiori, it would amount to criminal contempt. 56. It was further argued that since the respondents were doing nothing more than remaining at the Area, ie given their mere presence, they were at most only liable for disobedience of Term (a) of the Amended Injunction order, hence civil contempt. In any given date prior to the 26th, I accept that that may be the case. However, this argument in my view ignores (i) the factual matrix against which the injunction was granted in the first place and that people had continued to occupy the Area; (ii) that at each stage of the injunction proceedings, the subject matter of the hearings and the result had been widely reported and advertised in the Local Media; (iii) that the clearance of obstacles on the previous day at Argyle Street had received extensive media coverage; (iv) that the imminent operation on the 26 November 2014 was widely reported likewise; and (v) that the warnings given by the Bailiffs and the Plaintiffs’ solicitors, in particular the demand made by the Bailiffs to the crowd in the Area to immediately pack their belongings and leave. The Bailiff’s demand in Chinese was: “請在上述範圍內嘅人士，立即收拾你地嘅物品離開。” 57. Whether each of the respondents intentionally chose to remain and did remain at the Area, again adopting the words of Wright J, would be a matter of fact and degree in every instance, whether or not the alleged contempt was made out. 58. It was somehow suggested that criminal contempt involved a “serious” interference with the administration of justice or a “serious” affront to the authority of the court. In other words, the interference and affront had to be serious, nothing less would suffice. In support of this legal proposition, the respondents relied on the Secretary for Justice v Cheung Kai Yin (No 2) [2016] 5 HKLRD 370. 59. Having read the judgment carefully, I do not think the Court of Appeal was adding an additional element into the offence. The word “serious” was used simply because in that case the Court of Appeal was contrasting criminal contempt with firstly the minor nature of the summary offence of Obstruction of Public Officer and secondly the mere disobedience of a court order, a less serious form of interference with the administration of justice. The word “serious” came into play under those legal contexts. Police Power of Arrest 60. Another issue raised during the course of the trial was the source of police power in arresting the respondents in this case. Having read paragraphs 126 and 127 of Au J’s judgment, it is clear that the police power originated from the Police Force Ordinance. Term (4) of the Amended Injunction Order did not confer onto the police extra power, rather it simply reinstated the origin of the police power in an unambiguous manner, so that, there was no confusion. This point was subsequently confirmed by the Court of Appeal (Cheung CJHC and Lam VP) in Chiu Luen Public Light Bus Company Limited v Persons Unlawfully Occupying and others (unreported HCMP 3028/2014 dated 21 November 2014): “19. We do not think there is any risk of confusion. As the judge had explained at paras 127 to 130 of the judgment, the arrest provision only set out what the law has already conferred on the police. The order itself is not the source of such power. Neither does the order confer any additional power on the police. Further, the arrest provision does not compel the police to make any arrest. As explained by Justice McLachlin in MacMillan Bloedel v Simpson, supra, such a provision only serves to alert persons being served with or given notice of the order the potential serious consequences for a breach of the order. In the end, it is still a matter for the police officer to decide whether he should exercise his power of arrest as a police officer. 20. At one stage, we were concerned about whether the arrest provision places undue restrictions on the police by setting the following prerequisites: (a) informing the person concerned as to the gist of the injunction; (b) warning him that his action is likely to constitute a breach of the injunction; and (c) warning him also the possibility of his being arrested if he does not desist. These requirements were obviously laid down for the protection of the defendants and persons against whom the bailiff takes action pursuant to the order. On a proper reading of the order, it is clear to us that those are steps to be taken by the bailiff in the execution of his duty under the order and the person concerned will be given a last opportunity to comply with the order voluntarily. In our view, these are steps a bailiff would have to take before a case of obstruction or interference with the carrying out of his duties can be made out. Whether a police officer would like to repeat any of these steps before he exercises his power of arrest is entirely a matter for him to decide. Viewed thus, these requirements cannot be regarded as undue restrictions on the proper exercise of police power.” 61. Consequently, in so far as the arrest of each of the respondents by the police is concerned, its relevance lies in the fact that by the time the police took arrest action, the respondents were still present at the Area (subject to the allegation that they were not allowed to leave). The circumstances pertaining to their actual arrest bore little relevance to whether they had committed criminal contempt. The Pleaded Case 62. In the course of the trial and more particularly during final submission, it was put forward by the respondents that the applicant had departed from their pleaded case and attempted to introduce new grounds and caused unfairness to the respondents in meeting their cases. 63. The alleged new grounds contained descriptions of the respondents’ other movements and conducts when they were present at the Area, for example in Chu Pui Yan’s case, the inclusion of her presence at an earlier time and posing for photographs with other people. 64. The purpose of Order 52 statement is simply to provide the respondents with enough information so as to enable them to meet the charge. It is not required to set out all the relevant evidence relied upon to establish the contempt.[2] 65. The case pleaded by the applicant was a lengthy one. It started with background information and ended with the arrest of each respondent. Paragraphs 4 to 43 stated above contain most of the pleaded case except the details in respect of the arrest of each of the respondents. In the pleaded case, the crowd was however referred to as the “obstructing crowd” and the allegation of the presence of each respondent was made in that light. The applicant then described the evidence that they would essentially rely on, namely video evidence that captured the acts and movements of the respondents. Given the length of the video evidence, particulars of their acts and movements were given. The pleaded case then concluded with the following paragraphs: “In the premises, the respondent: (1) was in the Area for an extended period of time and must have heard the repeated requests and warnings made by the Bailiffs / police officers prior to the arrest; (2) had ample opportunities to leave the Area prior to the arrest but decided against doing so; and (3) the actions taken by the respondent constituted a breach of the Amended Injunction Order. By reason of the aforesaid, the respondent must have been aware of this and the breach was intentional. The deliberate acts of the respondent identified, coupled with his determination to continue flouting the Amended Injunction Order constituted interference with and / or impediment to the due administration of justice by obstructing or resisting the police and / or Bailiff Officers in executing the Amended Injunction Order, and therefore amounted to criminal contempt of court.” 66. In my view, the pleaded case was clear. It was the presence of the respondents and their acts and movements which had caused interference and / or obstruction to the clearance operation. One should have little difficulty in understanding the applicant’s case on any objective reading. If that was not clear enough, the opening of the applicant put the applicant’s position beyond any dispute. At paragraph 85 of the opening it was stated that the essential allegation against each respondent was that they were found present in the Area during the execution of the Amended Injunction Order. 67. Subject to fairness, addition of new evidence is not unacceptable. In the case against Chu Pui Yan, the applicant was permitted to add fresh particulars which indicated her presence at much earlier times. The ground relied upon by the applicant remained the same, namely that the respondent was found to be present in the Area as part of the crowd despite numerous warnings being given. Her presence at earlier times was only revealed because video footage (unused by the applicant initially) had been used by the respondent in cross‑examining witnesses for the applicant. It was further argued by Mr Lok that this court should not be looking at any other video footage, either before or after the one used for cross examination, even though they were taken by the same police officer within the same video. I disagree. Matters must be judged in their context. So long as they are relevant to the determination of an issue in this case, I do not see any rule precluding the court from viewing them. In any event, all the video evidence, whether used or unused video, had been provided to the respondents prior. No challenge had been made as to the authenticity of the video. No witness was required to be recalled and as such I fail to see any unfairness. The same can be said to be applicable to other respondents. 68. Bearing in mind the elements of the offence, I now proceed to evaluate the evidence against each of the respondents. 69. These proceedings are brought by the Secretary of Justice against the nine respondents, the burden of proof rests on the former throughout and the standard of proof required is beyond reasonable doubt. Although a number of witnesses had been called by the applicant, with the exception of a few, their evidence was practically unchallenged. This was so because the most crucial evidence came from the video footage taken by different teams of police officers. Generally speaking, very little challenge was made against witnesses’ credibility. In view of the rather chaotic situation on that day, reliability issues were taken in respect of some witnesses. Again these issues were not seriously challenged. Given the existence of the unchallenged video evidence, the facts of this case, in particular the relevant conducts of each respondent, could therefore be readily ascertained with little difficulty. 70. I will first deal with Mr Lok’s three clients. HCMP 778/2015 Chu Pui Yan 71. Ms Chu was first spotted being present in the Area at about 8:48 am. Video footage depicted her wearing a mask and a T‑shirt “堅持到底 Never give up”. I mention the words showing on her T‑shirt not because I wish to endorse or condemn her decision in support of or against the “Occupy Movement”, it is simply because one of my duties requires me to make a determination in accordance with legal principle as to whether Ms Chu intentionally chose to remain in the Area or rather it was a mere presence. This is one piece of circumstantial evidence. 72. Ms Chu was also seen posing for photographs with a group of young people outside HSBC. This was after the Bailiffs and solicitors for the Plaintiffs had made their first announcement at the Nathan Road / Argyle Street junction at about 8:38 am and walked past her and the group of young people. 73. In fact, two further announcements were made between 8:45 am and 9:00 am. During these three announcements, Ms Chu was either sitting on or standing near a sofa outside HSBC. She remained outside HSBC till 10:29 am, not far away from where the second announcement was made. 74. After the Bailiffs’ request for police assistance and the arrival of police officers in front of the barricades, PW7 Chief Inspector Lam Chi Yuen gave his warning at around 10:03 am. Ms Chu at this juncture was present outside HSBC within a short distance from the barricades. I have no doubt that she was able to hear the police warning. Ms Chu could have left the Area as video evidence indicated that there was still ample opportunity for her to do. Yet Ms Chu chose to remain. Shortly before 10:29 am when police officers proceeded to dismantle barricades and obstacles, Ms Chu continued to stay in the middle of Nathan Road. 75. Ms Chu was seen again at about 10:46 am standing right in front of the police check line. By then she had put on her mask and goggles and continued to face the police check line while walking backwards along Nathan Road. 76. At about 11:07 am, the police check line had reached Nathan Road / Shantung Street junction. After another warning had been given, Ms Chu was arrested shortly afterwards. 77. As mentioned above, between the application for an injunction order in October and the Bailiffs and the Plaintiffs’ agents carrying out the operation in November, all the relevant matters had been widely reported in the Local Media for a month. On the 25th, there was a clearance operation in Argyle Street, a very short distance away, that too was widely reported. The Bailiff’s announcement was clear. It essentially informed the crowd to pack up their belongings and leave as the Bailiffs and the Plaintiffs’ agents would be soon carrying out an operation to clear up obstacles. 78. Given the much publicized event, the presence of strong contingent of police officers as well as local and international media and the announcements given by various parties at the time, it was highly unlikely that any citizen of Hong Kong would misapprehend the situation. In my view, those who had decided to remain and to stand in the middle of Nathan Road within the Area knew precisely what they were doing. 79. Given the length of time that Ms Chu chose to remain at the Area, the acts and movements she performed and her knowledge of the clearance operation by the Bailiffs and police officers in the Area, I am satisfied that Ms Chu had no desire to leave the Area. The presence of individual protestors like Ms Chu also lent support to other fellow protestors too. As is common in mass protests, fellow protesters banded together to fight for their beliefs. Video footage clearly indicated that there was no way the Bailiffs and the Plaintiffs’ agents were able to carry out the clearance operation by themselves given the presence of the large crowd. Their presence made the job of clearing and removing obstacles along Nathan Road substantially more difficult. Police assistance was eventually sought by the Bailiffs. Their presence and conducts were, not only inherently likely, but factually calculated, to delay if not to frustrate the Bailiffs and the Plaintiffs agents’ efforts. I have no doubt that Ms Chu’s conducts on that day amounted to a serious interference with the administration of justice. 80. Submissions were made that by the time the police took arrest action, Ms Chu was no longer able to leave the Area. To a certain extent, that is the case. But that is her own choosing. I have no doubt Ms Chu wanted to do her part till the last minute in support of what she believed to be the correct course, true to the words shown on her T‑shirt. As explained earlier, the arrest action bore little relevance to the commission of the offence. Having considered all the evidence, I am satisfied that Ms Chu is guilty of criminal contempt. HCMP 780/2015 Kwok Yeung Yuk 81. Mr Kwok was first seen at about 9:11 am standing outside the barricades. He was carrying with him a yellow/orange umbrella. I do not think anyone would dispute that the “Occupy Movement” was also known as the “Umbrella Movement”. At about 9:43 am he was captured on video again. Between 9:43 am and 10:15 am, he remained in the Area and moved around until his arrest. During his stay, Mr Kwok was seen most of the time staying close to the barricades in the middle of Nathan Road. 82. When the Bailiffs and the Plaintiff’s solicitors were making the fifth announcement at about 9:49 am, Mr Kwok was present behind the barricades, within the earshot of the announcement. I have no doubt that he was aware of the Bailiff’s warning. When the Plaintiffs’ agents started to remove the barricades, Mr Kwok was not far away and could have witnessed that. Even as late as this juncture in time, Mr Kwok still had ample opportunity to leave. Yet he remained. 83. When PW7 made the first police warning, Mr Kwok was standing immediately behind the barricades. Again I am sure he was fully aware of the warning. He chose to remain at the Area. The matters and observations mentioned from paragraphs 77 to 79 above are also applicable in Mr Kwok’s case. I have no doubt that Mr Kwok had made a deliberate decision to remain in the Area in support of the “Umbrella Movement”. His presence clearly flouted the Amended Injunction Order. Not only that, because of his presence and the presence of other protestors, their acts of defiance inevitably made the clearance much more difficult and were conducts not only inherently likely to but did factually cause prejudice to and interference with the due administration of justice. 84. Submissions were made that by 10:13 am, despite the fact that Mr Kwok wanted to leave the Area, he was not allowed to because he was by then surrounded by police officers and that all the pockets within the crowd practically had been blocked off and that he remained peaceful throughout and did not put up any resistance. Mr Kwok was then ordered to squat on the ground together with other protestors before his formal arrest. I accept that to be the case but again the circumstances of his arrest bore little relevance to the commission of the offence. Having considered all the evidence, I am satisfied that Mr Kwok is guilty of criminal contempt. HCMP 789/2015 Hung Cheuk Lun 85. Mr Hung was first seen at about 9:39 am. Video footage captured him carrying a backpack with a white helmet attached to it. Mr Hung was standing right behind the barricades, within earshot of the announcement when the Bailiffs and the Plaintiffs’ solicitors began to make the fifth announcement. 86. When the Plaintiffs’ agents began to dismantle the barricades, Mr Hung was in a position to witness all that. He did not leave the Area. At about 10:03 am, when PW7 made the first police warning, Mr Hung stood behind the barricades. Similarly, there could not be any question that he had not been made aware of the police warning. Although Mr Hung was never seen putting on the helmet, his intention was clear. He was ready to face the police. 87. Not only did he choose to remain behind, he was seen jeering and booing by one police officer (“PW41”). There was practically little cross examination of this issue. I am satisfied that PW41 was telling the truth despite the fact that it was not captured on any video footage. In fact, I would reach the same conclusion even without the evidence of PW41. The matters and observations mentioned from paragraphs 77 to 79 above are also applicable in Mr Hung’s case. His presence together with other protesters in the crowd not only constituted a flagrant breach of the Amended Injunction Order, but also created enormous burden for those responsible for clearing the obstacles. His conducts were conducts not only inherently likely to but factually did cause prejudice to or interference with the due administration of justice. 88. Similar submissions had been made that Mr Hung was not able to leave the Area by the time the police took action. For reasons stated earlier, I do not believe it bore any relevance to the commission of the offence. Had Mr Hung wanted to leave, he had ample opportunity to do so. In fact Mr Hung stayed almost all the time close to the barricades. Having considered all the evidence, I am satisfied that Mr Hung is guilty of criminal contempt. 89. I now proceed to deal with Mr Pun’s six clients. HCMP 788/2015 Kwan Siu Wang 90. Mr Kwan was first seen at around 9:43 am when he moved towards the barricades at Nathan Road / Argyle Street and joined up Mr Wong Ho Ming, Mr Lester Shum and Mr Wong Chi Fung (“the Group”). 91. Mr Kwan was seen staying in the Area close to the barricades for a significant period of time before his arrest by police officers. At about 10:12 am, he was seen having conversations with various members of the Group. 92. When the Bailiffs and the Plaintiffs’ solicitors began to make the fifth announcement, Mr Kwan was present right behind the barricades, within earshot of the announcement. I am sure not only was Mr Kwan aware of the warning, the presence of the large number of Bailiffs as well as the solicitors’ agents and police officers also clearly indicated to him that they were going to clear up the barricades. Again the matters and observations mentioned from paragraphs 77 to 79 above are applicable in Mr Kwan’s case. 93. When PW7 made the first police warning, Mr Kwan was again standing behind the barricades. I am sure he was able to hear the warning. After the first police warning, had Mr Kwan decided to leave, he still had ample opportunity to do so. In fact, Mr Kwan continued to stay in the Area, showing no sign of any intention to leave. His determination to stay with other protestors in the front inevitably created extra burden for the Bailiffs and the Plaintiffs’ agents in removing the barricades. I have no doubt that his conducts on that day not only were conducts inherently likely to prejudice or interfere with the administration of justice, but did actually produce that effect. Having considered all the evidence, I am satisfied that Mr Kwan is guilty of criminal contempt. HCMP 791/2015 Fung Kai Hei 94. Mr Fung was first seen at about 9:54 am. He emerged from a tent behind the barricades after the fifth announcement made by the Bailiffs. Mr Fung remained close to the barricades. Shortly afterwards, the Bailiffs and the Plaintiffs’ agents started to clear the barricades. 95. The emotions on both sides, the protestors and the Plaintiffs’ agents, were high. Both sides not only pointed at each other, but also exchanged unacceptable language. In fact, one person stood out on that day among the protestors, a male wearing only a shield without any upper garment on his upper body (referred to as the fake Captain America by the applicant). This male was seen carrying with him a small loudhailer and shouting across the barricades with rather unpleasant language. 96. At about 10:12 am, video footage depicted Mr Fung sending signals to the crowd further down Nathan Road regarding the police’s movement. Mr Fung was also clearly seen having a brief conversation with this fake Captain America. 97. When PW7 made the first police warning, Mr Fung was standing on Nathan Road (near Citibank) within earshot of the announcement. Mr Fung could have left the Area but decided not to do so. The matters and observations mentioned from paragraphs 77 to 79 above are applicable in Mr Fung’s case. His presence together with the presence of other protestors indicated to me that Mr Fung had no desire to leave the Area. No doubt their presence had made the Bailiffs and the Plaintiffs’ agents’ job in removing the barricades substantially more difficult. His conducts were not only conducts inherently likely to but did factually cause prejudice to or interference with the due administration of justice. Having considered all evidence, I am satisfied that he is guilty of criminal contempt. HCMP 783/2015 Chiu Chi Sum HCMP 784/2015 Chan Po Ying 98. Both Mr Chiu and Ms Chan can be dealt with together for the case against them overlapped considerably. Most of the time they were seen acting together. 99. Ms Chan was first seen at about 9:50 am standing outside HSBC. At about 9:53 am she walked towards the direction of Tsim Sha Tsui, away from the barricades. Had she wanted to leave the Area, she would have ample opportunity to do so. Instead she returned to the Area. She was seen again at about 11:07 am walking in the direction of Argyle Street. 100. PW7 gave another warning at about 11:07 am. At about 11:11 am, Ms Chan was seen holding a loudhailer passed to her by a male wearing a T‑shirt “公民抗命” while Mr Chiu was standing next to her. Thereafter, Ms Chan held the loudhailer whereas Mr Chiu acted as her guide, leading her in moving backwards against the advance of the police check line. Video footage showed that when Ms Chan was holding the loudhailer along Nathan Road, the male was making broadcast to the crowd. 101. Both Mr Chiu and Ms Chan were fully aware that they were required to leave the Area as the police check line advanced. They chose not to leave and had no desire to leave. The matters and observations mentioned from paragraphs 77 to 79 above are also applicable in their cases. They did their part in support of the movement. Their presence together with the presence of other protestors inevitably put extra burden on the people responsible for clearing the obstacles. Their conducts were conducts not only inherently likely to but factually caused prejudice to and interference with the due administration of justice. 102. Submissions were made that in fact they were not the targets of the police’s arrest and that they were retreating towards the direction of Tsim Sha Tsui. Whether or not they were the targets of the police was irrelevant to the case. They were walking backwards towards the direction of Tsim Sha Tsui, but no doubt with the intention of assisting others to slow down the advance of the police check line. 103. Having heard all the evidence, I am satisfied that they are guilty of criminal contempt. HCMP 796/2015 Mak Ying Sheung 104. Ms Mak was the only respondent who chose to testify in these proceedings. In so far as other respondents who decided to exercise their right of silence are concerned, no adverse inference had been drawn against them. In considering Ms Mak’s evidence, I also bear in mind that she enjoys a clear record in Hong Kong. That actually applies to all respondents in this case. 105. According to Ms Mak, she joined the Independent Media Hong Kong as an intern journalist at the beginning of November 2014. She was present on the 25 November 2014 during the execution of Amended Injunction Order at Argyle Street. Her task on the 25th was to interview protestors after the clearance. 106. Ms Mak was also aware that there would be another operation on the 26th and that the Bailiffs would be present to assist. She arrived at Mong Kok MTR station at 9:48 am. Her intention was again to interview the protestors. 107. Shortly after her arrival in the Area, Ms Mak overheard the police warning but was unable to hear the content in its entirety due to the noisy surroundings at the time. However, she could hear the warning requesting the crowd not to obstruct the Bailiffs in removing obstacles. Soon, blue team police officers moved in and she wanted to leave. However, as she had not been issued with a journalist identity card, she was stopped by police officers from leaving. 108. As such, Ms Mak returned to the front of the barricades and stood next to both Mr Wong Chi Fung and Mr Lester Shum. According to Ms Mak, she did not know Mr Wong Ho Ming, Mr Wong Chi Fung and Mr Lester Shum. Soon she was surrounded by police officers. All of a sudden, sensing that Mr Wong Chi Fung was about to lose his balance, she reached out her hands to help. According to Ms Mak, she had no idea that the police were in fact making arrest at the time. Soon she too was arrested. When being interviewed by the police under caution, she told the police her journalistic activity that day. 109. Ms Mak also called two witnesses, Mr Macro Mak, editor of the Independent Media Hong Kong, and Mr Lam Kin Hang, a freelance photographer, to testify. Similar to other witnesses, there was little challenge to their evidence. 110. The real issue in Ms Mak’s case is if she was wrongly arrested by the police at the time when she was doing no more than journalistic work. Having heard her evidence and viewed all relevant video footage, particularly, video footage that captured her conducts in support of the crowd, I do not think she was a credible witness. 111. I accept that Ms Mak had joined the Independent Media Hong Kong as an intern journalist in early November and that based on the Whatsapp messages appearing on her mobile phone, she had been asked by Mr Marco Mak to interview protestors on the 25th. 112. However, there was an absence of Whatsapp messages between Mr Marco Mak and Ms Mak regarding her works / tasks on the 26th, given Ms Mak was new to the journalistic field. There was no written instruction from Mr Marco Mak telling Ms Mak to return to the Area on the next day. On the contrary, the parties agreed to meet on Tuesday in the following week. On this point, I do not accept Mr Marco Mak’s evidence regarding the continuation of Ms Mak’s work on the 26th. Further we have not been presented with any photographs taken or indeed any journalistic works performed on the 26th. Ms Mak was seen most of the time looking at her mobile phone, instead of interviewing protestors. Before her arrest, she was seen standing next to the Group, yet no attempt had been made to interview any of them. 113. On the contrary, we have seen photographs depicting Ms Mak shouting and raising her arms along with other protestors. When she was cross‑examined on these photographs, her answers were unconvincing and evasive. When the police advised the media to leave the “operation area”, Ms Mak also told us that she did not know the extent of the “operation area”. I find that unconvincing too. At the time in front of Ms Mak was the large police presence, surely she should have no difficulty in realising that she was in the eye of a storm. I do not believe she was telling the truth. 114. I, of course, understand that Ms Mak was under no obligation to adduce any evidence. However, the absence of messages, photographs and journalistic work pointed to the inherent unlikelihood of her evidence. Having considered all the evidence, I come to the conclusion that whatever she did on 25th for the Independent Media Hong Kong, she was not doing that on the 26th. Ms Mak joined in the protest on the 26th. She took part in the protest with other protesters, despite the existence of an injunction prohibiting their continued occupation. The matters and observations mentioned from paragraphs 77 to 79 above are also applicable in Ms Mak’s case. 115. During her presence in the Area, there were pockets of time when she could leave without any difficulty. But for obvious reasons, she chose to remain. Her presence and conducts together with the presence of other protestors created extra burden on the people who were responsible for the clearance of the obstacles. Her conducts were conducts not only inherently likely to but did in fact cause prejudice to or interference with the due administration of justice. Having considered all the evidence, I am satisfied that Ms Mak is guilty of criminal contempt. HCMP 798/2015 Wong Ho Ming 116. Mr Wong was first seen in the footage at around 8:44 am. 117. Video footage showed that when the Bailiffs and the Plaintiffs’ solicitors were moving along Nathan Road and making announcements at different locations, Mr Wong had repeatedly asked questions on (i) whether people present in the Area amounted to obstacles; (ii) whether the Plaintiffs’ agents had been properly authorised; (iii) how to ensure the content of Amended Injunction Order being communicated to people present at the Area; and (iv) demanding the Bailiffs to use a bigger loudhailer. His challenges to the Bailiffs and the Plaintiffs’ solicitors were broadcast to the crowd. Nevertheless, the ways and manners adopted by Mr Wong were, on any objective viewing, desired to a certain extent, ridicule the Plaintiffs’ solicitors as well as to incite the crowd to continue their defiance. 118. The terms of the Amended Injunction Order were clear on their face. There were extensive legal arguments before Au J and the Court of Appeal. The proper way to clarify any purported uncertainty on the meaning and scope of the injunction should be to apply to the court to seek a variation or clarification. The same should also be applicable to the issue of authority of the Plaintiffs’ agents. Given the impracticality, it was therefore not surprising to see that nowhere in the terms of the Amended Injunction Order was there a requirement to prove authority to the protestors. There were hundreds if not thousands of them. 119. After the fifth announcement made by the Bailiffs and the Plaintiffs’ solicitors, the Plaintiffs’ agents began to clear the barricades. Mr Wong continued to challenge the Plaintiffs’ agents’ identity and authority and demanded the sight of written authorisation. His presence and challenges undoubtedly created enormous burden on the Bailiffs and Plaintiffs’ agents. The emotions on both sides were high. Accusations and counter‑accusations had been exchanged and given the chaos at the time, there was no way the Bailiffs and the Plaintiffs’ agents on their own could have carried out the clearance operation. The only resort was for them to seek the police’s assistance as contemplated. 120. At 10:02 am, the Bailiffs sought the assistance of the police. At 10:03 am PW7 gave the first police warning. Mr Wong decided to remain in the Area. When the police moved in effecting his arrest, Mr Wong stood in front of the barricades. Video footage showed that Mr Wong seemingly invited the police to arrest him and even signalled to the police that he had no weapon on his body. The matters and observations mentioned from paragraphs 77 to 79 above are likewise applicable in Mr Wong’s case. His presence and challenges to the Bailiffs and the Plaintiffs’ agents were conducts not only inherently likely to but did factually cause prejudice to or interference with the clearance operation on that day hence the due administration of justice. Having considered all the evidence, I am satisfied that Mr Wong is guilty of criminal contempt. 121. In conclusion, I am satisfied beyond reasonable doubt that all the respondents are guilty of the offence of criminal contempt. Mr Victor Dawes, SC leading Mr Jin Pao and Mr Derek Chan, instructed by Department of Justice, for the applicant in HCMP 778, 780, 783, 784, 788, 789, 791, 796 and 798/2015 Mr Lawrence Lok, SC leading Miss Yu Cheuk Wing, Christine, instructed by M/s Ho, Tse, Wai & Partners assigned by DLA, for respondents in HCMP 778, 780 and 789/2015 Mr Hectar Pun, SC leading Mr Deng, Jundi Earl, instructed by M/s JCC Cheung & Co assigned by DLA, for respondents in HCMP 783, 784, 788, 791, 796 and 798/2015 [1] Secretary for Justice v Ng Wai Bing (2012) 1 HKLRD 245 [2] Kao, Lee & Yip v Donald Koo Hoi Yan (2009) 12 HKCFAR 830 
Press Summary (English) Press Summary (Chinese) WKCC 3633/2021 [2023] HKMagC 2 IN THE WEST KOWLOON MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3633 OF 2021 Coram: Mr Peter Law, Principal Magistrate Date of Verdict: 4 March 2023 V E R D I C T CONTENTS Introduction The Facts Public Interest Immunity and Legal Professional Privilege Preliminary issues The Trial proper The Evidence Introduction 1. D1, D2, D5 and others are facing one count of failure to comply with notice to provide information. 2. The particulars of the offence: “… on the 8th day of September, 2021, being an office-bearer of Hong Kong Alliance in Support of Patriotic Democratic Movements of China in Hong Kong, or a person managing or assisting in the management of the said organization in Hong Kong, who has been served with a notice under section 3(1)(b) of Schedule 5 to the Implementation Rules for Article 43 of the Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region (Instrument A406A), failed to comply with the said notice[1].” 3. I have dealt with D1 before in other proceedings; as a professional Magistrate I will eliminate all irrelevant matters not relating to the present case and ensure a fair trial. 4. There are multiple Defendants in this case and each case must be considered separately. 5. Sch. 5 s1 states: “……………….. foreign agent (外國代理人)- (a) means a person who carries on activities in Hong Kong, and- (i) is directly or indirectly directed, directly or indirectly supervised, directly or indirectly controlled, employed, subsidized or funded by a foreign government or foreign political organization, or accepts monetary or non-monetary rewards from a foreign government or foreign political organization; and (ii) carries on all or part of the person’s activities for the benefit of a foreign government or foreign political organization; but (b) does not include a diplomatic agent, a consular officer, or an employee of a consular post, who is entitled to privileges and immunities in Hong Kong in accordance with the laws of Hong Kong, or any other person or body that is entitled to privileges and immunities in Hong Kong in accordance with the laws of Hong Kong;” 6. Sch. 5 s3 states: “Regulation of foreign or Taiwan agents (1) If the Commissioner of Police reasonably believes that it is necessary to issue the requirement for the prevention and investigation of an offence endangering national security, the Commissioner may from time to time, with the approval of the Secretary for Security, by written notice served on a foreign agent or Taiwan agent, require the agent to provide the Commissioner with the following information within the specified period in the specified way- …………………… (b) if the agent is an organization- (i) the personal particulars of the staff of the organization in Hong Kong, and of the members of the organization in Hong Kong (including name, age, type and number of identification document, occupation and residential address); (ii) the activities of the organization in Hong Kong; (iii) the assets, income, sources of income, and expenditure of the organization in Hong Kong. …………………… (3) If a foreign agent or Taiwan agent is an organization- (a) the obligations imposed on the agent by subsection (1)(b) is binding on every office-bearer in Hong Kong, and on every person managing or assisting in the management of the organization in Hong Kong, if the office-bearer or person has been served with the notice under subsection (1); and (b) if the agent fails to comply with a notice served under subsection (1)(b), every office-bearer and person who is mentioned in paragraph (a) and who has been served with the notice commits and offence and is liable on conviction on indictment to a fine of $100,000 and to imprisonment for 6 months unless the office-bearer or person establishes to the satisfaction of the court that the office-bearer or person has exercised due diligence and that the office-bearer or person has failed to comply with the notice for reasons beyond the office-bearer’s or person’s controls.” The Facts 7. The Hong Kong Alliance in Support of Patriotic Democratic Movements of China[2] (HKA) was incorporated in 1989 as a company under the Companies Ordinance until the recent winding up. At all material time, D1 was the vice-chairperson while D2 and D5 were committee members. 8. On 25 August 2021, the Commissioner of Police in pursuance NSL 43 and Sch. 5 of the Implementation Rules (IR) issued and served Notice on D1, D2, D5 and others, requiring for some specified information in writing with supporting documents within 14 days. 9. Before the expiration of the prescribed period, HKA held a press conference announcing their non-compliance and subsequently presented an open letter to the Commissioner on the last day of the period to show their dogged determination of non-compliance. 10. The Defence asserted that: (1) They were not foreign agent of any organization; (2) They will challenge the legality of the Notice; (3) They will take issues on numerous aspects; (4) Infringement of rights against self-incrimination[3] and the rights to a fair trial[4], freedom of association[5] and protection of privacy[6]. Public Interest Immunity and Legal Professional Privilege 11. The prosecution claimed Public Interest Immunity (PII) mainly on the grounds that the disclosure would jeopardise the ongoing investigation. 12. The investigation report[7] and the recommendations[8] (see attachments [1] and [2]) were not solely targeted at HKA, but also related to ongoing investigation into other organizations and persons (the other targets). 13. I have gone through all materials including the supporting affirmation of the Chief Secretary. I followed the steps laid down in HKSAR v Nyab Amin[9]. My ruling on PII was not final, I have monitored the developments throughout the trial and review when necessary to fit the best interest of the Defence. 14. The investigation is large-scale and still ongoing. National security is utmost importance to public interest. Leaking of secret information, such as identities, strategies and interim investigation results of others would definitely seriously jeopardise the ongoing investigation. 15. The key defence is the legality of the Notice, after the preliminary rulings, to prove as a matter of fact the subject organization was a foreign agent is not required. 16. I ordered a redaction on: i. The identities of entities/person are subject to ongoing investigation; ii. All acts, activities, roles and interim investigation results which could be reasonably be regarded as leading to the leaking of the identities; iii. All information relating to ongoing investigation, but not related to HKA and the Defendants; in order to minimize the risk of any sidetrack strategy leading to reasonably guessing as to the identities of the targets; especially the exact amount or details of monetary transactions. After balancing, I deployed some safeguards, only to disclose the figures in number of digits. 17. The focus is on the factual nexus involvement leading to the triggering of the measure rather than the identities of others. I do not see the non-disclosure of materials, other than which related to HKA and the Defendants would undermine a fair trial. 18. The PII is based on the said direction, I came to the view that the anonymity with limited disclosure of some factual nexus be sufficient for purpose of conducting the defence case and ensuring a fair trial. I am satisfied there is no possible detriment or disadvantage of any kind or degree to the Defence. 19. As to the legal advice given by the Department of Justice, it is protected by Legal Professional Privilege. Preliminary issues 20. Two issues have been posed for preliminary determination: “(1) Legality of the Notice Whether it is open to the defence to challenge the legality of the Notice in the context of Schedule 5 to the Implementation Rules in a criminal trial. (2) Elements of the offence Under Section 3(3) of Schedule 5 to the Implementation Rules, whether the prosecution need to prove the subject organisation was as a matter of fact a foreign agent.” 21. I deal with the second issue first. 22. The information required under Sch. 5 are related to privacy, any overriding of fundamental rights should take a narrow approach. 23. NSL 43 confers on the Police some power to obtain information by two means: (1) By a Police Notice under Schedule 5. (2) By a Production Order issued by a CFI Judge under Schedule 7. 24. There is a statutory definition of foreign agent under Sch. 5 s1, but silent on the criteria to identify a particular organisation as a foreign agent (the threshold). 25. Under Schedule 7: “Requirement to furnish information and produce materials (1) The Secretary for Justice may, for the purpose of an investigation into an offence endangering national security, make an ex parte application to the Court of First Instance for an order under subsection (2) in relation to a particular person or persons of a particular description. (2) The Court of First Instance may, if on such application is satisfied that the conditions in (4)(a), (b) and (d) or 4(a), (c) and (d) are fulfilled, make an order complying with subsection (3) in respect of the particular person or persons of the particular descriptions to whom the application relates.” 26. Under Sch. 7 s2(4): “…………………. (b) Where the application relates to a particular person that there are reasonable grounds for suspecting that the person has information, or is in possession of materials likely to be relevant to the investigation; (c) Where the application relates to persons of a particular description, that - (i) there are reasonable grounds for suspecting that some or all persons of that description have such information or in possession of such material, and” 27. An analogy can be drawn between Schedule 5 measure and Schedule 7 on some special features. 28. There is no doubt that the Sch. 5 measure is comparatively less stringent than that under Sch. 7. 29. NSL 3 states: “………………………….. It is the duty of the Hong Kong Special Administrative Region under the Constitution to safeguard national security and the Region shall perform the duty accordingly. The executive authorities, legislature and judiciary of the Region shall effectively prevent, suppress and impose punishment for any act or activity endangering national security in accordance with this Law and other relevant laws.” 30. Taking together both the effectiveness requirement under NSL 3, and the said special features of the Sch. 5; the IR is intended to provide an effective administrative procedure to facilitate the implementation of the National Security Law, in particular to NSL 43(5). 31. I have the following observations: (1) It is clear that measures taken under Sch. 5 are meant to be responsive and effective, which is the purpose of NSL 3. (2) To set up a foreign agent or any associate with it is not an offence. There is no mandatory scheme or mechanism for setting up a foreign agent. It can even be as simple as registered as a company. As for an individual, there is also no requirement for registration. (3) There is no mechanism or scheme for setting up a foreign agent, nor is there any list thereof. (4) A foreign agent is inevitably associated with overseas connections. To seek information, assistance from foreign authorities is unavoidable but, that would also be delayed with unexpected difficulties and even be unfeasible under the current climate. The lawmakers were well aware of that situation and the difficulties. (5) Sch. 5 measure is comparatively less stringent than Sch. 7 for reasons that: (i) At a more peripheral stage. (ii) Less severe on the maximum penalties upon violation. (iii) More simplified in procedure. 32. From the above observations, I conclude that: a) Lesser severity of the measure should be proportionate to lesser stringency of the threshold. b) The overall purpose of Sch. 5 is an effective measure for prevention and investigation of matters relating to national security. To be effective, the measure must be responsive and efficient. c) There is a deliberate silence on the threshold requirement for identification of a foreign agent; obviously the rule‑makers were minded to create some flexibility for the Commissioner to exercise his professional judgment at that juncture, regardless of the strict rules of evidence, the burden and standard required in a criminal trial. d) Identifying the foreign agent is the entry to and also part and parcel of the long process of effective prevention and investigation. Looking into the background and the purpose of the promulgation coupled with the deliberate silence on the threshold requirement; it is clear that the lawmakers and the rule-makers were not intended to create the proof of foreign agent as a matter of fact be an element of the offence upon non-compliance. e) Prevention and investigation is a state of surmise, the threshold must be relatively low, even not requiring prima facie proof. 33. I rule that as a matter of law: (1) The concept of foreign agent is the conclusion of an administrative decision at that juncture, not an essential constituent element that the prosecution has to prove at the trial. (2) The prosecution need not prove the subject organisation was as a matter of fact a foreign agent. 34. I move back to the first issue. (1) There is no existing mechanism for licensing nor registration of foreign agent. The norm is all entities are without obligation to disclose their composition, activities or means unless by stipulation. (2) I am not convinced that, as a matter of law, the defence is barred from challenging the legality of the Notice on aspects not been ruled by me. I rule that the defence is entitled to challenge the legality of the Notice at trial for matters not been dealt with before. The Trial proper 35. After my rulings on the preliminaries, the remaining issues for the trial proper is mainly on the legality of the Notice. The Evidence PW1 36. Superintendent Hung was in charge of the investigation team as well as the applicant to the Commissioner for the Sch. 5 Notice. He prepared an investigation report[10] refined the information therein and submitted his application with recommendations[11] to the Commissioner who subsequent endorsed them with final approval by the Secretary for Security. 37. In the investigation report and the application, HKA was amongst the other targets for the Sch. 5 measure. 38. For easy reference, Mr Cheung for the prosecution has prepared a chart[12] (see attachment [3]) on the key points of: (i) The relevant information with the corresponding contents in the investigation report which was confirmed in evidence; (ii) The information put forward to the Commissioner for consideration. 39. PW1 explained that even though some information could be retrieved from other sources, but still necessary for cross-checks on completeness, accuracy, mistakes and different interpretation. 40. Under the marathon and intensive cross examination by D1 on numerous aspects, mostly circling around on: (1) By using a lot of guessing tactics from all angles aiming at retrieval of the redacted information, in particular the identities of the organizations and persons; trying to establish HKA was not a foreign agent as PW1 had thought. (2) To distinguish the exact wordings of the political ends of HKA from other organization, to illustrate they were not on the same track; such as “to end one party ruling” as oppose to “end one party dictatorship” and “to rebuild a democratic China” as oppose to “to build a democratic China”; (3) To laid the basis for her subsequent testimony to establish or raise doubt on to HKA not being any foreign agent with some examples; (4) The query about the possible unreasonableness, that some information could be retrieved from other source; (5) Make suggestion that certain description and events could have more than one interpretations and highlight some participants with double status who participated in a capacity not related to HKA; (6) Suggestion of ulterior motive; bring out the disparity to illustrate possible fabrication as the Police had commenced strike off proceedings against HKA almost at the same time without mentioning HKA was a foreign agent. PW2 41. PW2 was the arresting officer upon instructions arrested D1 at her office on 8 September 2021. 42. All Defendants have clear record, I have to consider their propensity and also the credibility of D1 as she testified. D1 43. D1’s testimony was that she joined HKA after her graduation from university and had held different positions, including part time staff, volunteer, committee member and lastly became the vice-chairperson in 2015 till the compulsory strike off from the Companies Registry. She told about what she had experienced over the years as well as some of her observations. The principles, history and the operation 44. HKA was a joint organization of lots of bodies. D1 reiterated HKA had never an agent or a puppet of any foreign entity. It acted independently according to their own objectives. Their Memorandum and Article of Association stipulated the income and property be applied solely on the promotion of their objects. Under the mechanism, the highest decisions were made at the general meeting and the committee was responsible for the daily operation. Explanation on the $20,000 from Org 4[13] 45. HKA had launched an appeal for donations for the extension of their June 4 museum by purchasing a property at $8,000,000. They had been receiving donations from various bodies including a one-off donation (not funding) of $20,000 from an entity which was a stranger and never had and would not have other dealing with HKA. Explanation on the $3,000 from “Asia Democracy Network” 46. It was her personal reward for writing articles and nothing to do with HKA. Since the request was about HKA’s affair and it had nothing to do with her personal matters. Interaction with other entities 47. In the course of its normal business, HKA had met with other entities which shared common objectives and involved in some interactions, including cooperation and joint participation in some activities and yet, it had never been an agent or acting for others’ interest. Donations 48. HKA accepted general donations, but the donors had no say in HKA’s operation nor had there been any case that it had acted for their interest. The allegation of foreign agent 49. D1 alleged that it was the Government’s tactics for propaganda to shift the focus from the international eye. Incorrect construction by PW1 50. HKA never used such slogan: • rebuild a democratic China as oppose to build; • end one party ruling as oppose to directorship. 51. PW1 had misunderstood the status of some participants who had more than one capacities who shown up in HKA activities. 52. D1 criticised PW1’s suggestion of the alleged common purpose was not substantiated Unreasonableness 53. D1 criticised the requirement is too harsh, too much, and too far. 54. She contended some of the information could have been retrieved from other sources, such as the director’s list from the Companies Registry, the account ledger from auditor, activities from the yearbooks and activities pamphlets or open announcements. 55. To retrieve information of the staff even dated back to 1989 is unreasonable since it was so long ago and with lots of difficulties. They even lost contacts with some of the staff. Furthermore, asking for identity card numbers and date of birth would trigger privacy issues. Oppression 56. Due to insufficient manpower and resources to deal with such large amount of information required, some of which even stored in the warehouse causing hardship. However, D1 agreed no specific arrangement nor steps had been taken to retrieve it, they only had discussion. Legal issue 57. The requirement for provision under Sch. 5 s3(1) is “… to provide the Commissioner with the following information in writing, together with the relevant supporting documents, …”, however the extra requirements of “in writing” and “together with supporting documents” are outside the perimeter of Sch. 5. Other matters 58. D1 also asserted that: • the issue of the Notice was an abuse with an ulterior purpose of doxing information of all civil societies; • Political persecution; • HKA had no obligation to assist the Police. Evaluation of evidence 59. PW1’s testimony was focused on the information from his team’s investigation and his analysis. He was unshaken under intensive and marathon cross-examination. He was not narrow-minded, and was willing to accept suggestions of possible multi-interpretations or misunderstanding without hesitation or delay. 60. PW1 declined to give answer on some areas due to the possible risk of jeopardising the restrictions in PII. Since he was in-charge of the whole information and in the know of the ongoing investigations, he was in the best position to evaluate the potential risk of leaking information; in particular he had been at the centre of an guessing exercise. His refusal to answer was justified. 61. I am satisfied PW1 was an honest and truthful witness. I am satisfied when he evaluated the investigation result from his team members, he held an honest belief in its truthfulness. In this regard he performed an honest and bona fide analysis in his best endeavour and professionalism. I accept the flaws from the misinterpretation and misunderstanding would not affect his overall judgement. I accept his evidence in the factual aspect and also the facts that exhibits P19(1) and P20(1) are his honest and bona fide analysis and recommendations, being direct and not bent by the flaws. 62. However, this is not the end of the matter. As to his analysis, I still have to take a panoramic look into the operational and other legal aspects on that factual basis. 63. PW2 was honest and truthful. He was the arresting officer with insignificant role in the incident and there was no challenge on his credibility and truthfulness of his testimony. I accept his evidence. 64. D1’s testimony can be divided into: (i) a description of the nature of HKA; (ii) her role in HKA and her experience; (iii) the factual material about certain incidents; (iv) her complaint of the faults in the Notice; (v) her criticism of the analysis of PW1. 65. For the factual aspect, the prosecution was not in any position to challenge the truthfulness and the substance as PW1 was an outsider did not have any participation in HKA’s activities; on this basis I accept those facts for the purpose of this trial. With regard to her speculative observations, those were lack of foundation, I cannot accept that. 66. As to her observations related to legal aspects, I will deal with that in due course with the accepted factual basis. 67. Her suggestion of options for retrieving information from open source; such as the yearbook, pamphlets and open announcements may negate part of the challenge on privacy will also be considered in due course. 68. D2 and D5 elected to remain silent nor did they call any defence witness. No adverse would be drawn for exercising their legal rights. Finding of facts 69. There is not much dispute over the facts and my facts finding, inter alia, is: (1) Since the establishment in 1989, HKA had been active and engaged in multiple nexus activities and interaction with Hong Kong and non-Hong Kong entities and people as stated in the investigation report[14] and the recommendations[15]; also had been holding at least one mass rally annually; (2) Throughout the entire period concerned, direct and indirect flow of funds was recorded; (3) HKA had the five operational goals throughout the years, some, not all, were common in nature with other entities; (4) D1, D2 and D5 were amongst the office-bearers at the material time; (5) PW1 honestly relied on and evaluated the investigation information, and he bona fide compiled the investigation report[16] and made recommendations[17] to the Commissioner; (6) The Commissioner endorsed the whole of PW1’s recommendations without any query or clarification; (7) The Secretary of Security approved the measure; (8) HKA held a press conference and sent an open letter to the Commissioner with D1, D2 and D5’s cosignatory; (9) Up to the due date, none of the required information was provided; (10) The Notice was properly served on all Defendants; (11) The Defendants were aware of the requirement. Legality of the Notice 70. The objective of NSL 43 and Sch. 5 is for prevention and investigation of an offence endangering national security. The foundations are: (1) honest belief in the truthfulness of the information available at that time; (2) the analysis was in bona fides; (3) the Commissioner made his decision according to the information at that time, not in hindsight. The Requirement 71. The Notice: “According to the police investigation, the Commissioner of Police has reasonable grounds to believe that the “Hong Kong Alliance” is a “foreign agent’ specified in Section 1 of Schedule 5. Under Section 3(1) of Schedule 5, as the Commissioner of Police reasonably believes that it is necessary to issue the requirement of this notice for the prevention and investigation of an offence endangering national security, the Commissioner, with the approval of the Secretary for Security, by this notice served on the “Hong Kong Alliance”, requires the “Hong Kong Alliance” to provide the Commissioner with the following information in writing together with relevant supporting documents, within 14 days: [1] For the staff members in Hong Kong and the members in Hong Kong since the establishment of the “Hong Kong Alliance”, that is, the following persons: [i] directors, [ii] standing committee members (and) [iii] full-time staff members, Their personal information, including name, date of birth, type and number of the identification document, contact phone number, residential address, position and employment period shall be provided; [2] For the activities held in Hong Kong by the “Hong Kong Alliance” from 2014 to present, involving the following organizations or person [regardless of the organizations or person participating in the activities as joint organizers, co-organizers, sponsors or attendees, and including the activities conducted by contracting with the places outside Hong Kong with the use of the communication technology]: [i] “New School for Democracy” established in Hong Kong, [ii] “China Human Rights Lawyers Concern Group”, [iii] political parties outside the territory of the People’s Republic of China or in Taiwan or other organizations outside the territory of the People’s Republic of China or in Taiwan that pursue political ends and/or their branches in Hong Kong, including “Federation for a Democratic China”, “Asia Democracy Network”, “New School for Democracy”, “The National Endowment for Democracy” and the organizations that receive money from “The National Endowment for Democracy” [including “National Democratic Institute”, “International Republican Institute”, “Solidarity Center” and “Center for International Private Enterprise”], (and) [iv] Mark Herman SIMON[18], Information on the activities concerned, including name, purpose, date, time and place of the activity, details of the person-in-charge, source of fund and breakdown of the expenditure, and all records of contacting the aforesaid organizations or person for organizing, holding, sponsoring or attending the aforesaid activities [including the electronic communication records] shall be provided; [3] Minutes of the following meetings that the “Hong Kong Alliance” convened in Hong Kong from 2014 to present [including the meetings with the places outside Hong Kong by communication technology] shall be provided: [i] board meetings and standing committee meetings, [ii] meetings with political parties outside the territory of the People’s Republic of China or in Taiwan or other organizations outside the territory of the People’s Republic of China or in Taiwan that pursue political ends and/or their branches in Hong Kong [including “Federation for a Democratic China”], (and) [iii] meetings with Mark Herman SIMON; [4] The assets, revenue, sources of revenue and expenditure of the “Hong Kong Alliance” in Hong Kong from 2014 to present shall be provided, that is; [i] details of account(s) directly or indirectly held with the local bank(s) currently or previously, including the account number, the account holder and/or the ultimate beneficial owner, [ii] details of the assets like the property, vehicle(s), stock(s), etc., in Hong Kong directly or indirectly held currently or previously, [iii] all transactions and money dealing, including reason, purpose, date, amount, the bank account and the account book(s) involved, with “New School for Democracy” [whether it was established in Taiwan or Hong Kong or other places]; “China Human Rights Lawyers Concern Group”; “Federation for a Democratic China”; “Asia Democracy Network”; “The National Endowment for Democracy”; and the organizations that receive money from “The National Endowment for Democracy” [including “National Democratic Institute”, International Republican Institute”, “Solidarity Center” and “Center for International Private Enterprise”]; Mark Herman SIMON; and political parties outside the territory of the People’s Republic of China or in Taiwan or other organizations outside the territory of the People’s Republic of China or in Taiwan that pursue political ends and/or their branches in Hong Kong, (and) [iv] reason and purpose of CHOW Hang-tung, the vice-chairperson of the “Hong Kong Alliance”, receiving some HK$3,000 from “Asia Democracy Network” on 4th February 2021 shall be provided.” The Challenges 72. In Leung Kwok Hung[19], the Court of Appeal held: “182. Thus, the proportional analysis has to be applied on two different levels: (1) examining the systemic proportionality by reference to the legislation or rules in question: (2) examining the operational proportionality by reference to the actual implementation or enforcement of the relevant rule on the facts and specific circumstances of a case at the operational level. 183. In these appeals, it should be emphasized that we are only concerned with the first level of challenges. It remains for the court to assess the proportionality on the second level on the facts and circumstances in a particular case if a charge is brought against person.” Systemic 73. NSL 43 confers on the law enforcement authorities, including the police, certain power: “When handling cases concerning offence endangering national security, the department for safeguarding national security of the Police Force of the Hong Kong Special Administrative Region may take measures that law enforcement authorities, including the Hong Kong Police Force, are allowed to apply under the laws in force in the Hong Kong Special Administrative Region in investigating serious crimes, and may also take the following measures:” ………………. “… (5) requiring a political organisation of a foreign country or outside the mainland, Hong Kong, and Macao of the People’s Republic of China, or an agent of authorities or a political organisation of a foreign country or outside the mainland, Hong Kong, and Macao of the People’s Republic of China, to provide information;” The National Security Law 74. In HKSAR v Lai Chee Ying[20] Court of Final Appeal held: “37. In our view, in the light of Ng Ka Ling v Director of Immigration (No 2), the legislative acts of the NPC and NPCSC leading to the promulgation of the NSL as a law of the HKSAR, done in accordance with the provisions of the Basic Law and the procedure therein, are not subject to review on the basis of any alleged incompatibility as between the NSL and the Basic law or the ICCPR as applied to Hong Kong.” 75. In the light of the above context; NSL is not the subject of any challenges. The Implementation Rules 76. Under NSL 43: “… The Chief Executive shall be authorized, in conjunction with the Committee for Safeguarding National Security of the Hong Kong Special Administrative Region, to make relevant implementation rules for the purpose of applying the measures under the first paragraph of this Article.” 77. NSL 13 stipulates the composition of the Committee for Safeguarding National Security be: “The Chief Executive shall be the chairperson of the Committee for Safeguarding National security of the Hong Kong Special Administrative Region. The other members of the Committee shall be the Chief Secretary for Administration, the Financial Secretary, the Secretary for Justice, the Secretary for Security, the Commissioner of Police, the head of the department for safeguarding national security of the Hong Kong Police Force established under Article 16 of this Law, the Director of Immigration, the commissioner of Customs and Excise, and the Director of the Chief Executive’s Office. A secretariat headed by a secretary-General shall be established under the Committee. The Secretary-General shall be appointed by the Central People’s Government upon nomination by the Chief Executive.” 78. Under NSL 14: “The duties and functions of the Committee for Safeguarding National Security of the Hong Kong Special Administrative Region shall be: ………………… (2) advancing the development of the legal system and enforcement mechanisms of the Region for safeguarding national security; and ………………… No institution, organization or individual in the Region shall interfere with the work of the Committee. Information relating to the work of the Committee shall not be subject to disclosure. Decisions made by the Committee shall not be amenable to judicial review.” 79. Judicial review is not a term of art; it must be construed accordingly to ordinary language together with the purpose of promulgation. The only logical understanding is the Committee’s decisions not be amenable to any judicial proceeding or decision. 80. Also in the light of the above contexts, the Implementation Rules, are not the subject of any challenge. Operational 81. NSL 42 and Sch. 5 confer on the Police the power to issue the Notice with consequential penalty in case of non-compliance. At the same time it is expected the Police will exercise the executive power properly. Although the charge is failure to comply with the Notice, the essence of the trial is basically on the legitimacy of the Notice as the recipients contended they were not obliged to answer the purported Notice. (ab initio). 82. Being a foreign agent is not an offence, the impugned provision is the non-compliance of the Notice. Owing to that special feature in this case, the legality of the Notice requires some studies. 83. The application of operational proportionality analysis must be taken in full picture and done objectively. “Reasonable grounds to believe” it to be a foreign agent 84. The IR already set out in Sch. 1 and Sch. 7 the mandatory standard of the threshold for identification of the target[21] or recipient[22], but none is provided in Sch. 5, the Commissioner adopted the threshold of “reasonable grounds to believe”. 85. I have the following observations: (a) To identify the foreign agent is the initial step to the measure; (b) When multiple organizations, people and interactions were involved; some even overseas. Adopting a stringent standard of identification would be unrealistic; (c) Information obtained at the early stage would normally be in loose pieces; (d) Ensuring effectiveness is essential[23]; (e) National security is of cardinal importance; (f) There is no existing mechanism for registration nor is there any existing list of foreign agents; (g) The overall difficulties in the entire situations. 86. Unlike Sch. 1 or Sch. 7 which requires strict judicial scrutiny, the application of Sch. 5 is comparatively less rigid and a slightly lower threshold is to be expected. “Reasonable grounds to believe” is not in itself a low standard but, just slightly lower than “reasonable grounds for suspecting”; with which there is still a series of hurdles to overcome in the balancing exercise. 87. Having regard to the nature, purpose, necessities and to strike a balance between the measures and the rights concerned; the decision to adopt the threshold of “reasonable grounds to believe” can hardly be criticized. 88. With their backgrounds, political aims, activities and nexus with both locals and non-locals throughout the years, suffice to say is that “… the Commissioner of Police has reasonable grounds to believe that the ‘Hong Kong Alliance’ is a foreign agent’…”[24] was the correct approach. The material times 89. The material time relevant to legality is the point of time when the Commissioner made his decision, not in hindsight. Any subsequent information, no matter how significant it was, was irrelevant as the legitimacy had already been frozen at that point of time. Reasonable belief as to necessity 90. All measures taken for prevention and investigation on national security must be executed in a solemn way, and thoroughly, and with the highest standard of professionalism. A professional investigation body will definitely have its own judgement and strategy to act on the needs raising in different circumstance. Exploration and analysis of information is the usual course of exercise. Divergence in the deployed tactics per se cannot be criticised as unreasonable and rejected unless it is found to be obviously absurd. 91. To evaluate reasonableness of needs, one must not derail the purpose of NSL, the IR and the reality. 92. Foreign agent is a new concept in Hong Kong without any legal precedent. Being an agent per se is not an offence and there is no existing mechanism for registration, everything has to start from zero. Some agents are even hidden, so a comprehensive searching and screening exercise is necessary. 93. HKA was set up in 1989 and had been carrying out non-stop political activities mainly circulating around the June 4 incident throughout the years. Most of their activities were nexus of interactions with local and non-local organizations and people. Inevitably a comprehensive list of directors and full-time staff would be required from the outset to ascertain the background of the organization. Retrieving their personal information for the purpose of identification was essential to the exercise. 94. Information of recent years that needed to be ascertained included from the minutes, their interactions, affiliations, and also the monetary flows with some named specified subjects and entities. HKA’s full list of assets was also essential for the ascertaining of its backgrounds and functions. 95. Since HKA had been running actively with various entities and people aboard, it is necessary to explore their dealings, connections, monetary flows and asset in order to find out their affiliation and ultimate purpose. 96. Such requirement for information was nothing like a broad-brush fishing exercise but rather was constrained in terms of periods of time and nature. The police had taken an abstemious and self-restrained approach. Use of wordings 97. “Information” is a term with wide coverage of meanings, which includes all sort of facts or details about a situation, person and event. 98. According to the Notice: “To provide the Commissioner with the following information in writing, together with relevant supporting documents, …” 99. Under the interpretation section of the Personal Data (Privacy) Ordinance[25] states: “data（資料）means any representation of information (including an expression of opinion) in any document, and includes a personal identifier,” 100. “in writing” is the means of transmission of information, “supporting documents” is additional information requiring for the purpose of verification/corroboration. Personal data is a type of information contains personal details. 101. “Investigation” is a broad description of the exercise, including to retrieve information and verification of its correctness. 102. The enabling law require the information is NSL 42(5): “requiring a … to provide information;” Suffice it to say that NSL 42(5) embraces Sch. 5 s3(1) and the Personal Data (Privacy) Ordinance. I cannot see the information required is anything would go beyond the perimeter of NSL 42(5) and Sch. 5. Retrospective 103. In the present case, there are two aspects to explore: (1) At the date of the offence; (2) The calling for information of events took place before the promulgation of NSL. 104. The present charge is non-compliance of the Notice served on 25 August 2021; there is no room for argument on retrospective issue. 105. It has been suggested that some of the information required was dated before the promulgation of NSL and some even back to 1989 when it was a time of a different regime. 106. The concept of national security is not just limited to an outbreak at a particular point of time but instead is, as can usually be found in most of cases, a continuation of series of acts with accumulative and generative aim to an ultimate end, be it a part of the adventure under the same or another different regime. In this case, the claim for retrospective limitation is invalid. Alternatives 107. Sch. 5 is intended for prevention and investigation. Hence no matter the information required was obtained by direct approach or indirect research from their yearbooks and pamphlets. The more direct must be better; at least to minimize the risk of delay and omission. 108. It would be unrealistic to expect the Police first to obtain a full collection of yearbooks and pamphlets, then approach the target under investigation for its assistance in verification as to correctness and completeness at its discretion. 109. Sch. 1 confers on Magistrate the power upon application, to issue search warrant to enter (to use reasonable force if necessary), to search, examine, seize and can detain anyone found therein until the conclusion of the search. 110. Sch. 7 is about an ex-parte application by the Secretary for Justice to a judge of the Court of First Instance for a Production Order. The maximum penalty is one year imprisonment and a fine of $100,000, which is double the imprisonment in Sch. 5, upon violation. 111. From the above information, suffice to say the Sch. 5 measure is the mildest of all. Ulterior motive 112. Although the police had initiated another proceedings against HKA, striking off HKA from the Companies Registry is under a different mechanism and criteria. Any omission from mentioning foreign agent is nothing odd or skeptical which should lead to any reasonable doubt on the genuineness of the Sch. 5 measure taken before me nor was there any ulterior motive behind. 113. I am satisfied Superintendent Hung held an honest belief in the truthfulness of the investigation information and he acted bona fide on his analysis. Oppression 114. The assessment of oppression is not merely on theoretical concept; it must be practical by looking into the whole picture, including the capabilities, resources and the conduct of the recipient. 115. The requirement of large amount of information, some even aged, within 14 days, sounded tough on the face of it, but there are some additional features of this case: (1) A liaison contact point was provided in the Notice, which could form a channel for some constructive and potential relief if necessary; (2) The high-profile press conference and the open letter, was a clear message of total refusal; (3) That letter clearly stated that the refusal was due to some legal issues involved, which shows the recipients had considered all the information required; (4) None of the required information was provided in the end. 116. I have the following observations: (1) Most of the information required was not that aged, only back from 2014; (2) Some of the information required is actually the information that needs to be maintained. Examples are: (i) accounting records and the supporting documents for years for spot check by the Inland Revenue; (ii) the tax return and the provident fund documents with personal details of all the employees; (3) D1 offered an explanation to the Court about the details of some of the requested information which was supposed could have been given to the Police at that time. From what can be seen, it was not that difficult to retrieve the required information; (4) No constructive actual attempt had been made for any retrieval, not even on the easy ones. 117. Based on the above, it is difficult to justify the significance of hardship was in any sense compromised when the Defendants had provided none, not even some were not that hard to be accessed. There is no room for any claims of oppression. Balancing 118. National security is cardinal importance to public interest and the whole nation. For prevention and investigation, information is the core of the measure; any obstruction would defeat the whole process. 119. From the facts before me, the requirement was in an abstemious and restrained manner, all confined to reasonable necessary. 120. Given the close nexus, interactions amongst HKA and the others who shared common objectives and the monetary flows. I am satisfied all the required information is necessary for the prevention and investigation of an offence endangering national security. 121. Taken an objective, panoramic and complete evaluation of all evidence before me, I am satisfied an overall fair balance had been achieved. Issuing the Notice 122. Although the Commissioner only had the recommendations[26] from Superintendent Hung, it was a refinement of the investigation report[27] in simple and direct terms put in context from one professional to his superior, focusing on the issues for the Commissioner’s decision. 123. The Commissioner endorsed the recommendations without query or seeking clarification was a sign of his satisfaction to the sufficiency of information which enable him to make his professional decision. 124. There is no material faulty on the Commissioner’s decision. Conclusion 125. In this case, the legality of the Notice is strictly related to the time when it was served; therefore the Court’s consideration is confined to that material moment. 126. D1’s subsequent explanation with details in Court on some of the events, monetary transactions and nexus which is part and parcel of the information required; which should have been provided before the deadline, not in hindsight. 127. I am satisfied the Notice was sound and legal at the time when it was served. There is no grounds to exercise my discretion to reject it. 128. I rule the Notice was legal and the recipients were obliged to provide the information required. The Open letter 129. HKA submitted an open letter co-signed by D1, D2, D5 and others to Commissioner of Police: “ An Open Letter from the Hong Kong Alliance in Reply to the Commissioner of Police Regarding the letter dated 25th August 2021 to the Hong Kong Alliance from the Commissioner of Police [hereinafter referred to as ‘the Letter’] in accordance with Schedule 5 of the Implementation Rules for Article 43 of the National Security Law [hereinafter referred to as Schedule 5], the Alliance would like to make the following responses: 1. The Alliance is not a ‘foreign agent’. Therefore, the Commissioner of Police has no power to request the provision of information from the Alliance under Schedule 5. 2. The Commissioner of Police made an error of law by mistakenly assuming that as long as he ‘had reasonable grounds to believe’ that the Alliance was a ‘foreign agent’ specified in Section 1 of Schedule 5, he could make a request for the provision of information in accordance with Schedule 5. Schedule 5 is application only if the organization is in fact a foreign agent. Merely ‘having reasonable grounds to believe’ does not suffice. 3. The Commissioner of Police failed to provide any justification to explain: (1) On what reasonable ground did (he) believe that the Alliance was a ‘foreign agent’? (2) Why did (he) reasonably believe the issuance of the Letter was necessary for the prevention and investigation of an offence endangering national security? (3) The nature of the offence endangering national security to be investigated and how it was related to the required information; It is a violation of the rules of natural justice. In summary, the Alliance is of the view that there is no legal basis for the issuance of the Letter. Therefore, we will not provide any information as requested in the Letter.” 130. There is clear evidence of no intention to provide the required information whatsoever. Statutory defence 131. Under Sch. 5 s3 (3)(b): “unless the office-bearer or person establishes to the satisfaction of the court that the office-bearer or person has exercised due diligence and that the office-bearer or person has failed to comply with the notice for reasons beyond the office-bearer’s or person’s controls.” 132. In the present case, nothing of the above was done to support such. They took no actual steps to retrieve the required information. Instead they merely held discussions among themselves. The Verdict 133. The conclusion is: (1) the Notice was legal when it was served on D1, D2 and D5 respectively; (2) the Defendants were obliged to answer; (3) the Defendants’ non-compliances was unjustified. 134. I am satisfied beyond reasonable doubt the prosecution has established their case on every aspect. 135. I thereby convict D1, D2 and D5 accordingly. [1] Contrary to section 3(3)(b) of Schedule 5 to the Implementation Rules for Article 43 of the Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region (Instrument A406A). [2] 香港市民支援愛國民主運動聯合會 [3] BOR 11 [4] BOR 10 [5] BOR 18 [6] BOR 14 and Personal Data (Privacy) Ordinance, Cap. 486 [7] Exhibit P19(1), also attachment [1] [8] Exhibit P20(1), also attachment [2] [9] [2020] HKCA 196 [10] Exhibit P19(1) [11] Exhibit P20(1) [12] Exhibit MFI 3, also attachment [3] [13] The identity of Org 4 is subject to PII and redacted. D1 presents her case on her own analysis. [14] Exhibit P19(1) [15] Exhibit P20(1) [16] Exhibit P19(1) [17] Exhibit P20(1) [18] According to the information accessed, Mark Herman SIMON is the personal assistant to LAI Chee-ying. [19] Leung Kwok Hung v Secretary for Justice (No 2) [2020] 2 HKLRD 771 [20] [2021] HKCFA 3 [21] Sch. 1 IR 2(2) “A magistrate may issue a authorizing a police officer with such assistants as may be necessary to enter and search any place if the magistrate is satisfied by information on oath that there is reasonable ground for suspecting that any specified evidence is in the place.” [22] Sch. 7 IR 2(4)(b) “where the application relates to a particular person – that there are reasonable grounds for suspecting that the person has information, or is in possession of material, likely to be relevant to the investigation;” [23] NSL 3 [24] The Notice [25] s2 of Cap. 486 [26] Exhibit P20(1) [27] Exhibit P19(1) 
The Court: Introduction 1. All major political transformations are haunted by the ghosts of the previous regime. There are few matters on which traditional habits are more persistent than the system of land tenure, which has a profound effect on the customs and usages of any settled community. This appeal is concerned with the impact of successive political transformations in the New Territories on its system of land tenure. 2. The Small House Policy is a non-statutory administrative policy operated by the Lands Department which authorises grants of land and building licences to the indigenous male population of certain villages in the New Territories on more favourable terms than those available generally. It was formalised in substantially its present terms by a decision of the Executive Council on 14 November 1972, which also gave it its current name. But it has existed in one form or another since the beginning of the 20th century, when the New Territories were being incorporated into Hong Kong as an extension to the British colony. The question at issue on this appeal is whether it is constitutional under the Basic Law which governed the transition from colonial rule when the People’s Republic of China[1] resumed the exercise of sovereignty over Hong Kong in 1997. The Appellant contends that it is invalidated by the anti-discrimination provisions of the Basic Law and the Hong Kong Bill of Rights[2]. It is common ground that it is prima facie discriminatory on grounds of both sex and social origin. But the Respondents and the Interested Party say that it is validated by a provision of the Basic Law, Article 40,[3] protecting the lawful traditional rights and interests of the indigenous population of the New Territories. The Small House Policy 3. It is common ground that the Small House Policy as currently applied is accurately described in a pamphlet entitled “How to Apply for a Small House Grant” published by the Lands Department in December 2014. For present purposes, the following summary is sufficient. The beneficiaries of the Policy are “indigenous villagers” of the New Territories. This means adult males descended through the male line from a resident in 1898 of a recognised village. Recognised villages are villages included in a list approved by the Director of Lands. There are currently 642 recognised villages. The Policy relates to land in the village or its immediate environs which are not affected by any impending development or future planning or development proposals. Its object is to enable an eligible villager to build, once in his lifetime, a small house in his own village for his own occupation. It allows him to apply for three kinds of grant: (i) a free building licence, which is a licence to build on private land at a nil premium in the case of pre-1898 villages or a full premium in the case of villages recognised since that date; (ii) a private treaty grant of government land at a reduced premium; and (iii) an exchange, involving the surrender of an existing title in exchange for the grant of a new title at a nil premium so far as it was private land and a reduced premium so far as it was government land. In any of the three cases there must be no “substantiated” local objection to the proposal. Typical objections include objections based on the indigenous villager status of the applicant, as well as those based on environmental or technical issues, or politically sensitive issues such as village boundary disputes or feng shui. Over the years, benefits under the Policy have come to be known as “Ding rights”. Historical background 4. As is well-known, the British colony of Hong Kong comprised three blocks of territory occupied at different times. The island of Hong Kong was ceded to Great Britain by the Treaty of Nanking[4] in 1842 and the southern part of the Kowloon peninsula by the Convention of Peking[5] in 1860. Under the Second Convention of Peking, in 1898, the Qing Government leased part of the administrative district of San On to Great Britain for 99 years with effect from 1 July 1898, and the district thereafter became known as the New Territories. 5. Shortly after the cession of Hong Kong Island, the colonial administration introduced a new system of land tenure which has subsisted in its essential respects to this day. All land became the property of the Crown, which granted only leasehold interests to others. In principle, leases were granted at a premium reflecting the market value of the land, subject to strict contractual and statutory controls over development. Conversion of agricultural or garden land to building use, where it was available, was granted at a premium reflecting the enhancement of the market value. The same system was applied to southern Kowloon after its cession in 1860. In the case of the New Territories, an Order in Council of 20 October 1898 (the so-called First Proclamation) provided that with limited exceptions all laws and ordinances in force in the colony of Hong Kong were to take effect in the New Territories until amended or repealed. Accordingly, the New Territories (Land Court) Ordinance[6] provided by section 15 that “all land in the New Territories is hereby declared to be the property of the Crown” during the term of the New Territories lease from the Qing Government. All persons in occupation of land were to be deemed trespassers as against the Crown unless their occupation was authorised by grant from the Crown, by title allowed by the Land Court, or by licence from the Crown. 6. At the time of their cession to Great Britain, Hong Kong Island and southern Kowloon had been thinly populated and undeveloped. The New Territories were very different. They comprised an extensive area of about 356 square miles on the mainland (including northern, or “New” Kowloon) and surrounding islands, with a population of about 90,000 people living in a large number of villages. The colonial authorities were therefore faced for the first time with the application of a land settlement originally devised for an almost empty territory to one which was already extensively settled by a substantial population with existing rights over land. Some adaptation of the new land system was therefore necessary if it was to accommodate existing entitlements and avoid significant social disruption. Upon taking possession of the New Territories in April 1899 the Governor, Sir Henry Blake, declared that the government’s policy would be to protect the existing commercial and landed interests and the usages and good customs of the inhabitants. 7. Chow J (as he then was)[7] heard expert evidence about the system of land tenure in force before the cession, which these arrangements replaced.[8] He found that in theory land tenure before 1898 was governed by the Qing Code.[9] Under the Code, all land was nominally the property of the Emperor and its occupation by others was permissible only by grant of the Imperial authorities in return for the payment of land tax. By the late 19th century, however, land tax was rarely collected, the Code had fallen into desuetude and a system of customary tenure had taken its place. The basic features of this system were that plots were treated as belonging to the owner of the subsoil, whose title was derived from leases from the Emperor and whose obligation was to pay the land tax. But the exclusive right of use and occupation belonged to the owner of the topsoil, whose title derived from leases granted by the subsoil owner. Both the subsoil leases and the topsoil leases created interests in the land which were perpetual, heritable and partible. This left the topsoil owner with an unlimited dominion over the land subject to payment to the subsoil owner of a rentcharge. There was also no legal restriction on building. 8. The effect of the new system of land tenure was that customary titles under the pre-colonial system were abolished and replaced by a system of Crown leases for a limited term not exceeding 99 years, subject to various covenants and statutory provisions restricting development. Between 1899 and 1903, land in the New Territories was surveyed, cadastral plans were drawn up, and a system of registration of title was established. All claims to land were required to be presented to a Land Court created in 1900, whose function was to receive the claims and determine those which were disputed. The New Territories were divided into blocks, in respect of each of which a Block Crown Lease was issued for a term of 99 years less 3 days[10] corresponding to the term of the lease from the Qing Government. Each Block Crown Lease contained a Schedule listing lots comprised within the Block. It recorded title to each lot, where title had been established, described its current use, and contained covenants against building without government consent. Existing title to 354,277 lots was established under this procedure. A Crown rent was fixed for each of them. These lots were commonly referred to as “Old Schedule lots”. 9. The Land Court fixed a terminal date for presenting claims in each district. Land unclaimed by the terminal date was deemed to be Crown land available to be sold. At some time between 1901 and 1904, the colonial government began to grant leases of Crown land. Lots disposed of by the Crown in this way after 1901 were referred to as “New Grant lots”. Free building licences 10. By two Gazette notices of 1906, Nos. 191/1906 and 192/1906, the Governor authorised Assistant Land Officers to grant building licences in respect of agricultural or garden land in the New Territories and to approve buildings erected pursuant to the licences. The introduction of a regime of building controls created potential difficulties in relation to Old Schedule lots. They were addressed in a memorandum in January 1906 by Mr. Cecil Clementi (then a judge of the Land Court) to the Colonial Secretary.[11] Mr. Clementi pointed out that before 1898 there had been no restrictions on use and no distinction between agricultural and building land. Landowners had been entitled to build on their land, subject only to a theoretical obligation (in practice never enforced) to report the change of use to the officers of the Imperial government so that an increased land tax could be charged reflecting its enhanced value. For this reason, it became the practice of Land Officers to grant “free” licenses to build village-type houses on Old Schedule lots. These licences were granted at an increased Crown rent but without premium. By comparison, a full market premium was charged for building licences issued in respect of New Grant lots. 11. The application of the free building licence policy was straightforward enough when the relevant proprietor was the original owner of an Old Schedule lot recorded in the relevant Schedule, or his descendant. After about 1920, however, outsiders steadily bought up Old Schedule agricultural land. A number of these purchasers applied for free building licences to build large luxury villas. From time to time the law officers pointed out that the original policy of 1906 had not distinguished between inheritors and purchasers of Old Schedule lots, and suggested it should make no difference who the owner of a lot was. In general, however, District Land Officers did not act on that view. They treated free building licences as a concession available only to villagers and their successors in title by descent. They generally refused to grant free building licenses to outsiders and even to indigenous villagers who acquired Old Schedule lots by purchase. Moreover, they strictly enforced the limitation of free building licences to village-type houses (“native cottages”) on lots in or close to recognised villages. Some of them applied this limitation to Old Schedule lots wherever located. 12. Differences between the practices of different Land Officers regarding free building licences ultimately led in 1957 to the standardisation of the Policy regarding free building licences. This was achieved by the Executive Council approving the policy as described in a memorandum dated 2 April 1957.[12] The effect of the 1957 Memorandum was to lay down that for Old Schedule lots: “A bona fide villager will in general be permitted, subject to planning and fung-shui considerations, to build a village-type house for his own occupation, and such permission (by building licence) will be free of premium.”[13] 13. For this purpose, the 1957 Memorandum abolished the distinction between villagers deriving their title from inheritance and those deriving it from purchase: “It is not proposed to make a distinction in applying these conditions between applicants who have inherited Old Schedule lots and those who have purchased them. While it would be theoretically correct to make such a distinction, it is not thought that it would be practicable and that if the advantages of the concessionary terms proposed for the owners of Old Schedule lots were restricted to those who had acquired them by inheritance the way would be open for other applicants to seek the same benefits by fraud or mis-representation.”[14] The restriction to applicants descended in the male line from a resident in 1898 continued. 14. In 1960, the system of free building licences was further modified so as to deal with the increase in the population of the New Territories, which had led to the geographical expansion of many villages. Free building licences were made available for New Grant lots in respect of which grants had been made before the Second World War. The reason for the limitation to pre-war grants was that these were thought less likely to have been made to outsiders. 15. The 1957 Memorandum referred to a note prepared by the District Commissioner for the New Territories, which was annexed to it. This identified five policy considerations underlying the revised policy. They included “the need to honour our undertakings towards the country people”,[15] along with ordinary considerations of town planning, development control and government revenue. Expanding on this point, the District Commissioner observed: “We have said many times that the grantees of old schedule lots and their descendants are not required to pay premium for conversion of their agricultural land to build traditional village houses for their own occupation; and although in town layout areas this 'right' (which derives simply from the fact that the Chinese government, who provided no services, also charged no premia for grants of land, only a land tax; and did not exercise any control on buildings beyond charging increased land tax) has become modified, and in New Kowloon it has never been recognised at all, elsewhere it has been treated for so long as ‘entrenched’ that it would now be dangerous to deny it.”[16] 16. This is one of a substantial number of internal documents of the colonial authorities recording the government’s view that it was honour bound to respect the building rights which had formerly been enjoyed by those holding Old Schedule lots in 1898. This was never the only consideration, however. Other factors included development control, improving building standards, dealing with the squatter problem, catering for the larger needs of an expanding population, and maintaining government revenue. A desire to reflect at least some of the advantages enjoyed by the village population before the lease to Great Britain was one consistent strand of official thinking on this subject before the Policy was formalised in 1972. But although the historical origins of the Policy were never forgotten, it is clear that in due course it acquired a life of its own, to some extent independent of the considerations which had led to its adoption in the first place. It created entrenched vested interests and expectations which would have been politically difficult to displace. Sale of Crown land on concessionary terms 17. The availability of Crown land for sale to indigenous villagers has always been an essential adjunct to the Small House Policy. As the population increased, more villagers found themselves without land on which to build a house, or without land of a size, shape or location suitable for building. 18. Various methods of sale of Crown land have been employed at different times, but all of them encountered a common problem when the land was located in a village area. The villagers arranged matters so that they did not compete against each other for the land. The existence of an effective market for village land therefore depended on applications by outsiders. They, however, were commonly frozen out by existing villagers. Those outsiders who did succeed in buying Crown land in village areas were habitually prevented from building on it by threats to the builders. It is not clear how long this problem subsisted, but the evidence shows that it was still a significant issue at the time when the Small House Policy was formalised in 1972. 19. Until 1909, New Grant lots were sold by public auction. In December 1908 the Governor, Sir Frederick Lugard, submitted a proposal to the Secretary of State that small lots in the New Territories (apart from New Kowloon and the shores of Junk Bay) should in future be sold by private treaty unless there was a likelihood of significant development in the area. This was because there was never any competition at auctions, as a result of which the premium rarely exceeded the upset price (reserve) and any profit was almost always swallowed up by advertising costs. In their joint report, the experts who gave evidence before the Judge agreed that “the villagers had shown themselves quite capable of ensuring that only one bidder would appear at an auction.”[17] The Secretary of State, Lord Crewe, accepted the proposal with modifications. As a result, from 1909 until shortly after the Second World War, sales were conducted by private treaty, generally at what would have been the upset price at an auction. Although there was no formal restriction to male indigenous villagers, in practice grants were made only to them. 20. Auction sales were reintroduced shortly after the Second World War. Initially, they were open auctions, but a practice quickly developed by which in practice they became “restricted” or “closed” village auctions. The auction was limited to indigenous villagers or non-indigenous villagers who had (or whose fathers had) lived in the village before the Japanese occupation. This was achieved by various administrative devices, such as restricting the publication of notice of the sale or holding the auction in the village. There would usually be only one bidder, put forward by agreement between the villagers, and the lot would be knocked down at the upset price. Lots were withdrawn if an outsider was seen to be bidding. These auctions can fairly be called a charade. But the evidence suggests that they reflected a realistic appreciation of the villagers’ ability to control the market for village land. 21. When the Policy was formalised by the Executive Council in 1972, private treaty sales were restored. But they were restored at a fixed premium of two-thirds of the market value, which corresponded to the upset price previously set for village auctions. “Exchanges” 22. Exchanges do not call for separate consideration. A villager who owned a plot of land which was not suitable for building on would commonly apply to surrender his plot in exchange for the grant of a plot, usually comprising all or part of his original plot plus some adjoining Crown land. Similar principles applied in these cases. The regrant was at a nil premium in the case of land which had been private and at a reduced premium in the case of Crown land. Exclusion of women 23. Grants of Crown land under the Small House Policy and its predecessors have always been available only to men. Whatever the method of disposal, female purchasers were excluded even if they were indigenous villagers. This was achieved by arrangement with the village elders, who invariably objected to sales to women unless they were buying to build houses for their infant sons. The reasons for the exclusion of women are explained in the expert evidence. The villages had generally originated in family settlements, and their inhabitants retained the character of clans. Before 1898 women had not been entitled to inherit land.[18] Traditional relationships within villages would have been seriously disrupted if a woman had been able to inherit land, because this would have enabled her and her children after her death to hold land in a village to which she no longer belonged and her children had never belonged.[19] The same consequences would have followed if a woman had been able to apply for a grant under the Small House Policy. In a report on the incorporation of the New Territories into the colony, laid before the Legislative Council by the Governor in 1912, it is stated: “A Chinese community like that of the New Territories is by its structure and its long habit of decentralised government very easy to administer. But its old established customs and institutions must not be lightly changed or affronted, and necessary innovations have to be introduced with the greatest delicacy. In the New Territories as elsewhere continuous descent in the male line is the paramount object in the life of the Chinese, and the necessity for this is the foundation for many of their habits and customs.”[20] 24. Although women are now entitled, as a result of changes in the law, to acquire land by will or upon an intestacy, the evidence is that in practice this rarely happens in the case of village land.[21] The statutory framework 25. On 19 December 1984 the governments of the PRC and the United Kingdom jointly declared that the PRC would resume the exercise of sovereignty over Hong Kong with effect from 1 July 1997 and that Hong Kong would become a Special Administrative Region[22] of the PRC. The Joint Declaration set out the “basic policies” of the PRC with regard to Hong Kong. The fundamental principle underlying the Joint Declaration was continuity. Paragraph 3(3) of the Joint Declaration provided that: “the laws currently in force in Hong Kong will remain basically unchanged”. Paragraph 3(5) provided: “The current social and economic systems in Hong Kong will remain unchanged, and so will the life-style. Rights and freedoms, including those of the person, of speech, of the press, of assembly, of association, of travel, of movement, of correspondence, of strike, of choice of occupation, of academic research and of religious belief will be ensured by law in the Hong Kong Special Administrative Region. Private property, ownership of enterprises, legitimate right of inheritance and foreign investment will be protected by law.” 26. The Joint Declaration stated that the “basic policies”, together with the elaboration of them in Annex I, would be embodied in a Basic Law. Its preparation was entrusted to a drafting committee comprising members from the Mainland and Hong Kong. In its final form, the Basic Law was adopted as a law of the PRC at the Third Session of the Seventh National People’s Congress on 4 April 1990 and promulgated by the President of the PRC on the same day. It laid down the “systems to be practised” in the Hong Kong SAR with effect from the handover on 1 July 1997. 27. The Basic Law provides, so far as relevant: “Article 25[23] All Hong Kong residents shall be equal before the law. Article 39[24] The provisions of the International Covenant on Civil and Political Rights… as applied to Hong Kong shall remain in force and shall be implemented through the laws of the Hong Kong Special Administrative Region. The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law. Such restrictions shall not contravene the provisions of the preceding paragraph of this Article. Article 40 The lawful traditional rights and interests of the indigenous inhabitants of the ‘New Territories’ shall be protected by the Hong Kong Special Administrative Region. Article 120[25] All leases of land granted, decided upon or renewed before the establishment of the Hong Kong Special Administrative Region which extend beyond 30 June 1997, and all rights in relation to such leases, shall continue to be recognized and protected under the law of the Region. Article 122[26] In the case of old schedule lots, village lots, small houses and similar rural holdings, where the property was on 30 June 1984 held by, or, in the case of small houses granted after that date, where the property is granted to, a lessee descended through the male line from a person who was in 1898 a resident of an established village in Hong Kong, the previous rent shall remain unchanged so long as the property is held by that lessee or by one of his lawful successors in the male line.” 28. After the promulgation of the Basic Law but before its coming into effect, the Hong Kong Bill of Rights Ordinance[27] and the Sex Discrimination Ordinance[28], were enacted in 1991 and 1996 respectively. Both remain in force. 29. The HKBORO gives statutory effect to the International Covenant on Civil and Political Rights[29]. Article 22 of the BOR[30], which gives effect to Article 26 of the ICCPR, provides: “Article 22 Equality before and equal protection of law All persons are equal before the law and are entitled without any discrimination to the equal protection of the law. In this respect, the law shall prohibit any discrimination and guarantee to all persons equal and effective protection against discrimination on any ground such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.” 30. The SDO also gave effect, although only in part, to the 1979 Convention on the Elimination of All Forms of Discrimination Against Women.[31] The SDO outlawed certain kinds of sex discrimination. Section 21 of the SDO prohibited discrimination by the government in the exercise of its powers, but was subject to a general exception in Schedule 5, Part 2, for: “that policy of the Government (a) known as the small house policy; and (b) pursuant to which benefits relating to land in the New Territories are granted to indigenous villagers who are men.” 31. Since both the United Kingdom and the PRC were parties to the CEDAW, this necessitated a reservation by both governments, on 14 October 1996 in the case of the United Kingdom and on 10 June 1997 in the case of the PRC. Both were in similar terms, providing that: “Laws applicable in the New Territories which enable male indigenous villagers to exercise certain rights in respect of property and which provide for rent concessions in respect of land or property held by indigenous persons or their lawful successors through the male line will continue to be applied.” The judgments below 32. Chow J held as follows: (1) The Ding rights could properly be regarded as “rights or interests” capable of being protected by BL40. On that footing, the critical questions were whether they were (i) lawful, and (ii) traditional.[32] (2) The word “lawful” was purely descriptive of those traditional rights which the indigenous inhabitants of the New Territories had enjoyed before 1 July 1997. All of the Ding rights were therefore lawful.[33] (3) “Traditional” rights were rights “traceable” to those which the indigenous villagers had enjoyed before the inception of the lease of 1898. The right to a free building licence for Old Schedule lots could be so described. The practice of granting such licences was therefore constitutional.[34] (4) But the practice of granting government land by private treaty sale at a concessionary premium to indigenous males only, and the practice of exchanging private land for government land at a reduced or nil premium with indigenous males only, were not traceable to rights existing before 1898 and were therefore unconstitutional.[35] (5) The exception of the Small House Policy from the SDO in Schedule 5, Part 2, was unconstitutional as being inconsistent with the Basic Law. 33. All parties appealed. The Court of Appeal[36] agreed with the Judge that Ding rights were relevant “rights and interests”, but held that the relevant date for determining both the lawfulness and the traditional character of the rights protected by BL40 was April 1990 when the Basic Law was promulgated.[37] On that footing, they held that the Small House Policy was constitutional in its entirety. They also considered that even if “traditional” rights were confined to those derived from rights enjoyed by indigenous inhabitants before 1898, all three forms of grant made under the Small House Policy satisfied that test. 34. There were issues in both courts below about the Appellant’s standing and as to his delay in bringing the proceedings, to which we will return at the conclusion of this judgment. Construction: the legal context 35. Before embarking on a detailed analysis of the language of BL40, it is necessary to ask what can be deduced about its purpose from the historical background and the circumstances in which it was enacted. As the Court of Appeal held, the relevant date for this exercise is April 1990, when the Basic Law was promulgated. In Director of Immigration v Chong Fung Yuen,[38] this court held that extrinsic material which threw light on the context or purpose of the Basic Law was available to construe it. This included the state of domestic legislation at the time of its promulgation in April 1990: “Because the context and purpose of the Basic Law were established at the time of its enactment in 1990, the extrinsic materials relevant to its interpretation are, generally speaking, pre-enactment materials, that is, materials brought into existence prior to or contemporaneous with the enactment of the Basic Law, although it only came into effect on 1 July 1997.” 36. Given that the Basic Law maintains in effect so far as consistent with its terms the existing laws of Hong Kong (Article 8), the existing rights and freedoms of residents (BL39) and the existing rights of leaseholders (BL120), BL40 must be addressed to rights and interests of the indigenous inhabitants of the New Territories which are (i) special to those inhabitants and not common to the generality of residents, and (ii) potentially open to challenge in the absence of a protective provision like BL40. One obvious basis for such a challenge was the anti-discrimination provisions of the ICCPR. The incorporation of the ICCPR into the domestic law of Hong Kong had not occurred in June 1990 but was required by BL39. 37. Shortly after the Joint Declaration, the Heung Yee Kuk,[39] the representative body of the New Territories villages, identified eight existing privileges of their constituents. In the course of the drafting process, discussions of the traditional rights or interests of indigenous villagers were conducted by reference to this list. The eight privileges were (i) the right to be represented to the Hong Kong government by the Kuk, a statutory body, and by the chairmen of rural committees; (ii) the right to build a small house in accordance with the Small House Policy formalised in 1972; (iii) a rent freeze while the property remained in the hands of a person descended in the male line from a pre-1898 resident, which was proposed in Annex III of the Joint Declaration and ultimately embodied in BL122; (iv) the right to a grant of alternative land when villages were compulsorily expropriated for industrial development or public housing; (v) exemption from rates for village houses; (vi) the right to bury deceased family members at the hillside near the village, instead of at the public cemetery, and to be compensated for the demolition of graves; (vii) the right to have a deceased estate distributed according to Chinese custom to the deceased’s male offspring or to have land (tso/t’ong land) held in a customary family trust for descendants in the male line; and (viii) the equal right of residence and entitlement to ancestral estates enjoyed by indigenous villagers living overseas if they return. Two points should be noted about these privileges. The first is that only (i), (iii), (v) and (vii) had a statutory basis. The others were benefits under non-statutory policies applied as a matter of administrative discretion. The second is that only (vi) and (vii), and arguably (ii), could be said to originate in some right enjoyed in pre-colonial times. 38. In a society where building land was scarce and expensive, the Small House Policy was by far the most important of the privileges identified by the Kuk, and by the same token much the most controversial. There were persistent demands before the promulgation of the Basic Law that it should come to an end along with the colonial government which had devised it. BL40 marked the rejection of these demands. “Rights and interests” 39. The starting point is to identify the nature of the “right” or “interest” which an applicant under the Small House Policy may be said to have. The existence of the Policy is implicitly acknowledged in a number of Ordinances, as well as the Basic Law itself. They included notably the Buildings Ordinance,[40] which applied to the New Territories since 1961 in the manner provided by the Buildings Ordinance (Application to the New Territories) Ordinance,[41] and exempted certain categories of houses there, including those which did not exceed the dimensions specified for the purpose of the Small House Policy; the provision for rating exemption in section 36(1)(c) of the Rating Ordinance;[42] and the provision for a rent freeze in BL122. But the Policy itself has never had a statutory basis. It is applied as a matter of administrative discretion. In those circumstances, it cannot give rise to a legal right in the ordinary sense of the word. The actual grant of a building licence or a lease gives rise to a right which is good against the world, but an application for such a right or interest does not. The relevant right is founded entirely on public law. We would define it as a right to have one’s application dealt with in accordance with the criteria laid down in the government’s statements of current policy, subject to the lawfully exercised discretion of the Lands Department. That discretion is not unlimited. It is governed by law. “Where a public authority has issued a promise or adopted a practice which represents how it proposes to act in a given area, the law will require the promise or practice to be honoured unless there is good reason not to do so”: R (Nadarajah) v Secretary of State for the Home Department;[43] c.f. Ng Siu Tung v Director of Immigration,[44] and Mandalia v Secretary of State for the Home Department.[45] This is therefore an inherently imperfect right. It depends on the availability of judicial review, a jurisdiction with highly flexible remedies. Moreover, unless BL40 makes the Small House Policy immutable (a question which we do not decide), it may change. But while the Policy remains in force in its current terms, it creates something which is clearly a “right” in the sense meant by BL40. Otherwise BL40 applies to very little. 40. In those circumstances the concept of an “interest” does not call for separate consideration, but it is clearly at least as broad as the concept of a “right”. “Lawful” 41. In April 1990, when the Basic Law was promulgated, there were no relevant statutory rules against discrimination in Hong Kong. Neither the HKBORO nor the SDO had been enacted, although BL39 expressly envisaged that effect would in due course be given by domestic legislation to the ICCPR “as applied to Hong Kong”, i.e. not necessarily in its entirety. At common law, it has been said that “treating like cases alike and unlike cases differently is a general axiom of rational behaviour”: Matadeen v Pointu.[46] As Lord Hoffmann went on to point out, this was “frequently invoked by the courts in proceedings for judicial review as a ground for holding some administrative act to have been irrational.”[47] However, the test of irrationality in this context is not the same as the test for unjustified statutory discrimination, and Mr Martin Lee SC[48] for the Appellant confirmed that the only basis of public law challenge to the Small House Policy that could then have been advanced was irrationality. It is difficult, however, if not impossible, to assess at this stage, on the hypothetical basis that this argument is now suggested, what prospect of success this challenge might have had. 42. Once the Basic Law came into force in 1997, the law of Hong Kong contained two relevant anti-discrimination provisions, BL25 and BOR22. BL25 is of course part of the same instrument as BL40, and must be read together with it. BOR22, although enacted separately by domestic legislation, also falls to be construed together with BL40. This is because by virtue of BL39, in particular its second paragraph, the BOR is incorporated into the Basic Law and given constitutional effect under it: see Comilang Milagros Tecson v Director of Immigration.[49] 43. The present argument turns on the relationship between BL40 on the one hand and the two anti-discrimination provisions on the other. The substance of the Appellant’s case is that consistency with the anti-discrimination provisions is a condition for the application of BL40. Otherwise, the Policy cannot be lawful and is not protected. In other words, the anti-discrimination provisions qualify and limit the scope of BL40. The substance of the Respondents’ and the Interested Party’s case is that BL40 is the dominant provision. It qualifies and limits the application of the anti-discrimination provisions, not the other way round. 44. In our judgment, the latter analysis is correct, for the following reasons: (1) The Basic Law is founded on the principle of continuity, with specific exceptions where this was inconsistent with the PRC’s basic policies for Hong Kong. Indeed, Article 5 of the Basic Law stated that “the previous capitalist system and way of life shall remain unchanged for 50 years.” One starts, therefore, with the expectation that a significant element in what paragraph 3(5) of the Joint Declaration calls “current social and economic systems” will remain unchanged. On the face of it, BL40 is a saving provision seeking to give effect to that principle by protecting an existing entitlement of a particular class of persons. (2) It is a principle of statutory construction that the specific prevails over the general. This is simply one aspect of the more general principle that legislative instruments must be read as a coherent whole: “The rule is, that wherever there is a particular enactment and a general enactment in the same statute, and the latter, taken in its most comprehensive sense, would overrule the former, the particular enactment must be operative, and the general enactment must be taken to affect only the other parts of the statute to which it may properly apply”: Pretty v Solly.[50] The same principle applies to constitutional interpretation. BL25 and BL39 and BOR22 are general provisions. BL40 is a specific provision dealing with the special position of the indigenous inhabitants of the New Territories. (3) Absent BL40, all the various advantages enjoyed by the indigenous inhabitants of the New Territories[51] would be inherently discriminatory unless they can be objectively justified as being necessary in pursuit of a legitimate aim: see Secretary for Justice v Yau Yuk Lung.[52] If consistency with the anti-discrimination provisions is treated as a condition of their being protected by BL40 (as the Appellant proposes), then either (i) the discrimination is justified, in which case BL40 is unnecessary; or (ii) it is unjustified, in which case BL40 applies to nothing. Yet BL40 was plainly intended to have some effect. (4) BL122, which deals with the level of rent payable on “old schedule lots, village lots, small houses and similar rural holdings”, assumes that grants will continue to be made under the Small House Policy to descendants in the male line of pre-1898 village residents, notwithstanding the discriminatory features of the Policy which are specifically referred to in that Article. It is fair to say that the assumptions of the legislator are not necessarily the same as his enactments. But this assumption casts a good deal of light on what the drafters of the Basic Law must have believed that they had provided for in BL40. 45. Both courts below described the word “lawful” as “descriptive”.[53] Mr Lee fairly submits that on that footing it is redundant and without legal effect. In our judgment, the word is neither purely descriptive nor redundant. It goes to the lawfulness of the way that the discretion is exercised as a matter of public law. We have defined the relevant right or interest of an applicant under the Small House Policy as a right to have his application dealt with in accordance with the criteria laid down in the government’s public statements of the current Policy, subject to a lawfully exercised administrative discretion of the Lands Department. That right or interest is lawful if the discretion to make a grant under the Policy is lawfully exercised as a matter of public law. It would not be lawful if, for example, the discretion was exercised in a manner contrary to some enforceable legitimate expectation of the applicant or was vitiated by corruption or bias. “Lawful” is not intended to refer to the absence of discriminatory features forbidden by BL25 or BL39, whose application in the special context of indigenous rights is addressed by BL40 and excluded. “Traditional” 46. This was the point on which the courts below differed. The Judge considered that to be “traditional”, a right or interest of the indigenous villagers had to be “traceable” to a right or interest which his ancestors had enjoyed before the lease of the New Territories in 1898. By “traceable” he meant based on, or capturing the essence of, the earlier right. In his view, the grant of a free building licence satisfied that test because there had been no building controls before the New Territories were incorporated into the colony of Hong Kong. By comparison, he thought that the right to a grant or exchange of Crown land had no equivalent under Qing rule. The Appellant’s case was substantially an adoption of that view, although he contended that not even free building licenses were “traceable” to a pre-1898 right or interest. The Court of Appeal, by comparison, considered that “whether a right or interest is traditional for the purpose of BL40 is to be determined by reference to the state of affairs in April 1990.”[54] “Traditional” rights were those which was recognised as traditional at that point. In our judgment, the Court of Appeal were right about this. 47. There is nothing in BL40 which requires a protected right or interest to be traceable to the period before 1898. BL40 does not say so in terms. Is there any ground on which such a principle might be implied? In our opinion there is not. The principle of traceability is not implicit in the concept of tradition as a matter of language. It is not necessary for the efficacy of BL40. And it is not consistent with the purpose of BL40. In view of the importance which has always been attached to purposive construction in interpreting the Basic Law, this last point calls for some expansion. 48. In the first place, it is clear from the history that we have summarised above that the Small House Policy did not replicate the essence of the old pre-1898 system. It was introduced by the colonial authorities in order to soften the transition from customary tenure, which conferred on the topsoil owner practically complete dominion over the land, to an utterly different system of intensively administered time-limited Crown leases and tightly controlled development. It was because the two systems were so completely different that managing the transition in this way was conceived to be politically necessary. 49. Secondly, the limitation of eligibility to those whose ancestors had held Old Schedule lots in the same village before 1898 ensured that families whose forebears had known the old system were the sole beneficiaries of the new one. But the connection inevitably became looser and more remote as time went on. Pre-1898 residents died off, to be succeeded by their sons, grandsons, and great-grandsons who had never known the old system. The Policy was extended to eligible purchasers on the same basis as it had applied to heirs. Once the restrictions on alienation had expired, the land was marketable. For this and other reasons, village land often found its way into the hands of outsiders. At the same time, some areas were designated for development, which transformed the landscape and way of life of villagers. The Small House Policy itself underwent substantial changes in the course of the 20th century. As we have already observed, the Small House Policy has acquired a life of its own. The reality is that the rights arising under it were new rights originally conferred by the colonial administration in the first decade after the inception of the New Territories lease of 1898, and maintained in various iterations and with some modifications thereafter. They were derived from the Policy itself. By 1990, they had become a tradition of the colonial administration. They were traditional in 1990 not because they were traceable to any rights or interests which had existed before 1898, but because the beneficiaries were confined to members of long-standing and relatively immobile village communities which had been treated as a separate category throughout the period of colonial government, and because they had by then passed through the male line in the same families for nearly nine decades. 50. Thirdly, it is right to remember that on the Mainland too, political, social and economic systems had fundamentally changed between 1898 and 1990. The Basic Law was addressed to the problem of continuity between the colonial regime and the system which would follow it. The problem of continuity between the Qing dynasty and the colonial regime was of no subsisting relevance by 1990. There was therefore no rational reason why the PRC, whose basic policies regarding Hong Kong were embodied in the Basic Law, should wish to make the preservation of indigenous rights which they would inherit from the colonial regime dependant on their similarity to rights which had existed before 1898. The fact that only descendants of pre-1898 villagers were eligible for small house grants and that the Small House Policy had been devised for an earlier transfer of power was part of the description of the system which the SAR inherited and which BL40 protected. But it was not an indication that between 1997 and 2047 the survival of those rights should depend on disputable antiquarian research into land rights which had been extinguished nearly a century before. The SDO 51. In this analysis, we have not referred to the SDO because of the express exclusion of the Small House Policy from its ambit by Schedule 5, Part 2. The Judge held that the exclusion was unconstitutional because it was inconsistent with BL25 and BL39. Once one concludes, as we have done, that BL40 takes the Small House Policy out of the ambit of BL25 and BL39, this issue falls away. Delay and standing 52. Our conclusions on the substance of the issues make it unnecessary to deal with the two procedural objections to the Appellant’s claim for relief. We do so because of their implications for public law generally and because we differ from the Court of Appeal. Delay 53. The present proceedings were begun by an application dated 28 December 2015, which was substantially amended in September 2016 and April 2019. In its amended form the decisions challenged were (i) the decision of the Director of Lands to “implement on and after 8 June 1991… the [Small House Policy], and his subsequent decisions to continue to implement the [Small House Policy]”, and (ii) the exemption of the Small House Policy from the SDO by Schedule 5, Part 2 of that Ordinance. Realistically, the relevant date would not have been 1991, but 1997. If the Appellant’s case had been accepted, that is the date from which the Small House Policy would have become unconstitutional. That was some eighteen years before the commencement of these proceedings. 54. Applications for judicial review vary greatly in their nature and their potential consequences. For this reason, the rule that they must be brought promptly has never been absolute. Section 21K(6) of the High Court Ordinance[55] provides that the Court may refuse leave to apply for judicial review or refuse to grant relief on the ground of delay “if it considers that the granting of the relief sought would be likely to cause substantial hardship to, or substantially prejudice the rights of, any person or would be detrimental to good administration.” Hardship or prejudice to individuals and disruption of good administration are more likely in cases where the relief would operate retrospectively to undo transactions on whose validity people will have relied. Where the object of the proceedings is to obtain the decision of the court on some general issue of legal or constitutional principle, these consequences are less likely and the public importance of having the issue resolved is greater. Delay is therefore likely to be a less significant factor. If a policy of the executive is unconstitutional or otherwise beyond its powers, it is in the public interest that the courts should say so. It cannot be right to allow the executive to continue to act unlawfully, simply because no one challenged it at the outset. The observations along these lines of Ma CJHC (as Chief Justice Ma then was) in Leung v Secretary for Justice[56] have considerable force as applied to cases like this one. 55. The present proceedings seek the decision of the court on a controversial constitutional issue of considerable public importance. The relief sought in the present proceedings is entirely declaratory. There is no claim to quash past decisions made pursuant to the Small House Policy. In any event, it is exceedingly unlikely that the courts would have been persuaded to quash past grants in favour of persons acting in good faith on the strength of the Lands Department’s publicly stated policies, even if they had accepted the Appellant’s case in principle. The Appellant having obtained leave to bring the proceedings, delay could only have operated as a ground for refusing to make the declarations sought, even if the court was satisfied that they were justified in principle. This would have been a surprising course for the Judge to have taken. It would have meant that the executive was unlawfully discriminating on the basis of sex and social origin from day to day, but the court, although seized of the issue, was disabled from intervening by the conduct of the particular individual who had brought the matter to its attention. We consider that the Judge’s decision to grant relief (on the view that he took of the merits) cannot be faulted in principle and was within his discretion. The Judge having decided the procedural issue, it would have been extraordinary for the objection to be taken by an appellate court charged with deciding whether he was right on the substantive issue. Standing 56. Section 21K(3) of the HCO and Order 53, rule 3(7) of the Rules of the High Court[57] both provide that the Court shall not grant leave “unless it considers that the applicant has a sufficient interest in the matter to which the application relates.” The rule is identical to that laid down for judicial review in the High Court in England by section 31(3) of the Senior Courts Act 1981. 57. Mr Kwok is a self-appointed public guardian with a long record of public law litigation. The objection on the ground of lack of standing is that he has no greater interest in the constitutionality of the Small House Policy than any other member of the public. He is not eligible or potentially eligible for a grant under the Policy. He does not own or occupy or intend to own or occupy land in any area affected by the Policy. Lord Pannick QC[58] submits that he is, in the time-honoured phrase, a “mere busybody”. The Judge disagreed. He held that Mr Kwok qualified because he was a permanent resident of Hong Kong and the case related to the use of land resources and revenue in the public interest. The question is whether that is a sufficient interest. 58. Where the decision of a public authority is challenged on account of its impact on some particular person or group of persons personally and directly affected by it, someone else who seeks to complain by way of judicial review may well lack standing. But the position is more complicated when the object of the process is not to redress some particular injustice but to vindicate the rule of law by raising a general legal or constitutional issue. An issue of that kind may be said to affect everyone equally. There is no general principle that it must affect the applicant more than others. What then is the general principle? 59. In Kwok Cheuk Kin v Commissioner of Police,[59] Chow J held that claims to act as a representative of the public interest called for careful scrutiny of the applicant’s good faith, but that the court should adopt: “a holistic approach by taking into account a host of relevant considerations including the merits of the application, the importance of vindicating the rule of law, the importance of the issue raised, the existence and absence of any other challengers who have a greater interest in the matter, and the nature of the breach of duty against which relief is sought.” 60. In AXA General Insurance v H.M. Advocate,[60] Lord Reed (with whom the rest of the Supreme Court of the United Kingdom agreed) observed: “A requirement that the applicant demonstrate an interest in the matter complained of will not however operate satisfactorily if it is applied in the same way in all contexts. In some contexts, it is appropriate to require an applicant for judicial review to demonstrate that he has a particular interest in the matter complained of: the type of interest which is relevant, and therefore required in order to have standing, will depend upon the particular context. In other situations, such as where the excess or misuse of power affects the public generally, insistence upon a particular interest could prevent the matter being brought before the court, and that in turn might disable the court from performing its function to protect the rule of law. I say "might”, because the protection of the rule of law does not require that every allegation of unlawful conduct by a public authority must be examined by a court, any more than it requires that every allegation of criminal conduct must be prosecuted. Even in a context of that kind, there must be considerations which lead the court to treat the applicant as having an interest which is sufficient to justify his bringing the application before the court. What is to be regarded as sufficient interest to justify a particular applicant's bringing a particular application before the court, and thus as conferring standing, depends therefore upon the context, and in particular upon what will best serve the purposes of judicial review in that context.” 61. Both of these statements of principle were adopted by the Court of Appeal in Kwok Cheuk Kin v President of Legislative Council.[61] We consider that they represent the normal principle to be applied in this jurisdiction. 62. The critical question in a public interest case is accordingly whether the purpose of judicial review, and in particular the rule of law, will be best served by allowing the applicant to proceed. In our judgment the Judge was right, on essentially this ground, to accept that the Appellant had standing. The decisive consideration is that the only people who can be said to have a manifestly greater interest in the constitutionality of the Small House Policy than the generality of the public are those who have obtained or hope to obtain grants under the Policy. They are actual or potential beneficiaries of the Policy with no interest in challenging it. Wider categories of applicants may have a direct personal interest in other aspects of the Policy, for example its planning or environmental consequences for neighbours, but not in its constitutionality. The reality is that the Policy will in practice be beyond challenge even if unconstitutional. Given the significance and controversial character of the issue, that state of affairs would do no service to the rule of law. Disposition 63. The appeal is dismissed. We make an order nisi that there be no order as to costs as between the Appellant and the Respondents, and as between the Appellant and the Interested Party. We further direct that any submissions that the parties may wish to make as to costs be submitted in writing within 14 days of the date of the handing down of this judgment and that, in default of such submissions, the order nisi stand as an order absolute without further directions. Mr Martin Lee SC, Mr Jeffrey Tam and Ms Isabel Tam, instructed by Ho Tse Wai & Partners, assigned by the Director of Legal Aid, for the 1st Applicant (Appellant) Mr Benjamin Yu SC and Mr Anthony Chan, instructed by the Department of Justice, for the 1st, 2nd and 3rd Respondents Lord Pannick QC, Mr Jat Sew-Tong SC, Mr Jin Pao SC and Mr Danny Tang, instructed by T.K. Tsui & Co., for the Interested Party [1] The “PRC”. [2] The “BOR”. [3] “BL40”. [4] Nanjing. [5] Beijing. [6] No. 18 of 1900. [7] [2019] HKCFI 867; [2020] 1 HKLRD 988 (“CFI Judgment”). [8] CFI Judgment, at [60]-[66]. [9] The “Code”. [10] That is, 75 years plus an option to renew for 24 years less 3 days. [11] Contained in CSO No 807/06. [12] Memorandum for Executive Council on Land Conversion in the New Territories XCC 27 for discussion on 9 April 1957 (the “1957 Memorandum”). [13] The 1957 Memorandum, at [9]. [14] Ibid., at [10]. [15] Ibid., Enclosure 2, at [1]. [16] Ibid., at [2]. [17] Joint Expert Report of Eric John Davison and Patrick Hugh Hase, at [20]. [18] CFI Judgment, at [62]; CA Judgment, at [41]. [19] Expert Report of Patrick Hugh Hase, at [101]. [20] Report on the New Territories 1899-1912 laid before the Legislative Council by command of the Governor on 22 August 1912, at [92]. [21] Expert Report of Patrick Hugh Hase, at [103]. [22] “SAR”. [23] “BL25”. [24] “BL39”. [25] “BL120”. [26] “BL122”. [27] Cap. 383 (the “HKBORO”). [28] Cap. 480 (the “SDO”). [29] The “ICCPR”. [30] “BOR22”. [31] The “CEDAW”. [32] CFI Judgment, at [38]-[40]. [33] CFI Judgment, at [129]. [34] CFI Judgment, at [115]-[116]. [35] CFI Judgment, at [117], [125]. [36] Poon CJHC, Lam VP and Au JA: [2021] HKCA 54; [2021] 1 HKLRD 737 (“CA Judgment”). [37] CA Judgment, at [93]. [38] (2001) 4 HKCFAR 211, 224. [39] The “Kuk”. [40] Cap. 123. [41] Cap. 121, formerly the Buildings Ordinance (Application to the New Territories) Regulations (Cap. 322). [42] Cap. 116. [43] [2005] EWCA Civ. 1363, at [68] (Laws LJ). [44] (2002) 5 HKCFAR 1, at [92]. [45] [2015] 1 WLR 4546, at [29]. [46] [1999] 1 AC 98, 109 (Lord Hoffmann). [47] Ibid. [48] Leading Mr Jeffrey Tam and Ms Isabel Tam. [49] (2019) 22 HKCFAR 59, at [24]-[35]. [50] (1859) 26 Beav. 606, 610 (Sir John Romilly MR). [51] See [37] above. [52] (2007) 10 HKCFAR 335, at [20]-[22]. [53] CFI Judgment, at [129]; CA Judgment, at [126]. [54] CA Judgment, at [90]. [55] Cap. 4 (“HCO”). [56] [2006] 4 HKLRD 211, at [39]. [57] Cap. 4A. [58] Appearing with Mr Jat Sew-Tong SC, Mr Jin Pao SC and Mr Danny Tang, for the Interested Party. [59] [2017] HKCFI 1852 at [31]-[34]. [60] [2012] 1 AC 868 at [170]. [61] [2021] HKCA 169; [2021] 1 HKLRD 1247, at [24]-[28]. 
Chief Justice Ma: A. INTRODUCTION 1. In this appeal, the appellant plaintiff (“the Assured")[1] sought to overturn the decision of the Court of Appeal[2] and restore the judgment given in their favour by the Court of First Instance[3]enabling it to recover the sum of US$1,555,209.00 (together with interest and costs) under a contract of marine insurance (“the Contract”) as against the respondent defendant (“the Insurer”).[4] Liability had been disputed by the Insurer primarily on the basis of an insurance warranty in relation to the deadweight capacity of the relevant vessel under the Contract. The essential question before this Court was whether the Insurer could successfully do so. After hearing submissions from Mr Man, we dismissed the appeal with costs, with reasons to be handed down. 2. The appeal was brought as of right under s 22(1)(a) of the Hong Kong Court of Final Appeal Ordinance.[5] 3. At trial, the Assured had also claimed against the broker (Courtesy Insurance Consultants Limited) (“the Broker”) who had arranged cover under the Contract for the Assured. In his judgment, Chung J found the Insurer (as first defendant) liable to indemnify under the Contract, but in the alternative, in the event that the Insurer had been able to deny liability, the Judge would have found the Broker (who was the second defendant) liable to the Assured. In view of this alternative finding, following the appeal of the Insurer being allowed, the Court of Appeal accordingly ordered judgment to be entered against the Broker.[6] The Broker had no involvement in the present appeal either. A.1 Facts 4. Although some of the findings of the trial Judge (and the treatment of them by the Court of Appeal) were contested before this Court, as I shall presently identify, many of the essential underlying facts were uncontroversial. 5. By a Marine Cargo Insurance Application dated 2 January 2008 (“the Application”), the Assured applied to the Insurer for marine insurance coverage in respect of the shipment of a cargo of Malaysian round logs from “Malaysian Port” to Zhangjiagang in the PRC. The relevant carrying vessel was named as the “MV Ho Feng No. 7” (“the Vessel”) and the amount sought to be insured was US$1,500,000.00. 6. The application was accepted and a Marine Cover Note (“the Cover Note”) was issued, confirming the insured interest as the logs valued at US$1,500,000.00. Three points are of note in relation to the Cover Note:- (1) Against the side heading “Ship”, there appeared the words “PER APPROVED VESSEL OR VESSELS TO BE DECLARED AND SUBJECT TO ANY ADDITIONAL SURCHARGE IF REQUIRED”. (2) Against the side heading “Conditions” was the clause “WARRANTED YEAR BUILT OF THE VESSEL NOT OVER 30 YEARS. WARRANTED DWT[7] NOT LESS THAN 10,000”. I shall refer to that part of this clause relating to deadweight as “the Deadweight Warranty”. This is the critical warranty to be considered in the present case. (3) The Cover Note also stated that the insurance cover would be subject to the terms, exceptions and conditions of the policy to be issued. 7. The policy replacing the Cover Note was dated 11 January 2008 (“the Policy”). The insured interest were the logs, but the value was stated to be US$1,555,209.00. Of note for present purposes were the following on the face of the Policy:- (1) In the box marked “Vessel” was typed “M.V. HO FENG No. 7 V.712S”, in other words the Vessel and the identification of the relevant voyage. (2) In the box to identify relevant clauses and conditions, there was typed out the Deadweight Warranty. 8. The Contract in the present case was contained in the Policy. As we shall see, the true construction of the Policy and the effectiveness of the Deadweight Warranty were critical matters in resolving this appeal. 9. I should mention at this point that there had been previous dealings between the Assured and the Insurer. As the trial judge found,[8] from 2004 to 2008, over 70 policies of marine insurance had been issued by the Insurer to the Assured. Only in 10% of such policies was there a deadweight warranty and in five policies issued by the Insurer for cargoes carried on board the vessel, there was no deadweight warranty at all. That said, it would appear that although the policies did not contain a deadweight warranty, the cover notes for all such policies did.[9] I shall refer to these facts again later in this judgment. 10. In or about mid-January 2008, in the course of the stipulated voyage from Kuala Baram Malaysia to Zhangjiagang in the PRC, the Vessel sank and the cargo of logs was totally lost. There is no doubt that, for the purposes of marine insurance, the cargo had been lost incidental to a marine adventure (in other words, a maritime loss). 11. A claim under the Contract was duly made by the Assured to the Insurer for the insured value of US$1,555,209.00. This claim was rejected by the Insurer on the basis that the Assured was in breach of the Deadweight Warranty: the vessel’s deadweight capacity was less than 10,000 tonnes.[10] A.2 The parties’ respective cases 12. It is convenient to begin with the Insurer’s case in denying liability under the Contract. It was put in simple terms: the Assured could not claim under the Contract owing to the breach of the Deadweight Warranty. The Vessel had a deadweight capacity of less than 10,000 tonnes. As an alternative argument, the Insurer also relied on the failure on the part of the Assured to make proper disclosure of a material fact, namely, the actual deadweight capacity of the Vessel. However, as we shall see, non-disclosure was not relevant in these proceedings. 13. The Assured’s case was more elaborate; in brief:- (1) Whatever otherwise might be the effect of the Deadweight Warranty, in the present case it was of no effect since at all material times, the Insurer actually knew or ought to have known that the deadweight capacity of the vessel was less than 10,000 tonnes. (2) In any event, on a true construction of the Contract, there was an obvious inconsistency between the actual naming of the Vessel as the relevant carrying ship of the insured cargo for the voyage and the Deadweight Warranty. The argument ran essentially along the following lines: the intention of the parties clearly being to have effective insurance cover for the carriage of the cargo of logs on board the Vessel from Malaysia to the PRC, it was simply inconsistent then to be able to deny liability on the basis that the very vessel that was identified under the Contract was somehow not covered by reason of the Deadweight Warranty. The intention of the parties that there should be cover should therefore prevail. (3) Insofar as maybe necessary, by reason of the above matters, the Assured also sought rectification of the Contract to delete the Deadweight Warranty, and also relied on waiver and estoppel. I shall deal with these arguments in Sections D to H below. B. THE MARINE INSURANCE ORDINANCE CAP 329 14. It is convenient first to refer to the Marine Insurance Ordinance Cap 329 (“the MIO”). The Ordinance uses the terms “assured” and “insurer”; and I have likewise done so in this judgment in describing the parties. The MIO was enacted in Hong Kong in 1961. Prior to that, although there was undoubtedly much marine insurance cover underwritten in Hong Kong, the insurance industry was content to follow the law and practice of marine insurance in the United Kingdom to govern the position here. In the United Kingdom, the law of marine insurance was codified by the Marine Insurance Act 1906. Underwriters and others in the marine industry in Hong Kong followed the Act but only as a matter of custom. Following Government initiatives in the 1950’s to enact laws relating to the insurance industry,[11] consideration was given as to whether there was a need to enact legislation regarding marine insurance. One of the perceived problems had been over the definition of constructive total loss in a marine adventure. Although the Marine Insurance Act 1906 largely codified the existing law in England, there was an apparent change in the definition of constructive total loss compared with what was arguably the position under common law[12]. In Hong Kong, it was considered by the Government that this potential discrepancy between the Marine Insurance Act 1906 (which was followed by the industry in Hong Kong as a matter of custom) and the common law might cause local insurers to be “embarrassed” and therefore it would be of advantage to the mercantile community to enact an ordinance dealing with marine insurance.[13] 15. Both parties referred to the MIO in their written Cases and rightly so, for it is the starting point (and often end point) in the consideration of any problem involving marine insurance.[14] The MIO follows almost word for word the provisions in the Marine Insurance Act 1906.[15] The MIO (and the Marine Insurance Act) state they are codifications of the existing law. It is perhaps a tribute to the skill of the draftsman of the Marine Insurance Act – Sir Mackenzie Chalmers (also the draftsman of the Bills of Exchange Act 1882 and the Sale of Goods Act 1893) – that the Act and our Ordinance have remained almost wholly unamended since enactment. 16. The MIO contains a part specifically addressing the nature of marine insurance warranties: ss 33 to 41. Only s 33 is relevant for present purposes:- “33. Nature of warranty (1) A warranty, in the following sections relating to warranties, means a promissory warranty, that is to say, a warranty by which the assured undertakes that some particular thing shall or shall not be done, or that some condition shall be fulfilled, or wherebyhe affirms or negatives the existenceofa particular state of facts. (2) A warranty may be express or implied. (3) A warranty, as above defined, is a condition which must be exactly complied with, whether it be material to the risk or not. If it be not so complied with, then, subject to any express provision in the policy, the insurer is discharged from liability as from the date of the breach of warranty, but without prejudice to any liability incurred by him before that date.” 17. A number of points ought to be made in relation to this provision:- (1) The provision defines warranties as promissory in nature and the definition of this term includes the affirmation of the existence of a particular state of facts: s 33(l). (2) Although no particular form of words is required before a marine insurance warranty is created[16]and the use of the word “warranted” does not conclusively mean that such a warranty exists,[17]nevertheless the use of the word does raise a presumption that a warranty is intended.[18] (3) Where a marine insurance warranty is breached, the insurer is without more discharged from liability; in other words there is an automatic discharge from liability: see Bank of Nova Scotia v Hellenic Mutual War Risks Association (Bermuda) Limited (The Good Luck).[19]. It provides a complete defence to any claim made under the policy.[20] There need not be any causal connection between the breach of warranty and any loss suffered by an assured for which a claim is made: see HIH Casualty and General Insurance Limited v New Hampshire Insurance Company.[21] 18. In the present appeal, reliance was placed by the Assured on waiver. A warranty will be ineffective if it is waived: see s 34(3) of the MIO. Waiver of course takes many forms and it may have different legal meanings depending on the context.[22] In the context of a waiver under s 34(3) of the MIO, in the light of the principle that the breach of a marine insurance warranty results in the automatic discharge of an insurer’s liability[23], the doctrine of waiver by election is inapplicable and the meaning of waiver is what has been termed waiver by estoppel:see Argo Systems FZE v Liberty Insurance (Pte) (The Copa Casino).[24] This type of waiver, the same as an equitable estoppel,[25] requires three elements to be established by the party relying on it: first, a clear and unequivocal representation by the person said to have waived rights (ie the Insurer in the present case), whether by words or conduct, that the representor’s legal rights will not be insisted upon;secondly, reliance by the representee (ie the Assured in the present case) on the representation; and thirdly, that it would be inequitable for the representor to go back on the representation. See: The Copa Casino at para 39 referring to the analysis of waiver in the shipping context by Lord Goff of Chieveleyin Motor Oil Hellas (Corinth) Refineries SA v Shipping Corporation of India (The Kanchenjunga)[26]; see also Kosmar Villa Holidays Plc v Trustees of Syndicate 1243.[27] 19. I have dealt with the aspect of warranties in some detail because it was key in the present case. Reference was made by the courts below and by theparties in thisappeal to the duty of disclosure on an assured in a contract of marine insurance, this duty being contained in s 18 of the Ordinance, in that part headed “DISCLOSURE AND REPRESENTATIONS”.[28] For the reasons which appear below, I do not believe that any issue regarding the aspect of an assured’s duty of disclosure under this provision arises at all in the present case. Reference can conveniently be made to s 18(3)(d) of the MIO: it is the consideration of the Deadweight Warranty in this appeal that was crucial. 20. Finally in the context of the MIO, I would just make reference to s 91.[29] Although the Ordinance is a codification of the law of marine insurance, common law principles continue to apply and this is understood to include principles of equity as well.[30] Relevant here are the rules regarding the rectification of contracts of marine insurance, a topic which will have to be considered later. C. THE DECISIONS BELOW C.1 Court of First Instance 21. It is somewhat curious that in the judgment of Chung J, no mention is made of the warranty provisions of the MIO (neither s 33 nor s 34).[31] The Judge dealt with the case as between the Assured and the Insurer under four main headings:- contractual construction, non-disclosure, rectification and estoppel.[32] Subsumed in these topics is the important aspect of the Judge’s findings of fact regarding the state of the Insurer’s knowledge of the deadweight capacity of the Vessel. 22. On the issue of construction of the Contract, the Judge was of the view that there was a “clear inconsistency” between the naming of the Vessel in the Contract and the stipulation regarding her deadweight capacity.[33] Relying on the passage in para 12-078 of Chitty on Contracts[34]to the effect where a contract contained inconsistent clauses, effect must be given to that part which was calculated to give effect to the parties’ intentions, the Judge took no account of the Deadweight Warranty. He was reinforced in this approach by the fact that in the Cover Note, reference was made to an “APPROVED VESSEL” which, according to the Assured, could only be a reference to the Vessel (she having been named in the Application). 23. No doubt because this formed a part of the Insurer’s defence to the claim, the Judge then dealt with the question of non-disclosure under s 18 of the MIO. The Insurer’s case was that the Assured had breached the duty of disclosure by failing to disclose the true deadweight capacity of the Vessel prior to entering into the Contract. In rejecting this defence, the Judge held that since the Insurer had as much, if not greater, access to information regarding the deadweight capacity of the-Vessel, this was at least information which was available to both parties. The Judge’s thinking appeared to be that as the Insurer was in the marine insurance business, it would have easy access to information regarding the Vessel’s deadweight.[35] Accordingly, the Judge held that the duty of disclosure was not breached by the Applicant. The Judge was influenced by the fact that the particulars of a vessel’s deadweight were readily available on the internet.[36] 24. In view of the Judge’s conclusions on the construction of the Contract and on non-disclosure, he was of the view that it was strictly speaking unnecessary to deal with the question of rectification. The Judge dealt with this issue nonetheless. 25. Here, the Judge was of the view that since the Deadweight Warranty rarely featured in the previous contracts of marine insurance between the parties[37] and by reason also of the view he had taken on the true construction of the Contract,[38] the insertion of the Deadweight Warranty in the Contract did not represent the common intention of the parties.[39] The Judge also took into account the fact that the person within the Assured’s employment who had been responsible for taking out the relevant marine insurance cover in the present case (a Ms Wong), was not “sufficiently sophisticated or experienced in marine insurance” to appreciate the significance of the Deadweight Warranty, even though she had read it.[40] For these reasons, the Judge would have been prepared to order rectification of the Contract by deleting the Deadweight Warranty. Important for the purposes of this appeal, in dealing with the issue of rectification, the Judge made certain remarks in paras 67 and 70 of the judgment. These remarks formed the basis of the Assured’s submissions before us regarding the issue of the Insurer’s relevant knowledge:- (1) In para 67(2) of the judgment, the Judge stated his disagreement with the Insurer’s submission “there is no evidence the Insurer was aware of [the Vessel’s deadweight capacity].” (2) In para 70, the Judge stated that if the Insurer had alleged mistake in the case (it does not matter for present purposes how such a claim could have been made in the first place), by reason of its failure to call evidence, the Judge would have drawn an adverse inference against the Insurer “regarding whether it was aware of [the Vessel’s deadweight capacity].” It was contended by the Assured that these two paragraphs in the judgment showed that findings of primary fact were made by the Judge to the effect that at all material times, the Insurer was actually aware of the Vessel’s deadweight capacity. 26. I shall deal with this aspect in greater detail below.[41] For the time being, it suffices to say that in arriving at these supposed findings of fact, the Judge does not identify any evidence to suggest that the Insurer had actual knowledge, nor is there any. Para 67(2) of the judgment merely makes references to the Insurer’s written submissions at trial setting out its case. In para 70 of the judgment, reference is made to two earlier passages in the judgment,[42] but these passages do no more than to make the point that the particulars of the vessel’s deadweight capacity are available on the internet and that such information would be available to a marine underwriter such as the Insurer. With respect, this went nowhere near supporting a finding of actual knowledge. 27. The fourth issue dealt with by the Judge was estoppel. I confess it is obscure just what the Judge intended to say on this aspect of the case. It seems to be suggested that since the Broker in the present case must be taken to have passed on to the Insurer the Application, the Insurer was somehow estopped from denying that it did not know that the Vessel was the relevant vessel for the purposes of the Contract. This was not an issue before the Judge: clearly the Contract recognized that the relevant vessel carrying the cargo of logs was the Vessel. 28. In view of the above, the Judge found against the Insurer and gave judgment in favour of the Assured. The Insurer appealed to the Court of Appeal. C.2 The Court of Appeal 29. The Court of Appeal allowed the Insurer’s appeal and set aside the judgment against the Insurer. Cheung JA, with whose judgment the other members of the Court agreed, dealt with the same four matters determined by the trial Judge. 30. On the issue of the construction of the Contract, Cheung JA did not see any inconsistency at all per se between the naming of the Vessel in the Contract and the Deadweight Warranty. 31. However, he saw it may be possible for the Assured to raise some sort of inconsistency argument if it could be established that at the time the Contract was made, the Insurer actually knew of the Vessel’s deadweightcapacity. It is not clear at all from Cheung JA’s judgment just what the legal analysis would be in such a situation and this was not really explained in the judgment. Reference was made to those paragraphs in the judgment of Chung J in which findings of knowledge were said (by the Assured) to have been made,[43] as well as to para 66 of the judgment.[44] Cheung JA could not discern from these paragraphs any proper findings by the Judge of actual knowledge. He concluded “But the Judge never held there was actual knowledge on the part of the Insurer.”[45] 32. The analysis in relation to knowledge did not end there. Cheung JA then considered the question of presumed knowledge on the part of the Insurer, as to whether the Insurer could be presumed to have knowledge of the Vessel’s deadweight capacity in the present case. Reference was made to s 18 of the Ordinance[46] and consideration was given to what matters could be presumed to be known to a marine insurer. It is not entirely clear from Cheung JA’s judgment precisely what issue he was addressing, but I can only assume that the argument went something like this: if a certain fact can be presumed to be within the knowledge of an insurer, then it might ill behove an insurer to be able to rely on an express warranty in the contract of marine insurance which indicated the contrary fact. No authority was cited to the Court of Appeal to support this proposition except that reference was made to s 18 of the Ordinance. Reliance was placed on an extract from Bennett: The Law of Marine Insurance,[47]where, at paras 4.96 to 4.97 (in dealing with the aspect of presumed knowledge under s 18), it is stated:- “4.96 .... A fortiori, insurers are not presumed to know something about insured property merely because it has been published in the media, even when they are marine insurers and the information is published in a specialist media forum targeted at the maritime community. 4.97 However, specialist information is increasingly made available through dedicated electronic products. Insurers can subscribe to databases that contain extensive information on, for example, vessels and their owners. Is a hull insurer presumed to know all the information on such a database if the insurer subscribes to it, or even if the insurer does not but a prudent hull insurer would? The ability to search for and retrieve information with ease distinguishes electronic repositories of information from paper files, strengthens the case for developing the scope of presumed knowledge, and may serve to distinguish Bates v Hewitt were a case with similar facts to arise in the modern era.” The Assured relies on the same passage in this appeal, but only in the context of non-disclosure under s 18 of the Ordinance. Mr Man placed no reliance on the argument that presumed knowledge may somehow be sufficient to offset the effect of an insurance warranty. This point has not been fully addressed by the parties in the present case but I am sceptical that presumed knowledge in the sense referred to in this case can ever be sufficient to negate the effect of an insurance warranty. 33. Cheung JA was not persuaded that the Insurer could be said to have acquired actual knowledge or presumed knowledge. The fact that information about the Vessel’s deadweight capacity could be obtained on the internet did not mean that the Insurer was to be fixed with knowledge, whether actual or presumed. The fact that such information could be obtained did not mean that as a matter of law, it should have been obtained.[48] 34. On non-disclosure, Cheung JA reversed the Judge’s conclusion on this, although he regarded the relevant question to be considered was the breach of the Deadweight Warranty.[49] 35. In view of his earlier analysis regarding, the construction of the Contract and the Insurer’s knowledge, Cheung JA said there was no need to deal with the rectification claim. In effect, he was of the view that the Contract, including the Deadweight Warranty, did reflect the parties’ intentions. 36. On waiver and estoppel, in view of the conclusion on knowledge, Cheung JA did not see any support for the view that the Insurer had waived reliance on the Deadweight Warranty, or was somehow estopped from raising it. C.3 This Appeal 37. As stated above, the Assured appealed to this Court as of right. It sought to challenge both the conclusions and reasoning of the Court of Appeal. In providing the reasons for dismissing the appeal, I shall deal with the following matters in turn:- (1) The legal effect of the Deadweight Warranty. (2) The construction of the Contract. (3) The state of knowledge of the Insured. (4) Rectification. (5) Waiver and estoppel. (6) Non-disclosure. D. THE LEGAL EFFECT OF THE DEADWEIGHT WARRANTY 38. I have already touched upon the law regarding marine insurance warranties.[50] Mr Man accepted that the Deadweight Warranty was a marine insurance warranty and that s 33 of the Ordinance was therefore engaged. There is no doubt that the Deadweight Warranty is a marine insurance warranty. By this warranty, the Assured affirmed a particular state of facts, namely, that the carrying vessel of the cargo (the relevant interest insured under the Contract) would have a deadweight capacity of over 10,000 tonnes. 39. The legal consequence of the Deadweight Warranty being a marine insurance warranty, as we have seen, is that in the event of breach, an insurer will generally be discharged from liability under the relevant contract of marine insurance.[51] I use the word “generally” because an insurer may be prevented as a matter of law from asserting its rights following a breach of warranty. 40. In the present case, the Assured contends that the Insurer was prevented from relying on the Deadweight Warranty by reason of any one or more of the following: on a true construction of the Contract, rectification and waiver/estoppel. The Assured maintained more or less the same arguments as it raised in the lower courts. E. THE CONSTRUCTION OF THE CONTRACT 41. It was apparent from Mr Man’s submissions at the hearing that reliance was still placed on the Judge’s reasoning that there was somehow an inherent inconsistency between naming a vessel in a contract of marine insurance and a warranty requirement regarding the deadweight capacity of that vessel (such as the Deadweight Warranty in the present case). Mr Man was right not to seek to advance this point with much enthusiasm; in my view, it is unarguable. The mere fact that a vessel is named in a contract of marine insurance does not mean in any way that an insurer is somehow prevented from insisting by way of warranty on that vessel possessing certain characteristics. Nothing in the Ordinance remotely suggests otherwise. Indeed, the references to express warranties[52] and the various other warranties in the MIO[53] all suggest that they would exist even where the relevant carrying vessel is known to an insurer. Just to give one illustration: it cannot seriously be contended that just because an insurer is aware of the name and voyage of a relevant vessel that the insurer is somehow prevented from insisting on compliance with the warranty of seaworthiness contained in s 39 of the Ordinance. 42. I see no inconsistency in the Contract between the identification of the Vessel and the existence of the Deadweight Warranty. Nor is the analysis any different by reason of the fact that the Vessel may have been an “APPROVED VESSEL”.[54] 43. Rather, the Assured’s position was that by reason of the Insurer’s knowledge in the present case of the Vessel’s deadweight capacity, it was prevented from relying on the Deadweight Warranty. The Assured submitted that the Judge made an express finding that at all material times the Insurer knew of the Vessel’s deadweight, that this finding should not be disturbed by this Court and it should not have been disturbed by the Court of Appeal. As the Assured’s Written Case stated, this finding of knowledge “should form the basis of the adjudication of this appeal”.[55] 44. What is clear from the Written Case and Mr Man’s submissions was that the question of knowledge was central to the success of this appeal. I was content in dealing with the issue of knowledge to assume that if the Assured had succeeded on this aspect, it would then likely succeed in the appeal. The Insurer was prepared to make this assumption as well. But it is right to point out that we have not been extensively addressed on the proper legal analysis to be adopted where an insurer is aware of facts indicating the contrary of what forms the subject matter of an insurance warranty. The legal analysis is not in my view at all straightforward and I daresay that much would depend on the particular facts in any given case. For my part though, contrary to the view taken by the trial judge, the Court of Appeal and Mr Man before us, I find it difficult to accept that the special knowledge of a party to a contract might affect what otherwise would be the true construction (as opposed to the effectiveness) of the terms of that contract. Much less, in the exercise of construing a contract, could a party’s knowledge result in a term of contract being ‘red pencilled’ into oblivion by its total deletion, as Mr Man submitted at the hearing (as far as the Deadweight Warranty was concerned). A party’s knowledge may, however, in some circumstances, result in some form of waiver on estoppel being applicable. Nevertheless, as indicated earlier, we did not hear full submissions on this and I was content to treat as the most important threshold question in the present appeal whether the Assured could succeed in establishing the requisite degree of knowledge on the Insurer’s part. It is to this aspect I now turn. F THE STATE OF KNOWLEDGE OF THE INSURER 45. The Assured contended first that the trial Judge made positive findings of primary fact to the effect that the Insurer did at all material times have knowledge of the Vessel’s deadweight capacity. Reliance was placed on those parts of Chung J’s judgment to which reference has already been made above,[56] but, as I have attempted to demonstrate, the Judge did not identify any evidence to support such a finding at all, if indeed he did make such a finding in the first place.[57] The Court of Appeal was of the view that no such finding of actual knowledge had been made by the Judge.[58] Nor was Mr Man really able to identify any evidence to justify what he submitted were the Judge’s findings of primary fact. 46. It was submitted, however, that the Assured having raised a prima facie case on the facts of actual knowledge, this prima facie position should be taken factually to be the established position in the absence of any contradictory evidence adduced by the Insured. In other words, a prima facie case on the facts having been raised, adverse inferences could be drawn from the failure to adduce contradictory evidence, particularly where a party could be expected to provide such evidence;[59] in such situations silence would be fatal. 47. However, before a prima facie factual situation can be said to exist, there must be evidence adduced of “sufficient cogency” to raise a prima facie case in the first place.[60] The Assured did not reach this threshold. In its Written Case, reliance was placed on the Statement of Claim in which there wasadmittedly a reference (without any particulars) to actual knowledge; reliance was also placed on the fact that information on the Vessel’s deadweight capacity could be found on the internet. These were insufficient by a long way to make out a prima facie case on knowledge. 48. There being no evidence of actual knowledge, the Assured then tried to make out a case to suggest that some form of constructive knowledge was sufficient. The trial Judge dealt with presumed knowledge on the part of the Insurer in the context on non-disclosure.[61] I have already referred to the way the Court of Appeal dealt with the issue of presumed knowledge in the context of the true construction of the Contract.[62] The Assured’s submissions before us took a rather different turn. As I understood Mr Man’s submissions, put simply, they amounted to this: since the exercise in contractual construction involves the Court taking into account the factual matrix of the relevant contract[63] and the factual matrix includes all facts which may be “reasonably available” to the parties,[64] the Vessel's deadweight capacity in the present case being “reasonably available” to the parties[65] meant that the Insurer was to be taken to have knowledge of the Vessel’s deadweight capacity; the result then is that the Deadweight Warranty should be given no effect. 49. I cannot agree with this somewhat elaborate argument. The account that one takes of the factual matrix of a contract is to assist in arriving at the true construction of the contract and its terms. It does not have some separate life of its own to undermine or nullify the effect of a clear term of the contract. The meaning and effect of the Deadweight Warranty in the present case is clear and no assistance can be derived by reference to the factual matrix of the Contract. 50. The Assured’s submissions on factual knowledge must fail, and insofar as some form of constructive or presumed knowledge is relied on, this did not advance the Assured’s case at all. Any information to which the Insurer may have had access did not affect the operation of the Deadweight Warranty. G. RECTIFICATION 51. Mr Man submitted (and I agree) that the correct approach to rectification of a contract where it does not accurately reflect the parties’ true agreement, is that set out in Agip SpA v Navigazione Alta Italia SpA (The Nai Genova and Nai Superba):- “As the law stands, the conditions which must be satisfied if rectification is to be granted on the grounds of common mistake may, in my opinion, be summarized as follows: First, there must be a common intention in regard to the particular provisions of the agreement in question, together with some outward expression of accord. Secondly, this common intention must continue up to the time of execution of the instrument. Thirdly, there must be clear evidence that the instrument as executed does not accurately represent the true agreement of the parties at the time of its execution. Fourthly, it must be shown that the instrument, if rectified as claimed, would accurately represent the true agreement of the parties at that time: (see generally Snell's Equity, 28th ed. (1982) at pp. 612-614).”[66] 52. In support of its case here, the Assured referred to that part of Chung J’s judgment in which he concluded that the insertion of the Deadweight Warranty must have been an error and did not represent the parties’ common intention.[67] As mentioned above, the Judge arrived at this conclusion by reference to two matters: the fact that the Deadweight Warranty had featured only rarely in the contracts of marine insurance previously entered into by the parties, and the view that the Judge took regarding what he saw as an inconsistency in the Contract between the naming of the Vessel and the Deadweight Warranty.[68] The first reason, however, provides no basis at all for saying that the Deadweight Warranty was therefore somehow ineffective as a term of the Contract. This is notwithstanding even the fact that although previous policies may not have contained a deadweight warranty, the cover notes in relation to such policies did.[69] The second reason is based on the Judge’s construction of the Contract, and that construction, for the reasons set out above,[70] was in error. Further, as indicated earlier, the Judge also seemed to place reliance on the fact that the Assured’s employee (Ms Wong) was not sufficiently sophisticated to appreciate the significance of the Deadweight Warranty.[71] This only has to be stated to be rejected as a ground for rectification and in fairness, Mr Man did not advance any submissions based on this point. 53. The claim for rectification accordingly failed. H. WAIVER AND ESTOPPEL 54. If the Deadweight Warranty was valid, the Assured contended that there had been waiver on the part of the Insurer. Reference was made to s 34(3) of the Ordinance. Under that provision, as discussed earlier, the meaning of waiver is waiver by estoppel or equitable estoppel.[72] On the facts of the present case, the Assured could not satisfy the requisite conditions for this form of waiver to apply:- (1) It was the Assured’s case that the Insurer made a representation to it that notwithstanding the Deadweight Warranty and notwithstanding the fact that the Vessel’s deadweight capacity did not comply with that warranty, the Insurer would nevertheless accept the Vessel for the purposes of marine insurance cover under the Contract. According to the Assured, the representation arose by reason of the following matters: the Insurer’s knowledge that the relevant vessel for the voyage was in fact the Vessel, the ease by which the Insurer could have found out about the Vessel’s deadweight capacity (through the internet), the issuance of the Policy and the acceptance of the premium. The Assured is said to have relied on this representation by not taking out any other policy of marine insurance. It was submitted that, accordingly, it would be inequitable to allow the Insurer to rely on the Deadweight Warranty. (2) This argument fell at the first hurdle. None of the matters relied on, whether singly or cumulatively, could possibly amount to the requisite clear and unequivocal representation contended for. Further, the underlying weakness of the argument, which has pervaded most of the Assured’s contentions in this case, was the insistence there was some fundamental inconsistency between the naming of a vessel in a contract of marine insurance and a warranty going to the existence of a state of facts regarding that vessel (such as deadweight capacity). I. NON-DISCLOSURE 55. Both courts below dealt with the defence advanced by the Insurer that the Assured had breached the duty of disclosure under s 18 of the MIO.[73] In its written Case, the Insurer maintained this defence before us. However, as the Court of Appeal remarked[74]and as Mr Man accepted before us, the present case is really about breach of warranty. 56. I agree. I do not see the relevance of non-disclosure in the present case, given the existence of the Deadweight Warranty. There is a certain illogicality in defending a marine insurance claim on the basis that a material fact should have been, but was not, disclosed by the assured to an insurer,[75]when that very fact is the subject matter of a marine insurance warranty.[76] Indeed, s 18(3)(d) of the Ordinance states that in the absence of inquiry, any circumstance which is superfluous to disclose by reason of an express warranty, need not be disclosed. To be fair, Mr Coleman accepted at the hearing that non-disclosure was not a relevant issue in the present appeal. 57. The aspect of non-disclosure not being relevant to consider, it was unnecessary in this appeal to go into the intricacies of s 18 of the MIO. I would like, however, just to comment on one aspect here. There was some discussion in the lower courts and before us regarding the extent of knowledge which can be presumed on the part of an insurer for the purposes of s 18(3)(b) of the MIO. Reliance was placed by the Assured on passages contained in Prof Bennett’s textbook, The Law of Marine Insurance.[77] By not dealing with this aspect, I should not be taken to accept what is said in these passages. The question of what an insurer is presumed to know in the course of underwriting marine insurance is not altogether a straightforward one and there is a long history of case law.[78] For a quick reference to this area of marine insurance, one needs only to refer to leading textbooks on this subject matter.[79] J. CONCLUSION 58. For the above reasons, this appeal was dismissed. Quite simply, the Deadweight Warranty was breached and there was no answer to that. Mr Justice Ribeiro PJ: 59. I agree with the judgment of the Chief Justice. Mr Justice Tang PJ: 60. I agree with the judgment of the Chief Justice. Mr Justice Fok PJ: 61. I agree with the judgment of the Chief Justice. Lord Neuberger of Abbotsbury NPJ: 62. I agree with the judgment of the Chief Justice. Mr Bernard Man, instructed by S.K. Lam, Alfred Chan & Co., for the Appellant Mr Russell Coleman SC, instructed by Reed Smith Richards Butler, for the Respondent [1] The Assured was represented by Mr Bernard Man. [2] Judgment dated 12 August 2013 (Cheung, Chu and Barma JJA). [3] Judgment dated 6 August 2012 (Chung J). [4] The insurer was represented in this appeal by Mr Russell Coleman SC. [5] Cap. 484. [6] The broker chose not to attend the appeal. [7] DWT is a reference to the deadweight capacity of a vessel. Generally speaking, the deadweight capacity of a vessel (measured usually in metric tonnes) represents the weight of cargo, fuel, stores and crew which a vessel may carry. It reflects the size and earning capacity or potential of a vessel. Deadweight is often used to calculate premiums due under a marine insurance policy or in the calculation of harbour dues and taxes. Technically, it is calculated by the displacement of water when a vessel is fully loaded compared with when she is unloaded. [8] Judgment of Chung J at para 20. [9] See Judgment Court of Appeal at para 7. [10] She only had a deadweight capacity of about 8,960 tonnes. [11] Such as the Motor Vehicles Insurance (Third Party Risks) Ordinance Cap 272 and the Third Parties (Rights against Insurers) Ordinance Cap 273 (both enacted in 1951 one immediately after the other). [12] The main change was contained in s 60(2)(iii) of the Act – our s 60(2)(c) – relating to constructive total loss in the case of damage to goods. This was said to have amended the law represented by Farnworth v Hyde (1866) LR 2 CP 204: see Arnould: Law of Marine Insurance and Average (18th ed) at paras 29-54 fn 288, 29-59. [13] See Hong Kong Legislative Council: Official Report of Proceedings for 10 May 1961. [14] The Ordinance only applies to contracts of marine insurance, being contracts whereby an insurer undertakes to indemnify an assured against marine losses, that is to say, losses incident to marine adventure: s 1. There is no dispute the Contract was a contract of marine insurance. [15] The only differences are minor and cosmetic, mainly the use of different headings or the use of terminology to make the statute a Hong Kong one. Section 92 of the Ordinance is the Hong Kong equivalent of s 1 of the Marine Insurance (Gambling Policies Act 1909), complementing s 4 of the Ordinance (in relation to wagering and gambling). [16] See s 35(1) of the MIO. And of course a warranty may be express or implied: s 33(2). [17] See Arnould at para 19-04. [18] Arnould at para 19-06. [19] [1992]1AC 233. [20] Moussi H. Issa NV v Grand Union Insurance Company Limited [1984] HKLR 137, at 142 A-B (per Sir Alan Huggins VP). [21] [2001] 2 Lloyd's Rep 161 at para 124 (per Rix LJ). [22] Kammins Ballrooms Company Limited v Zenith Investments (Torquay) Limited [1971] AC 850, at 882H (per Lord Diplock); Mardorf Peach and Company Limited v Attica Sea Carriers Corporation of Liberia (The Laconia) [1977] AC 850, at 871C (per Lord Wilberforce). [23] See para 17(3) above; The Good Luck. [24] [2012] 1 Lloyd's Rep 129, at para 38 (per Aikens LJ). [25] Whereby contractual rights are effectively renounced: see Equity: Sarah Worthington (2nd ed) at Pg 245 fn 56. [26] [1990] 1 Lloyd's Rep 391, at 397(1) to 399(2). [27] [2008] 1 CLC 307, at paras 36 to 38 (per Rix LJ). [28] Section 18 states:- “18. Disclosure by assured (1) Subject to the provisions of this section, the assured must disclose to the insurer, before the contract is concluded, every material circumstance which is known to the assured, and the assured is deemed to know every circumstance which, in the ordinary course of business, ought to be known by him. If the assured fails to make such disclosure, the insurer may avoid the contract. (2) Every circumstance is material which would influence the judgment of a prudent insurer in fixing the premium, or determining whether he will take the risk. (3) In the absence of inquiry the following circumstances need not be disclosed, namely- (a) any circumstance which diminishes the risk; (b) any circumstance which is known or presumed to be known to the insurer. The insurer is presumed to know matters of common notoriety or knowledge, and matters which an insurer in the ordinary course of his business, as such, ought to know; (c) any circumstance as to which information is waived by the insurer; (d) any circumstance which it is superfluous to disclose by reason of any express or implied warranty. (4) Whether any particular circumstance, which is not disclosed, be material or not is, in each case, a question of fact. (5) The term “circumstance” (情況) includes any communication made to, or information received by, the assured.” [29] Section 91 states:- “91. Saving The rules of the common law including the law merchant, save in so far as they are inconsistent with the express provisions of this Ordinance, shall continue to apply to contracts of marine insurance.” [30] See Howard Bennett: The Law of Marine Insurance (2nd ed) at para 1.47. [31] Although mention is made in passing of s 17 of the Ordinance: Judgment of Chung J para 59. [32] Judgment of Chung J at para 40. [33] Judgment of Chung J at Para 45. [34] 31sted, 2012. [35] Judgment of Chung J at para 63. [36] Judgment of Chung J at para 38. [37] See para 9 above. [38] See para 22 above. [39] Judgment of Chung J at para 66. [40] Judgment of Chung J at paras 34(c), 68. [41] Section F below. [42] Paras 38 and 63 of the Judgment. [43] Paras 67(2) and 70 of the judgment of Chung J. See para 26 above. [44] See para 25 fn 39 above. [45] Judgment of the Court of Appeal para 16.11. [46] Set out in para 19 fn 28 above. [47] Earlier referred to in para 20 fn 30 above. [48] Judgment of the Court of Appeal at paras 16.14 to 16.24. [49] Judgment of the Court of Appeal at paras 17 and 18. [50] Paras 16 to 18 above. [51] See: s 33(3) of the Ordinance; para 17(3) above. [52] Section 35 of the MIO. [53] Contained in ss 36 to 41. [54] See para 6(1) above. [55] Para 54 of the Appellant’s Case. [56] Para 67(2) and 70 of Chung J’s judgment; para 25 above. [57] Para 26 above. [58] Para 31 above. [59] See Nina Kung v Wang Din Shin (2005) 8 HKCFAR 387, at paras 367 to 369 (per Ribeiro PJ); R v Inland Revenue Commissioners ex parte TC Coombs & Company [1991] 2 AC 283, at 300 F-G (per Lord Lowry). [60] See Nina Kung at para 369 (per Ribeiro PJ). [61] See Para 23 above. [62] See para 32 above. [63] See Prenn v Simmonds [1971] 1 WLR 1381, at 1385H (per Lord Wilberforce); Reardon Smith Line Limited v Yngvar Hansen-Tangen [1976] 1WLR 989, at 995G to 997D (per Lord Wilberforce); Investors Compensation Scheme Limited v West Bromwich Building Society [1998] 1 WLR 896, at 912F to 913F (per Lord Hoffmann). These well-established principles of construction apply equally to marine insurance contracts: see Pratt v Aigaion Insurance Company SA (The Resolute) [2009] 1 Lloyd’s Rep 225, at paras 9 to 14 (per Sir Anthony Clarke MR). [64] The words “reasonably available” appear in the speech of Lord Hoffmann in Investors Compensation Scheme Limited, at 912(H). [65] By reason of the availability of such information on the internet. [66] [1984] 1 Lloyd’s Rep 353, at 359 (per Slade LJ). This is an authority dealing with common mistake. [67] Judgment of Chung J at para 66. [68] Para 25 above. [69] Para 9 above. [70] Para 41 above. [71] See para 25 above. [72] See para 18 above. [73] See Paras 19, 23 and 34 above. [74] See para 34 above. [75] Section 18 of the MIO. [76] Under s 33 of the MIO. [77] See para 32 above. [78] Beginning with Carter v Boehm (1766) 3 Burr. 1905 (per Lord Mansfield). [79] See, for example, Arnould at Ch. 16; Marine Insurance: Law and Practice (2nd ed) by Prof FD Rose at paras 5.23 to 5.87. 
Mr Justice Fok PJ: 1. This uncontested appeal is a salutary reminder that, in order properly to exercise the discretion whether to permit a defendant to reverse his plea from guilty to not guilty, a magistrate has a duty to inquire into and adjudicate upon a defendant’s request to do so before proceeding to pass sentence. 2. This is the judgment of the Court and arises from the following facts. The facts 3. The appellant is a cross-border truck driver. On 21 June 2010, he was driving a 45-foot container truck from the Mainland into Hong Kong via the Man Kam To Control Point. Upon his cargo being examined, it was discovered that it contained: (1) 12,030 batteries, 6,000 chargers, 140 handbags, 2,000 headphones, 400 mobile phones, 479 pairs of sports shoes, 118 suits, 61 ties, 40 pairs of pants, 58 T-shirts and 6,305 watches to each of which a forged trade mark was applied; (2) 570 pirated video game cartridges; and (3) 12,795 batteries, 6,040 chargers, 7,270 watches, 570 video game cartridges, 20 converters and 20 remote controls that were not included in the manifests submitted by the appellant. 4. The appellant was thereupon arrested and charged with the following offences: (1) Importing goods to which a forged trade mark was applied, contrary to sections 12(1) and (2) as read with section 18(1) of the Trade Descriptions Ordinance, Cap.362; (2) Importing an infringing copy of a copyright work into Hong Kong otherwise than for private and domestic use without the licence of the copyright owner of the work, contrary to sections 118(1)(b) and 119(1) of the Copyright Ordinance, Cap.528; and (3) Importing unmanifested cargo, contrary to section 18(1)(a) of the Import and Export Ordinance, Cap.60. 5. Upon arraignment before the Magistrate[1] on 9 June 2011, the appellant, represented by counsel, Mr Steven Liu, pleaded not guilty to each of these charges and the case was adjourned to 18 August 2011 for trial. 6. On 18 August 2011, the appellant appeared, again represented by Mr Liu, before the Magistrate and pleaded guilty to each of the charges and admitted the facts. He was then duly convicted on his plea. The case was then adjourned for three weeks in order for background and community service order reports to be obtained before sentencing. He was granted bail in the meantime. 7. At the resumed hearing on 8 September 2011, Mr Liu informed the Magistrate that the appellant wished to reverse his plea and to apply for an adjournment of 7 days in order to instruct a new legal representative. Counsel informed the Magistrate that his instructions were limited to so informing the court and so the Magistrate then discharged counsel. 8. The Magistrate then asked the appellant, now acting in person, a series of questions to ascertain what he had done to secure the services of another lawyer to act for him. The Magistrate clearly doubted the genuineness of the appellant’s efforts to find another lawyer, principally because it would appear the appellant only tried to do so the night before the hearing notwithstanding having been on bail during the adjournment of three weeks. Consequently, the Magistrate refused to entertain the appellant’s application to change his plea and to adjourn the hearing. After exchanges with the appellant as to anything he wished to add by way of mitigation in the light of the reports, he proceeded to sentence the appellant to concurrent sentences of 6 months’ imprisonment on each charge. 9. After he had served his term of imprisonment, the appellant applied for leave to appeal out of time against his conviction and leave was granted on 14 September 2012. His application for legal aid was refused on 12 November 2012 and he acted in person at the appeal on 6 December 2012. The appeal[2] was heard by Deputy High Court Judge J. Yau and dismissed by him in a judgment dated 16 January 2013. The Judge did not consider the issue of whether the Magistrate had inquired adequately or at all into the appellant’s application to reverse his plea. 10. On 22 April 2013, the appellant applied to the Appeal Committee for leave to appeal to this Court on the ground of substantial and grave injustice. The Appeal Committee[3] granted that application on 24 September 2013. The duty to inquire into an application to reverse a guilty plea 11. It is a self-evident proposition that a plea of guilty should be entered voluntarily. A guilty plea made under duress, inducement or misrepresentation is a nullity: see, for example, The Queen v Lam Yin [1995] 2 HKCLR 124 at p.127 line 20-24; R v Li Yuen Chu [1994] 2 HKC 621; and HKSAR v Wong Chi Yuk [2000] 3 HKLRD 125 at p.135. 12. A magistrate has a discretion to allow a defendant to change a plea of guilty to one of not guilty at any time prior to passing sentence. This is the case even where the guilty plea is unequivocal. The discretion must be exercised judicially. For these propositions, see Chan Wah v The Queen [1967] HKLR 254; S (An Infant) v Recorder of Manchester [1971] AC 481 (HL) esp. per Lord Reid at pp.488H and 491A-E and Lord Upjohn at p.507E-H; and Eric Henry Dodd and Others (1982) 74 Cr.App.R. 50 at p.57. 13. Where a defendant applies to change his plea from guilty to not guilty, it is incumbent on a magistrate to make sufficient inquiries to ascertain the basis of the defendant’s wish to reverse his plea and to decide if that basis is sound in fact and in law. Authority for this proposition can be found in the judgment of Huggins JA (as he then was) in Wong Ching-wah v The Queen [1976] HKLR 412 where he said, at pp.413-414: “What, however, worries me is that the learned magistrate did not apparently make any proper enquiry into the facts at the time that [the defendant’s solicitor] made his application. The question was then one of fact, whether or not the plea was her plea or whether it was a plea made under duress. That, in my view, was a matter which should have been investigated as a matter of fact and probably on oath. No enquiry was made: the magistrate merely says that in all the circumstances he was satisfied that the plea should not be allowed to be withdrawn. His conclusion may have been right, but in my view there was not sufficient enquiry. I am not going to embark upon an enquiry at this stage and justice requires that this girl be given the opportunity to reverse her plea and make her defence.” (Emphasis added) 14. In The Queen v Lee Kwok Sum, unrep., HCMA 1597/1983 (7 February 1984), Garcia J held, citing Wong Ching-wah as authority, that the discretion to allow a reversal of a plea cannot be exercised judicially without a hearing to inquire into allegations of duress. 15. There are other examples of the application of this principle where the issue is not one of duress but of inducement and misrepresentation: see, Law Hau On v The Queen, unrep., CACC 578/1982 (3 August 1982); R v Li Yuen Chu [1994] 2 HKC 621; and The Queen v Lam Yin [1995] 2 HKCLR 124. As Sir Alan Huggins VP held in Law Hau On v The Queen (at p.4): “It is easy to be wise after the event but it must be said that in the event it is unfortunate that the magistrate did not inquire into the allegation, for we would then know the real facts. As it is, there is a possibility that the Appellant pleaded Guilty when he would not have done so had he not been induced by circumstances which he could now reasonably believe to have amounted to a trick. However remote that possibility I think that justice requires that his plea be declared null and void and that there should be a venire de novo. It matters not that the magistrate may well have been right when he said that on the basis of the admitted facts it would have been difficult for anyone to advise the Appellant to do otherwise than enter a plea of Guilty: the Appellant was entitled, if he wished, to have the case formally proved against him.” (Emphasis added) 16. What will amount to sufficient inquiries will depend on the circumstances of the case and the basis on which it is alleged the defendant should be permitted to reverse his plea. In The Queen v Lam Yin [1995] 2 HKCLR 124 Keith J (as he then was) considered that it would be necessary for the magistrate to adjourn the application for leave to change the pleas to a date upon which the appellant and immigration officer could give evidence about the alleged conversation said to be the basis of the misrepresentation inducing his guilty pleas (see pp.127-128). In that case, whether the immigration officer actually said what the appellant claimed, and what its impact on the appellant actually was, were issues of fact which had to be determined before the magistrate could decide if the original pleas were nullities. In other cases, the magistrate’s duty to inquire may not require the adjournment of the case if any factual issues that need to be resolved in order for the magistrate to exercise a fully-informed discretion can fairly be determined there and then. 17. But, as the underlined passages from the judgments cited above show, what the magistrate must not do is anticipate the outcome of the inquiry he is obliged to undertake as to the reason for the change of plea and, on that basis, refuse to make any or any sufficient inquiries. 18. It should not, however, be thought that the discretion to permit a defendant to reverse his plea should be exercised lightly and magistrates will have to be wary of applications to reverse pleas which are merely being used as a delaying tactic. As Lord Upjohn said in S (An Infant) v Recorder of Manchester at p.507G: “… this discretionary power is one which should only be exercised in clear cases and very sparingly.” Discussion 19. As indicated at the beginning of this judgment, the appeal is uncontested by the respondent and the parties have filed a Joint Case. It is jointly submitted that by failing to inquire into the appellant’s application for a reversal of his plea of guilty, the Magistrate departed from an established legal norm to the appellant’s disadvantage and that justice requires that the appeal be allowed. It is further submitted that, on the Magistracy Appeal, the Judge failed to correct the error of the Magistrate and did not direct his mind to the failure of the Magistrate to inquire into the appellant’s application to change his plea. 20. The Joint Case asserts that, at the hearing before the Magistrate on 8 September 2011, the appellant, acting in person, tried to explain to the court that he was misled by his former lawyers but the court stopped him. 21. The following parts of the transcript of the hearing on 8 September 2011[4] are important: 22. The appellant made it clear to the Magistrate that he wished to plead “not guilty” and, in view of his earlier plea, this was clearly a change of plea. In addition to saying he wished to plead not guilty, he expressly disavowed his guilty plea and asked for a retrial. In the context of the case, his request for a retrial was a clearly request for his plea of not guilty to be tried and his conviction, which had earlier been entered on his guilty plea to each of the charges, to be reconsidered. 23. Despite the basis of any complaint of misrepresentation against the appellant’s former lawyers not being clearly asserted, it is apparent from the transcript that the appellant informed the Magistrate that the guilty plea was not voluntary (“he has told me to plead guilty, but I don’t”) and that the appellant considered that his former lawyers had represented to him that they would defend him (“they said they would defend for me, this lawyer. After engaging this lawyer, and then he said my probation”) but this was self-evidently no longer the case. 24. In those circumstances, particularly since the appellant was acting in person and had indicated difficulty to the Magistrate in expressing himself (“Please listen to me, my Hakka is not good – my Punti is not very clear”) it was incumbent on the Magistrate to make further inquiries to ascertain precisely what it was that the appellant was trying to say. Clearly, he was indicating a desire to change plea and the Magistrate was aware of that. What was critical was the reason for that desire to reverse his plea and this was what the Magistrate was duty bound to inquire into. 25. Although the allegation of misrepresentation or improper pressure was not raised as clearly as it could have been at the hearing before the Magistrate on 8 September 2011, when the appellant’s Magistracy Appeal was heard, the appellant, again acting in person, did clearly express to the Judge that: “It was the lawyer engaged for him by the company who owned those goods to induce him to plead guilty.”[5] This stance was also repeated in the appellant’s letter to the High Court requesting an extension of time for the Magistracy Appeal and in his affirmation seeking leave to appeal to this Court. 26. The appellant may or may not have been induced by misrepresentation or improper pressure to enter his guilty plea. But if he was, the plea was a nullity and should not have been accepted. That factual question is one which the Magistrate was obliged to inquire into and resolve, if necessary, by hearing the evidence of the appellant and his former lawyers. Depending on the scope of the issue and the evidence relevant to it, the Magistrate may or may not have been able to conduct that inquiry then and there. However, he did not make that or any inquiry into the circumstances of the appellant’s wish to change his plea and instead thought the appellant was simply employing delaying tactics. That may or may not have been the case but, without a sufficient inquiry into the appellant’s reason for wanting to change his plea, the Magistrate was not in a position properly to exercise his discretion to permit or refuse the plea reversal. Disposition 27. For the reasons set out above, the appellant’s appeal should be allowed and his conviction on each charge quashed. 28. In the Joint Case, the respondent takes the stance that, in view of all the circumstances of the case and the appellant, including the nature of the offences and the fact the appellant has already served his term of imprisonment, it is not in the public interest to have the case remitted to a magistrate for a re-trial. It is therefore submitted that his conviction should be quashed without requiring him to stand another trial. In response to a request from the Court, further written submissions on the question of a remitter were sought from the parties and duly provided, the parties indicating that they were content for the Court to resolve the question of whether there should or should not be a remitter on the basis of the further written submissions alone and without an oral hearing. 29. The further joint submissions maintain that it is appropriate, if the appeal is allowed, to dispose of the appeal without ordering a remitter to the magistrate to inquire into the appellant’s application to change plea. It is submitted that the long lapse of time since the original trial and Magistrate’s refusal to permit the appellant to reverse his plea and the fact that the appellant has fully served his sentence constitute special circumstances justifying a different outcome to that in other cases where either a remitter was ordered or an inquiry held by the judge hearing the magistracy appeal. It is submitted that the lapse of time and circumstances of the appellant would result in his being materially prejudiced in any inquiry into the plea reversal. Furthermore, it is submitted that it is not in the public interest to have a re-trial. 30. The authorities are not entirely consistent as regards disposition where there has been a failure to make proper inquiry on an application to change plea. In R v Li Yuen Chu, Woo J remitted the case to the magistrate to inquire into the application to withdraw the guilty plea (see p.623). In The Queen v Lam Yin, Keith J adjourned the magistracy appeal before him to a date when he would himself hear evidence to decide the issues of fact relevant to the change of plea application (see p.128). The report of the case in Wong Ching-wah v The Queen suggests that Huggins JA ordered a re-trial (see p.414). However, it would seem from the judgment that his intent was that “this girl be given the opportunity to reverse her plea and make her defence”. It is not clear, therefore, whether the re-trial was in fact a remitter of the issue of duress in relation to the original plea. Finally, in Law Hau On v The Queen, Huggins VP declared the original plea null and void and ordered a venire de novo, i.e. (it would seem) a re-trial on what would presumably be a fresh plea of not guilty (see p.4). 31. As between the choices of remitter or re-trial, in ordinary circumstances the more appropriate disposition may be remitter of the issue of whether the appellant should be permitted to change his original plea of guilty on the basis it was, in law, a nullity. The remitter of the issue of whether the change of plea should be permitted does not assume the nullity of the original plea of guilty and is appropriate because the conviction is not quashed because the appeal court considers the appellant should have been acquitted after trial. Rather it is because the appellant was deprived of the opportunity to change his plea and to have the case against him formally proved. In contrast, an order for re-trial makes the assumption of nullity of the original plea. 32. In ordinary circumstances, therefore, the appropriate disposition where there has been a failure to make proper inquiry on an application to change plea is an order remitting the matter to a magistrate to inquire into that application. The remitter will clarify whether the original plea should stand, in which case a conviction will follow, or whether a trial should proceed on the basis of a not guilty plea. There would ordinarily be a public interest in having a proper determination of a defendant’s guilt or innocence of the offences with which he has been charged. That is all the more so in the present case where the offences in question are serious offences attracting (as in this case) a custodial sentence. 33. In the present case, however, in the light of the parties’ joint further submissions, the Court is persuaded that it should, exceptionally, not order a remitter of the change of plea issue to a magistrate. There has now been a lapse of nearly three years since the original plea was entered on 9 June 2011 which will inevitably make the trial of the remitter and, if the change of plea is permitted, of the charges against the appellant more difficult. More significantly, the appellant has already served his prison sentence and so, even if the change of plea were not permitted on the remitter (so that the original guilty plea were to stand), the appellant would be convicted on his guilty plea and the matter would then proceed to sentencing at which time the magistrate would take into account the fact that the appellant has already served a 6-month term of imprisonment for the offences. 34. Finally, the costs being agreed between the parties, the Court makes an order that the respondent pay the appellant’s costs of the appeal, to be taxed if not agreed, and shall pay the appellant an agreed sum of $1,392 in respect of disbursements for the application for leave to appeal. The appellant’s own costs of the appeal are to be taxed in accordance with the Legal Aid Regulations. The practice and procedure for uncontested appeals 35. Following the grant of leave to appeal, which was contested, the solicitors for the appellant informed the Registrar that the Department of Justice had agreed that the appeal should be allowed by way of the Mok Kin Kau procedure.[6] Subsequently, the solicitors for the appellant informed the Registrar that the parties had agreed the contents of the Joint Case and duly lodged the same. 36. As the discussion above shows, the Court was not initially persuaded by the Joint Case as to the appropriate disposition of the appeal and therefore invited the parties to make further submissions on that question. This is therefore a timely opportunity to provide some guidance as to the practice and procedure to be adopted when parties may be minded to invite the Court to apply the Mok Kin Kau procedure. 37. Where a party is minded to concede an appeal, therefore, the following points of practice and procedure should be noted: (1) It is axiomatic that the decision whether or not to allow an appeal is one which lies with the Court. Similarly, even where the parties do not contest an appeal, the question of whether to grant leave to appeal in the first place is one which lies with the Appeal Committee. The parties’ agreement either that leave to appeal should be granted or that an appeal should be allowed does not in any way bind either the Appeal Committee or the Court. (2) On the other hand, in order to avoid or reduce the incidence of costs, the Appeal Committee or the Court (as the case may be) may exercise a discretion to dispense with the need for an oral hearing of either the application for leave to appeal or the appeal itself where the application or appeal is uncontested and it is readily apparent that leave should be granted or the appeal allowed. (3) It is likely that uncontested appeals will be a small proportion of the cases which come before the Court. It would appear that, since Mok Kin Kau’s case was decided in January 2008, there have only been four uncontested appeals.[7] (4) Where a respondent to an application for leave to appeal wishes to consent to the grant of leave whether with or without a view to conceding the appeal if leave is granted, the Registrar should be informed of this by joint letter from the parties upon or as soon as practicable after the filing of the Form B[8] and a consent summons submitted for approval. The respondent’s consent may be forthcoming at that stage, or after the applicant’s written submissions have been filed, or after the respondent’s written submissions have been filed. If the consent involves a change of stance on the part of the respondent from that set out in written submissions opposing leave to appeal, a brief explanation for that change should be given by the respondent or respondent’s solicitor. (5) If the Appeal Committee agrees to grant leave on such consent, it will do so on the papers and without the need for an oral hearing. If the respondent has conceded the application for leave with a view to conceding the appeal, it will direct a Joint Case to be filed by the parties in respect of the appeal. The parties should endeavour to agree any ancillary orders and costs consequent on the appeal being allowed and include such agreement, or state their respective positions on those matters, in the Joint Case. If necessary, further procedural directions prior to the determination of the appeal by the Court itself may be made by a single permanent judge of the Court or the Registrar (as the Court thinks fit). (6) When leave to appeal has been granted in the ordinary way upon an oral hearing of the leave application, should a respondent to an appeal then wish to consent to the appeal being allowed, the parties should by joint letter inform the Registrar of this fact and submit a consent summons containing any agreed directions for approval. This should be done as soon as practicable after the grant of leave but may occur at any stage, for example, following the filing of the appellant’s written case or even following the filing of the respondent’s written case. In any event, a brief explanation for the change of stance should be given by the respondent or respondent’s solicitor. A single permanent judge of the Court or the Registrar (as the Court thinks fit) will give directions for the further conduct of the appeal, which will normally include a direction for the filing of a Joint Case. The parties should endeavour to agree all ancillary orders and any costs consequent on the appeal being allowed and include such agreement, or state their respective positions on those matters, in the Joint Case. (7) If satisfied that the appeal should be allowed, the Court will proceed to determine the appeal on the papers and without an oral hearing, handing down a judgment explaining its decision. Any outstanding ancillary orders or costs consequent on the appeal being allowed will be dealt with on the basis of the parties’ written submissions. The Registrar will notify the parties of the date when judgment will be handed down. The notice of handing down will usually state that it will be unnecessary for counsel to appear but that it is expected that a representative for each party be present in court. In cases involving an appellant on bail, however, the appellant will normally be required to attend. (8) At any stage the Court (or Appeal Committee) may, by direction of a single permanent judge, rescind a direction for the appeal (or application for leave to appeal) to proceed on an uncontested basis without a hearing and direct that the appeal (or leave application, as the case may be) be listed for an inter partes hearing in the normal way. 38. It is anticipated that the above practice and procedure will apply to all future appeals, whether civil or criminal, which are uncontested. There may be exceptional cases in which, notwithstanding the fact the appeal is uncontested, the interests of justice will require an inter partes hearing be held or the Court may, for some other reason, decline to exercise its discretion to dispense with an oral hearing. In such cases, the Court’s usual practice and procedure will apply. Order 39. Accordingly, the Court: (1) Allows the appeal and quashes the appellant’s conviction on each charge; (2) Makes the orders as to costs in paragraph 34 above. Joint Written Submissions by: Mr Edmond Lee SADPP, of the Department of Justice, for the Respondent Ms Margaret Ng, instructed by ONC Lawyers, assigned by the Director of Legal Aid, for the Appellant [1] Mr Tang Siu-hung, Daniel, then sitting as a Deputy Magistrate [2] HCMA 686/2012 [3] Chan Ag CJ, Tang PJ and Bokhary NPJ (FAMC 16/2013) [4] “Judge” refers to the Magistrate and “D1” refers to the appellant [5] English translation of Judgment in HCMA 686/2012, 16 January 2013, at §11 [6] The reference to the Mok Kin Kau procedure being a reference to the case of Mok Kin Kau v HKSAR (2008) 11 HKCFAR 1 [7] Chiu Hoi Po v Commissioner of Police (2009) 12 HKCFAR 597; Mahabobur Rahman v HKSAR (2010) 13 HKCFAR 20; Van Huijstee v HKSAR (2011) 14 HKCFAR 489; and Lai Kwok Fai v HKSAR (2012) 15 HKCFAR 450 [8] Containing the notice of application for leave to appeal: see Hong Kong Court of Final Appeal Rules (Cap.484A), Rule 4 and Schedule 1 
The Court: A. Introduction 1. These appeals arise from the convictions in the Magistrates’ Court of the three appellants for, in the case of the 1st and 3rd appellants, taking part in an unlawful assembly,[1] and, in the case of the 2nd appellant, for inciting others to take part in an unlawful assembly.[2] They were each sentenced to community service orders[3] of various lengths. On the application of the Secretary for Justice for a review of the sentences,[4] the Court of Appeal substituted substantially increased sentences of imprisonment ranging between 6 and 8 months. Leave to appeal to this Court was granted by the Appeal Committee. 2. In its judgment on the review application, the Court of Appeal took the opportunity to provide guidance to sentencing courts in the future regarding the sentences for unlawful assemblies, particularly emphasising the need to take a much stricter view where disorder and any degree of violence was involved. The Court of Appeal, consistent with its responsibilities for providing guidance in sentencing matters, was fully entitled to provide this guidance for the future and accordingly note should be taken of this new approach. Like the Court of Appeal, we specially draw attention to the importance of taking a much stricter view where disorder or violence is involved. Naturally, it will be incumbent on the sentencing court to take into account the extent of the participation or involvement of the convicted person but where disorder or violence is involved, these are serious aggravating features. Hong Kong is on the whole a peaceful society and these elements are to be deterred. 3. The appeals before us raise four particular questions (which we will presently identify) concerning sentencing principles and approach, the most relevant to the outcome for the three appellants being the question of whether the Court of Appeal, in providing guidance for the future in relation to unlawful assemblies, ought to have applied the new approach to these appellants. The other three questions relate to the approach regarding facts and factual findings that should be adopted in a review before the Court of Appeal, the taking into account of motives for the commission of offences such as the present (and in particular the aspect of what is known as civil disobedience) and lastly, the relevance of section 109A of the Criminal Procedure Ordinance (dealing with offenders who are minors). 4. The particular unlawful assembly out of which the convictions arose took place on 26 September 2014, shortly before the mass demonstrations known as “Occupy Central” took place mainly in and around the streets surrounding the Central Government Offices in Admiralty. The convictions and sentences of the three appellants have given rise to widespread publicity and intense, sometimes heated, public discourse. In particular, since the actions leading to the appellants’ convictions arose from the political debate on the content of proposed constitutional reforms of the process for the election of the Chief Executive, strong expressions of opinion have been voiced and feelings on both sides of the debate have run high. 5. It is important to state at the outset of this judgment that it is not the role or function of the courts of the Hong Kong Special Administrative Region (“HKSAR”) to enter into this or any other political debate. Instead, the duty of the courts is, through an independent judiciary, to administer the law of the HKSAR, including the Basic Law, and to adjudicate on the legal issues raised in any case according to the law. In reaching a decision in any given case, a court exclusively applies the applicable legal principles to the relevant facts and thereby reaches a decision on the appropriate disposition of the case, explaining its reasons in its judgment. That is the sole task of this Court in these appeals. B. The background facts 6. The following statement of facts is derived principally from the Reasons for Verdict (“RV”) of the trial magistrate in the Magistrates’ Court. They were not controversial. B.1 The appellants 7. The 1st appellant, Joshua Wong Chi Fung, was born in 1996. He was aged 17 at the time of the offence of which he was convicted, 19 at the time of trial before the magistrate and 20 at the time of the sentence review in the Court of Appeal. He is a politically active young person and was the convenor of a pro-democracy secondary school student movement called Scholarism, which opposed the Government’s proposal to introduce “Moral and National Education” as a compulsory school subject. After Scholarism was dissolved, the 1st appellant and others founded a new political party called Demosistō in April 2016. 8. The 2nd appellant, Nathan Law Kwun Chung, was aged 21 at the time of the offence of which he was convicted and was 23 when convicted and sentenced by the magistrate. He was a standing committee member of the Hong Kong Federation of Students (“HKFS”) and is a politically active young person. Since April 2016, he has been the chairperson of Demosistō. 9. The 3rd appellant, Alex Chow Yong Kang, was aged 24 at the time of the offence of which he was convicted and was 25 when convicted and sentenced by the magistrate. Also politically active, he was the secretary-general of the HKFS. 10. As the magistrate found, each of the appellants came from what she described as grass root or humble families and they each performed well in school. The 1st appellant was a student of the Open University, studying Politics and Public Administration. The 2nd appellant was Chairman of the Student Union of the Lingnan University of Hong Kong, where he was studying a degree course in Cultural Studies. The 3rd appellant was Chairman of the Hong Kong University Student Union and, at the time of the offence, was studying for a bachelor’s degree in Comparative Literature. He has since been accepted for a master’s degree programme at the London School of Economics and Political Science. B.2 The Forecourt of the Central Government Offices 11. The Central Government Offices (“CGO”) at the Tamar site in Admiralty have been in operation since October 2011. In front of the East Wing of the CGO is a forecourt, also known as Civic Square (“the Forecourt”). The Forecourt was previously an area open to the public without any boundary surrounding it and members of the public were permitted, upon prior application, to hold organised assemblies there to demonstrate on political and social issues on Sundays and public holidays. In August 2012, as part of the opposition to the proposed “Moral and National Education” programme, Scholarism organised an assembly in the Forecourt during which police did not clear protesters away. Following the opposition to it, the Government abandoned its plan to introduce the “Moral and National Education” programme. The Forecourt has since then become a place of significance for those who took part in opposing those Government policies. 12. After the anti-national education protests, the Forecourt was closed from July 2014 to early September 2014 during which time a perimeter fence around the boundary of the Forecourt was constructed. This perimeter fence is three metres high and has three access gates, Gates 1 to 3. Construction was completed and the Forecourt re-opened on 10 September 2014 but with restricted access in accordance with arrangements set out in notices displayed on Gates 1 to 3. The notices stated that the gates would be closed from 11pm to 6am and that members of the public could not enter, except for those with permission from the CGO and the Legislative Council (“Legco”). In special circumstances, such as when there was a danger to security, the Forecourt would also be closed and only those with permission could enter. Members of the public would require permission from the Director of Administration if they wished to assemble in the Forecourt. Such application could only be made in relation to specific times, between 10am and 6.30pm, on Sundays and public holidays. Mondays to Saturdays were excluded. 13. On the day of the offences, 26 September 2014, which was a Friday, the gates to the Forecourt were closed. B.3 The assembly in Tim Mei Avenue on 26 September 2014 14. In September 2014, critics of the Government’s proposed constitutional reforms of the process for election of the Chief Executive were expressing opposition to those proposals through various means. The HKFS wished to express their opposition to the proposals and had applied on two occasions to the Department of Administration to open the Forecourt for the purposes of public activities between 23 September and early October 2014 but both applications were refused. 15. On 26 September 2014, Scholarism and the HKFS held an assembly in an area off Tim Mei Avenue outside the Forecourt in relation to the proposed constitutional reforms. Prior to holding the assembly, a Notice of No Objection[5] had been obtained from the police and it was valid until 10pm that day. For security reasons, the gates to the Forecourt were closed, security guards were on duty inside and outside the gates and Mills barriers had been erected outside. A stage had been set up in the area where the assembly was being held and the 1st appellant and others addressed those in attendance from the stage by means of a microphone and amplification system. Several hundred people were in attendance. 16. Before the conclusion of the assembly, between about 6pm and 8pm on 26 September 2014, the HKFS held a meeting attended by the three appellants. A discussion took place regarding how those attending the assembly might enter the Forecourt and it was proposed that protesters should seek to enter the Forecourt through one of the gates when a member of staff or a reporter was entering. The possibility of criminal liability was discussed and copies of a document entitled “Points to Note When Under Arrest” which gave telephone numbers for legal assistance were distributed. B.4 The commission of the offences 17. At about 10.20pm on the evening of 26 September 2014, when participants in the assembly were beginning to leave, the 1st appellant was on the stage addressing the crowd. He appealed to them to stay and called on them to go into the Forecourt, saying “Now, here we call on you, we hope you all enter the Civic Square together with us now.” At that point, he left the stage and ran towards the Forecourt. Many people were gathered at Gate 2 in front of the Forecourt. Security guards were trying to stop them from pushing the gate open. Others had already decided to enter by scaling the fence. About three minutes later, the 1st appellant climbed over the fence and jumped down into the Forecourt, ignoring police officers who shouted to him to stop. The 1st appellant landed on the ground in front of Police Sergeant 52877, Yam Ho Chung, and was immediately arrested. 18. One minute after that, the 3rd appellant also climbed over the fence and ignored the police who were shouting at him to stop. He evaded the police and entered the Forecourt and ran towards the area near Gate 2 where a throng was pushing to gain entry into the Forecourt. Some of the protesters were able to stop the guards from closing the gate and, due to their superiority of numbers, the protesters succeeded in pushing the gate open and entering the Forecourt. Only after police reinforcements arrived and several dozen protesters had entered the Forecourt was the gate closed. 19. In the meantime, after the 1st appellant left the stage at Tim Mei Avenue, the 2nd appellant took over from him as master of ceremonies and appealed to the crowd to enter the Forecourt together. He provided a commentary on the events taking place at the Forecourt, including the intervention of the police and the clashes occurring at the scene. As recorded in the judgment of Poon JA in the Court of Appeal[6] (at [36]), the 2nd appellant said the following things at that time: (1) “I call on everybody to go into the Civic Square together. Now a spearhead team has begun to charge into the Civic Square!” (2) “We have lifted up a door. If you people are coming up from LegCo, just turn right after coming up from LegCo and that’s the main entrance of the Civic Square.” (3) “Everybody, go into the Civic Square now. Let us recapture the public space which belongs to us.” (4) “Every time you close it down, we bust it!” (5) “… Tonight, we shall enter and station in the Civic Square…” (6) “Recapture the Civic Square!” (7) “We shall recapture the Square that belongs to us by using a method of non-active attack” (8) “Give us back the Civic Square!” (9) “Surround the Central Government Offices!” (10) “Pepper spray is now being used at the scene.” (11) “The police have now deployed three lines of human chain hoping to stop us from entering the Civic Square.” (12) “Encircle the police in return.” (13) “The police are doing political oppression. Shame on them!” (14) “Everybody, please use Facebook to keep appealing for more reinforcement”, “Everybody, please use Facebook, all social media to call on others to come to Civic Square to support us; recapture the Square which belongs to the people.” (15) “Some friends have got injured at the scene. We are very sorry for that. This action of ours comes hastily. No information could be leaked out…” (16) “30 odd of our demonstrators are now encircled by the police. Inside this circle, they are brutally treated and there are massive clashes.” (17) “Some of our friends have been arrested ...” (18) “The convenor of Scholarism, Wong Chi-fung, has been arrested, and was accused of assaulting a police officer.” (19) “Someone is suffering from heart attack but the police do not allow the ambulance to come in”, “Now (a) nurse/s has/have reached the gate but the police refuse his/her/their entry. Let the people go, open the gate!” 20. The 2nd appellant asked students in school uniform, secondary school students and those underage to leave but asked them to go online to call more people to come. He told the participants to be peaceful and sensible and to exercise restraint, asking them to raise their arms in the air. He warned them to be prepared to be arrested and prosecuted. A telephone number was given to participants to which they could send their personal information if arrested. 21. Several hundred people tried to enter the Forecourt and, ultimately, several dozen successfully did so. Some of them pushed over the Mills barriers placed at the bottom of the flagpole at the centre of the Forecourt. Subsequently, those participants who were left, including the 3rd appellant, gathered under the flagpole, hand-in-hand, and chanted slogans. The period of time that elapsed from when the 1st appellant appealed to the crowd to enter the Forecourt until the time the participants who managed to enter the Forecourt gathered under the flagpole was about 12 minutes. 22. As a result of the crowd’s attempts to gain entry to the Forecourt, either by pushing through the gate or jumping off the fence, 10 of the security guards on duty, who tried to prevent them doing so, were injured. Most of the injuries were minor, such as tenderness, bruising and swelling. However, one security guard, Chan Kei Lun, had bruising and swelling of his left big toe and a mild fracture near the base of the phalanx of his toes.[7] He gave evidence of being pushed from behind and injuring his left elbow and left big toe. Of the 10 injured security guards, 5 had to take sick leave for between 4 and 6 days and Mr Chan had to take sick leave for 39 days. C. The proceedings against the appellants C.1 The charges 23. As a result of the events described in Section B.4 above: (1) The 1st appellant faced two charges: Charge (1) – inciting others to take part in an unlawful assembly, contrary to section 18 of the Public Order Ordinance[8] and section 101I of the Criminal Procedure Ordinance;[9] and Charge (2) – taking part in an unlawful assembly, contrary to section 18 of the Public Order Ordinance. (2) The 2nd appellant was charged with inciting others to take part in an unlawful assembly (Charge (3)). (3) The 3rd appellant was charged with taking part in an unlawful assembly (Charge (4)). 24. The particulars of Charges (1) and (3) were materially the same and alleged that the 1st appellant, on 26 September 2014, and the 2nd appellant, between 26 and 27 September 2014, in Hong Kong: “… unlawfully incited other persons to take part in an unlawful assembly by assembling together, conducting themselves in a disorderly, intimidating, insulting or provocative manner intended or likely to cause any person reasonably to fear that the persons so assembled would commit a breach of the peace, or would by such conduct provoke other persons to commit a breach of the peace.” 25. The particulars of Charges (2) and (4) were, similarly, in materially the same terms and alleged that the 1st appellant, on 26 September 2014, and the 3rd appellant, between 26 and 27 September 2014, in Hong Kong: “… and other persons, took part in an unlawful assembly in that they, assembled together, conducted themselves in a disorderly, intimidating, insulting or provocative manner intended or likely to cause any person reasonably to fear that the persons so assembled would commit a breach of the peace, or would by such conduct provoke other persons to commit a breach of the peace.” C.2 The trial before the magistrate 26. The appellants each pleaded not guilty to the respective charges against them and were tried before Ms June Cheung Tin Ngan, sitting in the Eastern Magistracy on 29 February, 1 to 4 March and 13 May 2016.[10] On 21 July 2016, the magistrate delivered her RV and acquitted the 1st appellant on Charge (1) but convicted him on Charge (2) and convicted the 2nd appellant on Charge (3) and the 3rd appellant on Charge (4). 27. Since they are central to one of the issues raised in the appeals before this Court, it is necessary to summarise at some length the findings made by the magistrate in her RV concerning the commission of the various offences by the respective appellants and, in the case of the 1st appellant, his acquittal on the charge of incitement. 28. As to the 1st appellant’s acquittal on Charge (1), the magistrate was of the view that the 1st appellant’s words, and his leaving the stage after addressing the crowd, amounted to encouraging, persuading and advising those assembled to go to the Forecourt together with him.[11] However, she believed that at the moment the 1st appellant made the call to the crowd to follow him, this was at an early stage of the action and that, when he was still on the stage, he might not have been able to see the situation at the gate or the Forecourt, and so he might have believed that the security guards would not actively stop people from entering the Forecourt.[12] She found there was no evidence that, during the meeting in the early evening and when making the call on the stage, the 1st appellant already knew that the security guards would actually block people from entering and that the participants would insist on doing so by pushing the gates or scaling the fence.[13] Accordingly, she concluded that, regarding the 1st appellant’s conduct when he was on the stage encouraging others to enter the Forecourt, the prosecution had not proved this charge beyond reasonable doubt.[14] 29. In relation to the 2nd appellant’s conviction on the charge of incitement (Charge (3)), the magistrate found that the content of his speech on the stage involved calling on people to enter the Forecourt, announcing that the police had intervened and that there were conflicts and injuries at the scene, and recognising that the action involved potential legal liabilities and that some people had been arrested.[15] She found that the act of making calls on the stage and the content of his speech obviously amounted to encouraging, persuading or advising participants of the assembly to enter the Forecourt together with the “spearhead team” and that he intended to encourage, persuade or advise the assembly participants to enter the Forecourt together.[16] She considered the key point to be whether, when the 2nd appellant made the call on stage, the security guards or police had opened the gate or imposed any physical intervention on entry into the Forecourt.[17] On this point, she found that, soon after the crowd entered the Forecourt, the 2nd appellant had already been informed that the police had intervened to stop the crowd from entering the Forecourt.[18] 30. On this basis, the magistrate held that “the only way for the participants of the assembly to force themselves into the Forecourt … was by pushing the gate and scaling the fences” which “would obviously be disorderly conduct … and intimidating conduct”.[19] Hence, she found that when the 2nd appellant made his announcements on stage, he knew that those people entering the Forecourt were being opposed by the police so that if they heeded his calls to enter the Forecourt, “physical bumps” were certain and also possibly physical injuries. Their actions were likely to cause the police and security guards to fear that the persons assembled would commit a breach of the peace or reasonably to fear that the persons so assembled would provoke others to commit a breach of the peace.[20] 31. In relation to the conviction of the 1st appellant on Charge (2), for taking part in an unlawful assembly, the magistrate found that, when he passed Gate 2 he saw a large crowd of people and the situation was “chaotic”. The people were stopped at the gate and could not enter and some had started to scale the fence to jump into the Forecourt.[21] She found that the 1st appellant knew he would be stopped by security guards or police from entering the Forecourt, yet chose to ignore them and instead scale the fence to gain entry and that this conduct was obviously disorderly and intimidating.[22] The magistrate noted that the 1st appellant accepted that his act in jumping from a 3-metre tall fence was a risky act and found that his conduct in scaling the fence, and that of others pushing at the gate, “would definitely cause police officers and/or security guards, who were attempting to stop them, reasonably to fear injury by those who jumped off from the fences, causing a breach of the peace.”[23] She also found that police officers or security guards at the scene would reasonably fear that other people would be provoked, by the actions of the 1st appellant and others trying to push their way into the Forecourt, to try to push open the gate forcefully and to scale the fence to enter the Forecourt. This, she found, would cause injuries to the police officers or security guards leading to a breach of the peace.[24] 32. In relation to the conviction of the 3rd appellant on Charge (4) for taking part in an unlawful assembly, the magistrate found that, about one minute after the 1st appellant climbed onto the fence, the 3rd appellant also scaled the same fence to enter the Forecourt.[25] She found that, like the 1st appellant in respect of Charge (2), the 3rd appellant and those who climbed the fence, as well as those who tried to push their way in through the gate, acted together with a common purpose, namely “to use violence and enter the Forecourt [in a] disorderly [way] despite the interventions of the security guards and/or the police.”[26] The magistrate found that the 3rd appellant’s action in jumping off the fence was likely to injure the police officers next to the fence in the Forecourt and that the actions of the 3rd appellant and the others climbing the fence and pushing the gate would cause police officers reasonably to fear that they would be injured by the 3rd appellant, causing a breach of the peace.[27] Similarly, as with the 1st appellant, the police officers and security guards would reasonably fear a breach of the peace as a result of others being provoked by the 3rd appellant’s actions.[28] C.3 The sentences imposed by the magistrate 33. Having convicted the appellants, the magistrate adjourned for Community Service Order Suitability Reports to be obtained and, on 15 August 2016, she sentenced the 1st appellant to an 80-hour community service order and the 2nd appellant to a 120-hour community service order.[29] She considered that an 80-hour community service order would also be appropriate for the 3rd appellant but, since he was due to commence a full-time master’s degree programme in London in September 2016, she considered that, in substitution for a community service order, the 3rd appellant should be subject to a term of imprisonment of 3 weeks, suspended for 1 year. 34. In so sentencing the appellants, the magistrate: (1) Took account of the fact that the appellants were first time offenders who had displayed concern for societal problems, were passionate about politics and were willing to take action to realise their ideals (RS at [1]); (2) Noted that the appellants’ enthusiastic participation in student movements had received the understanding and support of their families (RS at [1]); (3) Held that the Court ought to consider the appellants’ motives and purposes in committing the offences along with the seriousness of the offences (RS at [2]); (4) Accepted that the appellants were genuinely expressing their opinions and grievances out of their political convictions and concern for society and that their purpose and motives were not for their own profit or to injure others (RS at [3]); (5) Noted that the case took place before Occupy Central and the more intense political incidents that followed, that the appellants’ actions were far more moderate than the incidents that followed, that the videos showed that the appellants consistently advocated that the action taken must be peaceful, rational and non-violent and that there was no evidence the appellants caused the injuries that were sustained or intended that others would be injured and expressed regrets about those injuries (RS at [6]); (6) Considered that the appellants’ actions were reckless but not very violent or intended to injure security personnel or police officers, that they only wanted to enter the Forecourt, a place of symbolic meaning, to form a circle and chant slogans, and did not consider the appellants and other participants to be of a very violent type (RS at [7]); (7) Concluded, based on the facts found after trial, that the main impugned conduct of the 1st and 3rd appellants was their climbing over the fence to enter the Forecourt and that the 2nd appellant was more culpable than them because he had incited others to act in a disorderly manner, although he had continually reminded those present to remain orderly, be mindful of their personal safety and that taking part carried risks of legal liability (RS at [8]); (8) Noted that the 2nd appellant expressed regret that security personnel had been injured and that all three appellants cooperated during the process of arrest, investigation and trial, each of them expressing the probation officer that they were willing to bear the legal consequences and willing to be subject to community service orders (RS at [9]-[11]); (9) Considered all the circumstances of the case, including the probation officer’s view that probation orders were not appropriate and his recommendation of community service orders, and concluded that the sentences she imposed (set out in the preceding paragraph) were appropriate (RS at [12]-[14]). 35. Following the magistrate’s sentencing, the Secretary for Justice applied to the magistrate for a review of the sentences pursuant to section 104 of the Magistrates Ordinance.[30] On 21 September 2016, having considered the written and oral submissions of the prosecution and defence, the magistrate was not persuaded to vary the original sentences she had imposed and refused the application for review.[31] The magistrate reiterated that, when imposing the sentences, she had carefully taken into account (i) the circumstances of the offence and each of the appellants’ criminal acts, (ii) the consequence of the appellants’ criminal acts, (iii) the appellants’ motives for committing the offences, and (iv) the background of and display of remorse by each of the appellants.[32] C.4 The application by the Secretary for Justice to the Court of Appeal for review 36. Contending that the sentences imposed were manifestly inadequate and/or wrong in principle, the Secretary for Justice applied to the Court of Appeal, pursuant to section 81A of the Criminal Procedure Ordinance, for a review of sentence. It was contended that the sentences imposed failed to reflect the seriousness, and in particular the public order nature, of the offence of unlawful assembly. It was also contended that the sentences imposed failed properly to reflect the culpabilities of the appellants and that the magistrate imposed sentences which were unduly lenient. 37. The Secretary for Justice’s application was heard by the Court of Appeal[33] on 9 August 2017. By their judgment dated 17 August 2017, the Court of Appeal allowed the application, set aside the magistrate’s sentences and imposed terms of imprisonment of: 6 months for the 1st appellant in respect of Charge (2), 8 months for the 2nd appellant in respect of Charge (3), and 7 months for the 3rd appellant in respect of Charge (4).[34] 38. The main judgment was given by Poon JA, with whom both Yeung VP and Pang JA agreed. After examining the sentencing principles applicable to cases of unlawful assembly (which will be discussed further below), and recognising that the Court of Appeal may accede to an application to review and increase a sentence imposed by a lower court only where that sentence is wrong in principle or manifestly inadequate, Poon JA summarised the reasons why, in his view, the facts of this case were serious and that it constituted “a large-scale unlawful assembly, involving violence”,[35] namely (in summary): (1) The appellants’ decision to enter the Forecourt at the meeting before the incident showed their actions were not spontaneous or unpremeditated. This was reflected in the distribution to participants of the document “Points to Note When Under Arrest”.[36] (2) During the meeting, it was within the appellants’ reasonable expectation that, in entering the Forecourt, there was a serious risk of the crowd clashing with the security guards and police officers, from which clashes violence would inevitably arise.[37] (3) Once the action started, the appellants must have realised that the security guards and police were stopping protesters from forcing their way into the Forecourt and there were clashes, yet they persisted in their acts.[38] (4) Several hundred people took part in the unlawful assembly and several dozen managed to force their way into the Forecourt over a period of about 12 minutes, which was not a short period of time.[39] (5) 10 security guards were injured in the incident, most injuries being relatively minor but one being more serious, so that the degree of violence could not be described as slight.[40] (6) The appellants and other participants did not have an absolute right to enter the Forecourt and knew that their act of forcing their way in was illegal.[41] (7) The appellants, as student leaders of the protesters, were planners of the incident and took leading roles in it; the 1st appellant called on, and the 2nd appellant incited, others to join in and these were highly irresponsible acts.[42] (8) What the 2nd appellant did was highly dangerous, making use of sensational and even unfounded, inflammatory, language to incite the crowd.[43] 39. Poon JA considered that the appellants’ attitudes after their convictions showed that they had no genuine remorse for the offences they had committed.[44] He considered that the seriousness of the facts required the court to place more weight on the two sentencing factors of punishment and deterrence, and correspondingly less weight on their personal circumstances, motives and the sentencing factor of rehabilitation. Hence, he concluded that the appropriate sentences must be immediate custodial sentences.[45] In so concluding, despite the 1st appellant being 17 at the time of the offence and 20 at the time of the appeal, Poon JA noted that his counsel, Mr Randy Shek,[46] agreed that it was unnecessary for the court to consider other sentencing options. 40. Poon JA held that the magistrate erred in principle in passing the sentences she did in that (in summary): (1) The magistrate did not consider the factor of deterrence in the sentences but gave disproportionate weight to the appellants’ personal circumstances and motives.[47] (2) In regarding the case as not involving serious violence, the magistrate overlooked that this was an unlawful assembly on a large scale and there was a risk of violent clashes.[48] (3) The magistrate overlooked the fact that the appellants must have been reasonably able to envisage clashes between the participants and the security guards and police and that it was inevitable that at least some security guards would be injured.[49] (4) In taking into account the appellants’ desire to enter the Forecourt as a place of historical significance to protest, the magistrate overlooked the fact that, on the night in question, Scholarism and the HKFS had already held an assembly on the road outside the CGO and the Forecourt was closed and their forcing their way in was unlawful.[50] (5) The magistrate gave too much weight to the appellants’ alleged remorse in that, whilst they did not deny what they had done and indicated they respected the court and were willing to bear the legal consequences, their remorse was superficial and should not be given too much weight.[51] 41. Poon JA therefore concluded that it was wrong in principle for the magistrate to have imposed community service orders or a suspended sentence and that these sentences were manifestly inadequate so that the Court of Appeal had to interfere.[52] He determined that the appropriate starting points to be imprisonment of: 8 months for the 1st appellant on Charge (2); 10 months for the 2nd appellant on Charge (3); and 8 months for the 3rd appellant on Charge (4). To these starting points, he applied a discretionary discount of one month because the sentences came about by reason of an application for review by the Secretary for Justice and a further discretionary discount of one month for the 1st and 2nd appellants because they had each completed their community service orders. Accordingly, the sentences of imprisonment imposed were 6 months for the 1st appellant, 8 months for the 2nd appellant and 7 months for the 3rd appellant.[53] C.5 The grant of leave to appeal to the Court of Final Appeal 42. Following the review of sentence by the Court of Appeal, the appellants applied to this Court[54] for leave to appeal on the ground that substantial and grave injustice had been done and the 2nd and 3rd appellants also applied to the Court of Appeal for certification that points of law of great and general importance were involved in the CA Judgment. By its judgment dated 26 October 2017,[55] the Court of Appeal refused to so certify. 43. By further applications to this Court, the appellants sought leave to appeal on the additional ground that points of law of great and general importance were involved in the proposed appeal and by its Determination dated 7 November 2017, the Appeal Committee granted leave to appeal to the appellants on both the point of law ground and on the substantial and grave injustice ground.[56] The issues for which leave to appeal was granted were (in respect of all appellants): “(1) To what extent can the Court of Appeal on an application for review of sentence under s.81A of the Criminal Procedure Ordinance, Cap.221 reverse, modify, substitute or supplement the factual basis on which the original sentence was based? (2) To what extent should a sentencing court take into account the motives of a defendant in committing the crime of which he or she has been convicted, particularly in cases where it is asserted that the crime was committed as an act of civil disobedience or in the exercise of a constitutional right? (3) Insofar as the Court of Appeal was seeking to do so at all, in arriving at the appropriate sentences for the applicants, to what extent ought the Court of Appeal have made allowance for the assertion made by them that guidelines to sentencing courts for the future were being given?” In respect of the 1st appellant alone, leave was further granted in respect of the following issue, namely: “(4) To what extent should the Court of Appeal have taken into account s.109A of the Criminal Procedure Ordinance, Cap.221?” 44. Prior to the hearing before the Appeal Committee, on 24 October 2017, the 1st and 2nd appellants were granted bail pending the determination of their applications for leave to appeal[57] and, upon the grant of leave to appeal, bail was extended to them pending the determination of their appeals to this Court. When the Appeal Committee granted leave to appeal on 7 November 2017, the hearing of the appeal was fixed for 16 January 2018. At that leave hearing before the Appeal Committee on 7 November 2017, the 3rd appellant then also sought bail pending the determination of his appeal and this was granted. D. Review of sentence under section 81A of the Criminal Procedure Ordinance 45. Section 81A of the Criminal Procedure Ordinance[58] provides: “(1) The Secretary for Justice may, with the leave of the Court of Appeal, apply to the Court of Appeal for the review of any sentence (other than a sentence which is fixed by law) passed by any court, other than the Court of Appeal, on the grounds that the sentence is not authorized by law, is wrong in principle or is manifestly excessive or manifestly inadequate.” And section 81B provides: “(1) Upon the hearing of the application the Court of Appeal may, by order – (a) if it thinks that the sentence was not authorized by law, was wrong in principle or was manifestly excessive or manifestly inadequate, quash the sentence passed by the court and pass such other sentence (whether more or less severe) warranted in law in substitution therefor as it thinks ought to have been passed; (b) in any other case, refuse to alter the sentence.” 46. These provisions were introduced into Part IV of the principal ordinance[59] by the Criminal Procedure (Amendment) Ordinance in 1972 and, as the speech of the then Attorney General moving the bill shows, their purpose was to maintain uniformity of sentence and to avoid the injustice caused by substantial disparity in sentences imposed by different magistrates and judges dealing with similar offences.[60] A similar review of sentence jurisdiction exists in England and Wales under section 36 of the Criminal Justice Act 1988. 47. As the wording of section 81A shows, the grounds on which the Court of Appeal may interfere with a sentence passed by a lower court are restricted to the four grounds specified, namely that it is (i) not authorised by law, (ii) wrong in principle, (iii) manifestly excessive, or (iv) manifestly inadequate. If any of these grounds exists, the original sentence may be set aside, increased or reduced. Plainly, therefore, a review of sentence may be either advantageous or disadvantageous to the respondent to the application. 48. But the power to review a sentence on the application of the Secretary for Justice, in particular to seek an increase in sentence, does not confer on the Secretary for Justice an analogous right to that of a convicted person appealing against sentence.[61] As Rigby CJ held in Re Applications for Review of Sentences: “For my part, I am satisfied that there is a substantial distinction and I adhere to the view that … a far more stringent test should be applied by the Full Court in considering an application by the Attorney General for leave to apply for review of sentence on the grounds of manifest inadequacy than the test required or imposed by the same court in an appeal by a convicted person against sentence.”[62] 49. Procedurally, too, a review of sentence proceeds differently to an appeal by a convicted person whether against conviction or sentence.[63] An application for review of sentence requires the leave of the Court of Appeal and must be signed by the Secretary for Justice and accompanied by the documents specified in section 81A(2A). These are limited to the formal statement of the findings of fact (before a magistrate) or reasons for verdict (before a District Judge) and reasons for sentence or, in the case of a sentence imposed by a judge of the High Court, “the record of the whole of the proceedings before him other than the evidence given in any trial that took place in those proceedings”, together with the sentencing reports in relation to the convicted person. Section 81C provides that the review may not be heard whilst there is a pending appeal by the convicted person against his conviction or a pending application by him under sections 104 or 105 of the Magistrates Ordinance.[64] 50. The ambit of the power of review of sentence is subject also, as a matter of legal context, to the constitutionally protected right against double jeopardy. In this regard, Article 11(6) of the Hong Kong Bill of Rights[65] provides that: “No one shall be liable to be tried or punished again for an offence for which he has already been finally convicted or acquitted in accordance with the law and penal procedure of Hong Kong.” It has not, however, been suggested in these appeals that the power of review in section 81A is unconstitutional or that, in an appropriate case, a convicted person who has not himself appealed or who may have completed a sentence later found to be wrong in principle or manifestly inadequate may not be subject to an increased sentence upon review by the Court of Appeal. 51. Nevertheless, whilst an increase of sentence is not incompatible with the concept of double jeopardy, as a matter of practice the Court of Appeal customarily makes allowance for the concept when increasing a sentence on review by applying a discount to the increased sentence: see Secretary for Justice v Lo King Fat [2016] 2 HKC 230 at [109]-[116]; Attorney-General’s Reference Nos.90 & 91 of 2003 [2004] EWCA Crim 1839 at [11]. In particular, where a sentence, such as a community service order, has been completed prior to an increase of sentence on review, the reviewing court typically applies a discount for that factor in arriving at the new sentence to be imposed: see HKSAR v Chan Pak Hoe (2012) 15 HKCFAR 244 at [52]; Secretary for Justice v Buk Chui Ying [2008] 5 HKLRD 185 at [26]. D.1 The extent of the Court of Appeal’s power to review the facts on a review of sentence 52. The Court of Appeal, whether engaged in an appeal against sentence or a review of sentence, functions as a court of review and does not conduct a sentencing exercise of its own, so it ordinarily relies entirely, or almost entirely, on material before the sentencing court: see R v A and B [1999] 1 Cr. App. R. (S.) 52 per Bingham LCJ (as he then was) at p.56 (sub-para.(4)). It has been repeatedly recognised that an appellate court does not have the advantages of a trial court in having received the evidence at first hand and that this provides a strong rationale for limiting the basis on which it can interfere with primary findings of fact made by the trial court: see, e.g. Ting Kwok Keung v Tam Dick Yuen & Others (2002) 5 HKCFAR 336 at [35]. 53. Subject to those general considerations, there is no express limitation in section 81A on the matters which the Court of Appeal may examine in hearing an application for review of sentence. The limits of the review are defined by the statutory language which imposes threshold conditions for the exercise of the jurisdiction to substitute a different sentence, be it more or less severe. As noted above, those conditions are that the original sentence is (i) not authorised by law, (ii) wrong in principle, (iii) manifestly excessive, or (iv) manifestly inadequate. Unlike an appeal, which may proceed by way of rehearing, a review of sentence is limited to an examination of whether any of those four grounds of review are made out. If made out, only then is the Court of Appeal’s discretion to vary the sentence or refuse the application engaged. 54. Although, procedurally, an application for review of sentence is accompanied by the documents listed in section 81A(2A), these are not the only documents that may be examined by the Court of Appeal in a review. The list of documents in section 81A(2A) is specified “for the purpose of subsection (2)(b), which simply provides that “[a]n application under subsection (1) shall … (b) be accompanied by the documents, or copies of the documents, specified in subsection (2A)”. 55. The specification of the documents to be filed by the Secretary for Justice on an application for review of sentence in section 81A(2A) was introduced into the Criminal Procedure Ordinance by amendment in 1979.[66] Prior to that amendment, the record of the proceedings in the case was required. As the remarks of the Attorney General when moving the amending legislation show, the limitation on the documents to be filed was not to restrict the material which the Court of Appeal was entitled to review but rather to save time and costs since experience had shown that the Court of Appeal did not need the whole of the record.[67] However, section 81A(2A) does not suggest that in a review of sentence there is a bar to the Court of Appeal considering any evidence or exhibit duly proved, admitted or adduced in the sentencing court. 56. Section 81B deals with the review of sentence itself and provides in section 81B(3) that, for the purposes of the section, the Court of Appeal “may exercise any of the powers conferred by section 83V.” Section 83V contains, for the purposes of Part IV of the Criminal Procedure Ordinance, provisions relating to evidence and is in very wide terms as to the type of evidence that can be received by the Court of Appeal, even including new evidence. But section 83V(5) provides that, “In no case shall any sentence be increased by reason of or in consideration of any evidence which was not given at the trial.” Whilst that provision restricts the consideration of new evidence in a review of sentence by the Court of Appeal, it recognises by implication that the sentence may be increased by reference to evidence that was given at the trial. 57. That the Court of Appeal is permitted to have regard to all the evidence that was available to the sentencing court is also reinforced by reference to section 83T, relating to the preparation of a case for hearing by the Court of Appeal, which provides by subsection (1)(b) that the Registrar of the High Court shall “obtain and lay before the Court of Appeal in proper form all documents, exhibits and other things which appear necessary for the proper determination of the appeal or application.” 58. Textbook commentaries on the powers of the Court of Appeal in a review are also consistent with the proposition that the evidence available in the court below can be looked at: see Taylor on Criminal Appeals (2nd Ed.) at [13.47] and Archbold Hong Kong (2018 Ed.) at [7-445]. 59. It is plainly appropriate that the Court of Appeal should be able, in determining whether the sentencing court has committed any of the errors which permit interference under section 81A (erred in principle etc.), to look at any relevant evidence available to the court below. If the court below has made an error as to the facts on which it proceeds to sentence, it is only right that the Court of Appeal can correct these: see e.g. Attorney-General’s Reference Nos. 90 & 91 of 2003 [2004] EWCA Crim 1839 at [10], where the English Court of Appeal substituted a higher sentence for an unduly lenient one which had been imposed on the mistaken assumption of the judge that the offender’s plea was on the basis he believed the drugs of which he was charged with possession were amphetamine rather than ecstasy, which belief was not accepted by the prosecution. Moreover, the review mechanism can be for the benefit of the convicted person and not just to his disadvantage and, if an error of fact has been made by the sentencing court, it is right that the Court of Appeal should be able to take the correct fact into account in reviewing the sentence imposed. 60. The correct approach in principle is succinctly stated in the joint judgment of Dixon J, Evatt J and McTiernan J in the High Court of Australia in House v The King (1936) 55 CLR 499 at p.505: “If the judge acts upon a wrong principle, if he allows extraneous or irrelevant matters to guide or affect him, if he mistakes the facts, if he does not take into account some material consideration, then his determination should be reviewed and the appellate court may exercise its own discretion in substitution for his if it has the materials for doing so. It may not appear how the primary judge has reached the result embodied in his order, but, if upon the facts it is unreasonable or plainly unjust, the appellate court may infer that in some way there has been a failure properly to exercise the discretion which the law reposes in the court of first instance. In such a case, although the nature of the error may not be discoverable, the exercise of the discretion is reviewed on the ground that a substantial wrong has in fact occurred.” 61. However, the Secretary for Justice is not permitted to rely on a factual basis different to that accepted by the prosecution at trial or to re-open the way in which a case has been put by the prosecution at trial and thereby to require the Court of Appeal to enquire into facts which have not been pursued: see Attorney General v Li Ah-sang [1995] 2 HKCLR 239. There, a magistrate passed sentence for the offence of employing persons not lawfully employable on the basis that the employees were not illegal immigrants. The prosecution did not object to this nor allege that the employees were in fact illegal immigrants. Upon the review of sentence, on the basis that the employees were illegal immigrants, the Court of Appeal held that it was not open to the prosecution to rely on that basis which was a different factual basis to that accepted at trial. A similar approach is taken in England and Wales: see Attorney-General’s Reference No.95 of 1998 (R v Highfield), Times LR, 21 April 1999; Attorney General’s References Nos. 114-116 and 144-5 of 2002 [2003] EWCA Crim 3374 at [27]-[28]. 62. It is also not open to the Court of Appeal in a review of sentence to ascribe a different weight to a factor properly taken into account by the sentencing judge in arriving at a sentence that is otherwise within the range of sentences appropriate for the offence. If the judge has failed to take a relevant matter into account or has taken into account an irrelevant factor, that is an error of principle. However, the relative weight the sentencing judge ascribes to each relevant factor is a matter within the judge’s discretion and, unless that exercise results in the imposition of a sentence that is manifestly inadequate, the relative weight attributed to each individual relevant factor is a matter for the judge. Save where it concludes that the sentence is manifestly inadequate, the Court of Appeal is not entitled to ascribe more or less weight to a relevant factor than did the sentencing court. 63. As Lane LCJ held in Attorney-General’s Reference (No.4 of 1989) [1990] 1 WLR 41 at pp.46A-C: “A sentence is unduly lenient, we would hold, where it falls outside the range of sentences which the judge, applying his mind to all the relevant factors, could reasonably consider appropriate. In that connection regard must of course be had to reported cases, and in particular to the guidance given by this court from time to time in the so-called guideline cases. However it must always be remembered that sentencing is an art rather than a science; that the trial judge is particularly well placed to assess the weight to be given to various competing considerations; and that leniency is not in itself a vice. That mercy should season justice is a proposition as soundly based in law as it is in literature.” Although a decision under section 36 of the Criminal Justice Act 1988, where the threshold for review is that the sentence was “unduly lenient”, the same considerations apply in the context of the inquiry under section 81A as to whether a sentence is “manifestly inadequate”. Unless the sentence imposed is manifestly inadequate or excessive, the Court of Appeal may not review the sentence on the basis that, had it been the sentencing court, it would have ascribed a different weight to a particular relevant factor. This limitation is simply a reflection of the statutory grounds for review of sentence laid down in section 81A. D.2 Civil disobedience and exercise of constitutional rights as motive 64. An offender’s motive for committing an offence is a relevant factor in a court’s decision as to the appropriate sentence to be imposed and can be a relevant mitigating factor.[68] As Spigelman CJ said in R v Swan [2006] NSWCCA 47 at [61]: “Motive is always a relevant factor. It affects the moral culpability of the offender, the weight to be given to personal deterrence and may affect the weight to be given to general deterrence.” 65. In the present cases, it is contended that it is relevant, as a mitigating factor, that the appellants committed the offences of which they have been convicted as acts of civil disobedience and in the exercise of their constitutional rights to freedom of expression and freedom of assembly under Article 27 of the Basic Law and Articles 16 and 17 of the Hong Kong Bill of Rights. 66. Article 27 of the Basic Law provides: “Hong Kong residents shall have freedom of speech, of the press and of publication; freedom of association, of assembly, of procession and of demonstration; and the right and freedom to form and join trade unions, and to strike.” And Articles 16 and 17 of the Hong Kong Bill of Rights[69] provide: “Article 16 Freedom of opinion and expression (1) Everyone shall have the right to hold opinions without interference. (2) Everyone shall have the right to freedom of expression; this right shall include freedom to seek, receive and impart information and ideas of all kinds, regardless of frontiers, either orally, in writing or in print, in the form of art, or through any other media of his choice. (3) The exercise of the rights provided for in paragraph (2) of this article carries with it special duties and responsibilities. It may therefore be subject to certain restrictions, but these shall only be such as are provided by law and are necessary– (a) for respect of the rights or reputations of others; or (b) for the protection of national security or of public order (ordre public), or of public health or morals. Article 17 Right of peaceful assembly The right of peaceful assembly shall be recognized. No restrictions may be placed on the exercise of this right other than those imposed in conformity with the law and which are necessary in a democratic society in the interests of national security or public safety, public order (ordre public), the protection of public health or morals or the protection of the rights and freedoms of others.” 67. Although they are not absolute rights and lawful restrictions may be placed on their exercise in the interest of public order and for the protection of the rights and freedoms of others,[70] the importance of these rights has been repeatedly recognised in previous decisions of this Court.[71] It follows therefore that the fact that an offence arises out of an occasion when constitutional rights to assemble and protest are being exercised is relevant to the background and context of the offending, particularly when those rights have been exercised peacefully and in accordance within the law up to the point when the offence was committed. 68. However, as Ribeiro PJ stated in HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837 at [39]: “Once a demonstrator becomes involved in violence or the threat of violence – somewhat archaically referred to as a ‘breach of the peace’ – that demonstrator crosses the line separating constitutionally protected peaceful demonstration from unlawful activity which is subject to legal sanctions and constraints. The same applies where the demonstrator crosses the line by unlawfully interfering with the rights and freedoms of others.” 69. For this simple reason, a submission in mitigation of the offence of unlawful assembly (and certainly in the case of incitement) that the act was committed in the exercise of the constitutional rights to freedom of expression and freedom of assembly will be unlikely to carry any significant weight. The fact of a conviction of the offence will necessarily mean that the offender has crossed the line separating the lawful exercise of his constitutional rights from unlawful activity subject to sanctions and constraints. In such a case, there is little merit in a plea for leniency on the basis that the offender was merely exercising constitutional rights since, by definition, he was not doing so at the time when the offence was committed. This is all the more so when the facts of the offending involve violence, in particular on the part of the offender himself, since there is no constitutional justification for violent unlawful behaviour. In such a case involving violence, a deterrent sentence may be called for and will not be objectionable on the ground that it creates a “chilling effect” on the exercise of a constitutional right, since there is no right to be violent. Quite simply the line of acceptability has been crossed. 70. Similar considerations apply to the contention that it is a mitigating factor that the offence was committed by way of an act of civil disobedience. As Lord Hoffmann noted in R v Jones (Margaret) [2007] 1 AC 136 at [89], “civil disobedience on conscientious grounds has a long and honourable history in this country.” Although that is a reference to the position in the United Kingdom, the concept of civil disobedience is one which is recognisable in any jurisdiction respecting individual rights, including Hong Kong. 71. Broadly, civil disobedience can consist of one of two types of behaviour. One type of behaviour consists of breaches of a particular law which is believed by the offender to be an unjust law. The other is law-breaking done in order to protest against perceived injustice or in order to effect some change in either the law or society. Either type of behaviour may be actuated by the offender’s conscientious objections and genuine beliefs and a sentencing court may properly take these into account as the motive for the offending, although the weight to be attached to that motive will necessarily vary depending on many other circumstances, including the facts of the offending and its consequences and the need for deterrence and punishment. 72. Regardless of the type of behaviour concerned, however, there are certain hallmarks or conventions common to all forms of civil disobedience. In A Theory of Justice (revised Edition, 1999) at p.320, John Rawls[72] defined civil disobedience “as a public, nonviolent, conscientious yet political act contrary to law usually done with the aim of bringing about a change in the law or policies of the government”.[73] To conform to civil disobedience as generally understood, therefore, the action carried out must be peaceful and non-violent.[74] An expectation of punishment is also inherent in the act of civil disobedience since it is by accepting the punishment that the protester seeks to draw attention to the alleged injustice against which he is demonstrating. As Lord Hoffmann observed, in R v Jones (Margaret) (supra) at [89]: “But there are conventions which are generally accepted by the law-breakers on one side and the law-enforcers on the other. The protesters behave with a sense of proportion and do not cause excessive damage or inconvenience. And they vouch the sincerity of their beliefs by accepting the penalties imposed by the law.” 73. In the present cases, the acts of civil disobedience relied upon were not directed towards section 18 of the Public Order Ordinance (Cap.245) as an unjust law. Instead, they were committed in the course of protesting against the Government’s proposals for constitutional reform. There is no injustice in the application of the Public Order Ordinance to the appellants and in the imposition of lawful punishment on them for breach of its provisions. Thus, as Lord Hoffmann said in Sepet v Home Secretary [2003] 1 WLR 856 at [33]: “… while the demonstrator or objector cannot be morally condemned, and may indeed be praised, for following the dictates of his conscience, it is not necessarily unjust for the state to punish him in the same way as any other person who breaks the law. It will of course be different if the law itself is unjust. The injustice of the law will carry over into its enforcement. But if the law is not otherwise unjust, as conscription is accepted in principle to be, then it does not follow that because his objection is conscientious, the state is not entitled to punish him. He has his reasons and the state, in the interests of its citizens generally, has different reasons. Both might be right.” 74. Where, therefore, in furtherance of an ostensibly peaceful demonstration, a protester commits an act infringing the criminal law which involves violence and is therefore not peaceful and non-violent, a plea for leniency at the stage of sentencing on the ground of civil disobedience will carry little (if any) weight since by definition that act is not one of civil disobedience. 75. Even where an act of protest may properly be characterised as one of civil disobedience, conforming to the hallmarks and conventions referred to above,[75] the court will not enter into an evaluation of the worthiness of the cause espoused. It was submitted, for example, on behalf of the 3rd appellant, that the fact that an offence was committed as an act of civil disobedience could be a significant mitigating factor where “(a) the ideals behind a course of civil disobedience are consistent with the promotion of a pluralistic, open and tolerant society or political order, and (b) the means undertaken by the offenders impose only minor injuries or damages to property”.[76] It is not, however, the task of the courts to take sides on issues that are political or to prefer one set of social or other values over another. 76. As Lord Hoffmann said in Sepet v Home Secretary (supra) at [34]: “… As judges we would respect their views but might feel it necessary to punish them all the same. Whether we did so or not would be largely a pragmatic question. We would take into account their moral views but would not accept an unqualified moral duty to give way to them. On the contrary we might feel that although we sympathised and even shared the same opinions, we had to give greater weight to the need to enforce the law. In deciding whether or not to impose punishment, the most important consideration would be whether it would do more harm than good. This means that the objector has no right not to be punished. It is a matter for the state (including the judges) to decide on utilitarian grounds whether to do so or not. As Ronald Dworkin said in A Matter of Principle (1985), at p 114: ‘Utilitarianism may be a poor general theory of justice, but it states an excellent necessary condition for a just punishment’.” D.3 Approach when Court of Appeal gives guidance for future cases 77. As a reflection of the principle of legal certainty, it is settled law that the sentence for an offence should be in accordance with the practice prevailing at the time of the commission of the offence: see HKSAR v Tsoi Shu & Ors [2005] 1 HKC 51 at [39], citing R v Chan Ka Wai, CACC 530/1988, unrep., 9 May 1989 at [6]-[7]. 78. The principle that an offender is to be sentenced on the existing or prevailing guideline or tariff of sentence existing at the time of the commission of the offence reflects the protection against retroactive criminal penalties conferred by Article 12(1) of the Hong Kong Bill of Rights[77] which relevantly provides: “... Nor shall a heavier penalty be imposed than the one that was applicable at the time when the criminal offence was committed. If, subsequent to the commission of the offence, provision is made by law for the imposition of a lighter penalty, the offender shall benefit thereby.” 79. A clear illustration of the approach of the Court of Appeal to this principle is the case of Secretary for Justice v Ma Ping Wah [2000] 2 HKLRD 312, which was an application for review of sentence in respect of a conviction for armed robbery. The Court of Appeal received further evidence as to the incidence of “head-bashing” robberies and the medical consequences of such attacks and determined that the existing guidelines for sentencing robbers armed with a knife or any other dangerous weapon (excluding firearms) which were displayed to the victim[78] should be changed to recognise head-bashing robberies “in a category of sufficient gravity to justify a particular band of guideline sentence which was not contemplated at the time when judgment was given in that case.”[79] The Court of Appeal, having given the new guideline as to sentencing in such cases, said: “This new guideline as to sentence cannot, of course, apply to the present case as the deterrent effect it is intended to achieve can only apply to offences committed after this judgment has been delivered.”[80] 80. Other examples of increased guidelines or tariffs of sentences are to be found in the context of trafficking in dangerous drugs,[81] “vote-planting” as a form of electoral fraud[82] and employing a person not lawfully employable.[83] 81. These categories of cases are, however, to be distinguished from sentencing cases concerning offences where there are no established guidelines or tariffs. The public order offences of which the appellants were guilty are such offences. There is neither an established set of guidelines nor any tariff of sentence for cases of unlawful assembly (the latter not being appropriate in any event since the nature of the offence does not lend itself to a tariff). 82. In the Secretary for Justice’s application for a review of sentence in respect of the appellants’ cases, it was submitted that there was an increase in the number of “cases involving offences committed in public order events” and hence the Court of Appeal was to be invited “to lay down guidelines in the sentences of offences related to such events.”[84] 83. In the event, although Poon JA said in the introductory paragraph of his judgment ([18]) that he was expounding on the principles on sentencing in unlawful assemblies that involve violence “to provide guidance to the sentencing courts in the future”, the Court of Appeal did not lay down any fixed starting point of sentence for this category of offence as such. Instead, as noted in paragraph [2] above, the Court of Appeal emphasised the need, when sentencing in cases of unlawful assembly, to take a much stricter view where disorder and any degree of violence was involved. This was consistent with the Court of Appeal’s responsibilities for providing guidance in sentencing matters and it was fully entitled to provide this guidance for the future.[85] D.4 The relevance of youth in sentencing and section 109A of the Criminal Procedure Ordinance 84. The age of an offender, whether youth or advanced age, is always a relevant mitigating factor in sentencing.[86] In the case of a young offender, it is recognised that “the ordinarily better opportunity for reformation and rehabilitation, likely thereafter to diminish, must assume greater significance”.[87] 85. Section 109A(1) of the Criminal Procedure Ordinance provides: “No court shall sentence a person of or over 16 and under 21 years of age to imprisonment unless the court is of opinion that no other method of dealing with such person is appropriate; and for the purpose of determining whether any other method of dealing with any such person is appropriate the court shall obtain and consider information about the circumstances, and shall take into account any information before the court which is relevant to the character of such person and his physical and mental condition.” 86. Section 109A was introduced by the Young Offenders (Miscellaneous Provisions) Ordinance in 1967 and is adapted from section 17(2) of the Criminal Justice Act 1948 in England and Wales. It is clear that the purpose of the provision is to ensure that, save in respect of certain excepted offences,[88] the imprisonment of young persons between the ages of 16 and 21 is a sentencing measure of last resort. 87. On its plain wording, the section places a duty on a sentencing court when considering the appropriate sentence to be imposed on an offender aged between 16 and 21, to obtain and consider information about the circumstances for the purpose of determining whether any method of dealing with the offender, other than a sentence of imprisonment, is appropriate. The “circumstances” refer to the circumstances of the offender, the offence and his suitability for particular types of punishment. 88. In addition to obtaining and considering the information, the sentencing court is also required, by the section, to take into account information relevant to the young offender’s character and his physical and mental condition. Each of these matters is plainly relevant to his suitability to other forms of punishment that may be imposed on a person aged between 16 and 21, be it probation, a community service order or committal to a detention centre or training centre. 89. Nevertheless, the requirement to obtain information is not absolute. As Salmon LJ (as he then was) observed in Morris v Crown Office [1970] 2 QB 114 at pp.129H-130A: “In my view the second part of section 17 of the [Criminal Justice] Act of 1948 is in any event directory only. If a judge fails to obtain or pay attention to the necessary information before sentencing, his sentence would not thereby be invalid. It would only mean that the appellate court would obtain the necessary information and review the sentence in the light of it.” 90. Moreover, there may be certain cases where the court is able to be satisfied that no other sentence other than imprisonment is appropriate for the offender. Yeung Ka Wah & Others v The Queen, CACC 357/1970, unrep. (27 March 1971) is an example of such a case. There, the young offender was jointly charged with others of the offence of demanding money with menaces. In his judgment for the Court of Appeal, Huggins J (as he then was) said (at pp.4-5): “In regard to two of the accused who were under 21 years of age and to whom s.109A of the Criminal Procedure Ordinance applied the learned judge expressly said that in view of the circumstances which he set out he was of opinion that the only appropriate method of dealing with those young accused was by way of imprisonment. We think a judge (but rarely, perhaps, a magistrate, for he deals with cases which prima facie are not so serious that a summary trial is inappropriate) is entitled, where the facts warrant it, to take the view that the serious nature of an offence so clearly overrides any considerations personal to the offender that he can properly exercise his discretion under s.109A without first obtaining a probation officer’s report.” It should be noted that this approach in the case of serious offences is now reflected in the exception of certain offences from the ambit of section 109A. Nevertheless, the general proposition that there may be some cases where the sentencing court can determine, without resort to obtaining information pursuant to section 109A, that the only appropriate sentence is one of imprisonment remains sound. E. Applying the principles to the present case 91. In applying the principles set out above, the first question to ask is whether the magistrate could be shown to have imposed sentences on the appellants that were wrong in principle or manifestly inadequate as contended by the prosecution on the application for review. 92. Even if that is not shown to be the case, it still remains to consider the approach of the Court of Appeal on the review and the guidance laid down by it for future cases of unlawful assembly involving violence. E.1 The range of sentences for unlawful assembly 93. On conviction on indictment for the offence of taking part in an unlawful assembly, a person is liable to imprisonment for 5 years and, on summary conviction, to a fine at level 2[89] and to imprisonment for 3 years.[90] 94. An examination of past sentences imposed for unlawful assembly demonstrates that a wide range of types and severity of sentence that have been imposed for the offence. This was acknowledged by Poon JA in the opening paragraph of his judgment in the Court of Appeal[91] and is not a matter of surprise since the offence of unlawful assembly can be committed in a wide variety of different ways and with different consequences. 95. As illustrations of the range of sentences imposed for the offence of taking part in an unlawful assembly, see: (1) In HKSAR v Wong Yuk Man [2015] 1 HKLRD 132, fines of HK$4,800 were substituted for suspended prison terms on appeal against sentence for two Legislative Councillors taking part in a protest who were convicted for an unlawful assembly consisting of a line of protesters moving slowly step by step in a line up to a police line and finally pressing their bodies against the police line to push it, but only for a short time. (2) In HKSAR v Wong Yeung Tat [2016] 4 HKLRD 445, an appeal against conviction was dismissed in respect of a conviction for unlawful assembly in respect of which a fine of HK$5,000 was imposed. The unlawful assembly consisted of disorderly behaviour as part of a group of about 10 who rushed towards one of the entrances of the Legco complex and bumped into police and Legco security officers, resulting in five officers sustaining injuries. (3) In HKSAR v Au Kwok Kuen [2010] 3 HKLRD 371, an appeal against conviction was dismissed in respect of a conviction for unlawful assembly by six defendants, who took part in a protest by a group of about 26 and who charged a police cordon set up to prevent the protesters entering private premises. Two of the defendants were each given community service orders for 60 hours and four of the defendants were each bound over to be of good behaviour (in the sum of HK$2,000 for a period of 18 months). (4) In HKSAR v Chung Kin Ping & Ors, unrep., HCMA 296/2015 (12 May 2017),[92] appeals against conviction by three defendants for unlawful assembly were dismissed in cases in which the magistrate had imposed community service orders of 80 hours. The unlawful assembly occurred when the defendants struggled with the police and tried to remove Mills barriers placed by the police to maintain order during a street march. (5) In HKSAR v Yip Po Lam [2014] 2 HKLRD 777, appeals against sentence by two defendants in respect of suspended sentences imposed for taking part in an unlawful assembly were dismissed. Sentences of four weeks’ imprisonment suspended for 12 months were imposed for their taking part in an unlawful assembly involving more than 150 persons who had been taking part in an unauthorised street protest march and who, on the two defendants’ direction, charged a police cordon designed to encircle the protesters. (6) In HKSAR v Tai Chi Shing & Ors [2016] 2 HKC 436, the defendants pleaded guilty to taking part in an unlawful assembly. The magistrate reviewed community service orders she had initially imposed and substituted custodial sentences based on a starting point of six months’ imprisonment. An appeal against sentence was dismissed. The unlawful assembly had involved an intimidating protest at the Legco complex by about 100 persons and acts by the defendants described as riotous in nature, if not a riot by legal definition. The three defendants had variously admitted damaging a glass door of the complex, charging the glass wall of the complex with a Mills barrier and throwing stones and kicking the glass doors of the complex. The costs of repair of the damage were about HK$600,000. The judge held that, in the circumstances, an immediate custodial sentence was appropriate, even for a first-time offender. (7) In Chan Sam v The Queen [1968] HKLR 401, an appeal against a sentence of 18 months’ imprisonment for unlawful assembly was dismissed. The conviction arose in the aftermath of the 1967 riots from the defendant’s actions as part of a crowd of about 200 protesters. The defendant had used a chair to attack one of the police officers who came to disperse the crowd. 96. In the Court of Appeal’s judgment, reference was made to various decisions in England and Wales where sentences in excess of 16 months’ imprisonment were imposed.[93] These cases should be read with caution, however, since they involved violence on a considerably greater scale than was present in this case. In R v Caird & Others (1970) 54 Cr App R 499, the offences involved conduct which was riotous in nature. Similarly, the offences in R v Gilmour [2011] EWCA Crim 2458 and R v Blackshaw [2012] 1 WLR 1126 arose out of a period of nationwide rioting and public disorder. E.2 Did the Magistrate err in principle or impose manifestly inadequate sentences? 97. As indicated above, the Court of Appeal considered that the magistrate erred in principle in sentencing the appellants to community service orders and a suspended sentence and that these sentences were manifestly inadequate. The five matters which the Court of Appeal considered constituted errors of principle are summarised in paragraph [40] above. 98. For the following reasons, we do not agree with the Court of Appeal’s conclusion that the magistrate erred in principle in the five respects identified. 99. As to the first matter (CA Judgment at [167](1)), the magistrate was plainly aware of the factor of deterrence in any sentence since she adverted to a deterrent sentence in paragraph [6] of her RS, albeit she rightly did not think it was fair to punish the appellants for the incidents arising in the Occupy Central protests which were subsequent to the events forming the offences of which they were convicted. It would be surprising if any court in Hong Kong passing sentence on a convicted person did not consider the factor of deterrence “at all” as the Court of Appeal concluded and we do not think the magistrate fell into this error. She reiterated, in her Decision on the Application on Review of the Sentence at [4], that she had carefully taken into account “the circumstances of the offence and each of the Defendants’ criminal acts” and cross-referenced this comment to the paragraphs in her RS where she referred to the possibility of a deterrent sentence. As to the Court of Appeal’s criticism of the disproportionate weight given by the magistrate to factors such as the appellants’ personal circumstances and motives, these were relevant matters for the magistrate to take into account and the weight to be accorded to them was a matter for her, subject only to the possibility that, if she imposed a sentence that was manifestly inadequate and out of line with the range of sentences imposed in practice because of the weight she gave those factors, that might provide an independent ground for reviewing the sentence. 100. As to the second matter (CA Judgment at [167](2)), the magistrate was aware of the large scale nature of the assembly since she variously referred to the fact that there were “hundreds of people” and “a large number of protesters and participants” involved: see RV at [50], [51] and [56]; RS at [7]. She was also well aware of the risk of violent clashes and clearly referred to this risk: see RV at [70]-[71] and [73]; RS at [7]. That it was the magistrate’s view, in her RS, that the level of violence was “not very violent” and the appellants and the other participants in the assembly were “not … the very violent type”, was an assessment that was open to her on the primary findings of fact she made in her RV. Furthermore, there is no reason to think that the magistrate did not have those primary findings of fact in mind when, shortly thereafter, she sentenced the appellants and then reviewed those sentences on the application of the prosecution. In this context, it is pertinent to bear in mind Lord Hoffmann’s comments[94] which Bokhary PJ quoted in Tam Dick Yuen at [41], namely: “The need for appellate caution in reversing the trial judge’s evaluation of the facts is based upon much more solid grounds than professional courtesy. It is because specific findings of fact, even by the most meticulous judge, are inherently an incomplete statement of the impression which was made upon him by the primary evidence. His expressed findings are always surrounded by a penumbra of imprecision as to emphasis, relative weight, minor qualification and nuance … of which time and language do not permit exact expression, but which may play an important part in the judge’s overall evaluation.” 101. As to the third matter (CA Judgment at [167](3)), it is not, with respect, correct to say that the magistrate “completely overlooked” the appellants’ knowledge of the likelihood of clashes between the participants and the security guards and police and the inevitability that at least some security guards would be injured. On the contrary, the magistrate plainly did take this into account in convicting the appellants: see RV at [57], [69]-[71], [76]-[77]. The magistrate adverted to the risk of injury but correctly noted that the injuries actually sustained were minor and there was no evidence the appellants themselves had caused those injuries: see RV at [81]; RS at [6]. Indeed, having seen the CCTV footage that formed part of the prosecution evidence at trial, we agree with her observation that the greater risk of injury was to the participants themselves who scaled the fence around the Forecourt. 102. As to the fourth matter (CA Judgment at [167](4)), there is no proper basis, in our view, for criticising the magistrate for overlooking the fact that the unlawful assembly followed a prior lawful assembly and that the protesters did not have an absolute right to enter the Forecourt and for taking the view that the intention of the protesters was only to enter the Forecourt to form a circle and chant slogans. The magistrate’s recital of facts included the fact of the prior lawful assembly and her rejection of the appellants’ defence of self-help shows that she was plainly aware of the fact that the appellants had no right to enter the Forecourt: see RV at [4], [9] and [86]-[96]. It is a fact that the intention of the protesters was to enter the Forecourt and that, after doing so, they did join hands under the mast in the centre of the Forecourt and chant slogans: see RV at [80]. The magistrate’s conviction of the appellants was necessarily premised on her view that their behaviour in executing their intention to enter and protest in the Forecourt, and (in the case of the 2nd appellant) to incite others to do so, was unlawful. 103. As to the fifth matter (CA Judgment at [167](5)), the expression of remorse by the appellants was a relevant matter for the magistrate to take into account. The appellants gave oral evidence before her at trial and it was a matter for her to form a view as to the veracity and sincerity of their claims of remorse and as to the weight to be given to that factor. Unless the sentence she imposed was manifestly inadequate, it was not for the Court of Appeal to question the weight the magistrate gave to that factor. 104. This leads to the question of whether the sentences were manifestly inadequate, as the Court of Appeal concluded. In the present case, in sentencing the appellants, the magistrate chose to emphasise certain positive factors in the appellants’ favour, namely their youth, the fact these were first offences, their stable family backgrounds, their good academic prospects, their idealism and having acted for a cause perceived to be just without any motive of personal gain. She chose to down-play the more serious aspects of the offence, namely that the appellants were the leaders of (and in the 2nd appellant’s case that he had incited) a large-scale unlawful assembly which involved confronting security guards and police giving rise to a high risk of some degree of violence and personal injury while pressing demands to be let into property from which they were lawfully excluded. 105. In doing so, the magistrate was carrying out a multifactorial assessment of the circumstances of the offending and of the offenders, in which assessment she was entitled to a degree of latitude in the weight to which she attributed to each individual relevant factor. As the discussion in Section E.1 above shows, the range of sentences for unlawful assembly includes the imposition of a community service order and, at the time the magistrate was sentencing the appellants, there was no appellate court guidance that required an immediate custodial sentence for a case of this nature. As Poon JA remarked,[95] a community service order was a sentence “frequently” passed in respect of unlawful assemblies. Other cases where custodial sentences had been imposed were far more serious and many of the mitigating circumstances in this case were not present. The case of HKSAR v Tai Chi Shing & Ors (supra), for example, in which a starting point of six months’ imprisonment was adopted was significantly more serious and involved considerably more violence than the present case which can properly be described as being at the lower end of the scale of violence. In the case of the 1st appellant, in particular, a sentence of imprisonment could only be imposed if the court was satisfied no other method of dealing with him was appropriate: see Section D.4 above. We do not consider that the community service orders and suspended sentence respectively imposed by the magistrate on these appellants were outside the reasonable ambit of the magistrate’s sentencing discretion. Since they were not manifestly inadequate, there was no proper basis for the Court of Appeal to ascribe different weights to the relevant factors taken into account by the magistrate. 106. In these circumstances, the applications for review of the particular sentences imposed on the appellants should have been refused. E.3 Was the Court of Appeal’s approach justified? E.3a The Court of Appeal’s findings of fact 107. The conclusion in Section E.2 above is sufficient to dispose of these appeals. However, the questions of law for which leave to appeal was granted raise wider issues as to whether the Court of Appeal’s approach to the facts of the case was justified. In particular, it was contended by the appellants that the Court of Appeal did not have power, on a review of sentence, to reverse, modify, substitute or supplement the factual basis on which the original sentence was based. 108. In this regard, the appellants jointly contended that the Court of Appeal had, in finding that the present case was one of “a large-scale unlawful assembly, involving violence”,[96] had taken account of factors which, it was submitted, constituted new findings of fact by the Court of Appeal. Those factors are the eight matters summarised in paragraph [38] above, together with the conclusion that the appellants’ attitudes showed they had no genuine remorse (CA Judgment at [165]). 109. Since it is not dispositive of the appeals, it is unnecessary to deal at length with these complaints. Nevertheless, the question of law underlying the appellants’ contentions is of general importance in the context of a review of sentence under section 81A and so it is necessary to deal briefly with them. 110. The extent of the Court of Appeal’s power to review the facts on a review of sentence is addressed in Section D.1 above. It is a proper exercise of the Court of Appeal’s jurisdiction under section 81A of the Criminal Procedure Ordinance to review the evidence before the sentencing court in order to ascertain whether one of the grounds on which the jurisdiction may be exercised is made out. Here, the Secretary for Justice was contending in the review that the magistrate had erred in principle and imposed a manifestly inadequate sentence. The Court of Appeal was entitled, and obliged, to consider the evidence before the magistrate to determine if those grounds of review were established. 111. We are satisfied that, save in two respects, the factors highlighted by the Court of Appeal were findings that were open to it to make on the evidence before the magistrate. Thus, on the evidence which the magistrate herself referred to in her RV, the Court of Appeal was entitled to find that: (1) The appellants’ plan to enter the Forecourt was not spontaneous and that it was premeditated (CA Judgment at [157]); (2) It was within the appellants’ reasonable expectation that there was a serious risk of the crowd clashing with security guards and police officers and that this would inevitably lead to violence (CA Judgment at [158]); (3) There was a point when the appellants must have realised that security guards and police were stopping protesters from entering the Forecourt yet they persisted in trying to get in (CA Judgment at [159]); (4) The appellants and other protesters managed to gain entry to the Forecourt by the persistence of their unlawful acts in the face of and despite resistance from the security guards and police (CA Judgment at [160]); (5) There is a self-evident correlation between the number and extent of the injuries sustained and the degree of violence involved (CA Judgment at [161]); (6) It is a fact that the Forecourt was a place from which the appellants were excluded and that they knowingly determined to gain entry even if that constituted a breach of the law (CA Judgment at [162]); and (7) In addition to the 2nd appellant who was convicted of incitement, it was open to the Court of Appeal to find that the 1st and 3rd appellants, by reason of their respective leadership roles in the student organisations to which they belonged and their activities that evening, encouraged the crowd of young people to take part in the unlawful assembly (CA Judgment at [163]). 112. The last-mentioned factor is not inconsistent with the 1st appellant’s acquittal on Charge (1) for incitement. The basis of the acquittal, as explained above, was the narrow ground that the prosecution did not prove to the satisfaction of the magistrate that, when he was speaking to the crowd on the stage, he knew that the security guards and police would block the protesters entry to the Forecourt. It remains a fact, however, that, given his high-profile in the student protest movement, the 1st appellant’s actions of leaving the stage, running towards the Forecourt and climbing the fence surrounding the Forecourt would have been seen by many of the crowd and would have encouraged them to join in the attempt to enter the Forecourt. There is no inconsistency between the 1st appellant’s acquittal and the Court of Appeal’s observation that he encouraged the crowd, in breach of the law, to enter a place from which they were excluded. 113. The eighth matter to which the Court of Appeal drew attention (CA Judgment at [164]) was, in our view, a matter which was open to it to find, save in one respect in which the court appears to have strayed beyond the evidence before the magistrate. It was open to the Court of Appeal to find that what the 2nd appellant did was “highly dangerous”. It was also consistent with the evidence for the Court of Appeal to find that he used sensational slogans and these are set out at paragraph [19] above. However, there would not appear to have been an evidential basis for the finding that the words “Someone is suffering from heart attack but the police do not allow the ambulance to come in”, “Now (a) nurse(s) has/have reached the gate but the police refuse entry” and “Let the people go, open the gate” were “unfounded”. There does not appear to have been any evidence before the magistrate to indicate whether any of the 2nd appellant’s statements were true or were false and certainly no finding by her that those particular statements were untrue. In the absence of evidence against which to test the veracity of the 2nd appellant’s statements, it was not open to the Court of Appeal to find that these statements were “unfounded”. However, as indicated, the more substantive point found by the Court of Appeal, namely that the 2nd appellant’s actions were highly dangerous was supported by the evidence as a whole and so this error by the Court of Appeal would not, on its own, have been material. 114. The Court of Appeal’s finding that the appellants’ attitudes “showed that they had no genuine remorse for the offences they had committed” (CA Judgment at [165]) would appear to contradict a clear finding of fact by the magistrate. She, it will be recalled, drew attention to what she considered to be remorse on the part of the appellants in her RS (at [9]-[11]) and in her Decision on the Application on Review of the Sentence (at [4(IV)]). Unless that finding of fact on the part of the magistrate was susceptible to being overturned on the usual grounds open to an appellate court (see Section D.1 above), the Court of Appeal should not, in our view, have made a finding which was contrary to that of the magistrate. In any event, for the reasons explained, it is not necessary to resolve this particular factual issue. E.3b The Court of Appeal’s guidance for future cases 115. The Court of Final Appeal is not a sentencing court and appeals to this Court on points of sentencing principle are among “the rarest of cases”.[97] The function of sentencing is primarily that of the convicting court of trial, subject to review by the Court of Appeal, whether on an appeal by the convicted person or on review on the application of the Secretary for Justice. The Court of Appeal is therefore the appropriate court to determine if there is a need for appellate guidance as to the levels of sentence for a particular offence and, if so, to set those levels of sentence. 116. The Court of Appeal correctly identified (CA Judgment at [108]) the factors relevant to any sentencing exercise, namely (1) protection of the public, (2) commensurate punishment, (3) societal disapproval, (4) deterrence, (5) compensation and (6) rehabilitation and reform. The weight to be given to these factors is generally a matter for the sentencing court, subject to the parameters of appropriate sentence fixed by statute and previous decisions of the courts. 117. When a particular offence is increasing in occurrence or frequency, or the consequences of an offence are more deleterious to the good of the community than previously understood, it is open to the Court of Appeal to give guidance to trial courts as to the appropriate level of sentences or as to which of the relevant sentencing factors should be given greater or lesser prominence in the weighing exercise carried out when determining the appropriate sentence for an offence. 118. In the present case, the Court of Appeal considered, in respect of sentencing principles generally, basic rights and the obligation to abide by the law (CA Judgment, Section H1), legal restrictions on the right of assembly (CA Judgment, Section H2), the gravamen of the offence of unlawful assembly (CA Judgment, Section H3), and disrupting public order with violence as an aggravating factor (CA Judgment, Section H4). Each of these matters were entirely appropriate factors for the Court of Appeal to stress in the context of sentencing for the offence of taking part in an unlawful assembly. 119. The Court of Appeal was justified in holding that, “In sentencing cases of disrupting public order, especially those which involve violence, the court must bear in mind the importance of preserving public order.”[98] Similarly, it was within its proper function as a court of review to hold that, “On the basic premise that public order must be maintained, and taking into account the gravamen of the offence of unlawful assembly, the court, in passing sentence, not only has to impose a penalty that is appropriate to the punishment of the offenders, but it also has to take into account the factor of deterrence.”[99] 120. In short, it was appropriate for the Court of Appeal to say that, in the circumstances now prevailing in Hong Kong including increasing incidents of unrest and a rising number of large scale public protests, it is now necessary to emphasise deterrence and punishment in large scale unlawful assembly cases involving violence.[100] In this context, the sentiments expressed by Starke J in the Court of Criminal Appeal in Victoria in R v Dixon-Jenkins (1985) 14 A Crim R 372 at p.379 are apposite: “There are large groups in present-day society of sincere, earnest but wrong-headed people who, because their convictions are so strong, or because they pretend their convictions are so strong, will stop at nothing in order to impose those views on the community, and this, in my opinion, just like hijacking, is calculated to become contagious, and if at the first step the courts do not show that such conduct, however well intended, will not be tolerated in this community, then it is unlikely that such behaviour will be stopped in its tracks. I therefore am of opinion that this is just the case where general deterrence has an overriding effect on the resulting sentence.” 121. We would therefore endorse the Court of Appeal’s list of factors relevant to a decision on the appropriate sentence for unlawful assembly involving violence (CA Judgment at [135]), namely: “(1) Whether the violent acts were spontaneous or premeditated; if it was the latter, how detailed and precise the plan was; (2) The number of people involved in the violent acts; (3) The degree of violence, including whether weapons were used and, if so, what kind and quantity of weapons; (4) The scale of violence, including the location, the number of places and the area in which violence took place; (5) The duration of violence, including whether the violent act was a prolonged one, and whether it still went on despite repeated warning by police or public officers; (6) The consequences of the violent act: for example, whether there was any loss or damage to properties and, if so, to what extent; whether anyone was injured and, if so, the number of injured persons and the degree of injury; (7) Even if there was no loss or damage to properties, nor any injury, what imminence and gravity of threat was caused by the violent acts; (8) The offender’s role and degree of participation; for instance, apart from taking part in the unlawful assembly, or using violence, whether he had arranged, led, summoned, incited or advocated others to take part in the unlawful assembly or use violence.” 122. Similarly, the Court of Appeal was justified in clarifying the principles on which it would be appropriate to impose a community service order (CA Judgment, Section H6) and the factors to be considered in determining if an offender is genuinely remorseful (CA Judgment at [147]). 123. The sentencing principles which the Court of Appeal laid down in cases of unlawful assembly involving violence (CA Judgment at [151]) were therefore entirely appropriate, namely: “(1) In accordance with general sentencing principles, the court will have regard to all the actual circumstances of the case and the seriousness of the facts pertaining to the commission of the offence. Appropriate weight will then be accorded to each applicable sentencing factor, and a sentence that is commensurate with the offence will then be imposed. The same principles apply to cases of unlawful assembly involving violence. (2) Although the definition of unlawful assembly in section 18 of the Public Order Ordinance is relatively simple, the range of factual situations covered is wide. The seriousness of the facts involved varies from case to case and may, depending on the actual circumstances, run from the extremely trivial to the extremely serious. Incidents involving violence are certainly much closer to the serious end of cases, but the facts of different cases still vary. So even for the more serious cases there will still be a spectrum of seriousness. Within the spectrum, the court will accord appropriate weight to the applicable sentencing factors based on the actual circumstances of the case and the seriousness of the facts pertaining to the commission of the offence. (3) On the basic premise that the public order must be maintained, and taking into account the gravamen of the offence of unlawful assembly, the court has to consider the factor of deterrence in sentencing. As to how much weight it should accord to this factor, the court has to have regard of the actual circumstances of the case. (4) If the case is of a relatively minor nature, such as when the unlawful assembly was unpremeditated, small in scale, involving very little violence, and not causing any bodily harm or damage to property, the court may give proportionally more weight[101]to such factors as the personal circumstances of the offender, his motives or reasons for committing the offence and the sentencing factor of rehabilitation while proportionally less weight to the sentencing factor of deterrence. (5) If the case is a serious one, such as when the unlawful assembly involving violence is large-scale or it involves serious violence, the court would give the two sentencing factors, namely punishment and deterrence, great weight and give very little weight or, in an extreme case, no weight to factors such as the personal circumstances of the offender, his motives or reasons of committing the offence and the sentencing factor of rehabilitation. (6) After the appropriate weight has been accorded to all the applicable sentencing factors, the court would then impose a sentence on the offender that is commensurate with the case.” 124. Having set out these factors in its judgment, the Court of Appeal went on to indicate an appropriate starting point for the appellants’ cases, namely 8 months’ immediate imprisonment.[102] This was not a tariff as such, since the offence of unlawful assembly does not lend itself to a tariff approach to sentencing, but it was a significant increase in the range of sentences compared with past cases (see Section E.1 above). Nevertheless, we are satisfied that it was right for the Court of Appeal to send the message that unlawful assemblies involving violence, even the relatively low degree of violence that occurred in this case, will not be condoned and may justifiably attract sentences of immediate imprisonment in the future, given the gravamen of the offence involving the instigation of a risk and fear of a breach of the peace by virtue of the number of protesters involved. 125. It should be noted, however, that the culpability of the offender may vary depending on his degree of participation in the unlawful assembly and the violence in question. In this regard, the “Guidelines on Freedom of Peaceful Assembly”[103] quoted by Tang PJ in his judgment in HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837 at [141] are relevant. A distinction must be drawn between a participant in an unlawful assembly who remains peaceful and one who himself engages in or encourages violence. But cases in which the defendant is shown to have actually participated in violent acts, or to have incited others to commit the offence (as in the case of the 2nd appellant here), will justify increased sentences. Similarly, where the offender is shown to have encouraged, if not actually incited, the unlawful assembly, for example by virtue of his status or leadership of others joining the assembly, this may justify attributing to him culpability for the actual violence involved. 126. In accordance with the principles discussed in Section D.3 above, it would not, however, have been appropriate to apply the Court of Appeal’s guidance to the appellants here. The increase in sentences intimated by the Court of Appeal represented a sentence significantly more severe than the range established by the courts’ existing sentencing practice and so, to avoid retrospectively imposing a more severe sentence based on a new sentencing guideline, the new level of sentence should not have been applied to them. 127. We would also add that we would respectfully disapprove the approach disclosed in paragraphs [6] and [7] of Yeung VP’s judgment. There, the learned Vice-President appears to have justified the increased sentences imposed on the appellants because their actions were thought, but not shown as a matter of evidence, to be the result of the influence of other people encouraging them to break the law. That is not a proper basis for sentencing since it ignores the culpability of the individual appellants and instead seeks to attribute the culpability of other persons to them. However, that approach was not reflected in the main judgment of Poon JA in which the guidance for future cases was set out, so our conclusions above as to the appropriateness of that guidance is not affected by this point. E.4 Did the Court of Appeal properly consider section 109A? 128. The question of whether the Court of Appeal’s consideration of section 109A was correct would arise only if it had been entitled to review the sentence of the magistrate and assume the role of sentencing court. At the review stage, the question is not whether the Court of Appeal applied section 109A correctly but whether the magistrate was entitled to take it into account. In the opinion of this Court, she plainly was and did. However, in sentencing the 1st appellant to a term of imprisonment in substitution for the community service order imposed by the magistrate, Poon JA observed (in parentheses): “The 1st respondent was only 17 at the time of the offence and now 20, his Counsel, Mr. Shek, nevertheless agreed that it was unnecessary for the court to consider other sentencing options.”[104] 129. It would appear that the submission of Mr Shek, who appeared for the 1st appellant in the Court of Appeal on the review of sentence, was misunderstood. Mr Shek’s submission was made in the response to the stance of the Secretary for Justice on the review, which was that the Court of Appeal, if it were minded to grant the application for review, “should call for the relevant reports (for example, detention center, training center and/or rehabilitation center reports) to assist in determining the appropriate sentence for the first [appellant].”[105] 130. Mr Shek agreed with the prosecution that section 109A established the principle that no court should sentence the 1st appellant to imprisonment unless of the opinion that no other method of dealing with him was appropriate. It was in that context that he submitted that the 1st appellant “takes the view that neither detention centre, training centre nor rehabilitation centre is a suitable sentencing option” and that “[t]he sentencing options which the prosecution asks the court to consider are fundamentally inappropriate for the 1st [appellant].”[106] 131. It is clear, therefore, that the submissions of Mr Shek to the Court of Appeal were not that it was not appropriate to consider alternatives to a sentence of imprisonment but that the alternatives of a detention centre, training centre or rehabilitation centre were not suitable and that any reports on the 1st appellant were unlikely to disclose that these were suitable for him. Implicit in Mr Shek’s submission was that the Court of Appeal must be of the opinion that a community service order was not appropriate before it could be satisfied that the only appropriate sentence was one of imprisonment. There was certainly no concession or waiver by Mr Shek that it was not necessary for the Court of Appeal to obtain information as required by section 109A. In any event, under section 109A, it would have been, if the Court of Appeal had been entitled to review the sentence of the magistrate, its duty as the sentencing court to consider all non-imprisonment sentencing options and not a matter for counsel. 132. As the discussion in Section D.4 above shows, there may be cases where the requirements of section 109A can properly be departed from since the circumstances will be such that it will be clear without the need to obtain further information that the only appropriate sentence is imprisonment. In the circumstances of the present offence of taking part in an unlawful assembly, this was certainly not one of those cases and the Court of Appeal erred in not following the requirements of section 109A. 133. In the circumstances, the Court of Appeal’s approach in dispensing with the need to consider other sentencing options was erroneous and provides an independent reason, in the case of the 1st appellant, for quashing the substituted sentence of imprisonment imposed on him by the Court of Appeal. F. Conclusion 134. For these reasons, we would allow the appeals of each of the three appellants, quash the sentences of imprisonment imposed by the Court of Appeal and reinstate those imposed by the magistrate. 135. As indicated in paragraph [2] above, the Court of Appeal was entitled to provide the guidance it did in relation to the appropriate sentences to be imposed in respect of large scale unlawful assemblies involving violence. Offenders in such future cases will therefore be subject to the new guidelines. Mr Philip Dykes SC and Mr Randy Shek, instructed by Bond Ng Solicitors,assigned by the Director of Legal Aid, for D1/the Appellant in FACC 8/2017 (the 1st Appellant) Mr Robert Pang SC, Mr Michael Chai and Ms Priscilla Chow, instructed by Ho Tse Wai & Partners, assigned by the Director of Legal Aid, for D2/the Appellant in FACC 9/2017 (the 2nd Appellant) Mr Edwin Choy and Mr Joe Chan, instructed by Ho Tse Wai & Partners, assigned by the Director of Legal Aid, for D3/the Appellant in FACC 10/2017 (the 3rd Appellant) Mr David Leung SC, DPP, Ms Annie Li, SPP (Ag.) and Mr Michael Chan, PP, of the Department of Justice, for the Applicant/Respondent in FACC 8, 9 & 10/2017 [1] Contrary to section 18 of the Public Order Ordinance (Cap.245). [2] Contrary to section 18 of the Public Order Ordinance (Cap.245) and section 101I of the Criminal Procedure Ordinance (Cap.221). [3] Under the Community Service Orders Ordinance (Cap.378). [4] Pursuant to section 81A of the Criminal Procedure Ordinance (Cap.221). [5] Pursuant to s.14(4) of the Public Order Ordinance (Cap.245). [6] See further below. [7] The basis phalangis digitorum pedis. [8] (Cap.245). [9] (Cap.221). [10] In ESCC 2791/2015. [11] RV at [35]. [12] RV at [40] & [97]. [13] RV at [41]. [14] RV at [42] & [97]. [15] RV at [45]. [16] RV at [48]. [17] RV at [51]. [18] RV at [55]. [19] RV at [56]. [20] RV at [57] & [99]. [21] RV at [63]. [22] RV at [67] & [98]. [23] RV at [69]-[70] & [98]. [24] RV at [71] & [98]. [25] RV at [72]. [26] RV at [73] & [100]. [27] RV at [77] & [100]. [28] RV at [78]. [29] ESCC 2791/2015, Reasons for Sentence, 15 August 2016 (“RS”). [30] (Cap.227). [31] ESCC 2791/2015, Decision on the Application on Review of the Sentence, 21 September 2016. [32] Ibid. at [4]. [33] Yeung VP, Poon and Pang JJA. [34] CAAR 4/2016, Judgment dated 17 August 2017 (“CA Judgment”) at [170]. [35] CA Judgment at [156]. [36] Ibid. at [157]. [37] Ibid. at [158]. [38] Ibid. at [159]. [39] Ibid. at [160]. [40] Ibid. at [161]. [41] Ibid. at [162]. [42] Ibid. at [163]. [43] Ibid. at [164]. [44] Ibid. at [165]. [45] Ibid. at [166]. [46] Appearing as junior counsel to Mr Philip Dykes SC on behalf of the 1st appellant in this court. [47] CA Judgment at [167(1)]. [48] Ibid. at [167(2)]. [49] Ibid. at [167(3)]. [50] Ibid. at [167(4)]. [51] Ibid. at [167(5)]. [52] Ibid. at [168]. [53] Ibid. at [169]-[170]. [54] By applications dated 11 September 2017 in the case of the 1st appellant and 14 September 2017 in the case of the 2nd and 3rd appellants. [55] CAAR 4/2016, Judgment dated 26 October 2017 (Yeung VP, Poon and Pang JJA). [56] FAMC 31-33/2017, Determination dated 7 November 2017 (Ma CJ, Ribeiro and Tang PJJ). [57] FAMC 31/2017 (1st appellant) and 33/2017 (2nd appellant), 24 October 2017 hearing, before Ma CJ. [58] (Cap.221). [59] Entitled “Appeals, Questions of Law Reserved and Referred and Review”. [60] Hansard, 15 March 1972, p.475. [61] Criminal Procedure Ordinance, sections 83G, 83H and 83I. [62] [1972] HKLR 370 at p.376. [63] Pursuant to sections 82 (in respect of conviction) or 83G or 83H (in respect of sentence) of the Criminal Procedure Ordinance. [64] (Cap.227). This limitation on the hearing of a review of sentence explains the lengthy delay in the disposal of the sentence review in the present cases since it was only after the appellants had withdrawn their appeals against conviction that the Secretary for Justice’s application under section 81A could be heard by the Court of Appeal. [65] By means of which the rights under Article 14(7) of the International Covenant on Civil and Political Rights are applied in Hong Kong and thereby given constitutional protection by Article 39(2) of the Basic Law. [66] By the Criminal Procedure (Amendment)(No.2) Ordinance 1979. [67] Hansard, 11 April 1979, p.712. [68] See, e.g., R v Bright [1916] 2 KB 441 per Darling J at p.444; Neal v R (1982) 42 ALR 609 per Brennan J (as he then was) at p.624; and Sentencing in Hong Kong (7th Ed.), Cross and Cheung, at p.383. [69] By means of which the rights under Articles 19 and 21 of the International Covenant on Civil and Political Rights are applied in Hong Kong and thereby given constitutional protection by Article 39(2) of the Basic Law. [70] HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837 at [2], [33]-[34]. [71] See, e.g. Yeung May Wan & Others v HKSAR (2005) 8 HKCFAR 137 at [1]; Leung Kwok Hung & Others v HKSAR (2005) 8 HKCFAR 229 at [1]-[3]; and HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837 at [31]-[32]. [72] Formerly James Bryant Conant University Professor at Harvard University. [73] See, too, the definition of “civil disobedience” in Black’s Law Dictionary (10th Ed.) at p.299: “A deliberate but nonviolent act of lawbreaking to call attention to a particular law or set of laws believed by the actor to be of questionable legitimacy or morality.” [74] See, e.g. Tabernacle v Secretary of State for Defence [2009] EWCA Civ 23 at [4] (peaceful protest against nuclear weapons); Taranenko v Russia (Application No.19554/05, First Section, 15 May 2014) at [91] (political protest against administration of President Putin by non-violent occupation of President’s Administration building); R v Krieger [1998] AJ No.1119 (19 October 1998) at [35]-[37] and R v Krieger [2009] MJ No.430 (21 December 2009) (peaceful and non-violent acts of possessing and supplying marijuana in furtherance of belief the drug should be legalised for medical purposes). [75] Para. [72] above. [76] Printed Case of the 3rd Appellant at [52]. [77] By means of which the rights under Article 15(1) of the International Covenant on Civil and Political Rights are applied in Hong Kong and thereby given constitutional protection by Article 39(2) of the Basic Law. [78] Laid down in Mo Kwong Sang v R [1981] HKLR 610. [79] Secretary for Justice v Ma Ping Wah [2000] 2 HKLRD 312 at p.320C-D. [80] Ibid. at p.320F-G. [81] Attorney General v Ching Kwok-hung [1991] 2 HKLR 125; Secretary for Justice v Hii Siew Cheng [2009] 1 HKLRD 1. [82] Secretary for Justice v Lai Wai Cheong [1998] 1 HKLRD 56. [83] Secretary for Justice v Ho Mei Wa & Anor. [2004] 3 HKLRD 270. [84] Secretary for Justice’s Application to the Court of Appeal for Review of Sentence under section 81A at p.9. [85] As was noted in Seabrook v HKSAR (1999) 2 HKCFAR 184 at p.194G-H, the Court of Appeal routinely deals with sentencing and in practice is always the final court dealing with sentence. [86] Sentencing in Hong Kong (7th Ed.), Cross and Cheung, at pp.341-344. [87] R v Chiang Sun Keung [1997] 1 HKLRD 24 at p.28C. [88] See section 109A(1A) and Schedule 3 of the Criminal Procedure Ordinance. [89] HK$2,001 to HK$5,000. [90] Public Order Ordinance (Cap.245), section 18(3). [91] CA Judgment at [18]. [92] Judgment only available in Chinese. [93] CA Judgment, Section H4. [94] In Biogen Inc v Medeva Plc [1997] RPC 1 at p. 45, repeated in Piglowska v Piglowski [1999] 1 WLR 1360 at p. 1372D-F. [95] CA Judgment at [136]. [96] CA Judgment at [156]. [97] Seabrook v HKSAR (1999) 2 HKCFAR 184 at p.186J. For examples, see Wong Chun Cheong v HKSAR (2001) 4 HKCFAR 12 (concerning the approach to imposition of a training centre order) and Lau Cheong & Anor v HKSAR (2002) 5 HKCFAR 415 (concerning the constitutionality of the mandatory life sentence for murder). [98] CA Judgment at [121]. [99] CA Judgment at [127]. [100] As Poon JA said at [18] of the CA Judgment, “In order to dispel any doubts that the public may have, and to provide guidance to the sentencing courts in the future, I find it necessary to expound on the principles on sentencing in unlawful assemblies that involve violence.” [101] This is not to say or to be misinterpreted that the court agrees with the motives or reasons of the offender for committing the offence. [102] CA Judgment at [169]. [103] 2nd Ed. published by the Organisation for Security and Co-operation in Europe/Office for Democratic Institutions and Human Rights, 25 October 2010. [104] CA Judgment at [166]. [105] CAAR 4/2016, Applicant’s Submission dated 12 July 2017 at [40]. [106] CAAR 4/2016, The 1st Respondent’s Skeleton Submission on the Prosecution’s Application for Review of Sentence dated 26 July 2017 at [35]-[37]. 
Mr Justice Ribeiro PJ: 1. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Tang PJ: 2. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Bokhary NPJ: 4. For the reasons which he gives, I would dispose of this appeal as Lord Hoffmann NPJ proposes. On its wording and in the circumstances, this insurance policy covers the injured employee. So, by statute, the insurers are obliged to pay him even if they are, as against the insured employers, entitled to repudiate and have a right of recourse. The insolvency of those employers renders any such right valueless. That is unfortunate. But it does not make it uncommercial or otherwise inappropriate to construe this insurance policy to cover the injured employee. An insurance policy is to be construed as a whole, purposively, in its factual and legal context and commercially. That is the construction which this insurance policy now receives. Lord Hoffmann NPJ: The Issue 5. On 19 March 2007 Mr Lo Siu Wa (“Mr Lo”), a carpenter employed by Nuovo Design Limited (“Nuovo”) suffered injury in an industrial accident. As required by law, Nuovo had taken out an insurance policy (“the Policy”) in respect of its liability to pay compensation to employees. The question in this appeal is whether, upon its true construction, the Policy covered Mr Lo. The Statutory Background 6. Section 40 (1) of the Employees’ Compensation Ordinance Cap 282 (as amended) (“the Ordinance”) provides, so far as material: “… no employer shall employ any employee in any employment unless there is in force in relation to such employee a policy of insurance issued by an insurer…in respect of the liability of the employer”. 7. Section 43 of the Ordinance provides that – “(1) …where in relation to an employee there is in force a policy of insurance issued for the purposes of this Part and the employer of the employee becomes liable to pay any sum under this Ordinance or independently of this Ordinance in respect of an injury to the employee arising out of and in the course of his employment, such sum shall forthwith become due and payable by the insurer, notwithstanding anything to the contrary in the policy of insurance. … (4) Where under this part an amount is paid by the insurer which would, but for this section, not be payable under the policy of insurance, the employer is liable to pay that amount to the insurer.” 8. The effect of this provision is that, provided a policy “in relation to an employee” is “in force” at the time of his injury, the insurer is liable to pay whatever compensation is due to the employee under the Ordinance or otherwise, notwithstanding that as a matter of contract with the employer the insurer would have been entitled to repudiate liability for misrepresentation or breach of a term of the policy or whatever: see King Tak-on v Lau Chun-yip [1987] HKLR 126. If the insurer would have been entitled to repudiate, section 43(4) gives it a right of recourse against the insured. But the risk of the insolvency of the insured is shifted from the employee to the insurer. 9. Section 43 therefore makes it important to distinguish between the cover provided by the policy and the representations or promises which induced the insurer to provide it. Ordinarily it might not matter whether the casualty falls outside the risk covered by the policy or whether the insurer is entitled to repudiate liability. But section 43 makes this a crucial distinction. The Policy 10. Nuovo’s principal business appears to have been interior design. That was how it was described in the application for insurance. It employed two Creative Directors and a Designer. In 2004, it took out the Policy with the Respondent AXA China Region Insurance Company (Bermuda) Limited (“AXA”). It consisted of a package of various forms of cover suitable for an office-based business and was sold under the name “Office Pak”. Section 1 was Office Contents, Section 2 Interruption of Business, Section 3 Money, Section 4 Public Liability and Section 5, with which this appeal is concerned, Employees’ Compensation. The Policy was renewed from year to year, the last renewal being for the period from 23 November 2006 to 22 November 2007. 11. Clause 3 of the General Conditions said: “The due observance and fulfilment of the Terms and Conditions of this Policy insofar as they relate to anything to be done or not to be done by the insured and the truth of the statements and answers in the Application shall be conditions precedent to any liability of the Company to make any payment under this Policy.” 12. The Policy contained a Schedule which clause 1 of the General Conditions declared to form “an integral part of the Policy.” The premium for Employees Compensation cover in Section 5 was calculated by reference to the salaries or “compensation” of the employees and the Schedule contained, under the heading “Employees’ Compensation”, a list of six posts in the company (“Creative Director”, “Designer” and so on) and the estimated earnings of the persons occupying each post. There were two Designers and the Schedule accordingly stated that there were seven employees. 13. The Special Conditions applicable to Section 5 provided for the insured having to provide information about the actual (in addition to estimated) earnings of the employees so that appropriate adjustment of the premium could, if necessary, be made. Records were to be kept of the identity and earnings of each employee. In more general terms, there was a requirement that the insured – “…immediately notify [Axa] in writing of any material change in the risk insured hereunder made by the Insured or any other person during the Period of Insurance including but not limited to: … iii) any material change in the nature of the Business or in the number of the Insured’s Employees.” 14. The “Scope of Cover” provided by Section 5 defined the risk covered as follows: “If any Employee in the insured’s immediate employ shall sustain bodily injury or death by Accident or Disease occurring during the Period of Insurance within the Geographical Area and arising out of and in the course of his employment by the Insured in the Business…” 15. The “Business” was defined as “the usual work and activities carried on by the insured pertaining to his business as specified in the Schedule and no others.” Section 5 contains several references to the Ordinance which makes it clear that the cover it provided was taken out pursuant to Nuovo’s obligation under section 40(1) of the Ordinance. The Facts 16. It appears from the statutory declaration of Colin Yue, the Creative Director of Nuovo, that the company did more than produce designs. It also undertook “design and build” contracts in which it executed its designs under the supervision of its Project Manager. For plumbing, electrical and plastering work it engaged sub-contractors. The company itself engaged painters and carpenters whom it provided with materials and some of their tools and paid fortnightly at a daily rate. The company regarded these workers as self-employed. Accordingly, it did not declare any such workers for listing in the Schedule and gave no notice to AXA that it had taken on more employees. 17. One such worker was Mr Lo, a carpenter who on 19 March 2007 sustained an injury while using the company’s circular saw. He made a claim against the company and it is now accepted that he was an employee for the purposes of the Ordinance. He sued the company for compensation under the Ordinance and for damages at common law. The company failed to pay the sums awarded and on 16 February 2011 it was wound up on the grounds of insolvency. The Litigation 18. Mr Lo made a claim against AXA under section 43 of the Ordinance. AXA refused to pay on the ground that he was not covered by the Policy. It did not apply “in relation to” him. He therefore turned to the Employees Compensation Assistance Fund (“the Fund”), to which, under section 16 of the Employees Compensation Assistance Ordinance Cap 365, an employee with an unpaid claim for compensation may apply for payment. The Employees Compensation Assistance Fund Board (“the Board”) which administers the Fund took the view that AXA was liable and Mr Lo brought proceedings against both AXA and the Fund. These were consolidated and when the proceedings came before Deputy High Court Judge Burrell in the High Court, Mr Lo had been paid and the only issue was which of the two defendants was liable. 19. The judge took the view that the Policy did not cover Mr Lo and accordingly the Fund was liable. His decision was upheld by the Court of Appeal (Yuen and Kwan JJA, Cheung JA dissenting). The Board appeals to this Court. Construction of the Policy 20. The scope of the cover under Section 5 of the Policy seems to me perfectly clear. It covers “any Employee”. There are only two qualifications. The first is that he must be “immediately employed”. I take it to be in contrast with “mediately” or “indirectly” employed, for example, by another employee or a sub-contractor. Mr Lo satisfied that requirement. His employment was by Nuovo itself. 21. The other requirement was that the business in which he had to be employed had to be, according to the definition, “the usual work and activities carried on by the insured pertaining to his business as specified in the Schedule and no others.” The Schedule described the business as “Interior Design” or “Office (Interior Design)”. The reference to “Office” was probably to identify the business as suitable for an “Office Pak” policy, in which, for example, the Contents section was limited to the contents of an office. But other parts of the Policy, such as the reference to overseas activities, show that the company’s activities were not thought to be confined to an office. 22. As for “Interior Design”, it was the case that the company’s principal business was producing interior designs. But it also carried on the ancillary activity of having its own designs executed. It is clear from Mr Yue’s declaration that such a service was part of the company’s “usual work and activities”. But the construction work was not a separate business. No one would have said that Nuovo was a construction company. It did not carry out construction work to anyone else’s designs. Of course, one could give a more elaborate description of the business, listing all its activities, but that does not appear to be customary. In Law Lai Ha v Zurich Insurance Co. [2011] 2 HKLRD 450 the business of a company manufacturing and selling machinery was described as “Trading”. At that level of generality, it seems to me entirely accurate to describe Nuovo’s business as Interior Design. 23. In Susic v Sunset Pty Ltd (1998) 145 FLR 363, an Australian case on similar legislation, the business was described in the policy as “Garden and Park Furniture”. The employer manufactured and sold such items as park benches and planter tubs, mainly made from concrete. It also made tombstones (“monumental masonry”) and this was the work upon which the employee was engaged when he was injured. The Supreme Court of the Australian Capital Territory (Miles CJ) said: “In my view, while the description [of the business] is not comprehensive, it describes [the employer’s] business in general terms. It cannot be said that [the employer] was not engaged in the manufacture of garden and park furniture or that such activity was such an insignificant part of [its] activity that the description does not fit the nature of its business. The plaintiff’s injury was not sustained when he himself was engaged in the actual manufacture of park or garden furniture, but…[t]he question is whether the policy is to apply to [the employer’s] liability to the plaintiff where that liability results from injury to the plaintiff arising out of, or in the course of employment in that business described in the schedule. There was only one business and it is the business so described.” 24. Likewise, it seems to me, Nuovo had only one business and that was the business correctly, if economically, described as Interior Design. I do not think one should be too finicky or pedantic about the interpretation of a brief description of the business of the insured. Otherwise an employer might be found to have committed the criminal offence of failing to insure (section 40(2) of the Ordinance) because he did not provide a sufficiently detailed description of his business. 25. Ordinarily, whatever the description of the business, the insurer will obtain a detailed description of its activities from the particulars of the occupations and salaries of the employees such as appear in the Schedule in this case. What seems to me to have happened was not that anyone considered that Nuovo’s construction work involved carrying on a different business but that it did not regard the people engaged in that work (apart from its Project Manager) as employees. That view is admitted to have been wrong. It means that Nuovo’s representations about the number and remuneration of its employees were inaccurate and that it was in breach of its obligation to give AXA correct information for the purpose of assessing the risk or notify AXA of any change in the number of its employees. As against Nuovo, these matters may have entitled AXA to repudiate liability under the policy and given it a right of recourse under section 43(4). But section 43(1) of the Ordinance prevents it from doing so as against Mr Lo. The judgments below 26. It remains for me to consider why three out of the four judges who have so far heard this case came to a different conclusion. 27. Deputy Judge Burrell said that if Mr Lo was covered by the policy – “every casual labourer doing plumbing, electricity work, air conditioning, carpentry etc in all Nuovo’s projects are covered. This must be wrong.” 28. On the contrary, it seems to me that, so far as such persons were also employees of Nuovo, it must be right. Of course, some of them would not be employees: the evidence was that “the work of plastering, installation of electric appliances and air-conditioning were sub-contracted to other companies” (Declaration of Lee Chun Fai, Nuovo’s Project Manager). Their employees would not be direct employees of Nuovo and therefore not covered. 29. The judge went on to say that, in context, “immediate employ” can only mean those clerical jobs listed in the Schedule, I do not see why this should be so. It would have been easy for the “Scope of Cover” clause to have been limited in this way, but it says “any Employee”. Furthermore, the obligation to notify AXA of changes in the number of employees indicates that the parties did not contemplate that cover would be confined to those on the list. 30. The judge also mentioned the language of Section 1 (Office Contents) which provided that the company’s office (the contents of which were insured) should be used “solely as an office…without any manufacturing work”. I cannot see how this throws any light on what work an employee within the cover of Section 5 could do outside the office. 31. Finally, the judge relied upon the decision of Sakhrani J in Law Lai Ha v Zurich Insurance Co. [2011] 2 HKLRD 450, which I have already mentioned. The employer manufactured and sold machinery and had a factory on the Mainland and an office in Hong Kong. The employee was engaged as production manager and worked at the factory. However, in the course of his employment he was sent to Kenya to supervise the installation of a plant manufactured by the company. There he contracted a disease from which he died. 32. The policy in Law Lai Ha’s case stated, as in this case, that it covered “any employee”. The judge referred to the schedule, which specified only two employees, at the office in Hong Kong. On this basis the judge concluded, without any further analysis or reasoning, that the cover was confined to those two employees. In my opinion he was wrong. 33. In the Court of Appeal, Yuen JA said that the words “any Employee” could not “be read in a vacuum”. That no doubt is true, but there was no context to require any qualification of the words used. She treated the condition that the insured should give notice of a change in the number of employees as justifying a construction limiting the cover to those in the Schedule. It seems to me, however, that it points in the opposite direction. If the parties did not contemplate that employees other than those listed in the Schedule might come within the scope of the cover, it is hard to see why the insurer needed to know about them. Mr Anthony Ismail, for AXA, suggested that the insurer might, after due consideration, agree to add them to the Schedule. But that, on his construction of the Policy, would have involved an amendment. It would not have affected the risk on the Policy as it stood. 34. Kwan JA stressed the fact that the Policy contained elaborate provisions about the insurer being informed about the precise numbers and salaries of employees and said that these did not suggest “that the insurer would have attached no or little significance to the particulars supplied as regards the employees”. That is undeniable, but the reason why the insurer was anxious to receive full particulars of all employees engaged by the company, whether mentioned in the Schedule or newly engaged, was that they would all be covered by the Policy. If they were not, the insurer would have no interest in who else was employed. Failure to provide such information would have been a breach of the insured’s obligations under the Policy and entitled AXA, as against Nuovo, to repudiate it. But it did not give AXA a defence to the claim by the employee under section 43(1). 35. For these reasons and those of Cheung JA in the Court of Appeal, with which I substantially agree, I consider that it was AXA and not the Fund which was liable to Mr Lo. I would allow the appeal. 36. The orders for costs in the Courts below must be set aside, with the exception of the Sanderson order in favour of the plaintiff at first instance. There will be substituted orders nisi (i) that AXA pay the costs of the Board in this Court and the Courts below; and (ii) that there be no order made in respect of the plaintiff's costs in this Court and in the Court of Appeal. His participation at those stages of the proceedings was unnecessary. The parties are at liberty, if so advised, to lodge written submissions as to costs within 14 days of the date of handing down of this judgment, in default of which, the orders nisi will stand as orders absolute without further direction. Mr Ruy Barretto SC and Ms Julia Lau, instructed by Liu, Chan & Lam, assigned by the Director of Legal Aid, for the Plaintiff (1st Respondent) Mr Horace Wong SC and Mr Clark Wang, instructed by Gallant, for the 1st Defendant (Appellant) Mr Anthony Ismail and Mr Justin Ismail, instructed by Clyde & Co., for the 2nd Defendant (2nd Respondent) 
Chief Justice Ma: 1. I agree with the Reasons given by Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. At the hearing, for reasons to be provided, the Court allowed the appeals, quashed the appellants’ convictions and ordered that they should have their costs here and below, including costs certified fit for three counsel in the Court of Appeal. The following are my reasons. 3. At the time of their arrest on 12 July 2012, the 1st appellant (“Mak”) was Secretary for Development and the 2nd appellant (“Tsang”) was an Assistant Director of the Highways Department. They were convicted in the District Court[1] of the common law offence of having conspired to defraud the HKSAR Government as well as having committed statutory offences[2] involving the use of documents as agents with intent to deceive their principal, being offences arising out of property transactions entered into between 1985 and 1988/1990 when they were both engineers in the civil service. They were sentenced to concurrent sentences of imprisonment for 8 months on the conspiracy charge and 6 months on the other charges, suspended for a period of two years. 4. On appeal, the Court of Appeal[3] found that the Judge had made findings of fact which were materially erroneous but nevertheless upheld their convictions, applying the proviso. Leave to appeal to this Court was granted by the Appeal Committee.[4] A. The initial acquisitions and tenancies 5. On 8 June 1985, Mak and his wife Wong Lai King (“Wong”) signed an agreement to purchase Flat 21E in a new property development known as City Garden in Electric Road, North Point (“Flat 21E”) for $925,800.00. On the same day, Tsang and his wife Pau Wai Ming (“Pau”) acquired Flat 22E in the same block of that development (“Flat 22E”) for $928,000.00. They each obtained mortgage finance from Wayfoong Credit Limited and duly completed the purchases, so that Mak and Wong became the registered owners of Flat 21E and Tsang and Pau, the registered owners of Flat 22E. 6. On 4 August 1986, Wong entered into a tenancy agreement with Tsang, letting Flat 21E to him at the monthly rental of $8,000 for a two-year term starting on 1 August 1986. On the strength of that tenancy, Tsang applied for and received from the Government, a Private Tenancy Allowance (“PTA”) in accordance with specified entitlements. 7. By a tenancy agreement dated 29 August 1986, Pau let Flat 22E to Mak for $8,000 per month for two years beginning on 1 September 1986. Mak applied for and received PTA in respect of that tenancy. B. Termination of tenancies and sale of the flats 8. Tsang resided in Flat 21E, receiving PTA payments as Wong’s tenant until 31 December 1990 (with an interval when the flat was tenanted by his sister Tsang Wai-wah, another government engineer, while Tsang was abroad on study leave). After he was allocated government quarters on 28 November 1990, he moved out and stopped paying rent and receiving PTA payments. Soon after he left, Flat 21E was sold with vacant possession, with Tsang executing the assignment dated 22 December 1990 on behalf of Mak and Wong under a power of attorney they had given to him dated 1 December 1990. The sale proceeds were kept by Tsang, a fact given prominence by the prosecution. 9. On 8 December 1990, Tsang and Pau gave a power of attorney to Mak authorising him to sell Flat 22E. It was, however, not sold until July 1992. Meanwhile, on 26 October 1988, Mak had been allocated government quarters and no longer needed to rent the flat and PTA payments ceased. After he left, Flat 22E was let by Tsang and Pau to Dennis Leung (“Leung”) who became its tenant as from 19 November 1988. 10. Pursuant to a fresh power of attorney dated 11 May 1992 given to Mak by Tsang and Pau, Flat 22E was sold on 20 July 1992 with Mak executing the assignment on their behalf. Mak kept the sale proceeds, a fact also highlighted by the prosecution. C. The Regulations and the charges 11. The Judge referred to two versions of the provision in the Civil Service Regulations (CSR 825(5)) dealing with eligibility for PTA. The first was effective as at 24th July 1985, stating: “(5) An officer will not be eligible for a private tenancy allowance if he resides in accommodation owned by himself, his spouse and/or a relation of either himself or his spouse. 'Relations' here include (a) parents, (b) brothers, sisters, and their spouses, and (c) children and their spouses.” 12. As at 13th March 1989, CSR 825(5) relevantly provided: “(a) An officer shall not claim private tenancy allowance in respect of - (i) accommodation owned by the officer himself and/or any relative of the officer; or (ii) accommodation in which the officer himself and/or any relative of the officer have/has a financial interest; or (iii) ... (iv) accommodation owned by a trust of which the officer and/or any relative of the officer is/are a beneficiary. For the purpose of this regulation, ‘relative’ includes - the officer’s spouse;...” 13. The differences between these two versions have not been treated as material. The regulations have throughout these proceedings been interpreted as excluding officers from receiving PTA if they or their relatives had a financial interest in the property they were renting. That interpretation is not in issue. 14. The prosecution case was that Mak and Tsang conspired to defraud the government by obtaining PTA which they knew they were not entitled to because they each had a financial interest in the respective flats which they were purportedly leasing from each other, concealing such interest by pretending merely to be tenants of each other’s wives. 15. The particulars of Charge 1 stated that between 8 June 1985 and 31 December 1990 the appellants conspired together to defraud the Hong Kong Government by dishonestly : “(a) (i) falsely representing that they did not have a financial or proprietary interest in the flats that they leased (MAK leased [Flat 22E] ... and TSANG leased [Flat 21E]... ) and that the lease in relation to each of the said properties was a genuine lease; (ii) concealing and not disclosing that they each had a financial or proprietary interest in the flat that they leased by stating the landlord to be their respective wives; and (iii) claiming and receiving Private Tenancy Allowance from the Government; or (b) causing a public officer or officers to act contrary to their public duty by paragraphs (a)(i), (ii) and (iii) above.” 16. The other charges[5] alleged that the appellants’ had used documents, in particular their PTA applications, with intent to deceive their principal, namely the Hong Kong Government. Like the conspiracy to defraud charge, those charges are premised on the appellants having had a financial interest in the leased premises. Charge 2 against Mak is illustrative, specifying that on 29 August 1986, with intent to deceive his principal, Mak : “... used a document, namely, the Application for Private Tenancy Allowance in relation to the leased property at [Flat 22E] ... in respect of which the Hong Kong Government was interested and which contained a statement which was false or erroneous or defective in a material particular, namely that the leased property was not owned by himself, his spouse and/or a relation of either his or his spouse and neither he, his spouse nor any of his or his spouse’s relations had a financial interest in it and which to his knowledge was intended to mislead the Hong Kong Government.” D. The defence case and the Judge’s decision 17. The appellants’ case appeared mainly from their cautioned statements to the ICAC, they having elected not to testify. It was that each had bought a flat which they had then leased to the other, quite properly claiming PTA in respect of the rental paid under such leases, having had no financial interest in the property which they were leasing. It was common ground, as the Judge noted,[6] that the practice they claimed to have engaged in – a practice known as “cross-leasing” – was not unlawful. 18. The Judge, however, convicted the appellants on all charges. He identified as the “important question” the issue whether the prosecution was able to prove that when Mak and Tsang each applied for and obtained PTA (i) Mak and Wong “were holding Flat 21E on trust for” Tsang and Pau; and (ii) Tsang and Pau “were holding Flat 22E on trust for” Mak and Wong.[7] If so, the prosecution would have established that they each had a beneficial and thus a financial interest in the respective flats.[8] 19. His Honour referred to the fact that the two sale and purchase agreements had been signed on the same day to acquire flats in the same block, similar in size, view and price; that Tsang had retained a high degree of control over the flat in being able to arrange for his sister’s residence in it during his absence although he was ostensibly only its tenant; and that each had kept the proceeds of sale when the respective flats were eventually sold.[9] On this basis, the Judge stated that the “only compelling inference” he could draw was that “it was Mak or Mak and Wong who provided the purchase money for Flat 22E in 1985/1986” and that “[it] was Tsang or Tsang and Pau who provided for the purchase money for Flat 21E in 1985/1986”.[10] 20. The Judge thus decided (although he did not say so in terms) that since they had provided the purchase monies to acquire the flats which they were ostensibly leasing, a resulting trust had come into being giving each appellant a beneficial and financial interest in the property concerned. E. The Judge’s erroneous findings 21. Lunn VP (writing for the Court) held that His Honour’s finding that Mak or Mak and Wong had provided the purchase money for Flat 22E “was not only unsupported by the available evidence but also contrary to it”.[11] His Lordship listed[12] documents including correspondence with the conveyancing solicitors and loan and mortgage payment receipts and statements which demonstrated that Tsang and Pau had provided the purchase monies in question. While the evidence was not as extensive in relation to Flat 21E, Lunn VP held likewise that the Judge’s determination that Tsang and Pau had provided the purchase money for its acquisition “was against the weight of the available direct evidence”[13] which showed that the money had come from Mak and Wong. 22. The erroneous findings were plainly material since “...the issue of whether or not the respective appellants had a financial interest in the premises in respect of which they claimed a Private Tenancy Allowance went to the heart of the charges brought against them”.[14] F. The application of the proviso 23. The Court of Appeal nevertheless applied the proviso, holding that the Judge had found that the appellants had “agreed to a cross-holding of the flats”, meaning that the appellants “had agreed that [Mak] [would] have a beneficial interest in Flat 22E, notwithstanding that it would be and was registered in the name of [Tsang and Pau] and that, in similar circumstances, [Tsang] [would] have a beneficial interest in Flat 21E”.[15] 24. As the prosecution had put it, such purchases were made “on behalf of each other”,[16] with the appellants agreeing “to register them in the Land Registry, not in their own names, but in the names of each other and then pretend that they were renting from each other flats in which they did not have a proprietary or financial interest.”[17] 25. The Court of Appeal was therefore distinguishing between a “cross-leasing” arrangement which was permissible; and a “cross-holding” agreement which it held the Judge had found to exist, constituting the alleged conspiracy to defraud. The proviso was therefore applied on the basis that the Judge had found that each appellant (and his wife) had expressly agreed to hold the property of which they were the legal and registered owners on trust for the other appellant (and his wife), so that each appellant thereby acquired a beneficial interest – and therefore a financial interest – in the property he was leasing. 26. There was no documentary or any other direct evidence of such an agreement. The Court of Appeal, however, inferred its existence pointing to the fact that the Judge had “adverted on various occasions to the coincidence of the similarity of the conduct of the two applicants”[18] involving the sale and purchase agreements and mortgages being signed on the same day, and the purchase of flats in the same block, similar in size, view and price. Their Lordships also noted that the Judge had found that naming their respective wives (rather than themselves) as landlords and using addresses other than the addresses of the newly-acquired flats was misleading conduct and concluded that these matters made it proper to infer that the parties were acting pursuant to an unlawful “cross-holding” agreement: “[On] that evidence, the judge found:[19] The provision of misleading addresses could not be a coincidence and it shows that Mak and Tsang were acting in pursuance to an agreement. The conduct of Mak and Tsang between 1st June 1985, ie the day they signed the sale and purchase agreements for Flat 21E and 22E and 31st December 1990, ie the day that the tenancy agreement for Flat 21E was terminated, in particular the provisions of misleading addresses, the use of their wives as the landlords in the leases and the fact that both Mak and Tsang falsely stated in their PTA applications/memorandum for renewal that they and their wives did not have a financial interest in the property they leased, when considered together, point to the only compelling inference, ie Mak and Tsang had agreed to apply for and obtain PTA from the government by cross-holding and bogus cross-leasing between themselves Flat 21E and Flat 22E.” 27. The proviso was thus applied because “the inevitable consequence of the judge’s findings as to their agreement to cross-hold the flats was that each enjoyed a beneficial interest, and therefore a financial interest, in the flat registered in the name of the other applicant and his wife”.[20] G. The present appeal 28. The central issue on this appeal is whether, notwithstanding the unsustainable basis of the appellants’ conviction by the Judge, the Court of Appeal was justified in applying the proviso on the footing that it was properly to be inferred that the appellants had, from the time of acquisition of the flats, entered into an express “cross-holding” agreement (that is, an agreement to hold their respective units on trust for each other) and not merely a “cross-leasing” agreement (where they had let out flats which they owned to each other). 29. The test for application of the proviso is demanding. As this Court stated in Kissel v HKSAR:[21] “The test for the application of the proviso is well established: Whether a hypothetical reasonable jury, properly instructed, would on the evidence without doubt convict or would inevitably come to the same conclusion.” H. Inferring an agreement between the appellants 30. It is well-established that the requirement of proof beyond reasonable doubt in criminal cases can be satisfied by drawing an inference of fact, but that such an inference “must be compelling – one (and the only one) that no reasonable man could fail to draw from the direct facts proved.”[22] As was stated in this Court in Winnie Lo v HKSAR: “There are three requirements for drawing such an inference. First, it must be grounded on clear findings of primary fact. Secondly, the inference must be a logical consequence of those facts. Thirdly, beyond being logical (since more than one inference might logically be drawn), in a criminal case the inference must be ‘irresistible’, that is, it must be the only inference that can reasonably be drawn on the basis of those facts.” [23] 31. It is not in dispute that each appellant was renting a flat of which the other was the registered owner. The crucial question was whether the prosecution had proved beyond reasonable doubt that, pursuant to an agreement to hold their respective units on trust for each other, they each had a beneficial interest, and thus a financial interest in the flat being leased, disqualifying them from PTA. Put another way, the central issue is whether the evidence compels the court to infer as the only reasonable inference that the appellants had entered into a cross-holding – and not simply a cross-leasing – agreement when purchasing the flats. H.1 The prima facie position 32. The Court of Appeal set aside the Judge’s findings regarding provision of the purchase monies because their Lordships accepted that the evidence showed that each appellant had paid for the flat of which he and his wife were the registered owners. It follows that when considering the appellants’ criminal liability, the starting-point must be that each appellant (and his wife) had duly acquired and was the legal owner of the flat which was rented out to the other appellant, so that the arrangement was on its face a permissible cross-leasing arrangement. 33. There was certainly no direct evidence that the parties had expressly agreed to a cross-holding arrangement constituting themselves trustees of the flats they owned in favour of each other. Indeed, the suggestion that the appellants might have chosen to enter into such an agreement is inherently implausible. Since it was common ground that a cross-leasing arrangement with a view to claiming PTA was permissible, it is difficult to detect any possible motive for the appellants to choose instead to enter into an unlawful cross-holding arrangement with the identical objective of claiming PTA. It is inherently improbable that the appellants should have (as the Judge thought it proper to infer[24]) “agreed to apply for and obtain PTA from the government by cross-holding and bogus cross-leasing between themselves...” If PTA could legitimately be obtained by a cross-leasing arrangement which they were going to set up in any event, why should they make that a “bogus” arrangement and engage in unlawful behaviour by entering into a surreptitious cross-holding agreement? 34. Mr Peter Duncan SC[25] was unable to suggest any benefit that they might gain by entering into an illicit instead of a lawful arrangement as the basis for claiming PTA. Reference was made to the Judge’s suggestion[26] that a possible motive for a cross-holding arrangement was that this would enable each couple securely to spend money decorating and improving their own flats. That was purely speculative since, as Mr Duncan accepted, there was no evidence of what, if any, money each appellant had spent or intended to spend on improving their respective units. Moreover, the evidence showed that they each had real prospects of obtaining government quarters on promotion and that, upon such prospects being realised, they had sold off the flats, indicating they had not held them as long-term investments on which significant sums were likely to have been invested on improvement. 35. The inference that the appellants had entered into a cross-holding agreement is not a natural inference for another reason. The appellants had access to legal advice and, as we shall see, sought legal advice in relation to a proposed declaration of trust in connection with the disposal of the flats. If similar legal advice had been sought as to the legal consequences of an informal agreement to hold the respective flats as trustees for each other, they would have discovered that in the absence of a written and signed document, such an arrangement had little to recommend itself since it would have been invalid for want of formality by virtue of section 5(1) of the Conveyancing and Property Ordinance.[27] The Judge’s speculative suggestion that a cross-holding arrangement would have secured to each appellant the ability to spend freely on improving a beneficially owned apartment also lacks substance for this reason. H.2 The evidence relied on in support of the inference 36. The broad considerations mentioned above weigh substantially against inferring the existence of a cross-holding agreement. What then is the evidence relied on by the prosecution for upholding the Court of Appeal’s application of the proviso? It must be evidence which compels the inference that the parties entered into a cross-holding agreement as the only reasonable inference to be drawn. It must justify application of the proviso on the footing that conviction of the appellants based on such an irresistible inference was inevitable. 37. The evidence relied on by Mr Duncan is set out in the respondent’s printed case as follows :[28] “... the following factors (found by the Judge and referred to by the Court of Appeal) led irresistibly to such an inference being drawn : (a) The odds against A1 [Mak] meeting A2 [Tsang] when they queued up to buy the flats. (b) The two flats were in the same block, one directly above the other and were very similar in size, view and price. (c) The sale and purchase agreements and the equitable mortgages for both flats were signed by the two couples on the same day. (d) The similarities in the manner in which the lease agreements were effected: i. The leases of the two flats stated the landlords to be the Appellants’ wives only and left out the Appellants’ names. ii. The PTA applications of the two flats provided misleading addresses of the landlords which were not the addresses of the two couples. (e) A2 retained a very high degree of control over Flat 21E when he departed HK on study leave. A2 and PW5 were able to break their lease of Flat 21E without a break clause. (f) A2 retained the proceeds from the sale of Flat 21E; A1 retained the proceeds from the sale of Flat 22E.” H.3 Matters listed in sub-paragraphs (a) to (e) 38. It is readily apparent that the matters referred to in sub-paragraphs (a) to (e) are at least as consistent with the existence of a permissible cross-leasing agreement as with that of a cross-holding agreement, making the inference urged by the prosecution far from irresistible. Sub-paragraph (f), concerning the sale proceeds, is separately dealt with below. 39. Sub-paragraph (a) records that the Judge rejected the suggestion that the appellants had met by coincidence in the queue formed by would-be purchasers of units in the development. At its highest, this means that His Honour found that they had met by design and not by chance – a finding equally consistent with a permissible cross-leasing arrangement. 40. The same applies to the finding that the two flats were in the same block, similar in size, view and price (§(b)); that the sale and mortgage documentation were signed on the same day (§(c)); that the tenancy agreements named the wives as landlords and gave addresses other than the addresses of the newly-acquired flats (§(d)); and the ability of Tsang to arrange for his sister to step in as tenant while he was away on study leave (§(e)). 41. Mr Duncan laid particular emphasis on sub-paragraph (d), criticising the appellants for acting in a misleading way by naming their wives rather than themselves as landlords when leasing to each other, and giving other addresses for the respective landlords even though they had, by the time PTA was applied for, moved into the newly-acquired flats. 42. It is fair to say that this conduct suggests a desire not to draw attention to the existence of a cross-leasing scheme. However, it does not compel inference of a cross-holding agreement. Contrary to Mr Duncan’s submission, the facts do not indicate an attempt at concealment of the cross-leasing arrangement (therefore suggesting, so Mr Duncan submitted, the existence of an underlying cross-holding arrangement). Each appellant and his wife were the registered owners of each flat. Anyone who did a land search to ascertain who owned the flat in respect of which PTA was claimed would see that they were named as owners and that one of them (the wife) was the landlord from whom the person claiming PTA was renting the flat. A cross-leasing arrangement was therefore readily discoverable from public records. None of this logically suggests that there must have been a cross-holding arrangement. H.4 Sub-paragraph (f): retention of the sale proceeds 43. The point sought to be made at sub-paragraph (f) is that Tsang, while purportedly merely a tenant, had kept the proceeds after Flat 21E was sold; and similarly, Mak had kept the proceeds after Flat 22E was sold, indicating that Tsang and Mak had all along been the beneficial owners of each of the flats in question. 44. The appellants’ case was that in 1990, after they had been allocated quarters and stopped being each other’s tenants (and stopped claiming PTA), they each agreed effectively to swap flats (or, more precisely, as the evidence indicates, to grant each other the right to receive the sale proceeds of each other’s flats), resulting in each appellant retaining such sale proceeds. 45. The admitted facts included the fact that government quarters were allocated to Mak in October 1988 and to Tsang in November 1990;[29] and that Flat 21E was sold in December 1990 and Flat 22E in 1992.[30] 46. In cautioned interviews, Mak told the ICAC that in November 1990, Tsang had wanted to sell Flat 22E, but since there was an existing tenant (Leung) in it, a sale would be difficult, so he had asked Mak to swap flats and Mak agreed. He said that he believed that ownership in the respective flats was exchanged with the help of a solicitors’ firm Lawrence Ong and Co (“LOC”) and that he thought there ought to have been a trust document or IOU prepared by LOC.[31] The arrangement had led to Tsang selling Flat 21E and keeping the proceeds in 1990, and to Mak doing the same with Flat 22E in 1992.[32] 47. In his cautioned interviews, Tsang likewise explained to the ICAC that he had wanted to sell Flat 22E but was hampered by a sitting tenant and persuaded Mak to agree to a swap of the flats to enable him to effect a sale.[33] He said that he believed that a “trust document” had been prepared to facilitate the swap, which accounted for the mutual retention of the sale proceeds, but that after the lapse of over 20 years, the relevant document could not be found.[34] 48. The Judge, who was upheld by the Court of Appeal on this point, rejected the appellants’ explanation essentially because that they had stated that the flats had been swapped pursuant to some “trust document” prepared by LOC; whereas the Judge accepted the testimony of Andus Lai Sai-on (“Lai”) of LOC that he had declined to prepare such a trust document because, among other things, he did not think it a viable option because of anticipated problems with the mortgagee. It was accordingly found that there had been no swap of the properties to explain retention of the sale proceeds. 49. With respect, this was not an adequate approach to the evidence. It is true that, asked to account for the cross-retention of the sale proceeds at a remove in time of over 20 years and without the benefit of contemporaneous documents, both appellants (laymen, not lawyers) stated their belief that there had been a swap of the flats effected by some trust document. In the light of Lai’s evidence the Judge was clearly entitled to find that no trust document had in fact been executed. However, that finding did not diminish the importance of the available evidence regarding an intended “swap” as the explanation for the way the sale proceeds were dealt with. The significance of that evidence was evidently overlooked. 50. The undisputed evidence was that (i) by a fax dated 27 November 1990 (whose faded and illegible contents were only recovered by forensic scientists after the ICAC interviews) the appellants had given “clear instructions ... to create a document of trust”;[35] (ii) on their instructions, Lai drew up a power of attorney dated 1 December 1990, whereby Mak and Wong authorised Tsang to sell Flat 21E on their behalf; and a power of attorney dated 8 December 1990 whereby Tsang and Pau authorised Mak to sell Flat 22E on their behalf; (iii) after Lai declined to draw up the trust document, he refunded $1,000 which had been paid to his firm as costs on account;[36] (iv) a fresh power of attorney dated 11 May 1992 was prepared by Lai whereby Tsang and Pau authorised Mak to sell Flat 22E on their behalf; (v) the sales which took place on 22 December 1990 (of Flat 21E) and on 20 July 1992 (of Flat 22E) were effected pursuant to those powers of attorney; and (vi) Tsang and Mak retained the sale proceeds respectively of Flat 21E and Flat 22E. 51. The faded fax dated 27 November 1990 conveys instructions to Lai in two parts. First, it asks him to prepare an “Authorization Document” in relation to Flat 21E owned by Mak and Wong in favour of Tsang. As noted above, a power of attorney dated 1 December 1990 authorizing Tsang to deal with Flat 21E was duly prepared and was in fact operated by Tsang on 22 December 1990. Secondly, the fax instructs Lai to prepare an “Authorization Document” and “a Declaration of Trust” to authorize Mak and Wong to sell Flat 22E and to make the “Existing Owners” Tsang and Pau the trustees of the same in favour of Mak and Wong. It also records that Flat 22E was then rented to Leung at $15,000 and mortgaged to the Wayfoong Finance Group for $788,000. 52. The fax provides clear contemporaneous support for the appellants’ case that they had sought to grant each other the right to sell their respective flats in 1990. It shows that it was intended that Tsang should be authorized to sell Flat 21E on behalf of Mak and Wong and, as the evidence establishes, to keep the proceeds. Since Flat 22E, owned by Tsang, would not be sold until later, Lai was asked to prepare a power of attorney authorizing Mak and Wong to sell it in due course and, in the meantime, to prepare a declaration of trust so that Tsang (who would have availed himself of Flat 21E’s sale proceeds) would become trustee of Flat 22E in favour of Mak and Wong. 53. Lai complied with the instructions regarding the powers of attorney and they were in fact utilised in implementing the “swap” arrangement. The fact that Lai advised against and did not draft the declaration of trust is not to the point. The evidence clearly shows that in 1990 there had in fact been an agreed “swap” – not, as the appellants had erroneously recollected, by exchanging legal title in the two flats, but by agreeing to give each other the right to sell and to receive the sale proceeds of each flat. The fax shows that the plan was for the registered owners to remain unchanged and for the flats to be dealt with using powers of attorney. Mak and Wong were intended to be given a measure of security by means of a declaration of trust in respect of Flat 22E while it remained in the registered ownership of Tsang and Pau. 54. The fact that the sale of Flat 21E went ahead without drawing up such a declaration of trust and without Mak and Wong obtaining such security does not detract from the significance of this evidence. It shows that there was a “swap” arrangement in the aforesaid sense and therefore a plausible explanation for the retention by Tsang and Mak of the relevant sale proceeds, consistent with them having previously entered into a cross-leasing arrangement. It indicates that the appellants were (as the Land Registry showed) the owners of their respective flats and that it was only in 1990 that any attempt was made to create an equitable interest in Flat 22E in favour of Mak and Wong as part of the swap arrangement. This evidence makes inference of the existence since 1985 of a cross-holding agreement on the basis of the appellants’ retention of the respective sale proceeds far from irresistible. H.5 Continued payment of rent 55. In his oral argument, Mr Duncan submitted that there ought to be added as a fact which militates against accepting that there was a “swap” arrangement in 1990, the fact that Leung, the tenant of Flat 22E, testified that he had continued to pay his rent after 1990 by delivering cheques to Tsang. The Judge commented: “If there was a swap of properties in 1990, one would expect [Leung] be informed that Mak and Wong had become the new owners of Flat 22E and henceforth [Leung] should pay his rents to Mak, not Tsang. ... The fact that Tsang continued to receive rent from [Leung] after 1990 is wholly incompatible with Mak’s and Tsang’s claim that they had exchanged the properties in 1990.”[37] 56. Since Leung was Tsang’s tenant, it is unsurprising that the rent continued to be paid to Tsang in accordance with the tenancy agreement. There was no evidence as to what Tsang did with the rent received. However, as the prosecution accepted and the Judge noted: “... it is perfectly possible that Tsang had been accounting to Mak for the rent received from [Leung] albeit there is no banking evidence available after the long lapse.”[38] 57. Oddly, acceptance of this possibility did not lead the Judge to question the cogency of the evidence as to rental payments for undermining the case that there had been a “swap”. Instead, the Judge reasoned tangentially that: “Such arrangement [ie, Tsang handing over the rent to Mak] is compatible with the Prosecution’s case that all along Mak and Tsang did not want others to know about their true arrangements.”[39] 58. But the premise of this approach is that their “true arrangements” did involve a “swap”. Retention of the sale proceeds therefore remained consistent with a permissible cross-leasing arrangement and did not provide any basis for inferring a cross-holding agreement. 59. Later in his Verdict, the point became merely a criticism for failing to inform Leung of a change of ownership, while accepting that Tsang might have been receiving rent on Mak’s behalf.[40] That criticism suffers from the erroneous assumption – which the faded fax dispels – that the intended swap involved a change in the ownership of the properties rather than granting each other the right to receive the respective sale proceeds. There was no change of ownership to inform Leung about and it was no concern of Leung’s whether upon sale of Flat 22E, the sale proceeds would go to Mak and Wong or to Tsang and Pau. I. Conclusion 60. The Court of Appeal rightly held that the appellants’ convictions based on the Judge’s finding that they had provided the purchase monies for the leased flats could not stand. However, the Court of Appeal erred in applying the proviso. In the absence of any direct evidence, it felt able to draw the inference that when acquiring the flats, the appellants had expressly agreed to hold them on trust for each other, thereby giving each of them a beneficial and thus a financial interest in the cross-leased flats, and making their claims for PTA dishonest. 61. Such an inference was inherently implausible and none of the matters relied on by the prosecution and by the Court of Appeal were capable of sustaining it. The contemporaneous evidence showed that each appellant (and his wife) had duly paid for and become the legal owners of their respective flats and that they had then entered into cross-leasing tenancy agreements. There was no cogent reason to infer that there was any illicit underlying “cross-holding” agreement. As the faded fax dated 27 November 1990 indicated, it was not until the end of 1990, after the appellants had stopped claiming PTA, that any steps were taken to create a beneficial interest in Flat 22E in favour of Mak and Wong as part of an agreement to swap the right to receive the sale proceeds of each unit. There was accordingly no basis for applying the proviso and the appellants were entitled to have their convictions quashed. Mr Justice Tang PJ: 62. I agree with the Reasons given by Mr Justice Ribeiro PJ. Mr Justice Fok PJ: 63. I agree with the Reasons given by Mr Justice Ribeiro PJ. Mr Justice Gummow NPJ: 64. I agree with the Reasons given by Mr Justice Ribeiro PJ. (Joseph Fok)				Permanent Judge (William Gummow)				Non-Permanent Judge Mr John Reading SC and Mr Kevin Li, instructed by David Hui & Co. for the Appellants Mr Peter Duncan SC, Counsel on fiat, and Mr Beney Wong, SPP of the Department of Justice for the Respondent [1] HH Judge Johnny Chan, DCCC 956/2012 (8 August 2013). [2] Under sections 9(3) and 12(1) of the Prevention of Bribery Ordinance (Cap 201). [3] Lunn VP, McWalters JA and D Pang J, CACC 309/2013 (14 November 2014). [4] Ribeiro Ag CJ, Tang and Fok PJJ, FAMC 75/2014 (14 May 2015). [5] Charges 2 and 3 against Mak and Charges 4 to 6 against Tsang. [6] Judge’s Reasons for Verdict (RFV) §36. [7] RFV§100. [8] RFV§§117-119. [9] RFV§§124-129. [10] RFV§130. [11] Court of Appeal at §63. [12] At §§55-63. [13] At §66. [14] At §67. [15] At §76 and §81. [16] At §50. [17] At §51. [18] At §69. [19] RFV§ 161-162. [20] At §82. [21] (2010) 13 HKCFAR 27 at §170. [22] R v Kwan Ping-Bong [1979] HKLR 1 (PC) at 5. [23] (2012) 15 HKCFAR 16 at §115; citing Nina Kung v Wong Din Shin (2005) 8 HKCFAR 387 at §185. [24] RFV§162. [25] Appearing with Mr Beney Wong, SPP, for the prosecution. [26] RFV§177. [27] Cap 219. Section 5(1): “Subject to section 6 [which is not presently relevant] - (a) no equitable interest in land can be created or disposed of except by writing signed by the person creating or disposing of the same, or by his agent thereunto lawfully authorized in writing, or by will, or by operation of law; (b) a declaration of trust respecting land or any interest therein shall be manifested and proved in writing signed by the person who is able to declare such trust or by his will.” [28] At §28 (with references to the Reasons for Verdict and Court of Appeal judgment omitted). [29] RFV§26(vi). [30] RFV§26(viii) and (ix). [31] RFV§29(xii). [32] RFV§29(viii) and (ix). [33] RFV§30(iii) and (iv). [34] RFV§30(i) and (ii). [35] Court of Appeal §40; RFV§113(e). [36] RFV§§69 and 113(b). [37] RFV§§49 and 50. [38] RFV§51. [39] Ibid. [40] RFV§113(g): “If there was a swap, one would expect [Leung], the tenant of Flat 22E be informed about the change of ownership and that he should henceforth pay his rents to the new owner, Mak and Wong. Even if it was the arrangement of Mak and Tsang that Tsang should continue to receive rent on Mak’s behalf from the tenant, one would still expect Tsang and/or Mak should inform the tenant about their agreement. It is clear from [Leung’s] evidence that he had never been informed that Mak had become the owner of Flat 22E; all along he took Tsang as his landlord.” 
Press Summary (English) Press Summary (Chinese) [2018] HKMagC 3 KCCC 3412-3432, 3496, 3629, 3969, 3971, 3974/2017, 502 & 503/2018 (Consolidated for Verdicts) IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3412 OF 2017 IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3413 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3414 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3415 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3416 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3417 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3418 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3419 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3420 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3421 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3422 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3423 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3424 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3425 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3426 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3427 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3428 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3429 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3430 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3431 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3432 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3496 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3629 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3969 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3971 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 3974 OF 2017 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 502 OF 2018 BETWEEN IN THE KOWLOON CITY MAGISTRATES’ COURTS OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 503 OF 2018 BETWEEN Before: Joseph To, Magistrate Dates of Hearing: 9 April, 28 May, 21-22 & 29 June 2018 Date of Reasons for Verdicts: 17 July 2018 Reasons for verdicts INTRODUCTION 1. Revolutionary as the invention of the internet was, it was not until the arrival of the immense access to the internet via mobile devices that our modern life became truly revolutionised. With movable access to the world-wide web, the world is no longer what it was a mere twenty years ago, and the rapid rise of Uber in the last ten years epitomises a world rapidly and relentlessly “algorithmicised”. 2. In early 2015, when Uber claimed a territorial reach of 260 cities in 45 countries, when it declared an increase of its scale and services in Toronto, and when it said it “expected to have 15,000 drivers signed up under the Uber X Driver App in Ontario by year-end,” the City of Toronto fought the software in court, fought it in equity, and lost.[1] 3. His Honour Mr Justice Dunphy of the Superior Court of Justice, Ontario, said that when Toronto was founded some two centuries ago, transport had been by foot or by horse, but a hundred years later, private automobile “marked a disruptive change in the technology of the era.”[2] His Honour framed the issue of the case in this way: Had the regulations, crafted in a different time, created a flexible firewall capable of fending Uber off, or a Maginot line behind which the taxi industry had retreated.[3] In this context, it is well-known in war history that the famous French line failed in no time. 4. That was not the first or the only case in which Uber’s operation in Canada was sought to be regulated and its expansion restrained. A couple of months before the Toronto case[4] was heard, the City of Edmonton confronted Uber in a court challenge. In that case, not only did Edmonton fail to bring to court the correct contestants to sue, but it failed also to show that Uber Canada had received “a fee in relation to the rides in Edmonton or at all.”[5] 5. In the course of His Honour’s ruling, Mr Justice Crighton of the Alberta Court of Queen’s Bench remarked: “It is not surprising … that legislation drafted to accommodate a more static, paper and people driven environment, sometimes lags behind the technological response to individual preferences and demands.”[6] 6. In our part of the world, instead of horses and carriages, it was litters, sedan chairs, and rickshaws, carried on broad shoulders, powered by strong legs, that were disruptively replaced by cars. In Hong Kong, what remains of these former forms of transport is a reference in our statute strictly prohibiting the use of rickshaws on our roads, for whatever purpose, unless the rickshaws are licensed.[7] 7. This court will say nothing of the heated debate whether our statute has created – to borrow Mr Justice Dunphy’s metaphor – a flexible penal firewall capable of providing perennial protection to a portion of our transport industry. The principal issue at trial before this court is, in essence, whether our legislative regime has constructed a Maginot Line, one that not even the most powerful of our prosecutors is allowed to bypass. In resolving this issue, the court must bear in mind the principal differences between the present 28 prosecutions and the two Canadian precedents so heavily relied upon by the defence in these proceedings. 8. The Toronto and Edmonton cases[8] were civil in nature. The cities sought equitable reliefs compelling Uber – unless otherwise licensed to continue – to cease and desist, which, to Uber, could very well have meant “to cease to exist.” In respect of all disputes, legal and factual, the parties in the two cases bore a lesser burden of proof than a prosecutor in a criminal case. 9. In the 28 criminal cases before this court, the prosecution must meet a much more stringent standard of proof of all necessary facts in support of a conviction. Further, the offence-creating provision concerned must, at common law, be given a restrictive construction. Further still, a criminal court must, when dealing with conflicting constructions of a penal provision and when in doubt, adopt a construction most favourable to the accused. PROSECUTION CASE 10. In none of the 28 prosecutions before this court is Uber implicated as a defendant; the defendants here are Uber drivers. They were each charged with the offence of driving for the carriage of passengers for hire or reward, the charges having been laid against them – with one exception – under sections 52(3) and 52(10) of the Road Traffic Ordinance, Cap. 374, the exception being a case in which the charge was preferred specifically under sections 52(3)(a) and 52(10)(a).[9] 11. It was alleged in the particulars of offences that each of the defendants drove, on a day specified, a private car for the carriage of passengers for hire or reward. The prosecution case is that the offences committed by the defendants were complete on – specifically – the respective day of the defendants driving their cars to carry passengers, and not on the day of the receipt by them of the rewards from Uber, and not on a day unknown within a period of time covering the offending driving of the cars and the illegal receipt of the monies. 12. The prosecutions proceed in this trial by way of facts and documents that are agreed. (1) The 1st to the 23rd cases The first 22 cases 13. In the first 22 cases, a total of 13 undercover police officers, acting on their own or with a colleague, conducted 22 operations between 28 April and 23 May 2017.[10] Some of the officers installed Uber’s Rider Apps a couple of days, about a week, or about a month before the operations,[11] while others used mobile phones with Rider Apps already installed.[12] The Apps were all registered in the officers’ own names with details of their own credit cards for payment.[13] 14. During the operations, the officers activated the Apps and requested for transport services, from various locations[14] to various destinations.[15] They were given three types of cars from which to choose, [16] with three corresponding prices, together with a map in which the journeys sought were described.[17] In each case, the officer chose only the type of car. The Apps then showed the make and model and the registration mark of a chosen car, together with the defendant’s nickname and photograph. A little later, the defendant showed up at the pick-up point in his car, where the officer (in some cases, with a colleague) got in the car; and the journey began. 15. The journeys were completed in most of the cases.[18] In several of them, however, the police took actions and arrested the defendants before they arrived at the destinations.[19] Whether or not the journeys were completed, the trips were paid by credit card transfers, and the officers were notified of the payments;[20] the charging parties – in banker’s terms, the acquiring parties – were either Uber,[21] Uber HK,[22] or Uber BV.[23] 16. Upon subsequent police enquiry, 17 defendants confessed to have worked as Uber drivers;[24] five remained silent.[25] Of the 17 defendants who confessed, 16 of them said that Uber would remit their shares in the fares to their bank accounts.[26] Of the 16 defendants who knew of the remissions, 11 of them[27] said that Uber would transfer to them 75%[28] or thereabouts[29] of the fares, four had no idea or could not recall how their shares were arrived at,[30] and the remaining one declined to say anything about how his share in the fares was calculated.[31] The 23rd case 17. The 23rd case before the court did not come to light as a result of a police undercover operation.[32] What took place was this. On the second day of his stay in Hong Kong, Mr G Bonifacio, a businessman from the Philippines, used his Rider App to request for transport services; a map describing the route and the fare were shown on the screen of his mobile phone. In response to Mr Bonifacio’s request, the defendant arrived at Tsuen Wan in his car to pick him up. En route to Wan Chai, the vehicle in front of the defendant’s car braked to a halt. The defendant failed to stop in time and ran into it. Mr Bonifacio was injured as a result and was taken to hospital for treatment. Upon discharge from hospital, he notified Uber of the accident and was offered a refund of the fare charged. (2) The 24th to the 28th cases 18. Subsequent to the close of evidence of the first 23 cases, five other prosecutions of Uber drivers came before this court for trial.[33] The five defendants were each prosecuted with an offence contrary to sections 52(3) and 52(10) of the Road Traffic Ordinance, Cap. 374, charging them for driving a private car for carriage of passengers for hire or reward. 19. In the 24th case, a civilian Mr Liu, an aged retiree, installed Uber’s Rider App on his mobile phone and requested for transport services from Shatin to Cheung Sha Wan.[34] The App responded by showing the route to Mr Liu’s destination and provided a choice of cars to be used, with the corresponding fares. Mr Liu made his choice and was shown on the screen of his mobile phone the defendant’s nickname and photograph, together with details of the chosen car. Shortly afterward, the defendant arrived in his car and provided Mr Liu with the requested conveyance. The journey was purportedly completed, the fare was duly paid, but Mr Liu realised only then that he had been driven to the wrong place. Dissatisfied, he made a complaint to the police of the matter. 20. Later that day, Mr Liu used the same method to arrange for transport from Shamshuipo to Fanling. In response, another Uber driver showed up in another car and provided him the services sought. The incident led to the prosecution of the driver in the 25th case[35] before this court. 21. The 26th case[36] involved also Mr Liu. On this occasion, he used his Rider App to arrange for transport from Shatin to Wanchai. A Uber driver responded and provided the services required. The transaction was completed with only one unusual feature: for reasons unknown, Mr Liu was charged a lesser amount than as initially shown on the screen of his mobile phone. 22. The 27th and 28th cases involved another civilian passenger Mr Doo. In the 27th case, he obtained transport services using his Rider App, in very much the same manner as shown in all the other cases before this court.[37] However, he was charged more than as initially shown on the screen of his mobile phone. Suspecting that he had been over-charged, he made a report to the police. The same evening, he acquired transport services by the same method again, which led to the prosecution of the 28th defendant before this court.[38] 23. The fares in the 24th to 28th cases were charged to Mr Liu’s and Mr Doo’s credit card accounts; the acquiring parties were either Uber HK or Uber TRIP. (3) The 28 cases before this court 24. In all but one[39] of the 28 cases, the acquiring parties’ references had the suffixes of either “NL”[40] or “NLD”[41] (which are commonly used to designate the Netherlands). In more two-third of them, settlement of accounts involved an overseas transaction fee being levied, showing that the payments were made in a foreign currency and that their receipt had taken place outside Hong Kong.[42] 25. It was admitted by the parties in all 28 cases that, first, the Transport Department had never issued any hire car permits in respect of the cars the defendants drove and, second, none of the defendants had applied for such a permit.[43] However, under reg. 15(5) of the Road Traffic (Public Service Vehicles) Regulations, Cap. 374D, a “hire car permit … may only be issued to the registered owner of the private car,” and not to the drivers involved, and not to the cars concerned. And in the Ordinance, “owner” in relation to a vehicle includes “the person by whom a vehicle is kept and used,” whereas “registered owner” is separately defined.[44] 26. In three of the 28 cases before the court, the defendants admitted under caution to be the registered owners of the cars concerned;[45] in light of the admitted facts that they had never applied for a hire car permit, the prosecution has in those three cases proved that the cars had been operated without such a permit. 27. The situation with respect to the remaining 25 cases is not so clear-cut. In 19 cases, the defendants have either made no cautioned responses in this regard[46] or said that the cars belonged to their friends,[47] younger brother,[48] wives,[49] mother,[50] or father-in-law.[51] In the remaining six cases, the defendants did say they owned the cars but nothing about who the registered owners were.[52] DEFENCE CASE 28. None of the 28 defendants elect to give evidence; and in contesting the charges, 21 of them rely on their clear records.[53] ANALYSIS Overview General considerations 29. The prosecution has the burden to prove the charges beyond all reasonable doubt; the defendants have no burden whatsoever to prove anything, for they are presumed innocent. It is the defendants’ right not to give evidence; the court cannot draw any adverse inferences arising from the manner in which they exercise their right. Some defendants rely on their clear records in their defence. A clear record shows that the defendant does not have any disposition to commit a crime, and it supports his credibility in all he says in his defence, in court or outside, under caution or otherwise. Material prejudicial to the defence is ignored.[54] The cases are consolidated for verdicts only, and the facts and evidence of each case are considered separately. 30. The 28 defendants were prosecuted under section 52(3) of the Road Traffic Ordinance, Cap. 374,[55] which provides insofar as is relevant to the present proceedings: “No person shall – (a) drive or use a motor vehicle; … for the carriage of passengers for hire or reward unless... (iii) a hire car permit is in force in respect of the vehicle.” Five major issues 31. In the past, the law in Hong Kong with respect to road traffic control had developed alongside its English counterpart. On 28 July 1982, Mr Alan Scott, Secretary for Transport, told the Legislative Council: “The present Road Traffic Ordinance (Cap. 220) was enacted in 1957… This legislation was largely based on practice in the United Kingdom.”[56] For a long period of time, there had been a problem in the administration of the regulatory regime on the Hong Kong as well as the English side, namely, a profound confusion in the law arising (i) from the statutory requirement – and the subsequent changes thereof – for insurance coverage in respect of third party risks and (ii) from the closely related prohibition against the use of private vehicles for carriage of passengers for hire or reward. 32. Cases concerning motor insurance liabilities were tried before the civil court. Prohibition against private car hire was enforced by way of criminal prosecution. In those days, a major dispute in the insurance cases concerned the interpretation of the phrase “a vehicle in which passengers were carried for hire or reward” and related phrases like “social and domestic occasions”, as they appeared in motor insurance policies. The criminal court, on the other hand, was repeatedly called upon to construe the phrase “drive or use a private vehicle for the carriage of passengers for hire or reward”, as it appeared in the relevant offence-creating provisions. Jurisprudence developed in the civil court of England with respect to the proper construction of “hire or reward” exerted considerable influence on the criminal bench of Hong Kong, resulting in much confusion. 33. The origin of the phenomenon was not difficult to trace. It was the famous dictum of Branson J in Wyatt v Guildhall Insurance Company Ltd,[57] a motor insurance judgment delivered in 1937, which had, for a long time that followed, been accepted as settled law of England.[58] His Lordship said: “I think that this sub-section is really dealing with vehicles normally or habitually used in the way mentioned in the exception, and that the mere fact that on an isolated occasion a man takes some reward – not necessarily a monetary reward – for the conveyance of a passenger in his car does not render him liable to a penalty for not having an insurance policy covering that passenger on that occasion.” [59] (emphasis supplied) The potential of applying His Lordship’s dictum to the prosecution of private car hire was clear and appealing. It was in this judicial environment that various Pak Pai cases came before our criminal bench in the 1960s and 1970s for adjudication on the proper meaning of “hire or reward” as an element of offence. 34. In the 1960s and 1970s, the English law on road traffic control was in a profound state of confusion, so much so that Karminski LJ lamented: “As things stand, it is or may be difficult for a passenger in a car to ascertain whether or not the owner or driver is covered by a policy of insurance if they meet with an accident,” [60] so much so that Lord Diplock complained: ‘[A]s to the meaning of eleven simple words: “a vehicle in which passengers are carried for hire or reward”, … [t]hree divergent views are held by your Lordships… Only one of these commands the support of even two of your Lordships,’[61] and so much so that Lord Denning pleaded: “I hope that Parliament will soon remedy the position.”[62] 35. Hong Kong responded to Lord Denning’s call for legislative intervention. A major revision of our road traffic regime was undertaken in 1982, which led to the enactment, that year, of our current Road Traffic Ordinance, Cap. 374. 36. The 28 defendants before this court were all charged with an offence under section 52(3) of the 1982 Ordinance. In this trial, the court is called upon to address the following major issues: (i) old cases, old regime, but new law; (ii) elements of offence under section 52(3) of the Road Traffic Ordinance, Cap. 374; (iii) constitutionality of the subject offence; (iv) strict liability; and (v) findings in respect of the present prosecutions. The 1st issue: Old cases, old regime, but new law Inappropriate use of Pre-1982 cases Pre-1982 regulatory regime 37. There is nothing wrong, generally speaking, with relying on old cases, and the prosecution has done just that in these proceedings. But the six 1960s and 1970s precedents[63] relied on by the prosecution were decided under the original 1957 legislation,[64] which was amended in 1964[65] and 1977,[66] and the 1979 Revised Edition[67] of which could not, in terms of scope and approach, possibly compare with the 1982 enactment.[68] The prosecution makes use of pre-1982 cases as if they are necessarily relevant, persuasive, or binding, still. No attempt has been made to explain why they are still relevant, still persuasive, or still binding. The problem is this. 38. Equal in stature, the Legislature and the Judiciary (together with the Executive) are pillars of modern governance, each discharging important constitutional functions. The Legislature concerns itself with appropriate enactments at appropriate time to verbalise popular will and vision. The Judiciary is charged with the duty to administer justice. Relying unquestioningly on old cases despite a major revision of the relevant regulatory regime is inappropriate and dangerous: inappropriate because there must have been a social problem so pressing, so prominent, so pervasive that the Legislature has taken steps to address; dangerous because it may resurrect and summon from its grave the very social evil that a replacement enactment has put to rest. 39. Judicial disinterment of social ills that the Legislature has taken actions to cure (eg., confusion in the law) must, unless there exist absolutely no alternatives, be avoided. Judicial intervention of this kind is plainly unwarranted. Such intervention can arrest the most serious of legislative endeavours, can have a long-lasting effect, and can be costly and time-consuming to correct; such intervention is contrary to the entrenched legal principle that legislative intent must be respected and must, wherever possible, be given full effect; such intervention can compromise the fundamental principle of separation of powers. The court must therefore be vigilant at all times, must exercise utmost care at all times, to ensure that there be no such judicial missteps. And in this exercise, the prosecution is expected also to do its best. Two central concepts contained in the definition of the word “public” 40. To understand the background to the issues raised in the old cases and the bases on which they were decided, a brief examination of the 1957 legislation[69] should suffice; but in view of the issues raised in this trial, both the previous and the current regimes have to be more thoroughly analysed. 41. The previous regulatory regime was founded upon two concepts. The two founding concepts had held the regime together throughout its entire history between 1957 and 1982. In this period of time, the two concepts were encapsulated in the definition of the generic term “public”, by reference to which a vehicle would be considered or classified as a public vehicle. The word “public” was defined in two phrases, each constituting a distinct concept of its own, each being the alternative to the other. The two phrases were (i) “standing or plying for hire or available and intended for hire” and (ii) “carrying goods or passengers for hire or reward.”[70] 42. Under the old law, vehicles with a public character were all referable to one or both of the two alternative concepts.[71] In the present proceedings, because of the defence submission[72] with respect to a “contract expressed or implied,” the two concepts have once again assumed prominence – more specifically, the first concept of “standing or plying for hire or available and intended for hire.” 43. To take the analysis one step further, there were in fact two limbs within the first concept in the definition of “public”. The two limbs were (a) “standing or plying for hire” and (b) “available and intended for hire”. The first limb was defined in the Ordinance, the second one was not. “Standing or plying for hire” meant a vehicle “being on any road…having a driver who is exhibiting any sign or signal indicating that [the vehicle]…was available for hire…”. 44. It must at once be noted that the first limb of “standing or plying for hire” already contained within it an explicit reference to the vehicle being “available for hire”. This was plainly different from “available and intended for hire” as that phrase was used in the generic definition of “public”. The only possible inference must be that the addition of the two words “and intended” in the definition of “public” was meant to cover situations other than “standing or plying for hire” on a road. This is a point of crucial importance, crucial because – together with the definition of “taxis” under the old law – it provides the basis on which the court rejects the defence submission with respect to a “contract expressed or implied”; this issue will be dealt with below. 45. Turning to the definition of “public cars” in the original 1957 Ordinance, the salient point is that it contained an additional contractual element. They were defined as cars, not being taxis, “which carry or are intended to carry passengers for hire and reward under a contract expressed or implied for the use of the vehicles as a whole at or for a fixed or agreed rate or sum.”[73] (emphasis added) “Hire omnibuses” were similarly defined, but this term was removed from the Ordinance in 1964.[74] 46. In 1964, by way of an amendment bill, the expression of “hire and reward” was removed, and the definition of “public cars” was expanded to cover two situations.[75] First, public cars could be operated by “stand[ing] or ply[ing] for hire and [being] used to carry passengers at separate and distinct fares…”[76] (emphasis supplied) Under this limb, the fare was not negotiable, and there was no requirement for a contract between the carrier and the passenger inter se. Second, the definition provided that public cars could also be “hired or [be] intended to be hired for the carriage of passengers under a contract expressed or implied, other than a hire purchase agreement, for the use of the vehicles as a whole at or for a fixed or agreed rate or sum.”[77] (emphasis suppled) In this latter situation, the rate or sum for the hiring of the public car was negotiable between the parties, and a contract of hire was necessary. A similar expansion took place in respect of “public omnibuses”, a term also referred to and relied on by the defence in closing submission.[78] Definition of “taxis” standing uniquely and conspicuously apart 47. As noted above, vehicles having a public character were all referable to one or both of the two central concepts, to one of which might also attach a contractual element, ie., when the vehicles were to be contracted out as a whole for the actual or intended use by somebody else under a contract of hire. The definition of “taxis” was unique in the regulatory regime. It contained not only the two central concepts but also – as a necessary qualifying condition – a reference to a contract express or implied. “Taxis” meant “any cars [1] which stand or ply for hire and [2] in which the passengers are carried for hire or reward under [3] a contract expressed or implied for the use of the vehicles as a whole at a rate indicated by the taximeter.” (numbers and emphasis added) 48. In respect of “taxis”, intended use was irrelevant, the definition targeted actual use only. And more importantly, this definition incorporated the two central criteria for designating a vehicle as a public car, plus the additional element, not of a contract of hire as in the case of “public cars”, but of a carrier contract “expressed or implied”. And even more importantly, despite the expansion of taxis into three categories in 1976, the gist of the original 1957 definition of “taxis” had remained intact[79] throughout the entirety of the old regime until the passing of the 1982 Ordinance. Salient features of the old law 49. Two things are clear under the old law. Firstly, “hire or reward” and “hire and reward” were – until the removal of the latter expression in 1964[80] – used in different context to mean different things. There exists a body of jurisprudence, not referred to by the parties in the present proceedings, concerning the conjunction “or” as it appears in “hire or reward”, which has a significant impact on the prosecutions before this court; this issue will be discussed below. 50. Secondly, public cars could be operated, but taxis must be operated, pursuant to “a contract expressed or implied.” In the case of taxis, where the fare was not predetermined, there had to be a carrier contract for “hire or reward”. For public cars, on the other hand, the consideration for a contract of hire must be fixed or agreed, and the contract must be one for “hire and reward”. Although the expression “hire and reward” was dropped in 1964, intact and maintained was the phrase “at or for a fixed or agreed rate or sum” as it appeared in the definitions of “public cars” and “public omnibuses”[81] in respect of a contract of hire. 51. In contrast, “private cars” were defined as “cars which are not public cars or taxis and are not used to carry passengers for hire or reward.” (emphasis added) This was, in the view of this court, where the problem began. Contractual demarcation under the pre-1982 legislation 52. Under the 1957 enactment, there was a sharp contractual demarcation between “private cars” and “taxis”, contractual in the sense that the characterisation of a vehicle as one or the other hinged upon, among other things, the existence or otherwise of a carrier contract. Carrying passengers for hire or reward simpliciter did not appear, and in motor insurance cases was certainly not regarded as, sufficient to deprive a private car of its private character (see Branson J’s dictum above). It was repeatedly argued, even before our criminal court, that there had to be a carrier contract between the driver and the passenger, the existence of which was needed – as was commonly understood in those days, it seems – to push a private car over to the “public” side of the regulatory regime. 53. It appears, and it is unnecessary for present purposes to decide, that a similar contractual demarcation might have existed in the English road traffic regime at the time. The English Court of Appeal made reference to “a contract express or implied” under the English Road Traffic Act (1930) in a civil case, namely, Coward v The Motor Insurers Bureau.[82] The court said: “In s.61 it is quite plain that the expression “vehicles carrying passengers for hire or reward”, is limited to vehicles in which passengers are carried for a monetary consideration legally recoverable by the carrier. The references … to “separate fares”, and … to contract express or implied, make this clear.’[83] (emphasis added) Pre-1982 cases relied on by the prosecution 54. Coming back to the cases on which the prosecution relies, the appellant in Kwong Pak-yam v The Queen[84] was convicted of two charges, one of which[85] was a charge of driving a private car for hire, a charge preferred against him under reg. 26(2) of the Road Traffic (Registration and Licensing of Vehicles) Regulations (1956). Relying on Coward’s case[86], he argued on appeal that there was no evidence of a carrier contract between him and the passenger. 55. The court in Kwong Pak-yam’s case[87] recognised that a carrier contract in breach of the criminal provision of reg. 26(2) would be unenforceable, meaning that the fare was not legally recoverable, but held, dismissing the appeal, that: “[I]t was necessary for the prosecution here to establish an express or implied agreement of carriage. … It was … inevitable that on this evidence the learned magistrate should come to the conclusion that this was something more than a friendly arrangement such as was found to exist in Coward’s case.”[88] (emphasis added) 56. The prosecution relies on a submission of counsel appearing for the Crown (as the prosecution then was) in Kwok Pak-yam’s case[89] that the journey had been made in the common expectation that a payment would be made,[90] and argues that the appropriate test for the subject offence should be more than “a friendly arrangement” or more than “a social and domestic occasion”.[91] With this proposition, this court disagrees. 57. First, the court in Kwok Pak-yam’s case[92] ruled specifically that the prosecution must prove an agreement of carriage, albeit unenforceable at law, to establish the “hire or reward” element of the offence. Second, it is clear that the court mentioned a “more than friendly arrangement” just to explain the outcome in Coward’s case.[93] A friendly arrangement, even paid and lasted for 18 months, so held the English Court of Appeal, remained just friendly. Kwok Pak-yam’s case[94] had never had an agreement of carriage substituted for the mere proof of an arrangement that was just more than friendly. 58. None of the other 1960s and 1970s cases cited by the prosecution have ruled that such an agreement was not required. They emphasised, rather, the business or commercial nature of the agreement between the driver and the passenger. In Reg v Ngai Kam-chung,[95] the court found that an intention to enter into an agreement that the work should not be gratuitous was necessary.[96] In Chan Kai-lap v The Queen,[97] it was held that proof of habitual use of the vehicle for “hire or reward” was not required. The cases of Cheng For Kar v The Queen[98] and of Li Wing-cheung v The Queen[99] were to the same or similarly effect, namely, an isolated incident could support an irresistible inference that a defendant was in the business of using his private car for hire.[100] 59. Next, the prosecution refers to HKSAR v Chan Chi Ho (transl.)[101] and HKSAR v Lam Shing-yin,[102] and submits that “local courts have since consistently upheld convictions for provision of illegal passenger carriage service based on the evidence of a single transaction.”[103] The prosecution submission is correct. But the defendants in those two cases were, by that time, proceeded against under section 52(3) of the Road Traffic Ordinance, Cap. 374, this being an obvious and important aspect of the two cases which the prosecution has nowhere in the present proceedings sought to address. What appears to have escaped the prosecution’s attention is this. Confusion in the law in England and in Hong Kong in the 1960s and 1970s 60. Significant regulatory changes were made by the 1982 enactment to address a major problem: confusion in the law. The confusion in this branch of the law in England had, as noted above, extended itself to Hong Kong. Various English civil cases were cited before our criminal bench – but this was far from surprising. The contractual demarcation in our former regulatory regime was, in the court’s view, responsible for much of the confusion. In conjunction with Branson J’s authoritative dictum, the contractual demarcation required the court, even a criminal court, to approach the issue of “hire or reward” from a contractual point of view. And the problem was this. 61. It is a fundamental principle of contract law that a contract would come into existence when promises supported by valid and valuable consideration are made in circumstances showing that the parties have intended to create a legally enforceable contract. In a business or commercial setting, such an intention can be readily inferred, but complications can arise when promises are exchanged between friends or family members. It was in this light that in Connell v Motor Insurers’ Bureau,[104]Sachs LJ said: “This is not merely one of those domestic or social occasions to which the learned author of Cheshire and Fifoot, 6th ed. (1964), refers as being those on which legally binding contracts are in appropriate circumstances found not to have been made.” [105] (emphasis added) 62. Various English motor insurance cases were relied on by appellants before our criminal court. It was held in Connell’s case[106] that a handful of isolated incidents were insufficient to make it mandatory for the driver of a private car to have his passengers insured on the basis of the vehicle being used for “hire or reward”, and that habitual use of the car for such a purpose was required. And Connell’s case[107] was relied on by the appellant in Chan Kai-lap’s case.[108] 63. Then came the case of Albert v Motor Insurers’ Bureau,[109] the leading authority on “hire or reward”. The House of Lords found that a carrier contract was not required for “hire or reward”, and that the test was “a systematic carrying of passengers for reward going beyond the bounds of mere social kindness.” Albert’s case[110] was cited by the appellant in Li Wing-cheung’s case,[111] but the court there decided that based on the evidence of even one solitary transaction, the driver could be found to be in a business arrangement, rather than a social one. The question arises: what has the 1982 enactment done to address an obviously unsatisfactory state of the law in regard to road traffic control? Removal of contractual demarcation by the 1982 enactment The present prosecutions before this court 64. Under the previous regulatory regime, to establish “hire or reward”, a carrier contract[112] was once thought to be essential; an unenforceable agreement of carriage[113] was considered necessary throughout; and in the end, bounds of mere social kindness[114] became relevant. But all this has changed with the passing of the 1982 Ordinance. 65. The contractual demarcation, which had been steadfastly maintained all along, [115] was decisively breached and broken at last. Contextual considerations such as social and domestic occasions, which were relevant to the issue of contractual intention and which in the language of the criminal law must be understood to mean motives for the commission of the offence, are no longer relevant. But the prosecution in the present proceedings continues to make reference to contextual considerations, such as social and domestic occasions, still. 66. Unlike the 1957 Ordinance, the following expressions are no longer defined in the Road Traffic Ordinance, Cap. 374: “public”, “public cars”, and “omnibuses”. Taxis are now defined by way of registration pursuant to the Ordinance,[116] rather than with reference to “a contract expressed or implied”. Gone therefore is the element of “a contract expressed or implied” as it appeared in the previous definitions of “public cars”,[117] “hire omnibuses”,[118] and “taxis”.[119] 67. More importantly, a “private car” under the 1982 legislation is no longer defined in contradistinction to “public cars” and “taxis” as it previously was.[120] Now it is defined primarily with reference to its passenger carrying capacity: “a motor vehicle constructed or adapted for use solely for the carriage of a driver and not more than 7 passengers and their personal effects but does not include … [a] taxi.”[121] So gone finally is the contractual demarcation between a “private car” and “taxis”. And most importantly in the context of the present proceedings, gone also are the contextual considerations such as social and domestic occasions. 68. The removal of the contractual demarcation in the 1982 Ordinance represented a most significant change in our regulatory regime. But there was another one, another more drastic, more fundamental, more problematic one, one that calls into question the constitutionality of section 52(3) of the Road Traffic Ordinance, Cap. 374; and the court will return to this issue later. 69. The evidence produced by the prosecution in these proceedings would appear to show that other than the physical car journey in which they were involved, there was no relationship between the driver and the passenger inter se. The prosecution evidence would appear to demonstrate that the driver was just the driver and the passenger the passenger. And this constitutes one of the major premises from which the defence attacks the prosecution case. The defence argues: “No contractual relationship or arrangement would appear to exist between the Defendants and the respective undercover officers or Mr Bonifacio.”[122] The obvious question now is, of course, “Does it matter anymore?” 70. To this question, the prosecution says yes, for reasons that have never been successfully articulated. The defence says yes, for reasons that have been presented with exemplary articulation. Prosecution position on a carrier contract or agreement inter se 71. The prosecution says: “[A]s long as the car is being used for the passage of passengers on a social and domestic occasion, there is no need to obtain a hire car permit,”[123] and “the driver…would not be caught by section 52(3) of the Ordinance for carrying members of his employer’s family on a social and domestic occasion.”[124] The prosecution submits: “Each case will be assessed according to its merits in accordance with the Prosecution Code.”[125] There are serious objections to the prosecution submission. 72. Who has the burden to prove social and domestic occasions? Burden of proof is a matter of law, and the Prosecution Code does not have the status of law. If the burden is on the prosecution, the obvious objection is that section 52(3) does not contain any requirement for the prosecution to prove “without reasonable excuse” as an element of the offence (cf. HKSAR v Lam Yuk Fai).[126] If the burden is on the defence, the obvious objection is that section 52(3) does not come with any statutory defence to that effect. The prosecution further argues that the proper test should be “the commercial nature of the journeys.” [127] By “commercial nature”, the prosecution must have meant activities undertaken with a view to profit, at the risk of loss. The obvious objection here is that section 52(3) never requires proof of the prohibited car hire having been committed “in the course of business,” never requires the court to turn its mind to the defendant’s profit or loss. In all these scenarios, the prosecution appears to invite the court to legislate, to fill the gap, a function that no courts of Hong Kong have ever had, do have, or ever will have. 73. When dealing with the purposive approach in the construction of statutory provisions, the prosecution suggests that a literal reading of section 52(3) would result in absurdities,[128] in that all types of motor vehicles would fall within the scope of its prohibition, except those mentioned in section 52(3)(i), (ii), (iia), and (iii),[129] eg., a licensed bus. The court considers the prosecution submission on absurdity misconceived. 74. Absurdity in the area of construction of statutes can arise in various ways. A classic example is where the language used in a statutory provision, when given a natural reading, is so contradictory that it is impossible to apply the provision in a sensible way. Read plainly and naturally, section 52(3) is clearly capable of sensible applications, although some of its potential applications might not be as the prosecution would desire. There is nothing unusual, nothing difficult to understand, nothing absurd in the language of section 52(3). What is unusual, difficult to understand, if not absurd, is the prosecution’s insistence on relying on Kwong Pak-yam’s case[130] which required proof of an unenforceable agreement of carriage for the purposes of a reg. 26(2) prosecution.[131] 75. Under the old law, proof of an unenforceable agreement of carriage was required because of Branson J’s dictum, because of the contractual demarcation identified above. But in criminal law, an agreement to perform an act amounting to the commission of a criminal offence is not only void and unenforceable inter se but also criminal and prosecutable as a conspiracy, whether or not the act is in fact performed. Successful proof of an illegal agreement of carriage, a necessity under the old law, is sufficient for a conviction of a conspiracy to commit the offence of private car hire. The scope of such a conspiracy is unlikely to go further than the overt acts of the substantive offence, plus the necessary mens rea. In consequence, the conspiracy would in most cases be co-extensive with the substantive offence. 76. It is, of course, within the power of our legislature to enact a substantive offence to include lesser offences; supremacy of our legislature reigns supreme, subject only to the Basic Law, the Bill of Rights, and the common law doctrine of absurdity. For example, under the Dangerous Drugs Ordinance, Cap. 134, manufacture of a dangerous drug includes the trafficking in it, and trafficking in a dangerous drug includes the possession of it. But the legislature must be aware of common law practices of long-standing, a well-known one of which is this: when the prosecution prefers, on the same indictment, a count of substantive offence together with a corresponding, co-extensive conspiracy count, the court will require the prosecution to justify the joinder, failing which the court will put the prosecution to an election as to which count to proceed. 77. In the court’s view, there is no reason for the legislature, having removed the contractual demarcation, to turn around to insist on proof of a conspiratorial agreement as a pre-condition for a conviction of the substantive offence of private car hire, disregarding an established common law practice thereby. The prosecution is incorrect to advocate an approach of interpretation of section 52(3) that requires proof of an illegal agreement of carriage in support of a prosecution of private car hire, as if the 1982 Ordinance had never been enacted, as if the contractual demarcation were still intact.[132] If there were anything absurd about the law on the prohibition against private car hire, it would arise not from section 52(3) itself, but possibly from a wrong interpretation of it. Defence position on a carrier contract or agreement inter se 78. The defence, on the other hand, submits: ‘[T]he reference to “contract expressed or implied” appears in the definitions of taxis, public cars and public omnibuses in the older versions of the RTO. The reference is used to distinguish between the vehicle standing or plying for hire (the first alternative), and a vehicle being hired out as a whole under a “contract” at or for a fixed [or] agreed rate or sum (the second alternative). The fact that the reference to “contract express[ed] or implied” is absent from the standing or plying for hire part of the definition cannot automatically mean that there is no contract of carriage when a fare paying passenger boards a vehicle standing or plying for hire –a contract of carriage clearly still exists in such circumstances. Therefore there is no interpretative significance in its absence. The only purpose for using a reference “contract express[ed] or implied” in the second alternative is to distinguishthe second alternative from the first, in that a vehicle can be used for hire or reward under a contract in circumstances where the vehicle was not standing or plying for hire.’[133] 79. The defence makes a further point relying on reg. 38 and then reg. 33 of the current Road Traffic (Public Service Vehicles) Regulations, Cap. 374D. “Public service vehicles” means “any motor vehicle registered as a public bus, public light bus or taxi, or as a private car in respect of which a hire car permit is in force.”[134] Under reg. 38, the first three types of public service vehicles can be hired out, as a whole, with or without the provision of a driver,[135] in accordance with what is described in reg. 33(4) as a “pre-arranged contract of hire.” The defence submits that “in the context of hiring a public vehicle as a whole”, a contract has not disappeared “in the current version of the RTO.”[136] 80. The defence further argues: ‘The prohibition against driving motor vehicles for hire or reward had not, in any previous versions, incorporated within it a reference to a “contract” or “contract of carriage”.’ However, the defence insists – ‘on an agreement … from which the driver was to be rewarded for his carriage service (or from a common expectation of payment for the carriage service), such that it can be said that the driver was driving for that purpose. This was the basis upon which old cases like Kwong Pak Yam … used the concept of “an express or implied agreement of carriage”, as opposed to being influenced by the used of the word “contract” in the definition of taxis, public cars or public omnibuses in the old RTO.’[137] 81. The defence’s first two points can be dealt with together. The defence is entirely correct in saying that under the old law, a “contract expressed or implied” insofar as it related to “public cars” applied only to vehicles being hired out or contracted out as a whole. It must be noted, however, that “public cars” did not include “taxis”, and “private cars” could not be “taxis” or “public cars”. In this context, the unique definition of “taxis” and its significance under the previous regulatory regime must not be ignored and must be taken into account. 82. Taxis in the old days must be operated by actual standing or plying for hire on the road, and passengers in a taxi must be conveyed under a carrier contract express or implied. “Private cars” were defined in contradistinction to “taxis” and “public cars”. In consequence, a rigid contractual demarcation was erected, one that was not removed until 1982. As noted above, under our current regime, to establish the “hire or reward” element of the offence of private car hire, it is no longer necessary to consider whether or not there has been a carrier contract or agreement between the driver and his passenger inter se. 83. The court disagrees with the defence that “when a fare paying passenger boards a vehicle standing or plying for hire –a contract of carriage clearly still exists… ,”[138] although the defence submission appears to be consistent with what Lord Denning said in Connell’s case:[139] “I should have thought that, in the ordinary way, when a man agrees to carry a man for payment, there is a contract, albeit informal, no matter whether the payment is by way of contribution to the petrol or a reward for the lift.”[140] Under the old regulatory regime, contextual consideration had to be considered. In the old days, if a passenger on a social and domestic occasion chose to pay his driver, the court could, but it did not necessarily have to, conclude that there had been a carrier contract or agreement between them. In our current regime, a carrier contract or agreement between the driver and the passenger inter se is, in the view of this court, simply irrelevant. 84. The defence is correct that in the prohibition against private car hire, there are no references to a contract in situation of standing or plying for hire, and that the lack of such references does not preclude a carrier contract from being entered into between the driver and his passenger. The objection to this argument is this. 85. The fact that reg. 33(4) and reg. 38 provide for the hiring of a taxi as a whole pursuant to a “pre-arranged contract of hire” is irrelevant to the use of the taxi by way of standing or plying for hire on the road. Now that even taxis are not subject to a carrier contract in its normal, regular operation on the road, what benefits, what justifications, what necessities are there for the court to still insist, in a prosecution of a section 52(3) offence, on proof of a carrier contract or agreement between the driver and the passenger inter se, a carrier contract or agreement that is unlawful, one that corresponds to and is co-extensive with the substantive offence? 86. With respect to the third point, the defence argument that Kwong Pak Yam’s case[141] required the proof of a carrier agreement in order to connect the act of driving with its illegal purpose must be rejected; that was not what happened in fact. 87. Our criminal court had never considered that an unenforceable agreement of carriage was required to provide the connection between the act of driving and the carriage for passengers for hire or reward. What in fact happened was that our criminal court had tried strictly and strenuously to apply the English civil cases to establish the “hire or reward” element of the offence in Pak Pai prosecutions. Our criminal court was unable to do so, however, because of the common law principle that an unlawful contract could not be enforced. In the end, our criminal court had the English approach relaxed and modified. A complete answer to the third point raised by the defence can be found in AG v Lun Chan,[142] in which the Full Court said: ‘… Upjohn, L.J., when delivering the judgment in the Coward case, went on to indicate that there could be no hire or reward, within the meaning of the statute, unless there was a “monetary reward legally recoverable by the carrier under a contract express or implied by the mere act of entering the vehicle”[143]… this statement of English law … appears … to enable anyone … to argue: “You cannot convict me of using my car for hire or reward, … since public policy precludes the courts from enforcing a contract for an illegal purpose …”. It is not apparent that Upjohn L.J. had such an argument in mind when propounding his proposition and it looks as if any such proposition, at least so far as Hong Kong is concerned, should be qualified by saying “legally enforceable unless debarred by public policy”. This would, however, still leave the requirement of a contract which would have been enforceable if public policy had not intervened.’[144] (underline supplied) 88. The answer to the defence’s third point was repeated in Kwong Pak-yam’s case[145] itself, the case on which the defence seeks to rely. The court there said: “Nevertheless, as the judges in Attorney-General v Lun Chan pointed out, the form of local legislation may make it necessary to adapt the language of Upjohn L.J. in Coward’s case because an agreement which contemplated a breach of regulation 26(2) of the Road Traffic (Registration and Licensing of Vehicles) Regulations, 1956, would not be legally enforceable.”[146] Conclusion on improper use of old cases 89. This court is called upon to adjudicate on the basis, not of the 1957 Ordinance or any amended versions of it, but of the 1982 Ordinance. 90. It is the judgment of this court that a contractual demarcation had existed in the previous regulatory regime, and that demarcation was removed in the 1982 enactment. In removing it, the legislature has sought to address a profound confusion of the law, one that had troubled our criminal court for decades. 91. The legislative initiative launched in 1982 has achieved considerable success over the last three and a half decades. Issues of a carrier contract, of an agreement of carriage, of habitual use for hire or reward, of more than a friendly arrangement, of going beyond the bounds of mere social kindness have not, it appears, been raised again before our criminal bench. Of post-1982 precedents in which these points were still pursued by the parties, were still pondered upon by the bench, the prosecution in the present proceedings has none to produce, and the defence has produced none. Proper use of pre-1982 cases Has anything survived the change in regulatory regime? 92. There is nothing wrong with relying on old cases when, in particular, a principle of law has arisen from them, a principle that has captured the essential reasoning of the decisions, a principle that has applications outside the particular factual context of the cases from which it was derived, provided always that the principle has not been made obsolete by a legislative enactment. Something did survive the change in legislative regime: the court’s approach in the construction of the conjunction “or” in “hire or reward”. [147] And this has a direct impact on the outcome of the present proceedings. 93. Under the old law, “hire or reward” and “hire and reward” were, until the 1964 amendments, used differently in different context to mean different things. When the fare depended on the distance travelled, as it did in the case of taxi journeys, the Ordinance used “hire or reward”, but when the contract rate or sum must be fixed or agreed, it was “hire and reward”.[148] In the opinion of this court, the interdependence between the fare and the distance traversed represented prima facie a key to understanding the different usage of the two expressions. While “hire and reward” was all but abandoned in 1964, “hire or reward” remains. In the instant cases, the evidence shows that the fares did depend on the journeys sought: the route was always shown, together with the fare, on the screen of the passenger’s mobile phone. 94. AG v Lun Chan[149] was decided by the Full Court in 1964. At trial, the magistrate found the defendant guilty of driving his private car, not for “hire or reward”, but specifically for reward alone, and acquitted him nevertheless of using a motor vehicle without third party insurance, on the basis that no evidence had been adduced to show habitual use of the defendant’s car for hire or reward. The Attorney-General appealed. 95. The Full Court allowed the appeal and reversed the acquittal. In the course of its rulings, the court noted that (i) in Bonham v Zurich General Accident & Liability Insurance Co., Ltd,[150] it was held (in the words of the Full Court) “a similar payment did not amount to hiring the car but did result in the car being used for reward”,[151] and (ii) the judgment in Bonham’s case[152] was powerfully endorsed by the Scottish High Court in Aitken v Hamilton.[153] The Scottish High Court there adopted Uthwatt J’s judgment in Bonham’s case[154] in which His Lordship said: “It appears to me that a distinction falls to be drawn between the word ‘hire’ and the word ‘reward’. The first word necessarily imports, I think, an obligation to pay. The inclusion of the second word is … for the purpose of bringing in a subject matter which does not include hire and including cases where there is no obligation to pay.”[155] 96. The Court Full in AG v Lun Chan[156] did not disturb the magistrate’s finding in respect of carriage of passengers for reward alone and said: “Had it been necessary…, we would have been disposed to follow the decision in Bonham’s case and to adopt the view of the Scottish Judges in Aitken’s case.”[157] In addition, while the case of AG v Lun Chan[158] dealt with the term “hire or reward” as it was used in an insurance policy, the court took the matter further and remarked: “It is … possible that the same expression may have different meanings when used in the context of an Act of Parliament and … of a commercial document; however much one may regret the misunderstandings that may arise if the same expression does not mean the same thing, particularly when parties are entering into obligations that are intended to be related to the requirements of the Act.” [159] 97. Of the case of AG v Lun Chan,[160] a subsequent Full Court in Chan Kai-lap’s case[161] had this to say: “The Full Court [in AG v Lun Chan][162] gave a very careful and lengthy judgment… Not only do we respectfully agree with this authority on the interpretation and construction of this section, but we are further of the opinion that it is binding upon us.” 98. These were considered opinions of high authority, and they coincided with the majority decision of the House of Lords in Albert’s case[163] in 1971, an opinion of even higher stature and authority. Lord Donovan and Lord Pearson held there that “hire or reward” must be read disjunctively; with their Lordships’ judgments Lord Diplock agreed. Lord Pearson provided powerful reasons for his judgment: “In my opinion, there is no need to say that in this composite expression the word ‘hire’ and the word ‘reward’ are synonymous. That would be a surprising statement, because it would mean that the words ‘or reward’ add nothing and might as well have been omitted. It is not the practice to insert unnecessary words in an Act of Parliament. The probable explanation of the composite phrase is that the words ‘for hire’ were used because they are the most familiar words to describe remuneration for carriage in some vehicles and the words ‘or reward’ were added because ‘reward’ is a wider word and apt to cover some forms of remuneration or some arrangements for which the words ‘for hire’ might not be appropriate.” [164] 99. On this issue, the defence points out that the focus of the English provisions and relevant insurance terms had been on “the vehicle” rather than “the driver” or “the act of the driving”; but the focus of section 52(3) is certainly not on the vehicle.[165] This point was made also by Lord Denning in Connell’s case,[166] adopted by Lord Donovan in Albert’s case,[167] but emphatically rejected by Viscount Dilhorne in the same case.[168] True it is that none of the cases examined in this part of the discussion concerned directly a prosecution of private car hire. However, the essential reasoning provided in these judgments for construing “hire or reward” disjunctively is compelling and self-sustained, self-sustained in that the validity of the proposition does not depend on the specific facts of the cases. On this issue, Lord Pearson’s reasoning must prevail. 100. Despite the passage of time, despite the revised regulatory regime, despite the tremendous technological tide, this court concludes that the principle propounded in these precedents concerning the proper construction of “hire or reward” survives; the principle is applicable to the expression “hire or reward” as it appears in section 52(3) of the Road Traffic Ordinance, Cap. 374; and the principle is binding upon this court. 101. In HKSAR v Tse Yee Ping,[169] the Court of Final Appeal said: “The fact that a statute renders a specified activity criminal is not … a reason for giving the definition of the specified activity an artificially narrow meaning. It is merely a reason … for giving the definition the narrower meaning, when it is, on a fair reading, otherwise equally capable of having a wider or narrower meaning.” The disjunctive construction of “hire or reward” has an obvious effect in enlarging the scope of the offence under section 52(3) of the 1982 Ordinance, an enlargement that has a clear and critical consequence on the constitutionality of the charge. An examination of the elements of the section 52(3) offence will provide the necessary introduction to the constitutionality issue that is to follow. The 2nd issue: Elements of the current offence General principle 102. Identification of elements of offence does not usually present much of an issue at trial, but in the instant cases, it turns out to be otherwise. In principle and in practice, the prosecution must prove all elements of the offence; if the offence is a statutory one, the elements must be ascertained in accordance with the words of the offence-creating provision; and the applicable principle is that a penal provision must be construed strictly in favour of the prosecuted. Prosecution submission on the law 103. In respect of the proper construction of an offence-creating provision, the prosecution refers to section 19 of the Interpretation and General Clauses Ordinance, Cap. 1, which provides: “An Ordinance shall be deemed to be remedial and shall receive such fair, large and liberal construction and interpretation as will best ensure the attainment of the object of the Ordinance according to its true intent, meaning and spirit.” 104. The prosecution submits that the court should adopt a purposive approach in regard to the interpretation of section 52(3). In particular, the prosecution stresses that both the context and the purpose of the enactment should be considered, even in the absence of any ambiguity;[170] the context should be taken in its widest sense, including other provisions in the same statute and the then existing state of the law;[171] social and technological changes since the enactment must be taken into account;[172] and the statute should be taken as always speaking.[173] Defence submission on the law 105. The defence highlights the following matters for the court’s consideration. In China Field Ltd v Appeal Tribunal (Buildings) (No.2),[174] the Court of Final Appeal said most forcefully: “[T]here has been a distressing development by the courts which allows them to distort or even ignore the plain meaning of the text and construe the statute in whatever manner achieves a result which they consider desirable. It cannot be said too often that this is not permissible… [I]ntention of the legislature … must be ascertained by a proper application of the interpretative process. This does not permit the Court to attribute to a statutory provision a meaning which the language of the statute, understood in the light of its context and the statutory purpose, is incapable of bearing.”[175] 106. In addition, the defence relies on a powerful judgment, namely, the dissenting judgment of Lord Wilberforce in Royal College of Nursing of the United Kingdom v Department of Health and Social Security,[176] a judgment which has since been regarded as authoritative:[177] “[I]t is proper, and indeed necessary, to have regard to the state of affairs existing, and known by Parliament to be existing, at the time… [W]hen a new state of affairs, or a fresh set of facts bearing on policy, comes into existence, the courts have to consider whether they fall within the Parliamentary intention… [T]here is one course which the courts cannot take… they cannot fill gaps; they cannot by asking the question ‘What would Parliament have done in this current case – not being one in contemplation – if the facts had been before it?’ attempt themselves to supply the answer, if the answer is not to be found in the terms of the Act itself.” Issue in dispute 107. None of the above principles are in dispute in this trial. The prosecution argues that section 52(3) is technologically neutral, and that it applies to private car hire via Uber. The defence urges the court to view section 52(3) in such a way as would have it disapplied when a driver carries passengers in discharge of a contractual duty towards his employer or the like, rather than in performance of a duty under a carrier contract with the passenger. Relevant provisions 108. Before examining the offence-creating provision in section 52(3), several important definitional provisions must be borne in mind. They provide the necessary context without which the offence cannot be accurately analysed. Many of these statutory definitions are relevant also to the constitutional issue, an issue that will be considered below. 109. Insofar as is relevant to present purposes, a “driver” is defined as any persons in charge of or assisting in the control of the vehicle.[178] A “passenger” means all persons on board a vehicle, except the driver.[179] A “motor vehicle” means all forms of road transport that are mechanically propelled.[180] A “private car” is defined primarily with reference to its passenger carrying capacity, but it does not include a taxi.[181] A “taxi” means a motor vehicle registered as such.[182] And a “hire car permit” can be issued only to registered owners of private cars authorising their use for the carriage of passengers for hire or reward.[183] And “hire or reward” is not defined anywhere in the Ordinance. 110. Section 52(3) provides: “No person shall – (a) drive or use a motor vehicle; or (b) suffer or permit a motor vehicle to be driven or used, for the carriage of passengers for hire or reward unless – (i) the vehicle is licensed as a public bus, public light bus or taxi; (ii) the vehicle is licensed as a private bus and the passengers are students, teachers or employees of an educational institution, disabled persons, or persons employed to assist disabled persons; (iia) the vehicle is licensed as a private light bus and is used – (A) as a school private light bus; or (B) exclusively for the carriage of persons who are disabled persons and persons assisting them; or (iii) a hire car permit is in force in respect of the vehicle. Prosecution position untenable 111. In regard to the issue of elements of offence, the prosecution position is not the easiest to understand. At one point, the prosecution says that the guilty conduct is just the driving, and that (i) the carriage of passengers and (ii) the “hire or reward” constitute the objective circumstances of the illegal driving.[184] At another point, the prosecution argues that payment of the reward is not an element of the offence,[185] and that its place in this offence is just to provide a characterisation of the journey as a business or commercial activity.[186] At yet another point, the prosecution submits that the focus should be on “hire or reward”. The prosecution insists that section 52(3) is silent as to mens rea (on this issue of mens rea, see paragraphs 170 to 172 below). 112. The prosecution must have had an offender’s criminal acts equated with actus reus of the offence. This is wrong. The editors of Textbook of Criminal Law (Glanville Williams, 2015) said in Chapter 10: “Although lawyers find the expression actus reus convenient, it is misleading in one respect. It means not just the criminal act but all the external elements of an offence. The term actus reus covers all the physical aspects of a crime including physical events that result from omissions when there is a duty to act.” 113. The prosecution submission in regard to elements of offence must be dismissed. By way of illustration, the court has pointed out during the trial that there is no evidence showing receipt by the defendants of any rewards. On this issue, the prosecution seeks to rely on Orr v Trafalgar Insurance Company Ltd[187] and argues that proof of an obligation to pay would suffice for the element of “hire or reward”. In Orr’s case,[188] the English Court of Appeal said: “[W]hen the question arises … whether … a motor car is being used for the purpose of private hire, that necessarily involves a decision as to whether … there was … an obligation, express or implied, to make payment for the journey in question.”[189] (emphasis supplied) 114. In closing submission, the prosecution asks the court not to attach weight to the Toronto case[190] because of its civil nature, not to attach weight to the Edmonton cases[191] because of its civil nature, and to attach weight to Orr’s case[192] despite its civil nature. The prosecution position is untenable, untenable not because it blows hot and cold, but because it is contrary to legal principles that are basic and fundamental. 115. In criminal law, assumption of an obligation to pay is synonymous with a subjective intention to make payment. No amount of a guilty mind can, without more, result in a criminal conviction unless the crime charged is one of conspiracy, for the prosecution of which offence a meeting of the mind would suffice and proof of overt acts is unnecessary. On the other hand, when an offender has full mens rea to commit an offence but has gone just beyond the preparatory stage of its commission, the proper charge must be one of attempt rather than the complete offence. And a criminal law court must accept that the legislature is, at all times, fully conversant with these principles of criminal law. 116. In the present proceedings, proof of the passenger’s full intention to pay goes nowhere near proving the physical, external element of “hire or reward”. In the court’s view, the prosecution’s reliance on Orr’s case[193] is completely misplaced. The prosecution appears to have – once again – brushed aside the important distinction in criminal law between substantive and inchoate offences (see also paragraphs 75 to 77 above), a consideration that is seldom relevant in a civil dispute, but a distinction that must have been within the knowledge of the legislature. It is entirely within the constitutional charter and competence of the legislature, one must remember, to make inchoate conduct or behaviour a full statutory offence. A good example relevant to the present discussion can be found in section 52(5) of the Ordinance, which reads: “No person shall solicit or attempt to solicit any person for hire or reward to travel in any vehicle which is licensed as – (a) a private car; …” In the court’s view, there is a strong presumption, rebuttable only by the clearest of necessary implications to the contrary, that the legislature legislates for substantive offences, leaving inchoate offences to the professional judgment of the prosecution. Defence position accepted 117. The defence position is that the prosecution is wrong about section 52(3) being silent on mens rea. In a clearly and carefully crafted submission, the defence argues that the subject matter of the offence is the driving of the vehicle by the defendant, and that the driving falls foul of section 52(3) only if it is done with the primary intention of carrying passengers and a secondary intention that the carriage of passengers must be for hire or reward. The defence stresses that this is the proper construction of section 52(3) as the word “for” is used, firstly, between the act of driving and the carriage of passengers and, secondly, between the carriage of passengers and the reward. 118. The defence argument on mens rea cannot be faulted, except that it is the court’s view that the focus of a section 52(3) offence is not on the act of driving but something else (see paragraph 164 below). In any event, the prosecution concedes in closing submission that no prosecution should be instituted against an employed truck driver who had no knowledge that a passenger had been carried, eg., where a person had hidden himself under the truck, or against a salaried chauffeur whose employer had lied to him, saying that the conveyance of passengers was free of charge. Conclusion on elements of offence 119. According to the natural reading of the words used, the physical aspects of a section 52(3) offence, in the context of the present proceedings, are: (i) the defendant must be the driver of a vehicle; (ii) the vehicle must be mechanically propelled; (iii) there must have been at least one passenger (rather than goods) on board the car; (iv) the car must have been driven on a journey (rather than, eg., sliding down a slope due to mechanical failures); (v) there must be a reward paid or received for the carriage of the passenger; and (vi) if the vehicle is a private car, its registered owner must not have, in force, a hire car permit in his favour. 120. Item (vi) is crafted in the form of a negative aversion, with respect to which the prosecution has the benefit of section 94A of the Criminal Procedure Ordinance, Cap. 221. The burden is on the defence to prove the existence of a hire car permit, a matter that is completely within the competence and convenience of the prosecution nowadays. In the present prosecutions, the non-existence of a hire car permit is not disputed. Still, the prosecution has failed to have the issue exhausted by way of a proper section 65C agreement (see paragraphs 25 to 27 above). As no decision on this issue is required of the court, the court will say no more of this issue. 121. In conclusion, except item (vi), the other physical, external aspects of the offence detailed in items (i) to (v) in paragraph 119 above, together with the offender’s full intention and knowledge of them, constitute the elements of the subject offence, which must be proved to the required standard of beyond all reasonable doubt. The 3rd issue: Constitutionality of the subject offence 122. The court is required by law to give a statutory provision a fair, large, and liberal construction to ensure the attainment of the legislative intent (see section 19 of the Interpretation and General Clauses Ordinance, Cap. 1, reproduced in paragraph 103 above).[194] In this formulation of the law, fairness not only comes first but also provides the overarching principle in the application of the section. No large and liberal construction can possibly survive if it is unfair. The issue is what is meant by fairness in this context. In the consideration of an offence-creating provision, there are at least three relevant aspects. 123. To the accused, fairness requires that the presumption in favour of positive proof of mens rea be applied. To the prosecution, fairness may require that failure to comply with prevailing safety standards and the like be strictly or absolutely criminalised. In due administration of criminal justice, fairness may require a penal provision that infringes fundamental rights and liberty be nullified. These three aspects of fairness are all engaged in the present prosecutions. 124. The Hong Kong Bill of Rights Ordinance, Cap. 383, was not promulgated until 8 June 1991. In the court’s view, enhanced vigilance with respect to potential issues of constitutionality is necessary when dealing with pre-1991 legislations like the Road Traffic Ordinance, Cap. 374. The degree of vigilance exhibited in Lam Yuk Fai’s case[195] exemplifies what is required. In that case, the Court of Final Appeal said of a constitutional point: “These points were not argued at the trial or before the Court of Appeal and so were not considered by those courts. They were certified by this Court because of their importance.”[196] Freedom of employment 125. During the trial, the issue is raised whether, in its application to Uber drivers, section 52(3) might infringe the fundamental right and freedom provided in art. 33 of the Basic Law. Art. 33 reads: “Hong Kong residents shall have freedom of choice of occupation.” 126. The defence indicates that the issue has been decided in the Court of Appeal case of Leung Sze Ho Albert v Bar Council of Hong Kong Bar Association.[197] The defence position on this issue is detailed in the written submission for the purpose of preserving the point for future consideration if the occasion presents itself. In consequence, it is unnecessary to take the matte further. Arbitrary arrest, detention or imprisonment (a) Uncertainty and overbreadth 127. Section 52(3) is not uncertain, section 52(3) is too certain and is most certainly too wide. In closing submission, the prosecution argues at length that the offence meets the requirement of legal certainty and is thus constitutional,[198] citing Winnie Lo v HKSAR,[199] Shum Kwok Sher v HKSAR,[200] and B v Commissioner of the Independent Commission Against Corruption.[201] In particular, Mo Yuk Ping v HKSAR[202]has been relied on in support of the submission that ‘The “proportionality test” … has no place in resolving issue of legal uncertainty.’[203] The paragraph cited by the prosecution reads: “79. There is no basis in either authority or principle for using the proportionality test as a criterion of legal certainty… [I]t was not applied in R v Rimmington ... (the public nuisance case). Nor does Committee for the Commonwealth of Canada v Canada ... a case relied upon by the appellant, support the use of the proportionality test otherwise than in the context of constitutional cases where a restriction on a fundamental right or freedom is in issue.” 128. In the court’s view, the prosecution must have missed the strategic use by the Court of Final Appeal of (i) Rimmington’s case[204]and (ii) Committee for the Commonwealth of Canada v Canada[205] to deal with two distinct branches of constitutional law. The former case was one of legal uncertainty, and the latter case “overbreadth”. The two branches of law were highlighted by Woolf J (as he then was) as early as in 1984 in R v Secretary of State for Trade and Industry, ex p Ford,[206] a case that was cited by our Court of Final Appeal in Noise Control Authority v Step In Ltd.[207] His Lordship said: “As was made clear by a series of authorities, uncertainty of language rarely creates the necessary degree of invalidity to cause the courts to intervene. It seems to me that a similar approach has to be applied to uncertainty of application, and especially in the area of consumer protection it is unfortunately inevitable that those who are responsible for supplying the public will at times be put in situation where they will have to make difficult decisions as to whether or not they are infringing the law.” (emphasis supplied) The two constitutional concepts were starkly explained in Mo Yuk Ping’s case[208] itself, the very case on which the prosecution relies: “77. The appellant’s case of overbreadth raises different considerations. Ms Montgomery QC acknowledges that vagueness and overbreadth are two distinct concepts – the intended breadth of a law may be clear yet its application may be overboard.” 129. A comparison between reg. 29(2) and (2A) of the former regime,[209] and section 52(3) of the present Ordinance (reproduced in paragraph 110 above) will show how much wider the scope of prohibition of section 52(3) is. Reg. 29(2) and (2A), long repealed, provided: (2) No person shall drive, use or suffer or permit to be driven or used, any motor vehicle which is registered as – (a) [Deleted, L.N. 161/77] (b) a private omnibus; (c) a goods vehicle; (d) a motor tricycle; or (e) a motor cycle, for the carriage of passengers for hire or reward. (2A) No person shall drive or use, or suffer or permit to be driven or used, any motor vehicle which is registered as a private car for the carriage of passengers for hire or reward unless there is in force in respect of that vehicle a hire car permit issued under regulation 29C, 29E or 29F. 130. Except for those under section 52(3)(i), (ii), (iia) and (iii), private hire of all types of motor vehicles, registered or not, are now prohibited; but under the old law, private hire of only five categories of registered motor vehicles were criminalised. The expansion of the prohibition is, on one view, needed and justified, for it brings within its scope motor vehicles that are not registered at all under the Ordinance. But the expansion must be viewed in the context of the wide definitions of “a motor vehicle” and “a passenger” (see paragraph 109 above). The four exceptions aside, section 52(3) has applications to all motor vehicles, as long as there is a passenger on board, as long as the journey is paid, in cash, in kind, or in kindness;[210] and the section applies regardless of the makes and models of the cars, regardless of their desired or designated use or purpose. 131. In the wordings of section 52(3), it matters not who pays or who receives the reward, whether the reward is paid before, during, or after the ride; the section applies so long as the reward is referable to the ride, and so long as the driver has the necessary knowledge and intention to commit the crime. Prima facie, employed drivers of all motor vehicles of all kinds (except those permitted under section 52(3) itself) who provide carriage to passengers are all caught by this prohibition. In ascertaining the scope of the section 52(3) prohibition, it is important to bear in mind the case of HKSAR v Tse Yee Ping[211] (see paragraph 101 above). 132. A side issue arises in the trial. Does section 52(3) apply to ambulance drivers, police car drivers, etc.? The parties refer the court to section 66 of the Interpretation and General Clauses Ordinance, Cap. 1, and section 3 of the Road Traffic Ordinance, Cap. 374. In brief, the former provision is to the effect that no Ordinances of Hong Kong shall affect the right of or be binding on the State, whereas the latter says certain Parts of the Road Traffic Ordinance, Cap. 374, apply to vehicles and persons in the public service of the State. 133. It is unnecessary for present purposes to examine the issue of state immunity in depth. Section 3 of the Road Traffic Ordinance, Cap. 374, contains a very limited form of vicarious liability on the government in situation where a driver in its employ is prosecuted with certain offences under the Ordinance. The vicarious liability ceases to apply if “it is shown … that only the driver was responsible.”[212] If it is necessary to decide the issue of state immunity, the court is inclined to find that section 52(3) applies to government drivers. It is unreasonable if a driver on government payroll uses a government car to conduct private car hire, he can claim state immunity on the basis that the car he has used happens to be a government car. 134. The extensiveness of the section 52(3) prohibition is indirectly recognised by the defence but dismissively marginalised by the prosecution. The defence says: “Take the example of the chauffeur who is paid to drive his employer’s senior officer and customers. Regardless of the arrangement among his employer and the officers/customers, the chauffeur himself would be driving pursuant to a “commercial arrangement”; he definitely would not be driving on a “social and domestic occasion.”[213] The prosecution for its part says: “Each case will be assessed according to its merits in accordance with the Prosecution Code.”[214] 135. The court is in no position to determine, without reference to the objective that the provision seeks to achieve, whether the section 52(3) prohibition is excessively extensive. What then is the objective of the prohibition against private car hire? The prosecution submits: “It is the government’s aim to provide a safe, efficient, reliable and environment-friendly transport system that meets the community’s economic, social and recreational needs, and that is capable of supporting sustainable development in Hong Kong.”[215] The court shall, for the time being, assume that the prosecution submission is correct, but this issue will be discussed in detail when the court deals with the rationality and proportionality tests below. 136. In this part of the discussion, the case of Canada (Attorney General) v Bedford[216] is instructive, a judgment that appears to this court to be a ground-breaking one, ground-breaking in respect of the common law principle of stare decisis, and ground-breaking in respect of the constitutional law principles of arbitrariness, overbreadth, and gross proportionality. On the issue of overbreadth, the Supreme Court of Canada held: “[101] Another way in which laws may violate our basic values is through what the cases have called “overbreadth”: the law goes too far and interferes with some conduct that bears no connection with its objective… [105] The overarching lesson that emerges from the case law is that laws run afoul of our basic values when the means by which the state seeks to attain its objective is fundamentally flawed, in the sense of being arbitrary, overbroad, or having effects that are grossly disproportionate to the legislative goal… [108] The case law on arbitrariness, overbreadth and gross disproportionality is directed against two different evils. The first evil is the absence of a connection between the infringement of rights and what the law seeks to achieve – the situation where the law’s deprivation of an individual’s life, liberty, or security of the person is not connected to the purpose of the law. The first evil is addressed by the norms against arbitrariness and overbreadth, which target the absence of connection between the law’s purpose and the s.7 deprivation… [112] Overbreadth deals with a law that is so broad in scope that it includes some conduct that bears no relation to its purpose… [113] Overbreadth allows courts to recognize that the law is rational in some cases, but that it overreaches in its effect in others… [123] All three principles – arbitrariness, overbreadth, and gross disproportionality – compare the rights infringement caused by the law with the objective of the law, not with the law’s effectiveness. That is, they do not look to how well the law achieves its object, or to how much of the population the law benefits. They do not consider ancillary benefits to the general population. Furthermore, none of the principles measure the percentage of the population that is negatively impacted. The analysis is qualitative, not quantitative…a grossly disproportionate, overbroad, or arbitrary effect on one person is sufficient…” 137. It appears from the Canadian jurisprudence that the analysis on the issue of overbreadth is a qualitative one. It involves a comparison between human rights infringement and the objective of the law concerned, rather than a consideration of the effectiveness of the law. The chief complaint in a case of overbreadth is that a legal measure has a reach – qualitatively – beyond its legitimate aim. 138. Section 52(3) constitutes, in the view of this court, a case of overbreadth: a significant sector of our driving public, namely, employed drivers, in their daily, income-deriving activities, is put at risk of a prosecution under section 52(3), a legal measure that does not have any connection to the objective of the prohibition (see paragraph 135 above). Various types of employed drivers for carriage of passengers for reward in the form of a salary are necessary for the normal functioning of society; not only should they not be banned from our roads, they must be allowed their full right to make a living by doing so. The next issue is whether a fundamental right is engaged. (b) Constitutionally protected rights and freedom 139. As a matter of logic, when a penal provision overreaches into conducts not having any bearing to its objective, it must necessarily engage the constitutional guarantee in art. 28 of the Basic Law, which provides: “The freedom of the person of Hong Kong residents shall be inviolable. No Hong Kong resident shall be subjected to arbitrary or unlawful arrest, detention or imprisonment. Arbitrary or unlawful search of the body of any resident or deprivation or restriction of the freedom of person shall be prohibited…” 140. At one point, the defence suggests that consideration of arbitrary or unlawful arrest and detention under art.28 goes no further than laws that regulates the power, and the exercise thereof, of arrest and detention, and it does not reach over to the offence for the investigation and prosecution of which the arrest and detention are effected. When a court orders a term of imprisonment, continues the defence, issues of arbitrariness or unlawfulness of the imprisonment can have a further reach than the court order that authorises the deprivation of the prisoner’s liberty; it reaches over to the arbitrariness and unlawfulness of the offence concerned. 141. The court is unable to accept the defence contention. Arrest, detention, and imprisonment are different forms of deprivation of freedom of the person, a freedom that is declared to be inviolable in the opening sentence of art.28. In argument, the court makes the observation that arbitrary or unlawful arrest, detention, and imprisonment are put pari passu in the second sentence of art.28. In further submission, the defence places before the court (for which this court is grateful), among other cases, Lau Cheong & Another v HKSAR,[217] which provides the legal framework for the consideration of arbitrariness or unlawfulness of arrest or detention reaching over to the offence for which the arrest and detention are made. 142. Lau Cheong’s case[218] held: “[41] The terms of BL art.28 differ from those of BOR art.5(1). Article 28 expressly provides a constitutional guarantee against arbitrary “imprisonment” and not just against arbitrary “arrest or detention”… Plainly, “imprisonment” covers incarceration pursuant to a sentence lawfully imposed by a court after a criminal conviction… [42] … Article 28 prohibits not merely “unlawful” imprisonment but “arbitrary or unlawful” imprisonment. It envisages that a term of imprisonment lawfully order may nonetheless be “arbitrary”. It follows that such arbitrariness may reside in the substantive rules of criminal liability whose breach led to the imprisonment ordered. … [44] In Fok Lai Ying v Governor in Council & Others[219] … Lord Cooke … noting: The expression arbitrary interference can also extend to interference provided for under the law. The introduction of the concept of arbitrariness is intended to guarantee that even interference provided for by law should be in accordance with the provisions, aims and objectives of the Covenant and should be, in any event, reasonable in the particular circumstances.” As the defence has helpfully pointed out, Lau Cheong’s case[220] was affirmed and applied by the Court of Final Appeal in So Wai Lun v HKSAR.[221] 143. In conclusion, the constitutional protection provided in art.28 of the Basic Law against arbitrary or unlawful arrest, detention, and imprisonment is engaged and infringed by section 52(3) on ground of overbreadth. To put the matter beyond doubt, apart from arrest and detention, arbitrary or unlawful imprisonment is also engaged as the subject offence does, on a first conviction, carry a maximum term of three months’ imprisonment.[222] (c) Rationality and Proportionality 144. It is a defence submission that “where a clearly prescribe[d] provision is challenged for being overbroad and therefore infringes another constitutionally protected freedom, a proportionality analysis is required.”[223] With this proposition of the defence, this court disagrees. In the court’s view, the rights-infringing provision in the present prosecutions cannot possibly be saved by the application of rationality and proportionality tests; there are several reasons for this. 145. First, subsequent to Bedford’s case[224], the Supreme Court of Canada in Carter v Canada (Attorney General)[225] was again called upon to consider the common law principle of stare decisis as well as the three constitutional principles. The court affirmed and applied the Bedford decision on both scores, and said with respect to overbreadth: “[46] The argument before the trial judge involved a different legal conception of s.7 than that prevailing when Rodriguez was decided. In particular, the law relating to the principles of overbreadth and gross disproportionality had materially advanced since Rodriguez. The majority of this Court in Rodriguez acknowledged the argument that the impugned laws were “over-inclusive” when discussing the principles of fundamental justice (see p.590). However, it did not apply the principle of overbreadth as it is currently understood, but instead asked whether the prohibition was “arbitrary or unfair in that it is unrelated to the state’s interest in protecting the vulnerable, and that it lacks a foundation in the legal tradition and societal beliefs which are said to be represented by the prohibition” (p.595). By contrast, the law on overbreadth, now explicitly recognized as a principle of fundamental justice, asks whether the law interferes with some conduct that has no connection to the law’s objectives (Bedford, at para. 101).[226] This different question may lead to a different answer.” (emphasis supplied) Quite obviously, the Supreme Court of Canada did not support the attempt made by the Rodriguez court in applying what was akin to the rationality and proportionality tests to justify the law in question. 146. It is the view of this court that when a legal measure is overbroad in its scope amounting to overbreadth, the attending infringement of a person’s fundamental rights and liberty cannot possibly be justified on the basis that the measure remains somehow rationally connected with its legitimate aim, or that it remains somehow no more than that which is absolutely necessary to pursue its legitimate aim. The well-known doctrine of margin of appreciation does not even begin to enter into the picture. In the court’s view, the use of the rationality and proportionality tests to justify an overbroad provision is inherently incoherent, and must be wrong in constitutional law. 147. Second, the editors of Archbold Hong Kong (2018)[227] said: “Article 39 of the Basic Law sets out two preconditions to the restriction of rights and freedoms. First, the rights and freedoms in the Basic Law and Bill of Rights may only be restricted as prescribed by the law: Basic Law, Article 39. The expression “prescribed by law” mandates the principle of legal certainty, that the restricting law must be formulated with a sufficient degree of precision so that the individual is given some indication as to how he may regulate his conduct: Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 [60]; Bahadur [228], at [34]. It follows that the court, having found the legislative scheme inconsistent with a Basic Law right (eg right to travel and right to enter), may not apply a proportionality test to uphold the decision against the applicant as this would be imposing a restriction not prescribed by law: Bahadur above, at [34].” (emphasis added) 148. The issue of the court’s jurisdiction to apply the proportionality test in the situation referred to in the paragraph above did not appear to have arisen directly in the two authorities cited therein. It is unnecessary here to make a final determination on the validity of the proposition put forward by the editors of Archbold Hong Kong; this court does not disagree with it but is of the view that another approach in respect of this issue is worth considering. 149. Art. 11 of the Basic Law provides, inter alia: “No law enacted by the legislature of the Hong Kong Special Administrative Region shall contravene this Law.” As noted above, on the basis of overbreadth, section 52(3) infringes the constitutional guarantee provided in art. 28 of the Basic Law, and it contravenes art. 11 by the same token. It goes without saying that no courts of Hong Kong have the jurisdiction to validate a contravention of the Basic Law; any judicial decisions to that effect, either directly or indirectly, must be null and void. In consequence, this court has no jurisdiction to conduct the rationality and proportionality tests to justify an infringement, occasioned by a legislative enactment, of the fundamental rights and liberty protected by the Basic Law. 150. Third, assuming contrary to the above discussion that the court can perform the rationality and proportionality analyses, the burden is upon the prosecution to justify the infringement. On this aspect of the issue, the prosecution has placed before the court material demonstrating the following. 151. First, Hong Kong experienced an expansion in its road system in the 1980s, but the rate of expansion has become stagnant since then. Second, there has been a sharp rise in the number of private cars in recent decades, putting significant stress on the loading of our road system. Third, private cars are inefficient as a means of mass transport. Fourth, the average urban car speed has been low and it is getting lower still. Fifth, severe traffic congestion is noted during peak hours at various locations. Sixth, the mass transit system in Hong Kong has been expanded and is continuing to expand. 152. Despite all the problems that Hong Kong has in its transport system, Hong Kong still “has one of the most efficient [public transport] systems in the world”.[229] In 2014, Hong Kong came “first in the world in the Urban Mobility Index”.[230] 153. It is of course in the interests of society to regulate road use as road systems are expensive and time-consuming to design, build, and maintain. No road systems can be conceived and constructed in vacuum, ie., independently of urban planning, of demographic changes, of financial constraints, and of environmental considerations. Control on road use is a legitimate social concern. The number of private cars has been on the rise. If a significant number of the private cars that would otherwise have stayed in carparks to await its next private use are to occupy our road surface for hire or reward, this can aggravate the presently strained loading of the road system. But the issue is by how much? 154. Uber claimed it expected to have, by the end of 2015, some 15,000 drivers in the Ontario area operating one of the several types of Uber cars alone (see paragraph 2 above). In Uber BV & ors v Aslam & ors,[231] it was not disputed that there were, towards the end of 2016,[232] around 30,000 Uber drivers in the London area and about two million passengers had registered themselves to use Uber’s services. The prosecution submits that the Uber cases decided in Canada and England are irrelevant. But these judgments pinpoint the fact that of evidence capable of showing the extent in Hong Kong of the use of Uber Apps, Driver and Rider, the prosecution has produced none. In conclusion, the prosecution has failed to discharge its burden to justify the infringement. (d) Remedial interpretation 155. The prosecution does not appear to have prepared for an examination of the constitutionality of the subject offence; that is a matter for the prosecution, with which this court has no quarrel. Throughout the trial, the court has repeatedly indicated that the section 52(3) prohibition might be excessively extensive and, subsequent to the close of evidence, invites the parties to address the court on the issue of constitutionality. The prosecution has in effect declined to do so; that again is a matter for the prosecution. 156. The prosecution has properly reminded the court, relying principally on the Court of Final Appeal case of Secretary for Justice v Yan Yuk Lung,[233] of the procedures to follow when the court is minded to declare the subject offence unconstitutional. 157. The effect of that case is, in brief, that where a magistrate makes a finding of unconstitutionality, the magistrate should consider the applicability of section 27 of the Magistrates Ordinance, Cap. 227, to amend the information, by substituting (if need be) the original charge with one that is constitutional valid; but before amending the information, the magistrate should afford the prosecution an adequate opportunity to challenge the finding by way of case stated, rather than proceed to dismiss the information. That approach was followed in Secretary for Justice v Latker.[234] 158. In closing submission, both the prosecution and the defence ask the court to abandon the constitutional analysis, on the basis that if this court is to find that the section 52(3) prohibition infringes a relevant constitutional guarantee, the court cannot, in the parties’ joint view, undertake a proportionality analysis to save the provision. And the defence goes even further. The defence says: ‘As Mo Yuk Ping and Lau Wai Ho makes clear, the analyses of rationality or proportionality do not feature in a legal certainty analysis, nor would this Court be able to exercise any “remedial function.”’[235] (emphasis supplied) 159. In principle, there can be no reasons why any courts of law should sidestep a constitutional issue, the fundamental duty of the court being to administer the law, not any laws, it must be admitted, but laws that are found or generally accepted to be constitutionally valid. Yan Yuk Lung’s case[236] has provided the procedures to follow at this level of the court if a constitutional challenge is sustained. There can be no reasons at all to refrain from making a declaration of unconstitutionality if the due administration of the criminal law requires such a declaration to be made, even if that would mean that the trial process has to be arrested pending a final determination in the higher courts. 160. The defence submission that this court has no jurisdiction to remedy the rights-infringement provision by adopting a Basic Law-compliant interpretation of it must be dismissed. The court’s jurisdiction comes from the Basic Law itself, according to the decision of HKSAR v Lam Kwong Wai and another[237], a case that neither the prosecution nor the defence has referred to in these proceedings. The Court of Final Appeal said: “[67] The next question is: does the Basic Law confer on the courts a power of remedial interpretation? … The function of the courts of the Region is described by or referred to, in the expressions “adjudicate cases” and “adjudicating cases” which are to be found in the Basic Law (arts 84 and 158). [68] The Basic Law neither sets out the powers of the courts nor the remedies which they may grant. The absence of provisions in the Basic Law dealing with these matters is not surprising. Article 83 of the Basic Law provides that the powers and functions of the courts “shall be prescribed by law”. No doubt this provision enables the legislature to confer powers and functions on the courts but it does not exclude the implication of powers and functions from the Basic Law itself. [69] In common law systems, courts enjoy wide-ranging inherent and implied powers and there is no reason to think that the courts of the HKSAR stand as an exception to the generality of this statement. The Basic Law recognizes that the courts of the Region (including this Court) are equipped with powers to grant appropriate remedies. In this respect, there is a distinction between inherent jurisdiction and jurisdiction by implication. When a statute sets up a court with a jurisdiction, it acquires by implication from the statute all powers necessary for its exercise (Grassby v The Queen (1989) 168 CLR 1 at 16-17, per Dawson J). As the courts are established by the Basic Law, the powers which they possess and the remedies which they may grant should be characterized primarily as implied, though some powers to be implied under the Basic Law may be ultimately traced back to the common law. [70] The grant of judicial power and, for that matter, the investing of jurisdiction in a court, carry with them all those powers that are necessary to make effective the exercise of judicial power and jurisdiction so granted. “Necessary”, in this context, means “reasonably required” … These powers will include power to grant and employ such remedies as may be appropriate. … [71] For the purpose of disposing of these appeals, it is necessary only to decide whether the powers of this Court and other courts in the HKSAR including the making of a remedial interpretation of a statutory provision in order to preserve its validity … Such an interpretation involves the well-known techniques of severance, reading in, reading down and striking out. These judicial techniques are employed by the courts of other jurisdictions whose responsibility it is to interpret and pronounce on the validity and compatibility of legislation which is challenged on ground that it contravenes entrenched or statute-based human rights and fundamental freedoms. … [73] In the context of the Basic Law, which arms the HKSAR with a modern constitution including entrenched rights and freedoms, the concept of judicial power necessarily includes the making of remedial interpretations in the sense already discussed. It is recognized as an incident of the exercise of judicial power in other jurisdictions.” 161. Is the magistrates’ court “a court of the HKSAR” equipped with the implied powers to adopt a remedial interpretation of section 52(3)? Many think not, because our legislature has never established such a court. But the magistrates’ court is, in fact, a court of the HKSAR; it has always been since July 1997, according to the Basic Law. Art. 81 provides: “The Court of Final Appeal, the High Court, district courts, magistrates’ courts and other special courts shall be established in the Hong Kong Special Administrative Region. …” 162. The court has not only the power but is in fact under a duty to perform the remedial analysis, to exhaust all possibilities of interpreting the provision in order to make it Basic Law-compliant. Sir Anthony Mason framed the issue in Lam Kwong Wai’s case[238] in these terms: “[57] In essence, the question is whether the courts of the Region have power or, indeed, a duty to so construe s.20(1) when read with s.20(3)(c) as to preserve its validity, even if the interpretation is one which would go beyond ordinary common law interpretation because it may involve the use of judicial techniques such as reading down, reading in and striking out. …” (emphasis supplied) 163. Turning to section 52(3), the offence seeks to prohibit an activity that is economic or financial in nature, and is best characterised as an economic crime. In respect of the types of motor vehicles and the kinds and timing of the payment or receipt of the rewards, there can be no objection for the prohibition to have as wide a coverage as the wordings of the section naturally and reasonably allow. 164. The key feature of this provision is, in the court’s view, the car journey (see item (iv) in paragraph 119 above). Under section 52(3), a driver is caught by this section because he is the driver of the car in the car journey, a passenger is a passenger within the scope of this section because he is the passenger in the car journey, and the reward to be paid or received is a reward within the scope of this section because of the car journey. 165. In common law, it is generally accepted that an act can be done to achieve multiple purposes, for more reasons than one. Where proof of a reason or purpose is required (see item (v) in paragraph 119 above), the prosecution does not have to prove that the illegal reason or purpose was the only reason or purpose for which the defendant committed the guilty act. The prosecution must prove, however, that the defendant has had an operating reason or purpose in mind, one that is prohibited by the law; nothing less would suffice. In this formulation of the common law approach, “operating” means a “significant” or “substantial” reason or purpose contributing to the commission of the guilty conduct; again, nothing less would suffice. 166. In the exercise of the court’s remedial function, it is the judgment of this court that in order to secure a conviction in respect of a section 52(3) offence, the prosecution must prove that (i) the driver has undertaken the journey for the sole reason or purpose that a reward will be paid or received and for nothing else, (ii) the passenger has accepted the carriage for no other reasons but a payment of the reward quid pro quo the journey, in that (a) not only has the reward come about as an incident of the specific car journey concerned and (b) but its calculation must also depend on the distance travelled. On point (b), there had been a difference between “hire or reward” and “hire and reward” in the regulatory regime between 1957 and 1964, in which period, “hire or reward” was used in situations where the reward had depended on the distance of the journey in fact made (see paragraphs 49 and 93 above). 167. The effect of taking this restrictive interpretation of section 52(3) is that if a defendant drives his car in performance of his duty under an employment contract, the section would not apply to him, unless the evidence is such as would allow the court to pierce the “contractual veil” to find beyond all reasonable doubt that the reward received is calculated per ride and by reference to the distance traversed. The burden of proof of a contractual veil, if the issue of an employment contract arises, rests on the prosecution as part of its case on the relationship between the ride and the reward. 168. Further, if a defendant drives his car for carriage of passengers for reward in situation where he would have provided the transport without payment in any event, due to the existence of some special relationship between him and the passenger, section 52(3) would have no application to his case, unless, again, the evidence before the court is such as would allow the court to pierce the “relationship veil” to find beyond all reasonable doubt that the relationship was relied on by the defence to mask the journey-dependent nature of the reward, per ride and as per the distance travelled. The burden to pierce the relationship veil rests likewise on the prosecution. (e) Conclusion on constitutional issue 169. In exercise of the court’s duty pursuant to Lam Kwong Wai’s case,[239] the court finds that section 52(3) is Basic Law-compliant and therefore constitutionally valid when interpreted in accordance with paragraphs 163 to 168 above. 170. The court’s jurisdiction to consider a section 27 amendment arises only when the court is poised to declare a penal provisional unconstitutional, but this court has specifically found the offence constitutional on a remedial interpretation. The charges as laid in the present prosecutions do not require amendment; they are not defective within the meaning of section 27 of the Magistrates Ordinance, Cap. 227, in that there is no defect in the formulation of the charges and there is no variance between the evidence and the information. 171. All that the court has done in these proceedings is to take a restrictive view of the offence-creating provision, and this is done in order to ensure that it is Basic Law-compliant. In this regard, there is no difference between the present prosecutions and a case in which the trial court adopts a restrictive interpretation of a penal provision under the common law. There are no reasons to arrest pronouncement of judgments in the instant cases. The 4th issue: strict liability 172. The issue of strict liability is straightforward for two reasons: the law is settled (see, in particular, HKSAR v Choi Wai Lun),[240] and the prosecution position is completely misconceived. 173. There is a common law presumption that proof of mens rea is required: “[I]nstead of demanding a justification for reading in a mens rea requirement, the modern starting-point is that mens rea is presumed to be an essential ingredient where the statute is silent on the mental element unless that presumption is displaced expressly or by necessary implication.” [241] 174. Modern approach in interpretation of statute never requires the legislature to adopt a formulaic formulation to signify that mens rea is expressly required. The use of words like “knowing”, “with intent”, etc., is never a must. A good example is provided in section 52(3)(b) of the Road Traffic Ordinance, Cap. 374 (see paragraph 110 above), which contains a permitting and suffering offence. The natural meaning of the words “permit” and “suffer”, by themselves and without more, imports the notion of knowledge, for how can anyone permit or suffer an act to be done without knowing that the act is to be or is being done? The same analysis applies to the word “for” which is twice used in section 52(3)(a). 175. The prosecution position that section 52(3) is silent on mens rea must be wrong (see also paragraphs 111 to 118 above). This conclusion is supported by an examination of the legislative history and by the existence of a reverse onus provision in the former as well as the revised regulatory regime. Under the former regime, the prohibition against private car hire was provided in reg. 26(2), or reg. 29(2A) as reg. 26(2) was later to become. To continue this discussion, reference is now made to the reverse onus provision in section 6 of the 1957 Ordinance, which section was first introduced into the regime in 1958, which read: “(1) Any regulation made under this Ordinance may provide that in any criminal proceedings for a contravention thereof – (a) it shall be for the person charged with such contravention to prove certain facts; or (b) facts may be presumed, with or without proof of other facts, until the contrary is proved. (2) Any regulation made in the exercise of the powers conferred by subsection (1) shall be subject to the approval of the Legislative Council. 176. When the road traffic regime was revised in 1982, the legislature reproduced section 6 of the 1957 Ordinance and had it re-enacted as section 14 of the 1982 Ordinance. Nothing can be clearer that the legislature has intended, and still intends, to reserve to itself the right to enact offences with a reverse onus of proof and with a rebuttable presumption of facts. There are two significant points here. First, not every Ordinance providing for regular as well as regulatory offences comes with a reverse onus provision. Second, in the 1982 enactment, the legislature has removed the prohibition against private car hire from the regulations and has it installed in the principal Ordinance instead. Nothing can be clearer that the legislature has never intended that the prosecution in a criminal trial on a section 52(3) charge is to have in its favour the benefit of a reverse onus, or is to have the benefit or of any presumptions of facts (except the negative averment in connection with the hire car permit, or more precisely, with the lack of it). 177. Assuming that the court is wrong on this issue, assuming that section 52(3) is silent on mens rea, is the presumption of mens rea displaced? On this issue, it is evident that the prosecution has misunderstood Hin Lin Yee v HKSAR.[242] The prosecution makes three major points. First, the prosecution argues that Hin Lin Yee’s case[243] emphasises that the application of the conventional guiding principles in imposing strict liability “necessarily depends on what alternatives exist to take the place of the mens rea requirement, should it be supplanted?”[244] There are altogether five such alternatives now. Second, there is “generally less need to feel inhibited about overriding the presumption in relation to what may compendiously be called ‘regulatory offences’.” [245] Thirdly, the prosecution says that imposing strict liability in respect of private hire car can assist in the enforcement of the section 52(3) prohibition, “for there are many kinds of case where putting on the prosecutor full burden of proving mens rea creates great difficulties and may lead to many unjust acquittals…”[246] 178. The prosecution has misunderstood the effects of Hin Lin Yee’s case.[247] The reason that all courts of Hong Kong must now consider the five alternatives is to avoid situations like those confronted by the English court in R v Warner[248] and Sweet v Parsley,[249] of the latter of which our Court of Final Appeal in Hin Lin Yee’s case[250] said: “[59] In the same case, Lord Pearce again expressed his regret at having no intermediate option. Having suggested that an approach requiring the defendant to persuade the jury that he did not know might in some cases be the best solution, he continued: ‘If it were possible in some so-called absolute offences to take this sensible halfway house, I think that the courts should do so. This has been referred to in Warner’s case … I see no difficulty in it apart from the opinion of Viscount Sankey LC in Woolmington v Director of Public Prosecutions … But so long as the full width of that opinion is maintained, I see difficulty. There are many cases where the width of that opinion has caused awkward problems.’ ” 179. The Court of Final Appeal repeated this point in Choi Wai Lun’s case:[251] “[35] … A court may recoil from imposing absolute liability for a particular offence and so hold against displacement, while it may be prepared to hold that the presumption is supplanted in favour of the second or third Kulemesin alternative. The decision in R v K[252] that the offence of indecent assault requires proof of full mens rea, should be viewed in that context. With no intermediate option, it is not surprising that the House of Lords ruled in favour of requiring mens rea instead of absolute liability.” 180. Nowhere in the judgments of our Court of Final Appeal cited before this court has it been ruled that the conventional guiding principles for finding strict liability are no longer relevant, nowhere in those judgments has it been decided that the prosecution no longer needs to identify the relevant matters referred to in Hin Lin Yee’s case:[253] “[44] … The case-law employs the same concepts and terminology in discussing the mental requirements, postulating the need, for example, to examine the statutory language and purpose; the nature and seriousness of the offence and its attendant penalties and social stigma; the utility of imposing sanctions; the prevailing societal conditions; and so forth.” (emphasis supplied) In fact, in Choi Wai Lun’s case,[254] the court specifically noted: “[37] However, given the manifest purpose of section 122(2) viewed in the light of the Hong Kong courts’ long-standing policy regarding age-related sexual offence, I am of the view that the presumption of mens rea is clearly displaced in respect of indecent assault on persons under the age of 16 in this jurisdiction.” (emphasis supplied) 181. When pressed in argument, the prosecution submits that the legislative aim of section 52(3) is to reduce improper road use, and that strict liability is required because of a perceived difficulty on the part of the prosecution to prove the defendant’s knowledge that the journey is paid. On the defence side, it is submitted that whatever the legislative purpose of section 52(3) is, it cannot be for the protection of the taxi industry. 182. To prevent improper road use so as to reduce traffic congestion is of course a legitimate aim to pursue, as is the protection of any given industry (eg., the legal services industry). But the ultimate objection to the prosecution position is that a section 52(3) offence does not have the nature and effect of a regulatory offence at all. The nature and effect of a regulatory offence are explained in Hin Lin Yee’s case,[255] the same case on which the prosecution relies: ‘[160] Regulatory offences do not as a rule involve conduct falling within the first of the aforementioned categories where absolute liability may be justified. The conduct being regulated generally forms an essential part of social life – the provision of food and drink, the operation of machinery, working on construction sites, selling pharmaceuticals and so forth. The policy of the law is not to say: “Proceed with these activities at your peril”, but to say: “When doing these things, you must meet proper standards aimed at promoting public health, safety and well-being.’ 183. A section 52(3) offence does not have any of the regulatory characteristics referred to in the above paragraph; it is just an ordinary economic crime. In the view of this court, it is a regular offence rather than a regulatory offence, and this explains why in 1982 the legislature moved the offence from the regulations to the main Ordinance. 184. As to the enforcement issue, it is a matter of inference arising from the evidence adduced whether a defendant has had the necessary guilty mind. In the context of private car hire, the only enforcement problem, as far as this court can see, arises from the lack of a victim of crime. Unless the passenger is dissatisfied with the services that the driver provides, neither a Uber driver nor his passenger would have much of a reason to report the crime. Imposing strict liability on the driver cannot be justified on the basis of the passenger’s unwillingness to make a report to the police. Regardless of the driver’s liability being full or strict, the passenger, a willing participant in the driver’s crime, will remain unwilling to report the crime. Conclusion on strict liability 185. Having considered the five available alternatives, in view of the legislative history, in light of the statutory language and purpose, it is the judgment of this court that the section 52(3) offence requires full proof of mens rea and is not a strict liability offence. The 5th issue: Findings in respect of the present prosecutions 186. The prosecution has provided the court (for which the court is grateful) with a detailed and well-referenced schedule containing a summary of evidence in each of the 28 prosecutions. With the defence consent, the schedule is made an annexure to this consolidated “Reasons for Verdict”, and the court orders that the schedule do form part of the judgment of the court. The court finds proved and established beyond all reasonable doubt the facts, acts, and circumstances arising from the agreed evidence contained in the schedule. 187. Before analysing the facts proved and drawing inferences from them, the court observes that according to the Bankers’ Affirmations placed before the court as part of the prosecution case, the payments and receipt of the rewards must have, in the first instance, taken place overseas. The Bankers’ Affirmation were compiled in accordance with the provision of Section 20 of the Evidence Ordinance, Cap. 8. 188. Section 20 is a powerful and useful provision of our law, but it is not powerful enough to make all inadmissible hearsay evidence admissible. One of the requirements of a Bankers’ Affirmation is for the deponent to vouch for the safe use and operational state of the computers involved in the processing and reproduction of the relevant bank records. Such a deponent is obviously in no position to make comments on a computer that is physically outside Hong Kong. Banks can of course exist and operate in one jurisdiction alone, but they all conduct overseas transactions. The prosecution might wish, for future purposes, to give further thoughts to the issue of proof of overseas bank records and transactions. 189. There is no evidence before the court showing any of the defendants had provided carriage to the passengers in fulfilment of a contractual duty owed to Uber. There is no evidence capable of severing the fares from the rides, eg., where a defendant was entitled to a salary not calculated by reference to the occurrences of, and to the distances to be covered in, the rides, or where the defendant would be entitled to his salary even though he was unable to provide the carriage services, eg., due to sickness or when he was granted paid leave. 190. In the context of the present prosecutions, the court finds (i) the conversations between the defendants and the undercover officers and (ii) the cautioned responses of the defendants do not take the respective cases of the parties any further. 191. On the facts found in these cases, the court draws the following only reasonable and irresistible inferences: (i) The 28 defendants responded to the passengers’ requests for transport services via Uber; they turned up at the pick-up points on purpose, by design, rather than as a matter of coincidence; (ii) The fares were journey-dependent, firstly, as to their calculation because the passengers were shown the routes and the fares before the selection of the cars to be used, before the commencement of the journeys, and, secondly, because the fares were paid almost immediately after the rides; (iii) The defendants and the passengers were strangers to each other; the passengers were passengers in the defendants’ cars solely because the journeys were intended to be paid; (iv) The fares were charged to the passengers’ credit cards accounts in Hong Kong on the day of the offences, so payments of the fares had made; (v) There is no evidence tending to show any of the defendants was at the time of the offences performing a contractual duty owed to Uber or anyone. Even if there had been such a contractual duty, the court would find, on the basis of the evidence adduced, beyond all reasonable doubt that the contract must have been entered into to mask the interdependence between the fares and the journeys; (vi) Based on the findings in (i) to (v) above, the court draws the following two further, only reasonable and irresistible inferences – (a) The defendants had allowed the passengers, who were strangers, to get in their cars solely for the reason and purpose of the car rides that were to be paid; and (b) The defendants must have known and intended that the journeys were to be paid. 192. Section 52(3) does not require the reward to be paid specifically by the passenger or received specifically by the driver, but specific knowledge and intention on the part of the driver that the journey would come or did come with a reward must be proved beyond all reasonable doubt; and where the reward was in fact paid, the full offence is complete and committed. CONCLUSION 193. Momentous technological breakthroughs never take account of prevailing public policies, but prevailing public policies must take account of momentous technological breakthroughs. Material placed before this court shows that the problem of personalised and point-to-point passenger services has caught the attention of the administration in the course of 2015.[256] There is no evidence before the court showing the current extent of the problem in Hong Kong, so the court will make no comments on this issue. In any event, this court hopes that the relevant authorities – to borrow a phrase from Lord Denning – “will soon remedy the position.” 194. Having considered the evidence, the extensive and in-depth submissions by the parties (for which the court is grateful), and in view of the findings in paragraphs 189 to 192 above, the court finds beyond all reasonable doubt that all the elements of the offences have been established against the 28 defendants, and the court finds them guilty of the offence they each face in their respective case. Summary of Evidence Part A. The 1st to the 23rd cases [N.B. “3412/PW1” refers to PW1 in KCCC 3412/2017 “3412/Exh. P1 at [3]” refers to paragraph 3 of Exh. P1 in KCCC 3412/2017] (1) KCCC 3412/2017 (Defendant: YUONG Ho-cheung) 1. 3412/Exh. P9 at [2] & [4] 2. At 3:28 p.m. on 28 April 2017, 3412/PW1 disguising as a passenger arrived at No. 3 Suffolk Road in Kowloon Tong. 3412/PW1 logged in the Rider App on her mobile phone and requested for “uberX” carriage service by setting her location as the pick-up location and Cheong Yin House of Nam Cheong Estate as the destination. The Rider App displayed a map showing 3412/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Ho Cheung”, a photograph depicting the Defendant, a vehicle registration mark “FX9812” and a description of “Honda Freed”. At all material times, the motor vehicle displaying registration mark FX 9812 (“Car 1”) was a Honda Freed private car. 3412/Exh. P9 at [4]- [6] 3412/Exh. P3(11) 3412/Exh. P1 at [1(a)] 3. At 3:47 p.m., 3412/PW1 received on her mobile phone a telephone call of number 91449120 from a male. The male told 3412/PW1 that he was the Uber driver and hadparked his vehicle on Suffolk Road near Tim Fuk Road behind a yellow lorry. Meanwhile, PW1 saw Car 1 behind a yellow lorry on the opposite side of Suffolk Road. 3412/Exh. P9 at [10] 4. Immediately thereafter, 3412/PW1 boarded Car 1 from the rear door. When the Defendant asked 3412/PW1 whether the destination was Cheong Yin House of Nam Cheong Estate, 3412/PW1 answered in the affirmative. Afterwards, the Defendant drove Car 1 off with 3412/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 1 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3412/PW1’s mobile phone kept updating. During the journey, the Defendant told 3412/PW1 that the income as an Uber driver was not bad. 3412/Exh. P9 at [10] 3412/Exh. P3(16)- (18) 3412/Exh. P9 at [10] 5. At 4 p.m. during the journey, Car 1 was briefly intercepted by a police officer. Thereafter, the Defendant continued the journey and further discussed with 3412/PW1 about Uber’s fare and taxi fare. 3412/Exh. P9 at [11]-[12] 6. By 4:06 p.m. the same day, Car 1 driven by the Defendant and carrying 3412/PW1 arrived outside Cheong Yin House of Nam Cheong Estate. Shortly afterwards, 3412/PW1 received an electronic bill showing a map of the journey, the fare of “$56.64”, “Yi Lo, thank you for choosing Uber” and an Uber logo. The Rider App on her mobile phone also displayed a page where 3412/PW1 could give a rating on a scale of five stars. Bank records confirm that a payment of HK$56.64 was made in favour of “UBER HK” the same day from 3412/PW1’s credit card. 3412/Exh. P9 at [13]-[15] 3412/Exh. P3(21)- (23) 3412/Exh. P8 7. At 7:57 a.m. on 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 91449120, which was registered by the Defendant. The phone also had the “Uber Driver” mobile application (“the Driver App”). 3412/Exh. P1 at [2]- [3] 3412/Exh. P7 3412/Exh. P4(3) & P5(3) 8. As at 28 April 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 1. 3412/Exh. P1 at [8] (2) KCCC 3413/2017 (Defendant: YIP Wai-ming) 9. Prior to the operation on 28 April 2017, DPC 6821 (“3413/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “Cc lai”, his email address, his mobile phone number and details of his credit card including the card number. 3413/Exh. P12 at [3] 10. At 4:42 p.m. on 28 April 2017, 3413/PW1 disguising as a passenger arrived at the roundabout at Aqua Marine near No. 101 Sham Mong Road in Cheung Sha Wan. 3413/PW1 logged in the Rider App on his mobile phone and requested for “UberBLACK” carriage service by setting his location as the pick-up location and Baptist Hospital as the destination. The Rider App displayed a map showing 3413/PW1’s location and that there was a small car in the vicinity. Below the map there was a vehicle registration mark “KR5929” and a description of “Toyota Alphard”. At all material times, the motor vehicle displaying registration mark KR 5929 (“Car 2”) was a Toyota Alphard private car. 3413/Exh. P12 at [5]-[6] 3413/Exh. P3(9) 3413/Exh. P1 at [1(a)] 11. Telephone number 95866352 appeared when 3413/PW1 tried to contact the driver through the Rider App. At 4:49 p.m., 3413/PW1 received on his mobile phone a telephone call of number 67490225 from a male. The male told 3413/PW1 that he was the Uber driver and he would arrive in a few minutes. 3413/Exh. P12 at [6] 3413/Exh. P13 at [4] 12. About 4:54 p.m. the same day, 3413/PW1 saw Car 2 arriving at the roundabout at Aqua Marine. 3413/PW1 boarded Car 2 and sat in the rear passenger seat. Inside Car 2, 3413/PW1 confirmed with the Defendant that the destination was Baptist Hospital. After that, the Defendant drove Car 2 off with 3413/PW1 on board. Throughout the journey, the small car which showed the real-time location of Car 2 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3413/PW1’s mobile phone kept updating. During the journey, the Defendant said that he was a full-time Uber driver and that driving for Uber could barely meet the ends. 3413/Exh. P12 at [8] 3413/Exh. P3(12)- (17) & P12 at [9] 3413/ Exh. P12 at [8] 13. At 5:17 p.m. during the journey, Car 2 was intercepted briefly by a police officer. Before Car 2 came to a stop, the Defendant asked 3413/PW1 for his surname and said 3413/PW1 was his friend. 3413/Exh. P12 at [9] 14. About 5:28 p.m. the same day, Car 2 driven by the Defendant and carrying 3413/PW1 arrived at Baptist Hospital. After alighting from Car 2, 3413/PW1 received an email, confirming the fare of the journey to be $173.68, which was charged on 3413/PW1’s credit card. 3413/PW1 also rated the journey on the Rider App. Bank records confirm that a payment of HK$173.68 was made to “UBER HK” the same day from 3413/PW1’s credit card. 3413/Exh. P12 at [11]-[12] 3413/Exh. P3(20)- (21) & P4(1)-(3) 3413/Exh. P3(18) 3413/Exh. P11 15. On 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 95866352, which was registered by the Defendant. The phone had the Driver App, in which there were records including fare record of the subject journey taken by 3413/PW1 on Car 2. 3413/Exh. P1 at [2]- [3] 3413/Exh. P9 16. Between about 9:35 a.m. and about 11:40 a.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver in March 2017 and that he would receive the fare of the journeys from Uber through a bank account. 3413/Exh. P1 at [9] 3413/Exh. P7 at Q&A 21 & 29 17. As at 28 April 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 2. 3413/Exh. P1 at [10] (3) KCCC 3414/2017 (Defendant: YIP Ka-shing) 18. Prior to the operation on 29 April 2017, WDPC 6684 (“3414/PW1”) successfully registered an account for using the Rider App on her mobile phone by providing an account identifier “Yi Lo Chan”, her mobile phone number, her email address and her credit card number. 3414/Exh. P10 at [2] & [4] 19. At 2:43 p.m. on 29 April 2017, 3414/PW1 disguising as a passenger was outside Cheung Sha Wan Plaza at No. 4 Cheung Shun Street. 3414/PW1 logged in the Rider App on her mobile phone and requested for “uberX” carriage service by setting her location as the pick-up location and Metro Harbour View as the destination. The Rider App displayed a map showing 3414/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Ka Shing”, a photograph depicting the Defendant, a vehicle registration mark “PB5366” and a description of “Toyota Alphard”. At all material times, the motor vehicle displaying registration mark PB 5366 (“Car 3”) was a Toyota Alphard private car. 3414/Exh. P10 at [4]-[5] 3414/Exh. P3(9) 3414/Exh. P1 at [1(a)] 20. At 2:47 p.m., 3414/PW1 received on her mobile phone a telephone call of number 65015116 from a male. The male told 3414/PW1 that he was the Uber driver and confirmed 3414/PW1’s exact location. The male told 3414/PW1 that he was on the way. 3414/Exh. P10 at [6] 21. At 2:55 p.m., the Defendant drove Car 3 to No. 7 Cheung Shun Street. Meanwhile, 3414/PW1 received on her mobile phone another telephone call of number 65015116 from the male, saying that his car had arrived. 3414/PW1 boarded Car 3 from the rear passenger door. 3414/PW1 acknowledged the destination to be Metro Harbour View. Afterwards, the Defendant drove Car 3 off with 3414/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 3 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3414/PW1’s mobile phone kept updating. During the journey, the Defendant told 3414/PW1 that the business fluctuated and it was relatively quiet in the afternoon. 3414/Exh. P10 at [7] 3414/Exh. P3(12)- (13) & P10 at [8] 3414/Exh. P10 at [7] 22. At 3:18 p.m. the same day, Car 3 driven by the Defendant and carrying 3414/PW1 arrived at Metro Harbour View. After alighting from Car 3, 3414/PW1 received an electronic bill showing a map of the journey, the fare of “$64.46”, “Yi Lo, thank you for choosing Uber” and an Uber logo. The Rider App on her mobile phone also displayed a page where 3414/PW1 could give a rating on a scale of five stars. Bank records confirm that a payment of HK$64.46 was made to “UBER HK” the same day from 3414/PW1’s credit card. 3414/ Exh. P10 at [10]-[11] 3414/Exh. P3(15) 3414/Exh. P3(19) 3414/Exh. P9 23. On 23 May 2017, the Defendant was arrested. The Defendant’s mobilephonewasseized fromhim. The phone contained a SIM card for telephonenumber 65015116, which was registered by the Defendant’s brother. The phone had the “Uber Partner” mobile application (“the Partner App”). 3414/Exh. P1 at [2]- [3] 3414/Exh. P1 at [1(c)] & P8 3414/Exh. P4(2)-(3) 24. Between about 9:47 a.m. and about 12:45 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver in October 2016 and that he would receive the fare of the journeys from Uber through a bank account. 3414/Exh. P1 at [8] 3414/Exh. P6 at Q&A 8, 9 & 19 25. As at 29 April 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 3. 3414/Exh. P1 at [9] (4) KCCC 3415/2017 (Defendant: CHOW Kwok-kwong Alan) 26. Prior to the operation on 29 April 2017, DPC 6821 (“3415/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “Cc lai”, his mobile phone number, his email address and his credit card number. 3415/Exh. P10A at [2] & [7] 27. At 3:47 p.m. on 29 April 2017, 3415/PW1 disguising as a passenger was outside Coronation Circle in Yau Ma Tei. 3415/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting his location as the pick-up location and Homantin Plaza as the destination. The Rider App displayed a map showing 3415/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “KWOK KWONG ALA[N]”, a photograph depicting the Defendant, a vehicle registration mark “TZ3408” and a description of “Volkswagen Golf”. At all material times, the private car displaying registration mark TZ 3408 (“Car 4”) was a Volkswagen Golf private car. 3415/Exh. P10A at [7]-[8] 3415/Exh. P3(10) 3415/Exh. P1 at [1(a)] 28. While waiting for Car 4 to come, 3415/PW1 received on his mobile phone two telephone calls of number 97890523 from a male. The male said that he was the Uber driver and confirmed the pickup location. The male also said that he needed some time to arrive at the pick-up location. 3415/Exh. P10A at [8] & P11 at [3] 29. At 4:04 p.m., the Defendant drove Car 4 to the drop-off area of Coronation Circle. 3415/PW1 raised his hand to signal the Defendant. Car 4 then stopped in front of 3415/PW1. Immediately thereafter, 3415/PW1 boarded Car 4 and sat in the rear passenger seat. Afterwards, the Defendant drove Car 4 off with 3415/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 4 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3415/PW1’s mobile phone kept updating. During the journey, the Defendant told 3415/PW1 that he purchased Car 4 for driving as an Uber driver, and he worked as a full-time Uber driver. The Defendant also said that he drove 8 to 10 hours per day and earned some $10,000 per month. 3415/Exh. P10A at [9] 3415/Exh. P3(14)- (16) 3415/Exh. P10A at [9] 30. At 4:30 p.m. during the journey, Car 4 was intercepted briefly by a police officer. Before Car 4 came to a stop, the Defendant told 3415/PW1 that they were friends. 3415/Exh. P10A at [10] 31. At 4:35 p.m. the same day, Car 4 driven by the Defendant and carrying 3415/PW1 arrived at the post office in Ho Man Tin Estate. After alighting from Car 4, 3415/PW1 received an email, showing a map of the journey, the fare of “$43.29”, “cc, thank you for choosing Uber” and an Uber logo. 3415/PW1 was also invited to give a rating on the Rider App. Bank records confirm that a payment of HK$43.29 was made to “UBER HK” the same day from 3415/PW1’s credit card. 3415/Exh. P10A at [11] 3415/Exh. P3(18)- (20) 3415/Exh. P3(17) 3415/Exh. P9 32. About 8:00 a.m. on 23 May 2017, Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 97890523, which was registered by the Defendant. The phone had the Driver App, in which there were records including fare record of the subject journey taken by 3415/PW1 on Car 4. 3415/Exh. P1 at [2]- [3] 3415/Exh. P8 3415/Exh. P5(16)- (17) 33. Between about 9:42 a.m. and about 12:05 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he would receive the fare of the journeys from Uber through a bank account and he had registered with Uber as a driver in July 2016. 3415/Exh. P1 at [8] 3415/Exh. P6 at Q&A 13 & 18 34. As at 29 April 2017, there was not any hire car permit within the meaning of section 2 of the Road Traffic Ordinance, Cap. 374 in force in respect of Car 4. 3415/Exh. P1 at [9] (5) KCCC 3416/2017 (Defendant: CHAN Pak-kay Andrew) 35. Prior to the operation on 2 May 2017, DPC 6821 (“3416/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “Cc lai”, his mobile phone number, his email address and his credit card number. 3416/Exh. P10 at [2] & [4] 36. At 11:43 a.m. on 2 May 2017, 3416/PW1 disguising as a passenger, arrived at Station Lane in Hung Hom. 3416/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting his location as the pick-up location and Golden Computer Arcade as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Andrew Pak Kay”, a photograph depicting the Defendant, a vehicle registration mark “GK8927” and a description “Volkswagen Polo”. At all material times, the private car displaying registration mark GK 8927 (“Car5”)was a Volkswagen Polo private car. 3416/Exh. P10 at [4]-[5] 3416/Exh. P3(12) 3416/Exh. P1 at [1(a)] 37. Afterwards, 3416/PW1 received on his mobile phone a telephone call of number 91930682 from a male. The male said that he was the Uber driver and confirmed the pick-up location. The male also said that he needed more time to arrive at the pick-up location. 3416/Exh. P10 at [6] 38. At 11:54 a.m., the Defendant drove Car 5 to Station Lane in Hung Hom. 3416/PW1 boarded Car 5 and sat in the rear passenger seat. After that, the Defendant drove Car 5 off with 3416/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 5 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3416/PW1’s mobile phone kept updating. During the journey, the Defendant told 3416/PW1 that he had been an Uber driver for 3 years and worked as a full-time Uber driver with monthly income of around $30,000. 3416/Exh. P10 at [7] 3416/Exh. P3(16), (18) & (19) 39. By 12:32 p.m. the same day, Car 5 driven by the Defendant and carrying 3416/PW1 arrived at Fuk Wa Street outside Golden Computer Arcade in Sham Shui Po. After alighting from Car 5, 3416/PW1 received an electronic bill showing a map of the journey, the fare of “$49.51”, “cc, thank you for choosing Uber” and an Uber logo. The Rider App on his mobile phone also showed a fare record of the journey. Bank records confirm that a payment of HK$49.51 was made to “UBER HK” the same day from 3416/PW1’s credit card. 3416/Exh. P10 at [9] 3416/Exh. P3(22)- (23) 3416/Exh. P3(20)- (21) 3416/Exh. P9 40. On 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 91930682, which was registered by the Defendant. The phone had the Driver App and the telephone record between the Defendant and 3416/PW1. 3416/Exh. P1 at [2]- [3] 3416/Exh. P8 3416/Exh. P5(2)-(4) 41. Between about 10:36 a.m. and about 12:35 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he would receive the fare of the journeys from Uber through a bank account and he had registered with Uber as a driver for two years. 3416/Exh. P1 at [8] 3416/Exh. P6 at Q&A 10, 11 & 20 42. As at 2 May 2017, there was not any hire car permit within the meaning of section 2 of the Road Traffic Ordinance, Cap. 374 in force in respect of Car 5. 3416/Exh. P1 at [9] (6) KCCC 3417/2017 (Defendant: LAM Yan-ming) 43. Prior to the operation on 2 May 2017, DPC 5844 (“3417/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “tong chan”, his mobile phone number, his email address and his credit card number. 3417/Exh. P8 at [3] 44. At 12:43 p.m. on 2 May 2017, 3417/PW1 disguising as a passenger arrived at Yen Chow Street near Apliu Street in Sham Shui Po. 3417/PW1 logged in the Rider App on his mobile phone and requested for “UberBLACK” carriage service by setting his location as the pick-up location and One Silversea as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Yan Ming”, a photograph depicting the Defendant, a vehicle registration mark “CR882” and a description “Mercedes-Benz E-Class”. At all material times, the private car displaying registration mark CR 882 (“Car 6”) was a Mercedes-Benz E-Class private car. 3417/Exh. P8 at [5] 3417/Exh. P3(13) 3417/Exh. P1 at [1(a)] 45. Meanwhile, the telephone number 55115678 appeared when 3417/PW1 tried to contact the driver through the Rider App. 3417/Exh. P8 at [5] 46. At 12:52 p.m., the Defendant drove Car 6 to Yen Chow Street near Apliu Street in Sham Shui Po. 3417/PW1 boarded Car 6 and sat in the rear passenger seat. After that, the Defendant drove Car 6 off with 3417/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 6 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3417/PW1’s mobile phone kept updating. 3417/Exh. P8 at [6] 3417/Exh. P3(17)- (20) 47. Later, Car 6 driven by the Defendant and carrying 3417/PW1 arrived at the roundabout of One Silversea. After alighting from Car 6, 3417/PW1 received an electronic bill showing a map of the journey, the fare of “$51.68”, “tong, thank you for choosing Uber” and an Uber logo. The Rider App on his mobile phone also displayed a page where 3417/PW1 could give a rating on a scale of five stars. Bank records confirm that a payment of HK$51.68 was made to “UBER HK” the same day from 3417/PW1’s credit card. 3417/Exh. P8 at [6] 3417/Exh. P3(23)- (25) 3417/Exh. P3(22) 3417/Exh. P7 48. On 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 55115678, which was registered by the Defendant. The phone had the Driver App, in which there were records including fare record of the subject journey taken by 3417/PW1 on Car 6. 3417/Exh. P1 at [2]- [3] 3417/Exh. P6 3417/Exh. P4(11)- (12) 49. Between about 10:18 a.m. and about 2:08 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he would receive the fare of the journeys from Uber through a bank account and he had registered with Uber as a driver for about half a year. 3417/Exh. P1 at [7] 3417/Exh. P5 at Q&A 10 & 18 50. As at 2 May 2017, there was not any hire car permit within the meaning of section 2 of the Road Traffic Ordinance, Cap. 374 in force in respect of Car 6. 3417/Exh. P1 at [8] (7) KCCC 3418/2017 (Defendant: WONG Wai-keung) 51. Prior to the operation on 2 May 2017, DPC 5844 (“3418/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “tong chan”, his mobile phone number, his email address and his credit card number. 3418/Exh. P9 at [3] 52. At 1:22 p.m. on 2 May 2017, 3418/PW1 disguising as a passenger arrived outside The Coronation, No. 1 Yau Cheung Road, Yau Ma Tei. 3418/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting his location as the pick-up location and Kwong Wah Hospital as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Wai Keung”, a photograph depicting the Defendant, a vehicle registration mark “UH3512” and a description “Suzuki Solio”. At all material times, the private car displaying registration mark UH 3512 (“Car 7”) was a Suzuki Solio private car. 3418/Exh. P9 at [5] 3418/Exh. P3(10) 3418/Exh. P1 at [1(a)] 53. Meanwhile, telephone number 92235727 appeared when 3418/PW1 tried to contact the driver through the Rider App. 3418/Exh. P9 at [5] 54. At 1:35 p.m., the Defendant drove Car 7 to The Coronation, No. 1 Yau Cheung Road, Yau Ma Tei. 3418/PW1 boarded Car 7 and sat in the rear passenger seat. After that, the Defendant drove Car 7 off with 3418/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 7 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3418/PW1’s mobile phone kept updating. 3418/Exh. P9 at [6] 3418/Exh. P3(16)- (19) 55. Later, Car 7 driven by the Defendant and carrying 3418/PW1 arrived at Kwong Wah Hospital. After alighting from Car 7, 3418/PW1 received an electronic bill showing a map of the journey, the fare of “$25.00”, “tong, thank you for choosing Uber” and an Uber logo. The Rider App on his mobile phone also displayed a page where 3418/PW1 could give a rating on a scale of five stars. Bank records confirm that a payment of HK$25.00 was made to “UBER HK” the same day from 3418/PW1’s credit card. 3418/Exh. P9 at [6] 3418/Exh. P3(22)- (24) 3418/Exh. P3(21) 3418/Exh. P8 56. On 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 92235727, which was registered by the Defendant. The phone had the Driver App, in which there were records of an account registered under the Defendant’s name. 3418/Exh. P1 at [2]- [3] 3418/Exh. P7 3418/Exh. P5(3)-(6) 57. At about 9:30 a.m. to 9:40 a.m. on 23 May 2017, DPC 8460 searched Car 7 and found a white card at the storage compartment at the driver’s door of Car 7 which stated “Uber invitation code: Sq8V8PHMVE”. 3418/Exh. P10A at [6] 58. Between about 12:00 noon and about 2:08 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he would receive the fare of the journeys from Uber through a bank account and he had registered with Uber as a driver in around November 2016. 3418/Exh. P1 at [8] 3418/Exh. P6 at Q&A 17 and 31 59. As at 2 May 2017, there was not any hire car permit within the meaning of section 2 of the Road Traffic Ordinance, Cap. 374 in force in respect of Car 7. 3418/Exh. P1 at [9] (8) KCCC 3419/2017 (Defendant: TSANG Kwok-ming) 60. Prior to the operation on 5 May 2017, DPC 5844 (“3419/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “tong chan”, his mobile phone number, his email address and his credit card number. 3419/Exh. P9 at [3] 61. At 12:48 p.m. on 5 May 2017, 3419/PW1 and DPC 6821 (“3419/PW2”) disguising as passengers arrived outside No. 1 Po Lun Street, Mei Foo. 3419/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting his location as the pick-up location and Tsim Sha Tsui YMCA at No. 41 Salisbury Road as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Kwok Ming”, a photograph depicting the Defendant, a vehicle registration mark “RL6961” and a description “Honda Stepwgn”. At all material times, the private car displaying registration mark RL 6961 (“Car 8”) was a Honda Stepwgn private car. 3419/Exh. P9 at [5] 3419/Exh. P3(10) 3419/Exh. P1 at [1(a)] 62. Meanwhile, telephone number 94394995 appeared when 3419/PW1 tried to contact the driver through the Rider App. 3419/Exh. P8 at [5] 63. At 1:05 p.m., the Defendant drove Car 8 to the location outside No. 1 Po Lun Street. 3419/PW1 and 3419/PW2 boarded Car 8 and sat in the rear passenger seats. After that, the Defendant drove Car 8 off with 3419/PW1 and 3419/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 8 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3419/PW1’s mobile phone kept updating. During the journey, the Defendant told 3419/PW1 and 3419/PW2 that he worked as a part-time Uber driver with monthly income of around $8,000 to $9,000. 3419/Exh. P8 at [6] 3419/Exh. P3(16)- (20) 3419/Exh. P9 at [4] 64. Later, Car 8 driven by the Defendant and carrying 3419/PW1 and 3419/PW2 arrived at Tsim Sha Tsui YMCA. After alighting from Car 8, 3419/PW1 received an electronic bill showing a map of the journey, the fare of “$51.74”, “tong, thank you for choosing Uber” and an Uber logo. The Rider App on his mobile phone also displayed a page where 3419/PW1 could give a rating on a scale of five stars. Bank records confirm that a payment of HK$51.74 was made to “UBER HK” the same day from 3419/PW1’s credit card. 3419/Exh. P8 at [6] 3419/Exh. P3(25)- (27) 3419/Exh. P3(22) 3419/Exh. P7 65. On 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 94394995, which was registered by the Defendant. The phone had the Partner App. 3419/Exh. P1 at [2]-[3] 3419/Exh. P6 3419/Exh. P5(2) 66. As at 5 May 2017, there was not any hire car permit within the meaning of section 2 of the Road Traffic Ordinance, Cap. 374 in force in respect of Car 8. 3419/Exh. P1 at [8] (9) KCCC 3420/2017 (Defendant: HUI Kwok-wai) 67. Prior to the operation on 5 May 2017, DPC 5844 (“3420/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “tong chan”, his mobile phone number, his email address and his credit card number. 3420/Exh. P8 at [3] 68. At 1:48 p.m. on 5 May 2017, 3420/PW1 and DPC 6821 (“3420/PW2”) disguising as passengers were outside MTR Tsim Sha Tsui Station on Middle Road. 3420/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting No. 15 Middle Road as the pick-up location and Laguna Verde as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Kwok Wai”, a photograph depicting the Defendant, a vehicle registration mark “UL5126” and a description of “Toyota Spade”. At all material times, the motor vehicle displaying registration mark UL 5126 (“Car 9”) was a Toyota Spade private car. The telephone number 90138831 also appeared when 3420/PW1 tried to contact the driver through the Rider App. 3420/Exh. P8 at [5] 3420/Exh. P3(11) 3420/Exh. P1 at [1(a)] 3420/Exh. P8 at [5] 69. At 1:54 p.m., the Defendant drove Car 9 to the pick-up location. 3420/PW1 and 3420/PW2 boarded Car 9 and sat in the rear passenger seats. The Defendant asked if 3420/PW1 and 3420/PW2 were police officers, and if so, whether or not they could let him go because he had earned only some $200 for picking up customers in the morning. 3420/PW1 and 3420/PW2 denied that they were police officers. The Defendant then drove Car 9 off with 3420/PW1 and 3420/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 9 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3420/PW1’s mobile phone kept updating. 3420/Exh. P8 at [6] 3420/Exh. P9 at [4] 3420/Exh. P3(15)- (18) & P8 at [6] 70. At 2:14 p.m. the same day, Car 9 driven by the Defendant and carrying 3420/PW1 and 3420/PW2 arrived at Laguna Verde. After 3420/PW1 and 3420/PW2 alighting from Car 9, the Rider App on 3420/PW1’s mobile phone displayed a page where he could give a rating on a scale of five stars. The Rider App also showed a receipt setting out the details of the fare. Shortly afterwards, 3420/PW1 received an electronic bill showing a map of the journey, the fare of “$28.48”, “tong, thank you for choosing Uber”, an Uber logo and that the fare had been charged on 3420/PW1’s credit card. Bank records confirm that a debit transaction of HK$28.48 was made to “UBER BV” the same day from 3420/PW1’s credit card. 3420/ Exh. P8 at [6] 3420/Exh. P3(20) 3420/Exh. P3(21)- (22) 3420/Exh. P3(23)- (25) 3420/Exh. P7 71. On 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 90138831, which was registered by the Defendant. 3420/Exh. P1 at [2]- [3] 3420/Exh. P6 72. As at 5 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 9. 3420/Exh. P1 at [9] (10) KCCC 3421/2017 (Defendant: LEE Hing-lung Gary) 73. Prior to the operation on 23 May 2017, DPC 9510 (“3421/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “Paul Chan”, his mobile phone number, his email address and his credit card number. 3421/Exh. P8 at [4]-[5] 74. At 7:15 a.m. on 23 May 2017, 3421/PW1 and PC 15059 (“3421/PW2”) disguising as passengers were outside No. 9 College Road in Kowloon City. 3421/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting No. 9 College Road as the pick- up location and Fire and Ambulance Services Academy as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Hing Lung Gary”, a photograph depicting the Defendant, a vehicle registration mark “GL1608” and a description of “BMW 1-series”. At all material times, the motor vehicle displaying registration mark GL 1608 (“Car 10”) was a BMW 1-series private car. 3421/Exh. P8 at [7]- [8] 3421/Exh. P3(7) 3421/Exh. P3(7) at [1(a)] 75. At 7:20 a.m., 3421/PW1 received on his mobile phone a telephone call of number 97707060 from a male. The male told 3421/PW1 that he was the Uber driver and confirmed 3421/PW1’s exact location. The male told 3421/PW1 that he almost arrived. 3421/Exh. P8 at [11] 76. At 7:25 a.m., the Defendant drove Car 10 to No. 9 College Road in Kowloon City. 3421/PW1 and 3421/PW2 boarded Car 10 and sat in the rear passenger seats. 3421/PW1 acknowledged the destination to be the Fire and Ambulance Services Academy. Afterwards, the Defendant drove Car 10 off with 3421/PW1 and 3421/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 10 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3421/PW1’s mobile phone kept updating. 3421/Exh. P8 at [12] 3421/Exh. P3(10)- (15) 77. At 7:39 a.m. the same day while Car 10 driven by the Defendant and carrying 3421/PW1 and 3421/PW2 on the way to the Fire and Ambulance Services Academy, it was intercepted by the police. Later, 3421/PW1 received an electronic bill showing a map of the journey, a breakdown of the fare of “$160.83”, “Paul, thank you for choosing Uber”, an Uber logo and that the fare had been charged on 3421/PW1’s credit card. The Rider App on his mobile phone also showed a record of the journey. Bank records confirm that a debit transaction of HK$160.83 was made to “UBER BV” the same day from 3421/PW1’s credit card. 3421/ Exh. P8 at [13]-[14] 3421/Exh. P3(16)- (19) 3421/Exh. P3(20)- (22) 3421/Exh. P7 78. Meanwhile, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 97707060, which was registered by the Defendant. The phone had the Driver App, in which there were records including details and charging rate of the subject journey taken by 3421/PW1 and 3421/PW2 on Car 10. 3421/Exh. P1 at [2]- [3] 3421/Exh. P6 3421/Exh. P4(3)-(7) 79. Between about 12:22 p.m. and about 3:23 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver for eight months and that he would receive the fare of the journeys from Uber. 3421/Exh. P1 at [7] 3421/Exh. P5 at Q&A 4, 11 & 20 80. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 10. 3421/Exh. P1 at [8] (11) KCCC 3422/2017 (Defendant: LAU Kin-fung Bosco) 81. Prior to the operation on 23 May 2017, DPC 33681 (“3422/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing his mobile phone number, his email address and his credit card number. 3422/Exh. P8 at [3] 82. At 7:09 a.m. on 23 May 2017, 3422/PW1 and PC 11875 (“3422/PW2”) disguising as passengers were outside Festival Walk at No. 80 Tat Chee Avenue in Kowloon Tong. 3422/PW1 logged in the Rider App on his mobile phone and requested for and chose a seven-seater carriage service by setting No. 80 Tat Chee Avenue as the pick-up location and Apple Daily Limited as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Kin Fung”, a photograph depicting the Defendant, a vehicle registration mark “UG801” and a description of “Toyota Estima”. At all material times, the motor vehicle displaying registration mark UG 801 (“Car 11”) was a Toyota Estima private car. 3422/Exh. P8 at [6]- [7] 3422/Exh. P3(4) 3422/Exh. P1 at [1(a)] 83. At 7:25 a.m., 3422/PW1 received on his mobile phone a telephone call of number 60182142 from a male. The male told 3422/PW1 that he was an Uber driver and confirmed the pick-up location. The male told 3422/PW1 that he would soon arrive. 3422/Exh. P8 at [9] 84. At 7:28 a.m., the Defendant drove Car 11 to the pick-up location. 3422/PW1 and 3422/PW2 boarded Car 11 and sat in the rear passenger seats. Afterwards, the Defendant drove Car 11 off with 3422/PW1 and 3422/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 11 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3422/PW1’s mobile phone kept updating. 3422/Exh. P8 at [10] 3422/Exh. P3(8)- (13) 85. At 7:50 a.m. the same day while Car 11 was on the way to Apple Daily Limited, it was intercepted by a police officer. The Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 60182142, which was registered by the Defendant. The phone had the Partner App, in which there were records including details and charging rate of the subject journey taken by 3422/PW1 and 3422/PW2 on Car 11. 3422/Exh. P8 at [10]-[11] 3422/Exh. P1 at [2]- [3] 3422/Exh. P6 3422/Exh. P4(3)-(5) 86. Later the same day, 3422/PW1 received an electronic bill showing a map of the journey, the fare of “$178.29”, “Man, thank you for choosing Uber”, an Uber logo and that the fare had been charged on 3422/PW1’s credit card. The Rider App on his mobile phone also showed a record of the journey and invited him to give a rating on a scale of five stars. Bank records confirm that a payment of HK$178.29 was made to “UBER” the same day from 3422/PW1’s credit card. 3422/Exh. P3(15) & P10 at [2] 3422/Exh. P3(16) & (18)-(21) 3422/Exh. P7 87. Between about 12:25 p.m. and about 4:05 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver for about one year and that he would receive the fare of the journeys from Uber through a bank account. 3422/Exh. P1 at [7] 3422/Exh. P5 at Q&A 8, 25 & 28 88. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 11. 3422/Exh. P1 at [8] (12) KCCC 3423/2017 (Defendant: TSE Kee-bo) 89. Prior to the operation on 23 May 2017, DPC 8023 (“3423/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing his mobile phone number, his email address and his credit card number. 3423/Exh. P8 at [3] 90. At 7:10 a.m. on 23 May 2017, 3423/PW1 and PC 14071 (“3423/PW2”) disguising as passengers were at MTR Kowloon Tong Station near Suffolk Road. 3423/PW1 logged in the Rider App on his mobile phone and requested for carriage service by setting Suffolk Road as the pick-up location and Tai Chik Sha Fire Station as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “KEE BO”, a photograph depicting the Defendant, a vehicle registration mark “UF1206” and a description of “Nissan Serena”. At all material times, the motor vehicle displaying registration mark UF 1206 (“Car 12”) was a Nissan Serena private car. 3423/Exh. P8 at [7]- [9] 3423/Exh. P3(1) 3423/Exh. P1 at [1(a)] 91. At 7:21 a.m., 3423/PW1 phoned 92215200. A male answered the call and confirmed that he was the driver of UF 1206, would arrive in a few minutes and would call on his arrival. At 7:29 a.m., 3423/PW1 received on his mobile phone a telephone call of number 92215200 from the male who said he had arrived outside a kindergarten on Suffolk Road. 3423/Exh. P8 at [10]-[11] 92. At 7:31 a.m., Car 12 driven by the Defendant was parked outside a kindergarten at No. 8 Suffolk Road. 3423/PW1 and 3423/PW2 boarded Car 12 and sat in the rear passenger seats. The Defendant confirmed the destination to be Tai Chik Sha Fire Station and drove Car 12 off with 3423/PW1 and 3423/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 12 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3423/PW1’s mobile phone kept updating. 3423/Exh. P8 at [12] 3423/Exh. P9 at [3] 3423/Exh. P8 at [12] 3423/Exh. P3(5)-(6) 93. At 7:55 a.m. the same day while Car 12 was on the way to Tai Chik Sha Fire Station, it was intercepted by a police officer. The Defendant was then arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 92215200, which was registered by the Defendant. The phone had the Uber application, in which there were records including details and charging rate of the subject journey taken by 3423/PW1 and 3423/PW2 on Car 12. 3423/Exh. P8 at [13] 3423/Exh. P1 at [2]- [3] 3423/Exh. P6 3423/Exh. P4(3)-(5) 94. Later the same day, the Rider App on 3423/PW1’s mobile phone displayed a receipt showing an itemized receipt of the journey taken on Car 12 and that the fare of HK$166.20 had been charged on 3423/PW1’s credit card. 3423/PW1 was also invited to give a rating on a scale of five stars. Bank records confirm that a payment of HK$166.20 was made to “UBER HK” the same day from 3423/PW1’s credit card. 3423/Exh. P8 at [15] 3423/Exh. P3(7)- (11) 3423/Exh. P7 95. Between about 12:25 p.m. and about 3:25 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver in November 2016 and that he would receive the fare of the journeys from Uber through a bank account. 3423/Exh. P1 at [7] 3423/Exh. P5 at Q&A 8, 12 & 13 96. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 12. 3423/Exh. P1 at [8] (13) KCCC 3424/2017 (Defendant: TONG Po-hin) 97. Prior to the operation on 23 May 2017, DSPC 52856 (“3424/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “Mike Lau”, his mobile phone number, his email address and his credit card number. 3424/Exh. P7 at [3] & [6] 98. At 7:16 a.m. on 23 May 2017, 3424/PW1 and PC 51798 (“3424/PW2”) disguising as passengers were outside St Teresa’s Hospital on Lomond Road in Kowloon City. 3424/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by No. 20 Lomond Road as the pick-up location and LOHAS Park as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Po Hin”, a photograph depicting the Defendant, a vehicle registration mark “UJ6399” and a description of “BMW 3-series”. At all material times, the motor vehicle displaying registration mark UJ 6399 (“Car 13”) was a BMW 3-series. 3424/Exh. P7 at [5]- [6] 3424/Exh. P3(5) 3424/Exh. P1 at [1(a)] 99. At 7:22 a.m., the Defendant drove Car 13 to the pick-up location. 3424/PW1 and 3424/PW2 boarded Car 13 and sat in the rear passenger seats. Afterwards, the Defendant drove Car 13 off with 3424/PW1 and 3424/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 13 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3424/PW1’s mobile phone kept updating. 3424/Exh. P7 at [7] 3424/Exh. P8A at [6] 3424/Exh. P3(9)- (13) & P7 at [7] 100. At 7:37 a.m. the same day while Car 13 was on the way to LOHAS Park, it was intercepted by a police officer. The Defendant was then arrested. The Defendant’s mobile phone was seized from him. The phone had the Uber application, in which there were records including details and charging rate of the subject journey taken by 3424/PW1 and 3424/PW2 on Car 13. 3424/Exh. P9A at page 2 3424/Exh. P1 at [2]- [3] 3424/Exh. P4(3)-(5) 101. Shortly afterwards, 3424/PW1 received an electronic itemized bill showing the details including the fare of the journey was “$146.51” which had been charged on 3424/PW1’s credit card. The Rider App on his mobile phone also displayed details of the journey and invited him to give a rating on a scale of five stars. Bank records confirm that a payment of HK$147.98 (including 1% credit card charge) was made to “UBER HK” the same day from 3424/PW1’s credit card. 3424/Exh. P1 at [2]- [3] 3424/Exh. P3(14)- (18) 3424/Exh. P3(19)- (21) 3424/Exh. P6 102. Between about 12:22 p.m. and about 3:30 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver in March to April 2017 and that he would receive the fare of the journeys from Uber through a bank account. 3424/Exh. P1 at [7] 3424/Exh. P5 at Q&A 7, 8, 16 & 17 103. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 13. 3424/Exh. P1 at [8] (14) KCCC 3425/2017 (Defendant: WONG Sai-ming) 104. Prior to the operation on 23 May 2017, DPC 58604 (“3425/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “wong lachi”, his mobile phone number, his email address and his credit card number. 3425/Exh. P10 at [2] 105. At 6:45 a.m. on 23 May 2017, 3425/PW1 and WDPC 9143 (“3425/PW2”) disguising as passengers were outside St Teresa’s Hospital near Prince Edward Road West in Kowloon City. 3425/PW1 logged in the Rider App on his mobile phone and requested for and chose a UberBLACK carriage service by setting 173 Boundary Street as the pick-up location and Concordia International School as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Sai Ming”, a photograph depicting the Defendant, a vehicle registration mark “RR2839” and a description of “Toyota Vellfire”. At all material times, the motor vehicle displaying registration mark RR 2839 (“Car 14”) was a Toyota Vellfire private car. 3425/Exh. P10 at [6] 3425/Exh. P3(12) 3425/Exh. P1 at [1(a)] 106. At 7:08 a.m., 3425/PW1 received on his mobile phone a telephone call of number 62898379 from a male. The male told 3425/PW1 that he was the Uber driver and confirmed 3425/PW1’s exact location. The male told 3425/PW1 that he needed about 15 minutes to arrive. At 7:15 a.m., 3425/PW1 received another telephone call from the male, asking 3425/PW1’s exact location and saying that he was on the way. 3425/Exh. P10 at [10] 3425/Exh. P10 at [12] 107. Shortly after 7:16 a.m., the Defendant drove Car 14 to the outside of St Teresa’s Hospital. 3425/PW1 and 3425/PW2 boarded Car 14 and sat in the rear passenger seats. 3425/PW1 confirmed the destination with the Defendant. Afterwards, the Defendant drove Car 14 off with 3425/PW1 and 3425/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 14 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3425/PW1’s mobile phone kept updating. During the journey, the Defendant told 3425/PW1 how to better set the pick-up location. 3425/Exh. P10 at [13] 3425/Exh. P3(15) & P10 at [14] 3425/Exh. P10 at [13] 108. At 7:26 a.m. the same day, Car 14 driven by the Defendant and carrying 3425/PW1 and 3425/PW2 arrived outside Concordia International School. 3425/PW1 and 3425/PW2 alighted from Car 14. Shortly afterwards, 3425/PW1 received an electronic bill showing a map of the journey, the fare of “$56.82”, “wong, thank you for choosing Uber”, an Uber logo and that the fare had been charged on 3425/PW1’s credit card. Bank records confirm that a payment of HK$56.82 was made to “UBER HK” the same day from 3425/PW1’s credit card. The Rider App on his mobile phone also displayed a page where 3425/PW1 could give a rating on a scale of five stars. 3425/ Exh. P10 at [15] 3425/Exh. P3(23)- (26) 3425/Exh. P9 3425/Exh. P3(17) 109. Shortly after the journey on 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 62898379, which was registered by a business ran by the Defendant’s friend. The phone had the Driver App, in which there were records including fare record of the subject journey taken by 3425/PW1 and 3425/PW2 on Car 14 and the telephone records with 3425/PW1. 3425/Exh. P1 at [2]- [3] 3425/Exh. P1 at [1(c)], P6 at Q&A 4 & P8 3425/Exh. P5(6), (7), (10), (18) & (19) 110. Between about 9:50 a.m. and about 12:20 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver in September 2016 and that he had been informed of how to receive remuneration. 3425/Exh. P1 at [8] 3425/Exh. P6 at Q&A 19 & 22 111. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 14. 3425/Exh. P1 at [9] (15) KCCC 3426/2017 (Defendant: LI Sui-leung) 112. Prior to the operation on 23 May 2017, DPC 6821 (“3426/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “Cc lai”, his mobile phone number, his email address and his credit card number. 3426/Exh. P10 at [3] & [5] 113. At 7:23 a.m. on 23 May 2017, 3426/PW1 and WDPC 6613 (“3426/PW2”) disguising as passengers were at the pick-up/drop-off area outside Olympian City 2 in Tai Kok Tsui. 3426/PW1 logged in the Rider App on his mobile phone and requested for “UberBLACK” carriage service by setting Olympian City 2 as the pick-up location and Concordia International School as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Sui Leung”, a photograph depicting the Defendant, a vehicle registration mark “PE660” and a description of “Toyota Estima”. At all material times, the motor vehicle displaying registration mark PE 660 (“Car 15”) was a Toyota Estima private car. 3426/Exh. P10 at [5]-[8] 3426/Exh. P3(14) 3426/Exh. P1 at [1(a)] 114. Telephone number 95673152 appeared when 3426/PW1 tried to contact the driver through the Rider App. About 7:28 a.m., 3426/PW1 received on his mobile phone a telephone call of number 95673152 from a male. The male told 3426/PW1 that he was the Uber driver and confirmed 3426/PW1’s exact location. The male told 3426/PW1 that he would arrive in about two minutes. 3426/Exh. P10 at [8] 115. At 7:32 a.m., the Defendant drove Car 15 to the pick-up/drop-off area outside Olympian City 2. 3426/PW1 and 3426/PW2 boarded Car 15 and sat in the rear passenger seats. 3426/PW2 confirmed the destination with the Defendant. Afterwards, the Defendant drove Car 15 off with 3426/PW1 and 3426/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 15 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3426/PW1’s mobile phone kept updating. During the journey, the Defendant told 3426/PW1 that he was a part-time Uber driver and the income from UberBLACK, but not uberX, could cover the costs of driving Car 15. 3426/Exh. P10 at [9] 3426/Exh. P11 at [7] 3426/Exh. P3(17)- (18) & P10 at [9] 3426/Exh. P10 at [9] 116. At 7:50 a.m. the same day, Car 15 driven by the Defendant and carrying 3426/PW1 and 3426/PW2 arrived outside Concordia International School. 3426/PW1 and 3426/PW2 alighted from Car 15. The Rider App on 3426/PW1’s mobile phone displayed a page where he could give a rating on a scale of five stars. The Rider App also showed records of the journey. Shortly afterwards, 3426/PW1 received an electronic bill showing a map of the journey, the fare of “$67.76”, “cc, thank you for choosing Uber”, an Uber logo and that the fare had been charged on 3426/PW1’s credit card. Bank records confirm that a payment of HK$67.76 was made to “UBER HK” the same day from 3426/PW1’s credit card. 3426/Exh. P10 at [10] 3426/Exh. P3(19)- (22) 3426/Exh. P3(23)- (28) 3426/Exh. P9 117. Shortly after the journey on 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 95673152, which was registered by the Defendant. The phone had the Driver App and the telephone record with 3426/PW1. 3426/Exh. P1 at [2]- [3] 3426/Exh. P8 3426/Exh. P5(2)-(3) 118. Between about 9:22 a.m. and about 12:45 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver in March 2017 and that he would receive fare of the journeys by bank transfers. 3426/Exh. P1 at [8] 3426/Exh. P6 at Q&A 3 & 12 119. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 15. 3426/Exh. P1 at [9] (16) KCCC 3427/2017 (Defendant: LEE Kwok-leung) 120. Prior to the operation on 23 May 2017, DPC 5844 (“3427/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “tong chan”, his mobile phone number, his email address and his credit card number. 3427/Exh. P12 at [3] 121. At 7:28 a.m. on 23 May 2017, 3427/PW1 and WDPC 8648 (“3427/PW2”) disguising as passengers were at Shining Heights, No. 83 Sycamore Street in Prince Edward. 3427/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service to No. 31 Begonia Road. The Rider App displayed a map showing the officers’ location and that there was a small car in the vicinity. Below the map there was the name “Kwok Leung”, a photograph depicting the Defendant, a vehicle registration mark “PR2801” and a description of “Nissan Tiida”. At all material times, the motor vehicle displaying registration mark PR 2801 (“Car 16”) was a Nissan Tiida private car. 3427/Exh. P12 at [6] 3427/Exh. P4(10) 3427/Exh. P1 at [1(a)] 122. At 7:32 a.m., 3427/PW1 received on his mobile phone a telephone call of number 95500834 from a male. The male told 3427/PW1 that he was the Uber driver and confirmed 3427/PW1’s exact location. The male told 3427/PW1 that he was on his way. 3427/Exh. P12 at [7] 123. At 7:47 a.m., the Defendant drove Car 16 to the outside of Shining Heights. 3427/PW1 and 3427/PW2 boarded Car 16 and sat in the rear passenger seats. 3427/PW1 confirmed the destination with the Defendant. Afterwards, the Defendant drove Car 16 off with 3427/PW1 and 3427/PW2 on board. Throughout the journey, the small car which showed the real-time location of Car 16 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3427/PW1’s mobile phone kept updating. 3427/Exh. P12 at [8] 3427/Exh. P7(17)- (20) 124. At 8:01 a.m. the same day, Car 16 driven by the Defendant and carrying 3427/PW1 and 3427/PW2 arrived at No. 31 Begonia Road. 3427/PW1 and 3427/PW2 alighted from Car 16. The Rider App on 3427/PW1’s mobile phone displayed a page where he could give a rating on a scale of five stars. The Rider App also showed records of the journey. Shortly afterwards, 3427/PW1 received an electronic bill showing a map of the journey, the fare of “$45.05”, “tong, thank you for choosing Uber”, an Uber logo and that the fare had been charged on 3427/PW1’s credit card. Bank records confirm that a debit transaction of HK$45.05 was made to “UBER HK” the same day from 3427/PW1’s credit card. 3427/Exh. P12 at [8] 3427/Exh. P4(21)- (24) 3427/Exh. P4(25)- (28) 3427/Exh. P11 125. Shortly after the journey on 23 May 2017, the Defendant was arrested. Under caution at the scene, the Defendant said he drove with Uber to see if he could make a living, as he was in debt. The Defendant’s mobile phones were seized from him. One of the phones (“3427/Phone A”) contained a SIM card for telephone number 95500834, which was registered by the Defendant. Another mobile phone (“3427/Phone B”) had the Driver App, in which there were records including fare record of the subject journey taken by 3427/PW1 and 3427/PW2 on Car 16. 3427/Exh. P1 at [2]- [3] 3427/Exh. P10 3427/Exh. P7(7)-(8) & (13) 126. Between about 9:15 a.m. and about 12:30 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had registered with Uber as a driver a week before and that he would receive fare of the journeys by bank transfers from Uber. 3427/Exh. P1 at [9] 3427/Exh. P8 at Q&A 3, 8 & 9 127. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 16. 3427/Exh. P1 at [10] (17) KCCC 3428/2017 (Defendant: CHUNG Tze-chun Ivan) 128. Prior to the operation on 23 May 2017, DPC 4810 (“3428/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “PO TONG CHENG”, his mobile phone number, his email address and his credit card number. 3428/Exh. P7 at [3] 129. At 7:07 a.m. on 23 May 2017, 3428/PW1 disguising as a passenger was outside Panda Hotel on Tsuen Wah Street in Tsuen Wan. 3428/PW1 logged in the Rider App on his mobile phone and requested for and chose “Ching Ying Yau Bo (transliteration)” carriage service to Terminal 2 of Hong Kong International Airport. The Rider App displayed a map showing 3428/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Ivan Tze Chun”, a photograph depicting the Defendant, a vehicle registration mark “TT8115” and a description of “Kia Morning”. At all material times, the motor vehicle displaying registration mark TT 8115 (“Car 17”) was a KIA Morning private car. 3428/Exh. P7 at [5]- [6] 3428/Exh. P3(2) 3428/Exh. P1 at [1(a)] 130. At 7:15 a.m., the Defendant drove Car 17 to Panda Hotel on Tsuen Wah Street. The Defendant alighted from Car 17 and asked 3428/PW1 if the latter had requested for Uber. 3428/PW1 answered in the affirmative. The Defendant then assisted to put 3428/PW1’s luggage into the baggage compartment of Car 17. 3428/PW1 boarded Car 17 and sat in the rear passenger seat. 3428/PW1 confirmed the destination with the Defendant. Afterwards, the Defendant drove Car 17 off with 3428/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 17 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3428/PW1’s mobile phone kept updating. 3428/Exh. P7 at [8]- [10] 3428/Exh. P3(4)-(9) 131. At 7:41 a.m. the same day, Car 17 driven by the Defendant and carrying 3428/PW1 arrived outside Terminal 2 of Hong Kong International Airport. 3428/PW1 alighted from Car 17. The Defendant assisted 3428/PW1 to unload the luggage from Car 17. 3428/Exh. P7 at [11] 132. The Defendant was arrested shortly afterwards. The Defendant’s mobile phone was seized from him. The phone had the Uber application, in which there were records of an account registered under the Defendant’s name. Records of an account registered under the Defendant’s name, details of Car 17 and an income record could also be accessed through a web browser on the phone. 3428/Exh. P1 at [2]- [3] 3428/Exh. P4(5)- (12) 133. Meanwhile, 3428/PW1 received an email receipt showing a map of the journey, the fare of HK$281.73 and an Uber logo. The Rider App on 3428/PW1’s mobile phone displayed records of the journey, including an itemized receipt containing a map of the journey, the fare of HK$281.73 and an Uber logo. Bank records confirm that a payment of HK$281.73 was made to “UBER HK” the same day from 3428/PW1’s credit card. 3428/Exh. P3(11)- (15) 3428/Exh. P6 134. Between about 9:20 a.m. and about 1:40 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had become an Uber driver for a few weeks and that he would be paid by Uber by bank transfers. 3428/Exh. P1 at [7] 3428/Exh. P5 at Q&A 4, 11 & 28 135. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 17. 3428/Exh. P1 at [8] (18) KCCC 3429/2017(Defendant: CHEUNG Yuk-fung Adrain) 136. Prior to the operation on 23 May 2017, WDPC 372 (“3429/PW1”) successfully registered an account for using the Rider App on her mobile phone by providing an account identifier “TSUI YIM LAM”, her mobile phone number, her email address and her credit card number. 3429/Exh. P9A at [3] 137. At 7:05 a.m. on 23 May 2017, 3429/PW1 disguising as a passenger was outside D‧PARK on Mei Wan Street in Tsuen Wan. 3429/PW1 logged in the Rider App on her mobile phone and requested for and chose “Jun Wing Yau Bo (transliteration)” carriage service by setting D‧PARK as the pick-up location and “Terminal 2 of Hong Kong International Airport” as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “YUK FUNG”, a photograph depicting the Defendant, a vehicle registration mark “TW6471” and a description of “Toyota Alphard”. At all material times, the motor vehicle displaying registration mark TW 6471 (“Car 18”) was a Toyota Alphard private car. 3429/Exh. P9A at [5]-[6] 3429/Exh. P3(4) 3429/Exh. P1 at [1(a)] 138. At 7:10 a.m., 3429/PW1 received on her mobile phone a telephone call of number 56916166 from a male. The male told 3429/PW1 that he was the Uber driver and said he would arrive after five minutes. 3429/Exh. P9A at [7] 139. At 7:21 a.m., 3429/PW1 received on her mobile phone another telephone call of number 56916166 from the male, who said he had arrived at D‧PARK. Meanwhile, the Defendant drove Car 18 to D‧PARK. The Defendant alighted from Car 18 and assisted to put 3429/PW1’s luggage into the baggage compartment of Car 18. 3429/PW1 boarded Car 18 and sat in the rear passenger seat. 3429/PW1 confirmed the destination with the Defendant. Afterwards, the Defendant drove Car 18 off with 3429/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 18 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3429/PW1’s mobile phone kept updating. 3429/Exh. P9A at [8] 3429/Exh. P5A at [1]-[3] 3429/Exh. P3(7)- (12) 140. During the journey, the Defendant explained to 3429/PW1 about the calculation of fare when using uberX, UberBLACK and 7-seaters. 3429/Exh. P5A at [6]-[28] 141. At 7:51 a.m. the same day, Car 18 driven by the Defendant and carrying 3429/PW1 arrived outside Terminal 2 of Hong Kong International Airport. 3429/PW1 alighted from Car 18. The Defendant assisted 3429/PW1 to unload her luggage from Car 18. 3429/Exh. P9A at [10] 142. The Defendant was arrested shortly afterwards. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 56916166, which was registered by the Defendant. 3429/Exh. P1 at [2]- [3] 3429/Exh. P7 143. Meanwhile, the Rider App on 3429/PW1’s mobile phone displayed records of the journey, including an itemized receipt containing a map of the journey and the fare of HK$338.36. Bank records confirm that a payment of HK$338.36 was made to “UBER” the same day from 3429/PW1’s credit card. 3429/Exh. P3(13)- (15) & P9A at [11] 3429/Exh. P8 144. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 18. 3429/Exh. P1 at [8] (19) KCCC 3430/2017 (Defendant: WONG Siu-pong) 145. Prior to the operation on 23 May 2017, WDSPC 56801 (“3430/PW1”) successfully registered an account for using the Rider App on her mobile phone by providing an account identifier “WUN SIM AU”, her email address and her credit card number. 3430/Exh. P10 at [3] 146. At 7:10 a.m. on 23 May 2017, 3430/PW1 disguising as a passenger was outside Citywalk 1 near Yeung Uk Road in Tsuen Wan. 3430/PW1 logged in the Rider App on her mobile phone and requested for “uberX” carriage service to Terminal 1 of Hong Kong International Airport. The Rider App displayed a map showing 3430/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Siu Pong”, a photograph depicting the Defendant, a vehicle registration mark “TK5297” and a description of “Honda Spike”. At all material times, the motor vehicle displaying registration mark TK 5297 (“Car 19”) was a Honda Spike private car. 3430/Exh. P10 at [6] 3430/Exh. P3(4) 3430/Exh. P1 at [1(a)] 147. At 7:21 a.m., 3430/PW1 received on her mobile phone a telephone call of number 53132985 from a male. The male told 3430/PW1 that he was the driver of Car 19, said he had stopped the car on Wo Tik Street due to parking restrictions and asked 3430/PW1 to go there. 3430/Exh. P10 at [7] 148. As 3430/PW1 entered Wo Tik Street, the Defendant waved at her and said he was the Uber driver. The Defendant assisted to put 3430/PW1’s luggage into the baggage compartment of Car 19. 3430/PW1 boarded Car 19 and sat in the rear passenger seat. Afterwards, the Defendant drove Car 19 off with 3430/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 19 on the map of the Rider App and the estimated arrival time were displayed on the Rider App on 3430/PW1’s mobile phone. 3430/Exh. P10 at [8] 3430/Exh. P11 at [2] 3430/Exh. P3(8) 149. At 7:47 a.m. the same day, Car 19 driven by the Defendant and carrying 3430/PW1 arrived outside Terminal 1 of Hong Kong International Airport. 3430/PW1 alighted from Car 19. The Defendant assisted 3430/PW1 to unload the luggage from Car 19. 3430/Exh. P10 at [10] 150. The Defendant was arrested shortly afterwards. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 53132985, which was registered by the Defendant. The phone had the Uber application, in which there were records of an account registered under the Defendant’s name and details and charging rate of the subject journey taken by 3430/PW1 on Car 19. 3430/Exh. P1 at [2]- [3] 3430/Exh. P8 3430/Exh. P5(3)-(5) 151. Later the same day, 3430/PW1 received an email receipt showing a map of the journey, the calculation of the fare of HK$358.97, an Uber logo and that the fare had been charged on 3430/PW1’s credit card. Bank records confirm that a payment of HK$358.97 was made to “UBER HK” the same day from 3430/PW1’s credit card. 3430/Exh. P4(1)-(2) 3430/Exh. P9 152. Between about 9:27 a.m. and about 11:35 a.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had become an Uber driver in November 2016 and that he would receive a share of the fare from Uber by bank transfers. 3430/Exh. P1 at [8] 3430/Exh. P6 at Q&A 15-16 153. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 19. 3430/Exh. P1 at [9] (20) KCCC 3431/2017(Defendant: CHOW Wing-ching Emma) 154. Prior to the operation on 23 May 2017, DSPC 33443 (“3431/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “KA MING CHAN”, his email address and his credit card number. 3431/Exh. P9 at [3] 155. At 6:56 a.m. on 23 May 2017, 3431/PW1 disguising as a passenger was at Tsuen Kwai Street near Tsuen Wan Garden in Tsuen Wan. 3431/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service to Terminal 1 of Hong Kong International Airport. The Rider App displayed a map showing 3431/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Wing Ching Emma”, a photograph depicting the Defendant, a vehicle registration mark “UK4832” and a description of “BMW 5-series”. At all material times, the motor vehicle displaying registration mark UK 4832 (“Car 20”) was a BMW 5-series private car. 3431/Exh. P9 at [6] 3431/Exh. P3(1) 3431/Exh. P1 at [1(a)] 156. At 6:58 a.m., 3431/PW1 received on his mobile phone a telephone call of number 53959464 from a female. The female told 3431/PW1 that she was the Uber driver, confirmed 3431/PW1’s exact location and said she would arrive in a few minutes. 3431/Exh. P9 at [7] & P11 at [2] 3431/Exh. P10A at [3] 157. At 7:07 a.m., the Defendant drove Car 20 to the pick-up location. The Defendant said she was the Uber driver and assisted to put 3431/PW1’s luggage into the baggage compartment of Car 20. 3431/PW1 boarded Car 20 and sat in the rear passenger seat. Afterwards, the Defendant drove Car 20 off with 3431/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 20 on the map of the Rider App and the estimated arrival time were displayed on the Rider App on 3431/PW1’s mobile phone. 3431/Exh. P9 at [8] 3431/Exh. P12 3431/Exh. P3(3) 158. At 7:34 a.m. the same day, Car 20 driven by the Defendant and carrying 3431/PW1 arrived outside Terminal 1 of Hong Kong International Airport. 3431/PW1 alighted from Car 20. The Defendant assisted 3431/PW1 to unload the luggage from Car 20. 3431/Exh. P9 at [10] 159. The Defendant was arrested shortly afterwards. The Defendant’s mobile phone was seized from her. The phone contained a SIM card for telephone number 53959464, which was registered by the Defendant. The phone had the Driver App, in which there were records of an account registered under the Defendant’s name. There were also fare records including that of the subject journey taken by 3431/PW1 on Car 20. 3431/Exh. P1 at [2]- [3] 3431/Exh. P7 3431/Exh. P4(3)- (5), (9)-(11) & (13) 160. Meanwhile, 3431/PW1 received an email receipt showing a map of the journey, the calculation of the fare of HK$295.51, an Uber logo and that the fare had been charged on 3431/PW1’s credit card. The Rider App of 3431/PW1’s mobile phone also showed an itemized receipt of the fare of the journey and invited 3431/PW1 to give a rating on a scale of five stars. Bank records confirm that a payment of HK$295.51 was made to “UBER BV” the same day from 3431/PW1’s credit card. 3431/Exh. P3(6)-(9) 3431/Exh. P3(4) 3431/Exh. P8 161. Between about 11:40 a.m. and about 2 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that she had become an Uber driver in August to September 2016 and that she would receive a share of the fare from Uber by bank transfers. 3431/Exh. P1 at [7] 3431/Exh. P5 at Q&A 12 & 18 162. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 20. 3431/Exh. P1 at [8] (21) KCCC 3432/2017 (Defendant: WONG Tak-ming) 163. Prior to the operation on 23 May 2017, DPC 11133 (“3432/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “NG KWUN LUN”, his email address and his credit card number. 3432/Exh. P8 at [3] 164. At 7:03 a.m. on 23 May 2017, 3432/PW1 disguising as a passenger was outside Wing Tai House, Fuk Loi Estate, Tsuen Wan. 3432/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service to Terminal 1 of Hong Kong International Airport. The Rider App displayed a page showing, inter alia, the name “Tak Ming”, a photograph depicting the Defendant, a vehicle registration mark “NS6318” and a description of “Toyota Alphard”. At all material times, the motor vehicle displaying registration mark NS 6318 (“Car 21”) was a Toyota Alphard private car. 3432/Exh. P8 at [5] 3432/Exh. P3(5) & P8 at [6] 3432/Exh. P1 at [1(a)] 165. At 7:15 a.m., the Defendant drove Car 21 to Wing Tai House, Fuk Loi Estate, Tsuen Wan. 3432/PW1 boarded Car 21 and sat in the rear passenger seat. Afterwards, the Defendant drove Car 21 off with 3432/PW1 on board. Throughout the journey, a small car which showed the real-time location of Car 21 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3432/PW1’s mobile phone kept updating. 3432/Exh. P8 at [6] 3432/Exh. P3(6)- (10) 166. During the journey, the Defendant told 3432/PW1 about his experience and his income as an Uber driver. 3432/Exh. P5A at [18]-[35], [46]-[97], [133]-[143] & [174]- [220] 167. At 7:43 a.m. the same day, Car 21 driven by the Defendant and carrying 3432/PW1 arrived outside Terminal 1 of Hong Kong International Airport. 3432/PW1 and the Defendant both alighted from Car 21. 3432/Exh. P18 at [10] 168. The Defendant was arrested shortly afterwards. Under caution at the scene, the Defendant said he drove as a part-time Uber driver to earn money and asked for a chance. The Defendant’s mobile phone was seized from him. The phone had the Uber application, in which there were records of an account registered under the Defendant’s name. There were also fare records including that of the subject journey taken by 3432/PW1 on Car 21. 3432/Exh. P1 at [2]-[3] 3432/Exh. P4(5)- (13) 169. Shortly afterwards, the Rider App on 3432/PW1’s mobile phone showed the calculation of the fare of $337.56 and that the fare had been charged on his credit card. Bank records confirm that a payment of HK$337.56 was made to “UBER HK” the same day from 3432/PW1’s credit card. 3432/Exh. P3(11) & P8 at [8] 3432/Exh. P7 170. Between about 11:25 a.m. and about 1 p.m. on 23 May 2017, the Defendant was interviewed. Under caution, the Defendant said that he had become an Uber driver for half a year and that he would receive a share of the fare from Uber by bank transfers. 3432/Exh. P1 at [8] 3432/Exh. P6 at Q&A 7, 12 & 19 171. As at 23 May 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car 21. 3432/Exh. P1 at [9] (22) KCCC 3496/2017 (Defendant: KONG Cheuk-lai) 172. Prior to the operation on 5 May 2017, DPC 5844 (“3496/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “tong chan”, his mobile phone number, his email address and his credit card number. 3496/Exh. P10 at [3] 173. At 11:50 a.m. on 5 May 2017, 3496/PW1 and DPC 6821 (“3496/PW2”) disguising as passengers were outside St Teresa’s Hospital at Lomond Road, Kowloon City. 3496/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting his location as the pick-up location and The Hong Kong Polytechnic University West Kowloon Campus as the destination. The Rider App displayed a map showing the pick-up location and that there was a small car in the vicinity. Below the map there was the name “Cheuk Lai”, a photograph depicting the Defendant, a vehicle registration mark “GP3230” and a description “Mercedes-Benz C-Class”. At all material times, the private car displaying registration mark GP 3230 (“Car 22”) was a Mercedes-Benz C-Class private car. 3496/Exh. P10 at [4]-[5] 3496/Exh. P3(11) 3496/Exh. P1 at [1(a)] 174. At 11:58 a.m., 3496/PW1 received on his mobile phone a telephone call of number 90960006 from a male. The male asked if 3496/PW1 had requested for Uber and confirmed the pickup location with 3496/PW1. The male also asked 3496/PW1 to wait for a while. 3496/Exh. P10 at [6] 175. At 12:05 p.m., Car 22 arrived outside St Teresa’s Hospital. 3496/PW2 waived to signal the Defendant. Car 22 then stopped in front of 3496/PW1 and 3496/PW2. Immediately thereafter, 3496/PW1 and 3496/PW2 boarded Car 22 and sat in the rear passenger seats. Afterwards, the Defendant drove Car 22 off with 3496/PW1 and 3496/PW2 on board. Throughout the journey, a small car which showed the real-time location of Car 22 on the map of the Rider App and the estimated arrival time displayed on the Rider App on 3496/PW1’s mobile phone kept updating. During the journey, the Defendant told 3496/PW1 and 3496/PW2 that vehicles registered under UberBlack could also take orders of uberX but passengers would only need to pay the fare of uberX. 3496/Exh. P10 at [7] & P11 at [4] 3496/Exh. P3(17)- (21) 3496/Exh. P11 at [4] 176. At 12:25 p.m. the same day, Car 22 driven by the Defendant and carrying 3496/PW1 and 3496/PW2 arrived outside The Hong Kong Polytechnic University West Kowloon Campus. 3496/PW1 and 3496/PW2 alighted from Car 22. Shortly afterwards, 3496/PW1 received an electronic bill showing a map of the journey, the fare of “$70.51”, “tong, thank you for choosing Uber”, an Uber logo and that the fare had been charged on 3496/PW1’s credit card. The Rider App on his mobile phone also showed a record of the journey and displayed a page where 3496/PW1 could give a rating on a scale of five stars. Bank records confirm that a payment of HK$70.51 was made to “UBER BV” the same day from 3496/PW1’s credit card. 3496/Exh. P10 at [7] & P11 at [5] 3496/Exh. P3(26)- (29) 3496/Exh. P3(22)- (25) 3496/Exh. P9 177. On 23 May 2017, the Defendant was arrested. The Defendant’s mobile phone was seized from him. The phone contained a SIM card for telephone number 90960006, which was registered by the Defendant’s wife. The phone had the Driver App and the telephone record with 3496/PW1. 3496/Exh. P1 at [2]- [3] 3496/Exh. P1 at [1(c)] & P8 3496/Exh. P5(2)-(4) 178. Between about 9:15 a.m. and about 11:08 a.m. on 23 May 2017, the Defendant was interviewed. Under caution, he admitted that his telephone number was 90960006. 3496/Exh. P1 at [7] 3496/Exh. P6 Q&A 6 179. As at 5 May 2017, there was not any hire car permit within the meaning of section 2 of the Road Traffic Ordinance, Cap. 374 in force in respect of Car 22. 3496/Exh. P1 at [9] (23) KCCC 3629/2017 (Defendant: LAU Siu-hang) 180. Prior to the subject journey on 17 May 2017, Gino BONIFACIO (“3629/PW1”) successfully registered an account for using the Rider App on his mobile phone by providing an account identifier “Gino Bonifacio”, his telephone number, his email address and his credit card number. 3629/Exh. P4A at [2] 181. At 10:00 a.m. on 17 May 2017, 3629/PW1 and Benjamin De Leon (“Benjamin”) were in Tsuen Wan. 3629/PW1 logged in the Rider App on his mobile phone and requested for carriage service to Hong Kong Convention and Exhibition Centre. The Rider App displayed that a person named “Ying Wai” driving a vehicle with registration mark “SG 1276” had accepted 3629/PW1’s request. 3629/Exh. P4A at [3] 182. Afterwards, the Defendant drove a private car displaying registration mark TH 5589 (“Car 23”) to the pick-up location. The Defendant showed 3629/PW1 a record of the latter’s request for carriage for service. Satisfied that the Defendant was the Uber driver, 3629/PW1 and Benjamin boarded Car 23 accordingly. Afterwards, the Defendant drove Car 23 off with 3629/PW1 and Benjamin on board. 3629/Exh. P4A at [3] 183. By around 10:20 a.m. the same day inside Western Harbour Crossing while Car 23 was on the way to Hong Kong Convention and Exhibition Centre, Car 23 had a traffic accident. 3629/PW1 sustained injuries in the accident and was sent to a hospital for medical treatment. The fare of the journey was HK$97.97 up to that time. 3629/Exh. P4A at [4]- [5] 184. 3629/PW1 later informed Uber of the accident. A representative of Uber enquired with 3629/PW1 about the accident and offered a full refund of the fare. The Rider App on 3629/PW1’s mobile phone showed a record of the journey. Subsequently, 3629/PW1 received a full refund of the fare of the journey. 3629/Exh. P4A at [4]- [5] 185. On 14 November 2017, the Defendant was arrested. 3629/Exh. P1 at [2] 186. As at 17 May 2017, there was not any hire car permit within the meaning of section 2 of the Road Traffic Ordinance, Cap. 374 in force in respect of Car 23. 3629/Exh. P1 at [6] Part B. The 24th to the 28th cases [N.B. “3969/PW1” refers to PW1 in KCCC 3969/2017 “3969/Exh. P1 at [3]” refers to paragraph 3 of Exh. P1 in KCCC 3969/2017] (24) KCCC 3969/2017 (Defendant: FAN Wing-yau) 187. Prior to the journey on 21 June 2017, LIU Chi-keung (“3969/PW1”) successfully registered an account for using the “UBER” mobile application (“the Rider App”) by providing an account identifier “LIU CHI KEUNG”, his email address, his mobile phone number and his credit card number. 3969/Exh. P7 at page 1 188. Around 7:49 p.m. on 21 June 2017, 3969/PW1 arrived at Lek Yuen Maternal and Child Health Centre in Shatin. 3969/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting his location as the pick-up location and Cheung Sha Wan Government Offices as the destination. The Rider App displayed the name “Wing Yau”, a photograph depicting the Defendant, a vehicle registration mark “UU5489” and a description of “Toyota Prius”. At all material times, the motor vehicle displaying registration mark UU 5489 (“Car A”) was a Toyota Prius private car. 3969/Exh. P7 at pages 2-3 3969/Exh. P2(2) 3969/Exh. P1 at [1(a)] 189. At 7:56 p.m., 3969/PW1 had a missed telephone call from the number of 96784195. At 7:57 p.m., 3969/PW1 saw the missed call on his mobile phone and therefore called back. A male answered the call and 3969/PW1 confirmed his pick-up location with the male. At 8:04 p.m., 3969/PW1 received a call from the said number again and the male told 3969/PW1 that he would soon arrive in UU 5489 to pick up 3969/PW1. 3969/Exh. P8 at [2] 190. At 8:23 p.m., 3969/PW1 saw Car A arriving towards him and thus waved at the Defendant to signal him. Car A stopped next to 3969/PW1. Immediately thereafter, 3969/PW1 boarded Car A and sat in the rear passenger seat. The Defendant apologized for arriving late to pick 3969/PW1 up and then confirmed the destination with 3969/PW1 to be Cheung Sha Wan Government Offices. Afterwards, the Defendant drove Car A off with 3969/PW1 on board. 3969/Exh. P7 at pages 4-5 3969/Exh. P7 at page 8 191. By around 8:35 p.m., Car A driven by the Defendant and carrying 3969/PW1 arrived outside Chiyu Bank near the junction of Lai Chi Kok Road and Shek Kip Mei Street. After alighting from Car A, 3969/PW1 received an email from Uber showing “Chi Keung, thank you for choosing Uber” and confirming the fare of the journey to be $88.30, which was charged on 3969/PW1’s credit card. Bank records confirm that a payment of HK$88.30 was made to “UBER HK” the same day from 3969/PW1’s credit card. 3969/Exh. P7 at pages 6-7 3969/Exh. P2(6)-(9) & P9 at [4] 3969/Exh. P6 192. On 15 December 2017, the Defendant was arrested. 3969/Exh. P1 at [2] 193. As at 21 June 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car A. 3969/Exh. P1 at [6] (25) KCCC 3971/2017 (Defendant: LAM Kin-fat) 194. Prior to the journey on 21 June 2017, LIU Chi-keung (“3971/PW1”) successfully registered an account for using the Rider App by providing an account identifier “LIU CHI KEUNG”, his email address, his mobile phone number and his credit card number. 3971/Exh. P6 at [2] 195. Around 10:51 p.m. on 21 June 2017, 3971/PW1 was near the junction of Fuk Wing Street and Yen Chow Street in Sham Shui Po. 3971/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting his location as the pick-up location and Fan Leng Lau in Fanling as the destination. The Rider App displayed a map showing 3971/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Kin Fat”, a photograph depicting the Defendant, a vehicle registration mark “HS221” and a description of “BMW 2-series Gran Tourer”. At all material times, the motor vehicle displaying registration mark HS 221 (“Car B”) was a BMW 2-series Gran Tourer private car. 3971/Exh. P6 at [3]- [4] 3971/Exh. P2(4) 3971/Exh. P1 at [1(a)] 196. Around 10:57 p.m., 3971/PW1 pressed the “call” button on the Rider App and made a call to the telephone number of 69925489. The male who answered the call told 3971/PW1 that he was on his way and asked 3971/PW1 to wait for a while. 3971/Exh. P6 at [5] 197. Around 11:06 p.m., 3971/PW1 saw Car B arriving towards him and thus waved at the Defendant to signal him. Car B stopped next to 3971/PW1 and the Defendant waved at 3971/PW1 to signal him to board Car B. Immediately thereafter, 3971/PW1 boarded Car B and sat at the rear passenger seat. The Defendant then confirmed the destination with 3971/PW1 to be Fan Leng Lau in Fanling. Afterwards, the Defendant drove Car B off with 3971/PW1 on board. 3971/Exh. P6 at [6] 198. By around 11:34 p.m., Car B driven by the Defendant and carrying 3971/PW1 arrived at Fan Leng Lau in Fanling. After alighting from Car B, 3971/PW1 received an email from Uber showing “Chi Keung, thank you for choosing Uber” andconfirmingthefare of the journey to be $264.40, which was charged on 3971/PW1’s credit card. Bank records confirm that a payment of HK$264.40was made to “UBER TRIP” the same day from 3971/PW1’s credit card. By around midnight, the Rider App on3971/PW1’s mobile phone also displayed a page where 3971/PW1 could give a rating on a scale of five stars. 3971/Exh. P6 at [7] 3971/Exh. P2(11)- (18) & P6 at [8] 3971/Exh. P5 3971/Exh. P2(19) & P6 at [8] 199. On 15 December 2017, the Defendant was arrested. 3971/Exh. P1 at [2] 200. As at 21 June 2017, there was not any hire car permit within the meaning of section 2 of the Ordinance in force in respect of Car B. 3971/Exh. P1 at [5] (26) KCCC 3974/2017 (Defendant: CHING Siu-kwok William) 201. Prior to the journey on 25 June 2017, LIU Chi-keung (“3974/PW1”) successfully registered an account for using the Rider App by providing anaccountidentifier“LIU CHI KEUNG”, his email address, his mobile phonenumber and his credit card number. 3974/Exh. P5 at [2] 202. About 4:33 p.m. on 25 June 2017, 3974/PW1 was at Lai Yan Lane, Lek Yuen Estate in Shatin. 3974/PW1 logged in the Rider App on his mobile phone and requested for “uberX” carriage service by setting hislocationasthe pick-up location and Fenwick Street in Wan Chai as the destination. The Rider App displayed a map showing 3974/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Siu Kwok William”, a photograph depicting the Defendant, a vehicle registration mark “HT8923” and a description of “Toyota Previa”. At all material times, the motor vehicle displaying registration mark HT 8923 (“Car C”) was a Toyota Previa private car. 3974/Exh. P5 at [3]- [4] 3974/Exh. P2(6) 3974/Exh. P1 at [1(a)] 203. Around 4:53 p.m., 3974/PW1 saw Car C arriving towards him and thus waved at the Defendant to signal him. Car C stopped next to 3974/PW1. The Defendant confirmed with 3974/PW1 that he was the Uber driver and asked 3974/PW1 to board Car C. Immediately thereafter, 3974/PW1 boarded CarCand satat the rearpassenger seat. The Defendant thenconfirmedthedestinationwith 3974/PW1 to be Fenwick Street in Wan Chai. Afterwards, the Defendant drove Car C off with 3974/PW1 on board. 3974/Exh. P5 at [5] 204. By around 5:22 p.m., Car C driven by the Defendant and carrying 3974/PW1 arrived at Fenwick Street inWan Chai. After alighting from Car C, 3974/PW1 received an email from Uber showing “Chi Keung, thank you for choosing Uber” and confirming the fare of the journey to be $90.67, which was charged on 3974/PW1’s credit card. Bank records confirm that a payment of HK$90.67 was made to “UBER BV” the same day from 3974/PW1’s credit card. 3974/Exh. P5 at [6] 3974/Exh. P2(10)- (13) & P5 at [7] 3974/Exh. P4 205. On 29 July 2017, 3974/PW1 received an email from Uber revising the fare of the journey as HK$143.16. Bank records confirm that an additional payment of HK$52.49 wasmadeto“UBER ADJ”from3974/PW1’s creditcard on 25 June 2017. 3974/Exh. P5 at [10] 3974/Exh. P4 206. On 15 December 2017, the Defendant was arrested. 3974/Exh. P1 at [2] 207. Asat25 June2017,there was not anyhire carpermit within themeaningofsection2of theOrdinance in forceinrespect of Car C. 3974/Exh. P1 at [6] (27) KCCC 502/2018 (Defendant: CHICK Wai-ho) 208. Prior to the journey on 18 August 2017, DOO Peter (“502/PW1”) successfully registered an account for using the Rider App by providing an account identifier “DOO Peter”, his email address, his mobile phone numberand his credit card number. 502/Exh. P6 at [2] 209. About10:35p.m.on18August2017,502/PW1wasat Shun Tak Centre in Sheung Wan. 502/PW1 logged in the Rider App on his mobile phone andrequestedfor “UberBLACK” carriage service by setting his location as the pick-up location and the Podium of Telford Gardens as the destination. The Rider App displayed a map showing 502/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Wai Ho”, a photograph depicting the Defendant, a vehicle registration mark “TV2825” and a description of “Tesla Model S”. At all material times, the motor vehicle displaying registration mark TV 2825 (“Car D”) was a Tesla Model S private car. 502/Exh. P6 at [3] 502/Exh. P2(3) 502/Exh. P1 at [1(a)] 210. Afterwards, 502/PW1 received on his mobile phone two telephone calls of number 62061608 from a male. In the first call, the male confirmed with 502/PW1 that the destination was Telford Gardens and told502/PW1that the registration mark of his car was TV 2825. He also asked 502/PW1 about the pick-up location and said thathe would arrive soon. In the second call, the male told 502/PW1 that he had arrived at Shun Tak Centre but could not find 502/PW1. 502/PW1 told him that he was waiting at the entrance of Rumsey Street Car Park. 502/Exh. P7 at Q&A 2 211. Around 10:40 p.m., Car D approached 502/PW1 and stopped next to him. Immediately thereafter, 502/PW1 boarded Car D from the rear door. Afterwards, the Defendant drove Car D off with 502/PW1 on board. 502/Exh. P6 at [4] 212. By around 11:16 p.m., Car D driven by the Defendant and carrying 502/PW1 arrived at the Podium of Telford Gardens. After alighting from Car D, 502/PW1 received an email showing“Peter,thank you for choosingUber”and confirming the fare of the journey to be HK$449.14,which was charged on 502/PW1’s credit card. Bank records confirm that a payment ofHK$449.14wasmadeto “UBER TRIP” the same day from 502/PW1’s credit card. 502/Exh. P6 at [5] 502/Exh. P2(4)-(5) & P6 at [6] 502/Exh. P5 213. Later, 502/PW1 made a report to the police about the subject journey. As a result on 29 August 2017, SPC 34445 issued a Notice Requiring Identification of Driver to theDefendant’selderbrother,who wastheregisteredowner of Car D. On 6 September 2017, there was a credit transaction of HK$449.14 made by “UBER TRIP” to 502/PW1’s credit card. 502/Exh. P6 at [7] 502/Exh. P8 at [3] 502/Exh. P1 at [1(c)] 502/Exh. P5 214. On 9 February 2018, the Defendant was arrested. 502/Exh. P1 at [2] 215. As at 18 August 2017, there was not any hire car permit within the meaning of section 2 of the Ordinanceinforce in respect of Car D. 502/Exh. P1 at [5] (28) KCCC 503/2018 (Defendant: TAM Chun-ming) 216. Prior to the journey on 18 August 2017, DOO Peter (“503/PW1”) successfully registered an account for using the Rider App by providing an account identifier “DOO Peter”, his email address, his mobile phone number and his credit card number. 503/Exh. P4 at [2] 217. About 6:18 p.m. on 18 August 2017, 503/PW1 was outside Cheung Sha Wan Government Offices at No. 303 Cheung Sha Wan Road. 503/PW1 logged in the Rider App on his mobilephoneand requestedfor“uberX”carriage service by setting his location as the pick-up location and Shun Tak Centre as the destination. The Rider App displayed a map showing 503/PW1’s location and that there was a small car in the vicinity. Below the map there was the name “Chun Ming”, a vehicle registration mark “TZ9918” and a description of “Nissan Tiida”. At all material times, the motor vehicle displaying registration mark TZ 9918 (“Car E”) was a Nissan Tiida private car. 503/Exh. P4 at [3] 503/Exh. P2(3) 503/Exh. P1 at [1(a)] 218. Around 6:31 p.m., Car E approached 503/PW1 and stopped next to him. Immediately thereafter, 503/PW1 boarded Car E and sat at the rear passenger seat. The Defendant then confirmed the destination with 503/PW1 to be Shun Tak Centre. Afterwards, the Defendant drove Car E off with 503/PW1 on board. 503/Exh. P4 at [4] 219. By around 6:51 p.m., Car E driven by the Defendant and carrying503/PW1arrived atShunTakCentre. Afteralighting from Car E, 503/PW1 received an email from Uber showing “Peter, thank you for choosing Uber” and confirming the fare of the journey to be HK$165.67, which was charged on 503/PW1’s credit card. Bank records confirm that apayment of HK$165.67 was madeto“UBER HK” the same day from 503/PW1’s credit card. 503/Exh. P4 at [5] 503/Exh. P2(4)-(5) & P4 at [6] 503/Exh. P3 220. On 9 February 2018, the Defendant was arrested. 503/Exh. P1 at [2] 221. As at 18 August 2017, there was not any hire car permit within the meaning of section2of the Ordinancein forcein respect of Car E. 503/Exh. P1 at [5] -END- [1] City of Toronto v Uber Canada Inc. et al. 126 OR (3d) 401 [2] City of Toronto v Uber Canada Inc. et al. 126 OR (3d) 401, [3] and [4] [3] City of Toronto v Uber Canada Inc. et al. 126 OR (3d) 401, [12] [4] City of Toronto v Uber Canada Inc. et al. 126 OR (3d) 401 [5] Edmonton (City) v Uber Canada Inc [2015] AJ No.357 [6] Edmonton (City) v Uber Canada Inc [2015] AJ No.357, [23] [7] See sections 26 and 52(2) of the Road Traffic Ordinance, Cap. 374 [8] City of Toronto v Uber Canada Inc. et al. 126 OR (3d) 401, and Edmonton (City) v Uber Canada Inc [2015] AJ No.357 [9] KCCC3629/2017 [10] Which resulted in the 22 prosecutions in KCCC3412 to 3432/2017, and KCCC3469/2017 [11] The undercover officers in KCCC3412 to 3426/2017, and KCCC3496/2017 [12] The undercover officers in KCCC3428 to 3432/2017 [13] See the officers’ witness statements produced to court under section 65B of the Criminal Procedure Ordinance, Cap. 221 [14] Kowloon Tong, Cheung Sha Wan, Yau Ma Tei, Hung Hom, Sham Shui Po, Mei Foo, Tsim Sha Tsui, Kowloon City, Olympic City, Sycamore Street, and Tsuen Wan [15] Sham Shui Po, Kowloon Tong, Tai Kok Tsui, Ho Man Tin, Yau Ma Tei, Tsim Sha Tsui, Hung Hom, Tseung Kwan O, Hong Kong International Airport, and Hoi Ping Road [16] Uber X, UberBLACK, Seven-seater [17] See photographs captured of the screens of the mobile phones [18] In 18 cases: KCCC3412 to 3420, 3425 to 3432, and 3496/2017 [19] In four cases: KCCC3421 to 3424/2017 [20] See the officers’ witness statements produced to court under section 65B of the Criminal Procedure Ordinance, Cap. 221 [21] See the Banker’s Affirmations in KCCC3422 and 3429/2017 [22] See the Banker’s Affirmations in KCCC3412 to 3419, 3423 to 3428, 3430, 3432 and 3496/2017 [23] See the Banker’s Affirmations in KCCC3420 and 3421, and 3431/2017 [24] KCCC3413 to 3418/2017, KCCC3421 to 3428/2017, and KCCC3430 to 3432/2017 [25] KCCC3412/2017, KCCC3419 and 3420/2017, KCCC3429/2017, and KCCC3496/2017 [26] KCCC3413 to 3418/2017, KCCC3421 to 3428/2017, and KCCC3430 to 3432/2017 [27] KCCC3414/2017, KCCC3417/2017, KCCC3421 to 3424/2017, KCCC3426 and 3427/2017, KCCC3431 and 3432/2017 [28] KCCC3421 to 3424/2017, KCCC3426/2017, and KCCC3431 and 3432/2017 [29] KCCC3414/2017 (about 75%), KCCC3427/2017 (75% to 80%), and KCCC3432/2017 (about 77%, which is arrived at by calculation only – see Q&A 19) [30] KCCC3413, 3418, 3428 and 3430/2017 [31] KCCC3425/2017 [32] KCCC3629/2017 [33] KCCC3969, 3971 and 3974/2017, and KCCC502 and 503/2018 [34] KCCC3969/2017 [35] KCCC3971/2017 [36] KCCC3974/2017 [37] KCCC502/2018 [38] KCCC503/2018 [39] See the Banker’s Affirmation in KCCC3425/2017 [40] See the Banker’s Affirmations in KCCC3412 to 3415, 3422 to 3424, 3426, 3428 to 3432, 3969, 3971, 3974/2017, and 502 and 503/2018 [41] See the Bankers’ Affirmations in KCCC3416 to 3421/2017, 3427 and 3496/2017 [42] See the Banker’s Affirmations in the following 16 cases: KCCC3412 to 3422, 3426, 3427 and 3496/2017, and 502 and 503/2018 [43] See the Admitted Facts and para.195 of the Prosecution Closing Submissions dated 7/5/2018 [44] See the definitions of “owner” and “registered owner” under section 2 of the Road Traffic Ordinance, Cap. 374 [45] KCCC3416, 3417, and 3426/2017 [46] KCCC3412, 3419, 3420, 3429, 3432, and 3629/2017, 3939, 3971 and 3974/2017, and KCCC502 and 503/2018 [47] KCCC3425, 3427, and 3430/2017 [48] KCCC3414/2017 [49] KCCC3428 and 3496/2017 [50] KCCC3424/2017 [51] KCCC3431/2017 [52] KCCC3413, 3415, 3418, and 3421 to 3423/2017 [53] See the admitted facts in KCCC3412 to 3415, 3418, 3420 to 3423, 3425 and 3426, 3428, 3430 to 3432, 3496, 3629, 3969 and 3974/2017, and KCCC502 and 503/2018 [54] Eg., the first sentence in para.5 of the Discussion Paper for LegCo Panel Meeting of 21/8/2016 at Tab 27 of the Prosecution Bundle of Authorities (Vol 2) [55] The full version of section 52(3) is reproduced at para. 110 below [56] LegCo Official Report of Proceedings dated 28 July 1982 (see Tab 2 of the Defence List of Authorities) [57] Wyatt v Guildhall Insurance Company Ltd [1937] 1 KB 633 [58] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1, per Lord Denning on p.3 [59] Wyatt v Guildhall Insurance Company Ltd [1937] 1 KB 633, at p.662 [60] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1, at p.6 [61] Albert v Motor Insurers’ Bureau [1972] RTR 230, at p.251B-F [62] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1, at p.6 [63] Kwong Pak-yam v The Queen [1965] HKLR 931, Ngai Kam-chung v The Queen [1965] HKLR 941, and Chan Kai Lap v The Queen [1969] HKLR 463, and Cheng For Kar v The Queen (unreported, CACC316/1972, 12 July 1972), Li Wing-Cheung v The Queen [1973] HKLR 98, and Lee Hing Cheong v The Queen (unreported, CACC866/1973, 14 December 1973) [64] Gazette (Suppl. No.3) dated 22 November 1957 (see Defence Suppl. List of Authorities at Tab 31) [65] Gazette (Suppl. No.3) dated 10 July 1964 (see Defence Suppl. List of Authorities at Tab 32) [66] Gazette (Suppl. No.3) dated 27 May 1977 (see Defence Suppl. List of Authorities at Tab 33) [67] Prosecution Bundle “Road Traffic Ordinance, Cap. 220,” supplied to the court at the court’s request after the close of evidence [68] Gazette (Suppl. No.3) dated 16 July 1982 (see Defence Suppl. List of Authorities at Tab 34) [69] Gazette (Suppl. No.3) dated 22 November 1957 (see Defence Suppl. List of Authorities at Tab 31). The parties have agreed that the court can use this Bill as if it was the enacted version of Cap. 220. The defence has during closing submission helpfully supplied the court with the 1964 Revised Edition of the Road Traffic Ordinance, Cap. 220. The definitions appear in section 2(1) and (2). [70] See the definition of “public” under section 2 in Gazette (Suppl. No.3) dated 22 November 1957 [71] See the definitions of “public goods vehicles”, “taxis”, “public cars”, “public omnibuses”, “hire omnibuses”, under section 2(2)(b), (c), (d), (f) and (g) in Gazette (Suppl. No.3) dated 22 November 1957 See also the definitions of “taxis”, “public cars”, “public omnibuses”, under section 2(2)(b), (c)(i) and (ii), and (e)(i) and (ii) in the 1964 Revised Edition of the Road Traffic Ordinance, Cap. 220 [72] See para. 3.10 of the Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018 [73] See Gazette (Suppl. No.3) dated 22 November 1957 (see Defence Suppl. List of Authorities at Tab 31). [74] But this definition of “hire omnibuses” was removed altogether when section 2(2)(g) was deleted in 1964. See Gazette (Legal Suppl. No.3) dated 10 July 1964 (see Defence Suppl. List of Authorities at Tab 32), and section 2(2) of the 1964 Revised Edition of the Road Traffic Ordinance, Cap. 220. [75] Gazette (Legal Suppl. No.3) dated 10 July 1964 (see Defence Suppl. List of Authorities at Tab 32), and section 2(2) of the 1964 Revised Edition of the Road Traffic Ordinance, Cap. 220. [76] See section 2(2)(c)(i) of the 1964 Revised Edition of the Road Traffic Ordinance, Cap. 220. [77] See section 2(2)(c)(ii) of the 1964 Revised Edition of the Road Traffic Ordinance, Cap. 220. [78] See para. 3.10 of the Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018 [79] See section 2(2)(b), (ba) and (bb) of the 1979 Revised Edition of the Road Traffic Ordinance, Cap. 220, in particular, section 2(d) [80] Gazette (Legal Suppl. No.3) dated 10 July 1964 (see Defence Suppl. List of Authorities at Tab 32), and section 2(2) of the 1964 Revised Edition of the Road Traffic Ordinance, Cap. 220. [81] See sections 2(2)(c)(ii) and (e)(ii) of the 1979 Revised Edition of the Road Traffic Ordinance, Cap. 220 [82] Coward v The Motor Insurers Bureau (1962) 1 All ER 531, where the driver of a private car agreed to give a friend a lift several times, each on payment of 10s and a drink. [83] Coward v The Motor Insurers Bureau (1962) 1 All ER 531 at p.538H-I [84] Kwong Pak-yam v The Queen (1965) HKLR 931, which was decided in 1965, by which time the definition of “public cars” had been amended but the contractual demarcation referred to in the preceding paragraphs of this judgment remained. See Gazette (Suppl. No.3) dated 10 July 1964, at section 2(d)(ii) of the amendment bill (see Defence Suppl. List of Authorities at Tab 32). [85] The other charge was one of driving a motor vehicle without third party insurance. [86] Coward v The Motor Insurers Bureau (1962) 1 All ER 531 [87] Kwong Pak-yam v The Queen (1965) HKLR 931 [88] Kwong Pak-yam v The Queen (1965) HKLR 931, pp.935-6 [89] Kwong Pak-yam v The Queen (1965) HKLR 931 [90] See Prosecution Closing Submission at para. 203, and Kwong Pak-yam v The Queen (1965) HKLR 931, pp.935-6. [91] See Prosecution Closing Submission at para. 233 [92] Kwong Pak-yam v The Queen (1965) HKLR 931 [93] Coward v The Motor Insurers Bureau (1962) 1 All ER 531 [94] Kwong Pak-yam v The Queen (1965) HKLR 931 [95] Reg v Ngai Kam-chung [1965] HKLR 941, where the driver and the passenger did not talk about money until the end of the journey. [96] Reg v Ngai Kam-chung [1965] HKLR 941, at p.950 [97] Chan Kai-lap v The Queen [1969] HKLR 463, by which time reg. 26(2) had been replaced by reg. 29(2) [98] Cheng For Kar v The Queen (unreported, CACC316/1972, 12 July 1972) [99] Li Wing-cheung v The Queen [1973] HKLR 98 [100] Albert v Motor Insurers’ Bureau [1972] RTR 230, the leading English case on “hire or reward”, was mentioned in the judgments. [101] HKSAR v Chan Chi Ho (transl.) HCMA1007/2005 [102] HKSAR v Lam Shing-yin HCMA559/2007 [103] Prosecution Closing Submission, para.209 [104] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1 [105] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1, at p.5 [106] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1 [107] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1 [108] Chan Kai-lap v The Queen [1969] HKLR 463 [109] Albert v Motor Insurers’ Bureau [1972] RTR 230, where, for about eight years, the driver took his co-workers to work on payment of irregular sums, or on payment in kind, or for free. [110] Albert v Motor Insurers’ Bureau [1972] RTR 230 [111] Li Wing-cheung v The Queen [1973] HKLR 98 [112] Coward v The Motor Insurers Bureau (1962) 1 All ER 531 [113] Kwong Pak-yam v The Queen (1965) HKLR 931, pp.935-6 [114] Albert v Motor Insurers’ Bureau [1972] RTR 230 [115] See eg., section 2(2)(b), (ba), (bb), (c)(i) and (ii), (d), (e)(ii) of the 1979 Revised Edition of the Road Traffic Ordinance, Cap. 220, in particular, section 2(d) [116] See the definition of “taxi” in section 2 of the Road Traffic Ordinance, Cap. 374 [117] See para.45 and 46 above [118] See para.45 above [119] See para.47 above [120] See para.51 above [121] See the definition of “private car” in section 2 of the Road Traffic Ordinance, Cap. 374 [122] Defence Closing Submission dated 21/5/2018 at para. 5.7, and see Defence Closing Submission dated 29/6/2018 at para.1.1(ii) [123] See para.233 of the Prosecution Closing Submissions dated 7/5/2018 [124] See para.234 of the Prosecution Closing Submissions dated 7/5/2018 [125] See para.12(iv) of the Prosecution’s Speaking Note dated 28/5/2018 [126] HKSAR v Lam Yuk Fai (2006) 9 HKCFAR 281 [127] See para.3(iii) of the Prosecution’s Speaking Note dated 12/6/2018 [128] See Prosecution’s Speaking Note dated 21/6/2018 [129] That is to say: a licensed public bus, public light bus or taxi, private bus used in taking students and teachers, etc., private light bus used for specified purposes, and private cars whose registered owners have in force a hire car permit in his favour. [130] Kwong Pak-yam v The Queen (1965) HKLR 931 [131] See para.2(i) and 3(iii) of the Prosecution’s Speaking Note dated 12/6/2018 [132] See para.2(i) and 3(iii) of the Prosecution’s Speaking Note dated 12/6/2018 [133] See para. 3.10 of the Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018 [134] See the definition of “public service vehicles” under section 2 of the Road Traffic Ordinance, Cap. 374 [135] See reg. 38(2) of the Road Traffic (Public Service Vehicles) Regulations, Cap. 374D, which is a permissive provision only. [136] See para. 3.11 of the Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018 [137] See para. 3.12 of the Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018 [138] See para. 3.10 of the Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018 [139] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1 [140] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1, at p.4 [141] Kwong Pak-yam v The Queen (1965) HKLR 931 [142] AG v Lun Chan [1964] HKLR 763 [143] AG v Lun Chan [1964] HKLR 763, at p.773 [144] AG v Lun Chan [1964] HKLR 763, at pp.776-7 [145] Kwong Pak-yam v The Queen (1965) HKLR 931 [146] Kwong Pak-yam v The Queen (1965) HKLR 931, at p.935 [147] See para. 49 above [148] See para. 50 above [149] AG v Lun Chan [1964] HKLR 763 [150] Bonham v Zurich General Accident & Liability Insurance Co., Ltd (1945) KB 292 [151] AG v Lun Chan [1964] HKLR 763, at p.772 [152] Bonham v Zurich General Accident & Liability Insurance Co., Ltd (1945) KB 292 [153] Aitken v Hamilton (1964) JCL 192 [154] Bonham v Zurich General Accident & Liability Insurance Co., Ltd (1945) KB 292 [155] Reproduced in AG v Lun Chan [1964] HKLR 763, p.778 [156] AG v Lun Chan [1964] HKLR 763 [157] AG v Lun Chan [1964] HKLR 763, at p.780 [158] AG v Lun Chan [1964] HKLR 763 [159] AG v Lun Chan [1964] HKLR 763, p.773 [160] AG v Lun Chan [1964] HKLR 763 [161] Chan Kai-lap v The Queen [1969] HKLR 463, pp.466-7 [162] AG v Lun Chan [1964] HKLR 763 [163] Albert v Motor Insurers’ Bureau [1972] RTR 230 [164] Albert v Motor Insurers’ Bureau [1972] RTR 230, at p.248D-G [165] Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018, at para.5.5-5.10 [166] Connell v Motor Insurers’ Bureau [1969] 2 Lloyd’s LR 1, at p.3 [167] Albert v Motor Insurers’ Bureau [1972] RTR 230, at p.237H-K [168] Albert v Motor Insurers’ Bureau [1972] RTR 230, at p.241F [169] HKSAR v Tse Yee Ping (2016) 19 HKCFAR 427, at [51] [170] HKSAR v Lam Kwong Wai(2006) 9 HKCFAR 574, at p.606E [171] HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568, at [13] [172] HKSAR v Wong Yuk Man (2012) 15 HKCFAR 712, at [27] [173] B v Commissioner of Independent Commission Against Corruption (2010) 12 HKCFAR 1, at [13], and HKSAR v Wong Yuk Man (2012) 15 HKCFAR 712 [174] China Field Ltd v Appeal Tribunal (Buildings) (No.2)[174](2009) 12 HKCFAR 342 [175] China Field Ltd v Appeal Tribunal (Buildings) (No.2)[175](2009) 12 HKCFAR 342, at [36] [176] Royal College of Nursing of the United Kingdom v Department of Health and Social Security [1981] AC 800, at p.822B [177] Regina (Quintavalle) v Secretary of State for Health [2003] 2 AC 687, at [10] [178] See the definition of “driver” under section 2 of the Road Traffic Ordinance, Cap. 374 [179] See the definition of “passenger” under section 2 of the Road Traffic Ordinance, Cap. 374 [180] See the definition of “motor vehicle” under section 2 of the Road Traffic Ordinance, Cap. 374 [181] See the definition of “private car” under section 2 of the Road Traffic Ordinance, Cap. 374 [182] See the definition of “taxi” under section 2 of the Road Traffic Ordinance, Cap. 374 [183] See the definition of “hire car permit” under section 2 of the Road Traffic Ordinance, Cap. 374 [184] See para.6 of the Prosecution’s Speaking Note dated 28/5/2018 [185] See para.5 of the Prosecution’s Speaking Note dated 12/6/2018 [186] See para.12 of the Prosecution’s Speaking Note dated 28/5/2018 [187] Orr v Trafalgar Insurance Company Ltd (1948-49) 82 Ll L Rep 1. In that case, a passenger Mr Moore sought a car for hire. The hired car that was to provide the services broke down; and in its place, a private car driven by Mr Prescot arrived, just as Mr Moore was thinking he might as well hitch-hike. There had been no discussion between Mr Prescot and Mr Moore concerning payment for the ride. An accident took place during the journey, resulting in injuries to Mr Orr. In the end, no rewards were paid by Mr Moore for what he had intended to be a ride in a car for hire. The English Court of Appeal found that Mr Prescot had driven his private car for hire or reward. [188] Orr v Trafalgar Insurance Company Ltd (1948-49) 82 Ll L Rep 1 [189] Orr v Trafalgar Insurance Company Ltd (1948-49) 82 Ll L Rep 1, per Tucker LJ at p.6 [190] City of Toronto v Uber Canada Inc. et al. 126 OR (3d) 401 [191] Edmonton (City) v Uber Canada Inc [2015] AJ No.357 [192] Orr v Trafalgar Insurance Company Ltd (1948-49) 82 Ll L Rep 1 [193] Orr v Trafalgar Insurance Company Ltd (1948-49) 82 Ll L Rep 1 [194] Section 19 of the Interpretation and General Clauses Ordinance, Cap. 1 [195] HKSAR v Lam Yuk Fai (2006) 9 HKCFAR 281 [196] HKSAR v Lam Yuk Fai (2006) 9 HKCFAR 281, at p.287C-F [197] Leung Sze Ho Albert v Bar Council of Hong Kong Bar Association [2016] 5 HKLRD 542 [198] Prosecution Closing Submission dated 7/5/2018, at paras. 236-246 [199] Winnie Lo v HKSAR (2012) HKCFAR 16 at [72]-[77] [200] Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 [201] B v Commissioner of the Independent Commission Against Corruption (2010) 13 HKCFAR 1 [202] Mo Yuk Ping v HKSAR (2007) 10 HKCFAR 386 [203] Prosecution Closing Submission dated 7/5/2018, at para. 242 [204] R v Rimmington [2006] 1 AC 459 [205] Committee for the Commonwealth of Canada v Canada (1991) 77 DLR (4th) 385 [206] R v Secretary of State for Trade and Industry, ex p Ford (1984) 4 Tr L 150 [207] Noise Control Authority v Step In Ltd (2005) 8 HKCFAR 113 at p.131J-132D [208] Mo Yuk Ping v HKSAR (2007) 10 HKCFAR 386 [209] Reg. 29(2) and (2A) of Road Traffic (Registration and Licensing of Vehicles) Regulations, Cap. 220H, in “Road Traffic Ordinance, Cap. 220,” supplied to the court at the court’s request after the close of evidence. See also para.205 of Prosecution Closing Submission dated 7/5/2018 [210] To borrow the description of “reward” from Willis J in the first instance judgment of Albert v Motor Insurers’ Bureau [1972] RTR 230, at p.233H-I, and 234A-B [211] HKSAR v Tse Yee Ping (2016) 19 HKCFAR 427, at [51] [212] See section 3 of the Road Traffic Ordinance, Cap. 374. [213] Defendants’ Reply Submissions dated 21/5/2018 at para.3.29 [214] See para.12(iv) of the Prosecution’s Speaking Note dated 28/5/2018 [215] Prosecution Closing Submission dated 7/5/2018, at para.256(a), citing Hong Kong 2016, HKSAR Government at p.205 of Tab 23 of the Prosecution Lists of Authorities [216] Canada (Attorney General) v Bedford[216][2013] 2 SCR 1101 [217] Lau Cheong & Another v HKSAR (2002) 5 HKCFAR 415 [218] Lau Cheong & Another v HKSAR (2002) 5 HKCFAR 415 [219] Fok Lai Ying v Governor in Council & Others [1997] HKLRD 810, (1997) 7 HKPLR 327 [220] Lau Cheong & Another v HKSAR (2002) 5 HKCFAR 415 [221] So Wai Lun v HKSAR (2006) 9 HKCFAR 530 [222] See section 52(10)(a) of the Road Traffic Ordinance, Cap. 374 [223] Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018, para. 4.18 [224] Canada (Attorney General) v Bedford[224][2013] 2 SCR 1101 [225] Carter v Canada (Attorney General) [2015] 1 SCR 331 [226] See paragraph 136 above [227] See paragraph 19-13 [228] Gurung Kesh Bahadur v Director of Immigration (2002) 5 HKCFAR 480 [229] Report on Study of Road Traffic Congestion in Hong Kong, Transport Advisory Committee (December 2014), para. 3.5.1 (Tab 22 of Prosecution Lists of Authorities) [230] Report on Study of Road Traffic Congestion in Hong Kong, Transport Advisory Committee (December 2014), para. 3.5.1, footnote no.4 (Tab 22 of Prosecution Lists of Authorities) [231] Uber BV & ors v Aslam & ors UKEAT 0056/17/DA, para. 2 (see Defence List of Authorities, Tab 22) [232] This information came from the first instance judgment delivered on 28 October 2016 (see Defence List of Authorities, Tab 21) [233] Secretary for Justice v Yan Yuk Lung (2007) 10 HKCFAR 335 [234] Secretary for Justice v Latker [2009] 2 HKC 100 [235] Defendants’ Responses to Questions Raised by this Court by way of letter dated 23 May 2018, para. 4.21 [236] Secretary for Justice v Yan Yuk Lung (2007) 10 HKCFAR 335 [237] HKSAR v Lam Kwong Wai and another (2006) 9 HKCFAR 574 [238] HKSAR v Lam Kwong Wai and anor (2006) 9 HKCFAR 574 [239] HKSAR v Lam Kwong Wai and anor (2006) 9 HKCFAR 574 [240] HKSAR v Choi Wai Lun [2018] HKCFA 18 [241] HKSAR v Choi Wai Lun [2018] HKCFA 18, [13] [242] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 [243] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 [244] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142, [44] [245] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142, [142] [246] Sweet v Parsley [1970] AC 132 at p.150, cited in HKSAR v Choi Wai Lun [2018] HKCFA 18, [21] [247] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 [248] R v Warner[1935] AC 462, at page 481 [249] Sweet v Parsley [1970] AC 132 [250] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 [251] HKSAR v Choi Wai Lun [2018] HKCFA 18 [252] R v K [2002] 1 AC 462 [253] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 [254] HKSAR v Choi Wai Lun [2018] HKCFA 18 [255] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 [256] The first sentence in para. 5 of the Discussion Paper for LegCo Panel Meeting of 21/8/2016 at Tab 27 of the Prosecution Bundle of Authorities (Vol 2) 
Chief Justice Ma: 1. At the conclusion of counsel’s submissions, the appeal was dismissed with costs. The reasons for dismissing the appeal are contained in the judgment of Lord Hoffmann NPJ, with which I agree. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Cheung PJ: 4. I agree with the judgment of Lord Hoffmann NPJ. Lord Hoffmann NPJ: 5. The following are my reasons for agreeing that the appeal should be dismissed. 6. The appellant Mr Joe Zhou is a venture capitalist. Between 2004 and 2006 he worked for a private equity fund known as SAIF Partners. The administration of the fund was structured in the form of a series of superimposed Cayman Islands Exempted Limited Partnerships. Under the Cayman Exempted Limited Partnership Law (2003 Revision) such a partnership has general partners and limited partners. A general partner is solely responsible for the administration of the business of the partnership and has unlimited liability for its debts. Limited partners have no powers or duties in respect of the business but are not liable for more than their contributions to the fund. 7. Mr Zhou was a limited partner in SAIF Partners II L.P. (“SAIF”), the first respondent, in which the general partner was SAIF II GP Capital Limited (“SAIF Capital”), the second respondent. SAIF Capital was controlled by Mr Yan, the founder of the business. SAIF was in turn the general partner in SAIF II GP L.P. (“SAIF GP”), in which there was one limited partner. Finally, SAIF GP was the general partner of SB Asia Investment Fund II L.P. (“the Fund”) in which the limited partners were a number of external investors. The effect of this structure was that SAIF, in which Mr Zhou was a partner, controlled (via its control over SAIF GP) the business of the Fund. It was also entitled to a management fee and 20% of the profits on investments made by the Fund. 8. Under the Cayman law, a limited partner does not, as such, owe any fiduciary duties to the other partners. But that is no reason why he should not enter into a separate agreement with one or more of the other partners which creates such duties. There was expert evidence at trial on Cayman law (which governed the position) to this effect. In this case the relations between Mr Zhou and the other partners in SAIF, all of whom worked in the business, were governed at the relevant time by an Amended and Restated Limited Partnership Agreement of 29 June 2005 (“the ALPA”). The judge found that the ALPA gave rise to a fiduciary duty on the part of Mr Zhou, owed to the other partners in SAIF, to perform his duties as a partner in the interests of the partnership and the Fund which it controlled and not to make a personal profit from investment opportunities available to the Fund. This finding was upheld by the Court of Appeal and leave to appeal against it has been refused. 9. The judge went on to find that in October 2005, while a partner in SAIF, Mr Zhou identified an investment opportunity in a PRC technology company named Show World Information Technology Co Ltd (“Show World”). On his recommendation the Fund in April 2006 invested US$8 million through a new Cayman company, My Show Group Limited (“My Show”) which established a PRC enterprise called Beijing Show World Science and Technology Co Ltd (a wholly owned foreign enterprise or “WOFE”) to which the assets and undertaking of Show World were transferred. Mr Zhou was appointed a director of My Show and the WOFE. 10. During 2006 an employee of the WOFE developed a new technology for internet advertising. The Respondents alleged, and the judge found, that this could have been profitably developed by the WOFE. Instead, Mr Zhou and the general manager of the WOFE diverted it to a company named Xinrui in which Mr Zhou afterwards acquired a personal interest. The judge held that this transaction was a breach of Mr Zhou’s fiduciary duty to SAIF, in consequence of which he was accountable for the profit he had made. This finding was also upheld by the Court of Appeal and leave to appeal against it was refused. 11. These two findings led to the conclusion that Mr Zhou was accountable to SAIF for any profit he had made on the Xinrui transaction. The judge made an order accordingly. But he went on to say although SAIF and its general partner were the only counter-claiming defendants, “[i]n substance, the ultimate beneficiaries to whom Mr Zhou owed the duty not to make a secret profit from his position as a Limited Partner were the external investors who contributed to [the Fund].” He said further that “[i]n order to avoid over-recovery by [SAIF] and to protect the interests of [the Fund], [SAIF] must, in my opinion, hold the profit …to be disgorged by Mr Zhou on trust for [the Fund]”. Counsel for the Respondents did not dissent from these propositions and consented to an order in those terms. 12. This part of the judgment, although as a matter of Cayman law a correct statement of the relationship between SAIF and the Fund, was unfortunate because it suggested (at any rate to Mr Barlow SC, counsel for Mr Zhou) that SAIF was suing “on behalf of” the contributors to the Fund and that the court did not have jurisdiction to make such an order unless they had authorised the proceedings or were parties to the action. Hence the certified question in this appeal: “Do our courts have jurisdiction (in either the literal sense or the broader sense), following the trial of an Action, to grant substantive remedies to a counterclaiming defendant, as a ‘trustee’ for non-parties whom the defendant has neither joined in the Action nor shown any specific authority to represent therein?”. 13. But the true position was that SAIF was not suing “on behalf of” anyone. It was asserting its own cause of action, as the party to whom the fiduciary duty created by the ALPA was owed. While SAIF, as indirect general partner of the Fund, did owe its limited partners a fiduciary duty,[1] I doubt whether Mr Zhou personally did so. He was a limited partner in the Fund but the Cayman statute expressly provides that limited partners do not as such owe any fiduciary duties.[2] He made no profit from his position as a limited partner in the Fund, the bottom rung of the fund administration structure. He made it from the knowledge he acquired as a partner in SAIF, the top rung where the decisions were made. On the other hand, if the WOFE had been a Hong Kong company, he would probably have been in breach of his fiduciary duty as a director by diverting an investment opportunity to Xinrui. There was no evidence of what the position was under PRC law. 14. But in my opinion none of this matters because SAIF had in its own right a complete cause of action for an account of profits. The judge found that the ALPA, which was an agreement between Mr Zhou and the other partners in SAIF, created a fiduciary duty and that Mr Zhou was liable to SAIF to account for any profit he had made. There is no question of “over-recovery” because it has been the law for at least three centuries that the purpose of the rule that a fiduciary must disgorge unauthorised profits is not compensatory: see Keech v. Sandford (1726) Sel. Cas. Ch. 61. As Viscount Sankey said in Regal (Hastings) Ltd v. Gulliver [1967] 2 AC 134 (Note) at pp.144-145: “The rule of equity which insists on those, who by use of a fiduciary position make a profit, being liable to account for that profit, in no way depends on fraud, or absence of bona fides; or upon such questions or considerations as whether the profit would or should otherwise have gone to the plaintiff, or whether the profiteer was under a duty to obtain the source of the profit for the plaintiff, or whether he took a risk or acted as he did for the benefit of the plaintiff, or whether the plaintiff has in fact been damaged or benefited by his action. The liability arises from the mere fact of a profit having, in the stated circumstances, been made. The profiteer, however honest and well-intentioned, cannot escape the risk of being called upon to account.” 15. Thus the question of whether SAIF, having received from Mr Zhou the proceeds of this action, ought to account for them to the Fund is res inter alios acta, a matter between SAIF and the Fund. It is not relevant to the liability of Mr Zhou to SAIF. If a trustee who has sold trust property sues for the price, he is not suing “as a trustee”. He is suing as vendor and, as between him and the purchaser, his trusteeship is irrelevant. Likewise, if a trustee of property employs an agent to sell it and the latter makes a secret profit, he is liable to his principal to account for the profit. The fact that the principal is a trustee and may have to account to the beneficiaries for whatever he receives from the agent is irrelevant. Mr Barrie Barlow SC and Mr Chan Pat Lun, instructed by MinterEllison LLP, for the Plaintiff (Appellant) Mr Charles Manzoni SC and Mr Alexander Tang, instructed by Fangda Partners, for the 1st and 2nd Defendants (Respondents) [1] Cayman Islands Exempted Limited Partnership Law, s.19(1) A general partner shall act at all times in good faith and, subject to any express provisions of the partnership agreement to the contrary, in the interests of the exempted limited partnership. [2] Ibid., s.19(2) Subject to any express provisions of the partnership agreement to the contrary, a limited partner of an exempted limited partnership in that capacity does not owe any fiduciary duty in exercising any of its rights or authorities or otherwise in performing any of its obligations under the partnership agreement to the exempted limited partnership or any other partner. 
Mr Justice Ribeiro PJ, Mr Justice Fok PJ and Lord Neuberger of Abbotsbury NPJ: 1. At the conclusion of the hearing of this appeal on 22 October 2019, the Court reserved its judgment and indicated that it would hand down its judgment on a date to be advised to the parties. On 5 November 2019, the parties wrote to the Registrar to inform him that they had reached an in principle agreement to settle the appeal. By that time, a draft judgment had already been completed and was in the final stages of preparation in readiness for handing down. The Registrar replied, telling them that the judgment had been prepared and would soon be ready for delivery, and indicating that if informed by the parties that they had concluded a settlement agreement, the Court would decide whether to proceed to hand down its judgment. 2. The parties subsequently confirmed that they had settled the appeal and submitted a consent summons for the appeal to be stayed on terms agreed by the parties. The Court duly made the order sought. For the reasons which follow, the Court has decided that notwithstanding the settlement, judgment should be handed down. 3. The question of the scope of a court’s discretion to deliver judgment in a case notwithstanding the settlement of a case after argument but before judgment is delivered has not previously arisen in any reported decision in Hong Kong. However, in England and Wales, the applicable principles were established by the Court of Appeal’s decision in Barclays Bank plc v Nylon Capital LLP,[1] stated by Lord Neuberger of Abbotsbury MR (as his Lordship then was) as follows: “73. I turn now to deal with a very different issue. After Thomas LJ had prepared his judgment in draft, and circulated it to Etherton LJ and me, the parties notified the court that they had reached agreement and effectively requested the court not to give judgment. 74. Where a case has been fully argued, whether at first instance or on appeal, and it then settles or is withdrawn or is in some other way disposed of, the court retains the right to decide whether or not to proceed to give judgment. Where the case raises a point which it is in the public interest to ventilate in a judgment, that would be a powerful reason for proceeding to give judgment despite the matter having been disposed of between the parties. Obvious examples of such cases are where the case raises a point of law of some potential general interest, where an appellate court is differing from the court below, where some wrongdoing or other activity should be exposed, or where the case has attracted some other legitimate public interest. 75. It will also be relevant in most cases to consider how far the preparation of any judgment had got by the time of the request. In the absence of good reason to the contrary, it would be a highly questionable use of judicial time to prepare a judgment on an issue which was no longer live between the parties to the case. On the other hand, where the judgment is complete, it could be said (perhaps with rather less force) that it would be a retrospective waste of judicial time and effort if the judgment was not given. 76. The concerns of the parties to the litigation are obviously also relevant and sometimes very important. If, for their own legitimate interests, they do not wish (or one of them does not wish) a judgment to be given, that request should certainly be given weight by the court. (Of course, in some cases, the parties may request a judgment notwithstanding the fact that there is no longer an issue between them). 77. Where there are competing arguments each way, the court will have to weigh up those arguments: in that connection, the reasons for any desire to avoid a judgment will be highly relevant when deciding what weight to give to that desire.” 4. Barclays Bank was adopted by the Supreme Court of Victoria in Clarke v Great Southern Finance Pty Ltd[2] and similar principles were applied by the Queensland Court of Appeal in Voss v Suncorp-Metway Limited (No.1).[3] Principles similar to those laid down in Barclays Bank, were also applied by the Supreme Court of New Zealand in Osborne v Auckland Council.[4] 5. We are satisfied that the principles set out in the citation from Barclays Bank at [3] above should be adopted in this jurisdiction in exercising the Court’s discretion whether or not to deliver judgment in an appeal notwithstanding the parties’ settlement. 6. The present case involves issues of law of general importance and, as was pointed out in argument, has attracted considerable public interest in Hong Kong and internationally. The draft judgment was completed and in the final stages of preparation for handing down when the parties informed the Registrar of the impending settlement. As will be seen in the reasons which follow, we are differing from the judgments appealed from and consider it necessary to correct certain erroneous propositions accepted below. Publication of this judgment will not impinge on any issues regarding confidentiality or privacy as it does not go beyond the detailed account of the parties’ dealings set out in the judgments below. The parties have not made any submissions indicating any opposition to the judgment being handed down. For all these reasons, we consider that in the proper exercise of our discretion, we should hand down our judgment notwithstanding the settlement. A. Introduction 7. This appeal arises out of claims made in respect of the administration of a trust established to hold the sole share of a private investment company which was used by the clients of a private bank to invest in various financial products. The issues on appeal relate to the two heads of claim which succeeded below. Those issues, raised by the appellants, who were the former trustees of the trust in question and the corporate services company which was the sole director of the private investment company through which the financial investments were made, relate to the bases on which the trial judge found liability established against them and on which he directed the assessment of equitable compensation, which were affirmed by the Court of Appeal. 8. The underlying facts are complicated and gave rise to litigation which the Judge described towards the end of his 378-page judgment as “complex, costly and prolonged” and we will in this judgment first provide a broad summary of the background before setting out some of the facts that may need to be examined in greater detail when addressing the issues raised. A.1 The parties and trust structure involved 9. The genesis of the trust arose from the introduction, in 2004, of Madam Ji Zhengrong (the 2nd plaintiff) (“Ji”) and her husband Zhang Hong Li (the 1st plaintiff) (“Zhang”) to DBS Bank (Hong Kong) Limited (the 1st defendant) (“DBS Bank”), the Hong Kong subsidiary of DBS Bank Limited of Singapore. Ji and Zhang had concerns about the protection of family assets from inheritance tax and were interested in setting up a trust for that purpose. The trust structure used by Zhang and Ji was DBS Bank’s most common trust structure and contemplated the use of a BVI company as the trust’s private investment company, with Ji acting as the company’s investment advisor. The Judge, before whom Ji testified at trial, was impressed by her high level of sophistication and intelligence. 10. Prior to the establishment of the trust in January 2005, through the assistance of DBS Corporate Services (Hong Kong) Limited (the 3rd defendant) (“DBS Corporate”), Ji and Zhang procured the incorporation of Wise Lords Limited (the 4th plaintiff) (“Wise Lords”), a BVI company of which Ji was sole director and shareholder. In the period from April 2004 until January 2005, with Ji as its sole signatory, Wise Lords made investments through a private banking account with DBS Bank. Ji and Zhang injected funds into Wise Lords’ account with the private banking division of DBS Bank (“DBS:PB”) and Wise Lords invested in various mutual funds and other investment products, including currency linked notes called Yield Enhanced Deposits (“YEDs”). 11. Ji and Zhang established the trust at the heart of this case, called the Amsun Trust (“the Trust”), by a Trust Deed dated 4 January 2005, the terms of which will be addressed in greater detail in Section B below. Ji and Zhang were the settlors of the Trust and the original trustee was DBS Trustee HK (Jersey) Limited (the 2nd defendant) (“DBS Trustee” or “the trustees”) and the beneficiaries were Ji, Zhang and their two minor sons, Anthony and Max. It was intended that Ji would be the investment advisor of the Trust. 12. Upon the establishment of the Trust, the sole share in Wise Lords was transferred to DBS Trustee and, apart from the sum of USD10 settled on the Trust as its initial property, became the Trust’s sole asset. In addition, Ji was replaced as the sole director of Wise Lords by DHJ Management Limited (the 4th defendant) (“DHJ Management”). By an Investment Advisor Agreement dated 4 January 2005, Wise Lords appointed Ji to be its investment advisor. Also on that date, Ji and Zhang executed a Letter of Wishes in respect of the Trust and directed to DBS Trustee which included the request that, whilst Ji was alive, DBS Trustee “should always consult her in the first place with regards to all matters and her recommendation should be final.” Further, by an Authorisation Letter dated 5 January 2005, DHJ Management granted authority to Ji to give investment instructions on behalf of Wise Lords. These arrangements were consistent with Ji and Zhang’s contemplated trust structure. 13. The services provided by DBS Corporate were governed by a Services Agreement dated 13 September 2005, entered into between DBS Corporate, Wise Lords and DBS Trustee, under which DBS Corporate provided services including the provision of a Nominee Director for Wise Lords, company secretary services, a correspondence address and bank authorised signatories. It was pursuant to this agreement that DBS Trustee nominated DHJ Management to act as director of Wise Lords. 14. After the relationship between Ji and DBS Bank deteriorated leading to this litigation, Bruno Arboit (“Arboit”) and Roderick John Sutton (“Sutton”) (the 3rd plaintiffs) became the trustees of the Trust. 15. Liu Hiu Hong, Linda (the 7th defendant) (“Linda Liu”) was employed as a vice president and relationship manager in DBS:PB and was the principal contact person between DBS Bank and Ji. Lim Leung Yau, Edwin (the 6th defendant) (“Edwin Lim”) was the executive head of DBS:PB. Lee Kwok Tai, Peter (the 5th defendant) (“Peter Lee”) was the head of DBS Bank’s Trust and Corporate Services division, executive head of DBS Corporate Services and supervisor of DHJ Management. A.2 Wise Lords’ investments 16. From January 2005 to April 2008, under Ji’s direction, Wise Lords invested principally, and successfully, in mutual funds of shares in the PRC, achieving profits in each of the financial years to 31 March 2006, 2007 and 2008 respectively and generating overall profits in that period in excess of HKD132.6m (USD17m). 17. In December 2006, on the recommendation of Ji and Zhang, Wise Lords applied for and was granted a credit facility of HKD78m (USD10m). Subsequently, Ji applied for increasing amounts of credit to be extended to Wise Lords to fund her investments through Wise Lords. Through 15 progressive increases between January 2007 and July 2008, Wise Lords’ credit facility was raised from HKD78m (USD10m) to HKD780m (USD100m). These credit facilities were used to fund the investments made by Wise Lords. Its investment in mutual funds grew from USD25.87m as at 31 December 2006 to USD106.37m as at 31 October 2007. 18. From about October 2007, the investments in mutual funds began to show diminished returns and so, in 2008, the investment strategy switched by Ji from mutual funds into foreign exchange transactions focusing on the Australian dollar (“AUD”), Euros (“EUR”) and YEDs. From the middle of 2008, Ji preferred a strategy of direct foreign exchange transactions rather than YEDs. By utilising Wise Lords’ USD credit facility, Ji borrowed in USD to buy AUD or EUR in order to earn higher interest. The strategy was successful when AUD and EUR appreciated against the USD but involved the risk of loss if those currencies depreciated against the USD. As at 18 August 2008, Wise Lords’ portfolio was concentrated to the extent of approximately 85% in foreign currency exposure (81% in AUD) and was leveraged to the extent of 272%, representing net assets of USD35.4m against borrowings of USD96.4m. From time to time, Linda Liu and other employees of DBS:PB recommended to Ji diversification away from such heavy concentration in AUD, but these recommendations were not accepted and Ji persisted in accumulating AUD. 19. From late July 2008 through to August 2008, the AUD began to fall against the USD. Ji was unwilling to sell Wise Lords’ AUD at a loss and Linda Liu introduced decumulators to Ji as an exit strategy, and she purchased three. The decumulators involved making (in this case) a fixed deposit for one year in one currency (say AUD or EUR) maturing in equal weekly instalments with possible conversion into another currency (say USD) dependent upon the movement of exchange rates on the market and the level of “strike” and “knockout” rates agreed by the parties over the course of that year. As a result of the continuing decline in the exchange rate of the AUD against the USD, the exchange rate fell below the rate at which USD would be received under the decumulators. Instead, Wise Lords was left with AUD at a declining exchange rate and faced possible margin calls because of the leveraging of its investments. To avoid these, in September 2008, at the height of the global financial crisis that year, Ji had to sell over AUD60m of Wise Lords’ AUD cash holdings for a significant loss. 20. The AUD exchange rate continued to fall in September and October 2008, resulting in DBS:PB issuing further margin calls to Wise Lords. Eventually, by 14 November 2008, two of Wise Lords’ three decumulators, denominated in AUD, were unwound involving aggregated termination costs of AUD1.5m and a loss of approximately USD15m. Wise Lords’ losses on its other decumulator, denominated in EUR, was approximately USD1.2m at this date. A.3 The proceedings 21. In December 2008, Ji and Zhang raised complaints through their solicitors claiming USD50m as losses caused by alleged breaches of duty by DBS Trustee and DBS:PB. On 31 January 2011, Zhang, Ji, DBS Trustee and Arboit and Sutton entered into a Deed of Appointment Retirement and Indemnity by which Arboit and Sutton replaced DBS Trustee as the trustees of the Trust in February 2011. 22. The plaintiffs commenced the proceedings leading to this appeal[5] on 28 February 2011 asserting claims: against DBS Trustee for both dishonest and negligent breach of trust; against DBS Corporate and DHJ Management for both dishonest and negligent breach of fiduciary duty; against DBS Bank, Peter Lee, Edwin Lim and Linda Liu for dishonest assistance of the asserted breaches of trust and fiduciary duty. 23. Following a 24-day trial, Bharwaney J dismissed the plaintiffs’ claims against DBS Bank, DBS Corporate, Peter Lee, Edwin Lim and Linda Liu. A cross-appeal by the plaintiffs to the Court of Appeal against the dismissal of those claims[6] was dismissed and a further application by the plaintiffs for leave to appeal to this Court was refused by the Appeal Committee.[7] 24. The Judge gave judgment for the plaintiffs against DBS Trustee and DHJ Management. He held that DBS Trustee were liable to the plaintiffs for negligent breach of trust and that DHJ Management were liable to the plaintiffs for negligent breach of fiduciary duty on the same reasoning. The basis on which the Judge found liability established will be further examined in Section B.1 below. He ordered DBS Trustee to pay Arboit and Sutton equitable compensation for breach of trust and DHJ Management to pay Wise Lords equitable compensation for breach of fiduciary duty. The basis and scope of the order to pay equitable compensation will be further considered in Section E below. 25. DBS Trustee and DHJ Management’s appeal against the judgment entered against them was dismissed by the Court of Appeal[8] as was their application to the Court of Appeal for leave to appeal to this Court.[9] The basis on which the Court of Appeal found liability established will be further examined in Section B.2 below. Leave to appeal to this Court was granted by the Appeal Committee by a Determination dated 26 April 2019.[10] The six questions for which leave to appeal was granted are set out in Section A.7 below. 26. The Judge also entered judgment for Arboit and Sutton against DBS Trustee in respect of various other claims. He ordered DBS Trustee to repay to the Trust USD68,825 and GBP13,325 (being amounts wrongfully paid out from the Trust for charges which were unrelated to the administration of the Trust) and interest thereon. He also ordered DBS Trustee to repay to the Trust USD1m (which the Judge held DBS Trustee had wrongfully retained as an indemnity pursuant to the Deed of Appointment Retirement and Indemnity) and interest thereon. There was no appeal from the first instance judgment against those orders. A.4 The terms of the Trust Deed 27. The Trust Deed provided that the Trust was to be governed by Jersey law. Relevant provisions of the 1984 Trust (Jersey) Law (“the 1984 Law”) are set out in Section A.5 below. 28. Relevant clauses of the Trust Deed provided as follows: (a) Clauses 1 and 2 provided definitions of key terms and the declaration of the proper law of the Trust. (b) Clauses 3 to 5 contained the declaration of trust over the Trust Fund which the Trustee was required to hold upon trusts over both the capital and income “for the benefit of ... the Beneficiaries” and which the Trustee was obliged to administer. (c) Clauses 6 and 9 to 11 conferred the Trustee’s powers of appointment, advancement, addition and exclusion. (d) Clauses 12, 13, 15 to 17, 22 and the First Schedule conferred various other specific trustee powers. (e) Clause 14 required the Trustee to exercise the trust powers “for the benefit of all or any one or more of the Beneficiaries”. (f) Clauses 19 and 20 provided for the release of an outgoing trustee from liability, except for liability in respect of: (a) any breach of trust arising from fraud, wilful misconduct or gross negligence; (b) any action to recover trust property in the possession of an outgoing trustee; or (c) any exercise of trust powers not made in good faith. A.5 Relevant provisions of Jersey Law 29. Evidence of Jersey law was adduced by the parties at trial. The plaintiffs’ expert was Jonathan Speck (“Speck”) and the defendants’ expert was Professor Paul Matthews (“Matthews”). The experts agreed that the duties owed by DBS Trustee as original trustee of the Trust from 2005 to 2011 were the statutory duties set out in the 1984 Law, the key duties thereunder being: (a) Subject to the 1984 Law, to “carry out and administer the trust in accordance with its terms” (Article 21(2)); (b) “Subject to the terms of the trust” and “so far as is reasonable”, to preserve and enhance the value of the trust property (Article 21(3)); (c) In the execution of its duties, and in the exercise of its powers, to act with due diligence, as would a prudent person, to the best of the trustee’s ability and skill, and to observe the utmost good faith (Article 21(1));[11] (d) “Subject to the terms of the trust and subject to the trustee’s duties under this Law, a trustee shall in relation to the trust property have all the same powers as a natural person acting as the beneficial owner of such property” (Article 24(1)); (e) To exercise the trustee’s powers only in the interests of the beneficiaries and in accordance with the terms of the trust (Article 24(2)). 30. Article 30(10) of the 1984 Law provided: “Nothing in the terms of a trust shall relieve, release or exonerate a trustee from liability for breach of trust arising from the trustee’s own fraud, wilful misconduct or gross negligence.” A.6 The First Schedule to the Trust Deed 31. The First Schedule to the Trust Deed contained a number of provisions which qualified the provisions of the 1984 Law. 32. Paragraph 1 of the First Schedule provided: “Subject always to any provisions or restrictions expressly contained in this Settlement the Trustees shall in relation to the Trust Fund and in particular but without prejudice to the generality of the foregoing the investment of the Trust Fund have all the same powers as a natural person acting as the beneficial owner of such property and such powers shall not be restricted by any principle of construction or rule or requirement of the Proper Law of this Settlement save to the extent that such is obligatory but shall operate according to the widest generality of which the foregoing words are capable notwithstanding that certain powers are hereinafter more particularly set forth.” 33. Paragraph 2 of the First Schedule relevantly stated: “(a) The Trust Fund may be invested or laid out in the purchase of (or at interest upon the security of) such property whether involving liability or not and whether producing income or not or upon such personal credit with or without security as the Trustees shall in their absolute discretion think fit including the purchase erection and improvement of any property as a residence for any person and the purchase of chattels for the use of any person to the intent that the Trustees shall have the same full and unrestricted powers of investing and transposing investments and laying out moneys in all respects as if they were absolutely entitled thereto beneficially and without regard to the requirements of the Proper Law of this Settlement save to the extent that these are obligatory. (b) The acquisition of any investment of a speculative nature shall be deemed to be an authorised investment of the whole or any part of the Trust Fund. ... (g) The Trustees shall be under no duty to diversify investments. ... (j) In the exercise of the powers herein contained the Trustees shall not be under any duty to see that the value of the Trust Fund or any part thereof is preserved or enhanced in any way nor shall they be liable for any failure in those respects whatsoever.” 34. The First Schedule of the Trust Deed also contained provisions commonly referred to as anti-Bartlett provisions, so named since they are provisions intended to disapply duties elaborated in Bartlett and others v Barclays Bank Trust Co Ltd (Nos.1 and 2).[12] These provisions were contained in paragraphs 4 and 5 of the First Schedule. (a) Paragraph 4 (headed “Trustees Not Bound to Interfere in Business of Company in which Settlement is Interested”) stated: “(a) The Trustees shall not be under any duty nor shall they be bound to interfere in the business of any company in which this Settlement is interested and in particular:- (i) the Trustees shall not be under any duty to exercise any control the Trustees may have over or to interfere in or become involved in the administration management or conduct of the business or affairs of any company in which this Settlement is or may be interested whether or not this Settlement holds the whole or a substantial proportion of the shares carrying the control of the company and without prejudice to the generality of the foregoing the Trustees shall not be under any duty to exercise any voting powers or rights of representation or intervention conferred on the Trustees by any of the shares in respect of such company; (ii) the Trustees shall leave the administration management and conduct of the business and affairs of such company to the directors officers and other persons authorised to take part in the administration management or conduct thereof and the Trustees shall not be under any duty to supervise such directors officers or other persons so long as the Trustees do not have actual knowledge of any dishonesty relating to such business and affairs on the part of any of them; and (iii) the Trustees shall assume at all times that the administration management and conduct of the business and affairs of such company are being carried on competently honestly diligently and in the best interests of the Trustees in their capacity as shareholders or howsoever they are interested therein until such time as they shall have actual knowledge to the contrary and so that the Trustees shall not be under any duty at any time to take any steps at all to ascertain whether or not the assumptions contained in this sub-clause are correct. (b) Without prejudice to the generality of the foregoing, the Trustees shall be under no duty:- (i) to exercise any rights or powers (whether available to them as shareholders debenture holders or otherwise) enabling them to appoint or elect or to remove a director officer or other person authorised to take part in the administration management or conduct of the business or affairs of such company and in particular the Trustees shall not be under any duty to take any steps to see that any Trustee or any officer or nominee of the Trustees becomes a director or other officer of such company; or (ii) to exercise any power to require the payment of a dividend or other distribution of profit and whether of an income or capital nature. (c) No Beneficiary shall be entitled in any way whatsoever to compel control or forbid the exercise in any particular manner of any powers discretions or privileges (including any voting rights) conferred on the Trustees by reason of any shares or other rights of whatsoever nature in or over such company. (d) The Trustees shall not be liable in any way whatsoever for any loss to such company or the Trust Fund or the income thereof arising from any act or omission of the directors officers or other persons taking part (whether or not authorised) in the administration management and conduct of the business or affairs of such company (whether or not any such act or omission by any such foregoing persons shall be dishonest fraudulent negligent or otherwise). (e) Without prejudice to the generality of the foregoing the Trustees shall not be rendered responsible in any way whatsoever for any default or other act or omission by the directors officers or other persons referred to in paragraph (d) above by any express notice or intimation of such default or other act or omission and the Trustees shall not be obliged or required to make and enforce any claim in respect of such a default or other act or omission and no person who is or may become entitled hereunder shall be entitled to compel the making of such a claim but the Trustees may be required to lend their names for the purpose of proceedings brought by a Beneficiary in respect of any such default act or omission upon being given a full and sufficient indemnity against all costs and expenses of such proceedings.” 35. Paragraph 5 (headed “Trustees Not Bound to Obtain Information Regarding Company in which Settlement is Interested”) materially stated: “(a) The Trustees shall not be under any duty to obtain or to seek to obtain in any way whatsoever any information regarding the administration management or conduct of the business or affairs of any company in which this Settlement is or may be interested (although this Settlement holds the whole or a majority of the shares carrying the control of the company) from the persons involved in the administration management or conduct or from the shareholders or other persons interested therein or any other matter relating to such company. (b) The Trustees shall assume that such information as is supplied to them by any person relating to such company is accurate and truthful unless the Trustees have actual knowledge to the contrary and the Trustees shall not be under any duty at any time to take any steps at all to ascertain whether or not the information is accurate and truthful. ... (d) The Trustees shall not be liable in any way whatsoever for any loss sustained by the Trust Fund or the income thereof arising from the Trustees not taking all or any possible steps to obtain any information referred to in paragraph (a) above or to verify the accuracy and truthfulness of such information as is supplied to the Trustees. ...” A.7 The questions for which leave to appeal was granted 36. The first four questions (of the six) for which leave to appeal was granted by the Appeal Committee concern the effect of the anti-Bartlett provisions and the correctness of the Judge’s holding (affirmed by the Court of Appeal) that DBS Trustee were liable for breach of trust. Those questions are: (1) Question 1: “Does a trustee owe any duty to supervise the investment decision making of an investment adviser appointed by an underlying company and/or to review investment decisions made by such investment adviser where the terms of the trust contain extensive and mandatory ‘anti-Bartlett’ provisions which expressly forbid the trustee’s interference with the management of the Company save where it has actual knowledge of dishonesty?” (2) Question 2: “If so, what is the nature and extent of such a duty?” (3) Question 3: “Does the de facto assumption by a trustee of an undefined role of supervision in respect of investment decisions made by the investment adviser to the company, and/or the de facto assumption by a trustee of a role involving the after-the-event review and/or the giving of after-the-event approvals of investments make any difference to the analysis?” (4) Question 4: “Whether the Court is entitled to interpret and/or add a gloss to the evidence of an expert on foreign law in a manner which is inconsistent with the express terms of the document under consideration by the expert when an interpretation or meaning which is fully consistent with such terms is available to it. Alternatively, is the Court itself permitted, entitled or required to construe the document and come to its own conclusions as to the meaning of such express terms in order to enable it to properly interpret, give effective meaning and make findings based on such evidence.” 37. The fifth question for which leave to appeal was granted concerns the Judge’s holding (affirmed by the Court of Appeal) that DHJ Management were liable for breach of fiduciary duty: Question 5: “What is the appropriate standard which should be adopted to assess the conduct of a corporate director of a company which has properly delegated investment management and investment decision making to an investment adviser as its agent or delegate? Specifically in this case what was the position of the 4th Defendant (‘DHJ Management’) as such director with regard to the decision-making of the 2nd Plaintiff (‘Ji’) as investment adviser and DHJ Management’s delegate in relation to investment strategy and decision making, in light of such standard and, in particular, the exonerations and indemnities available to DHJ Management.” 38. The sixth question for which leave to appeal was granted concerns the relief ordered by the Judge (and affirmed by the Court of Appeal). Question 6: “What is the correct and/or appropriate formula for assessing equitable compensation in the light of the proper approach to assessing causation?” A.8 The issues in this appeal 39. The first five questions raise the issue of whether, notwithstanding the terms of the Trust, there was a “high level supervisory duty” on the part of DBS Trustee and a like fiduciary duty on the part of DHJ Management, as held by the Judge and affirmed by the Court of Appeal. If so, the question arises as to whether the Judge and Court of Appeal were correct to hold that DBS Trustee and DHJ Management were each in breach of their respective duties. 40. The sixth question raises the issue of the proper approach to equitable compensation in the event that the judgments below in respect of duty and breach are upheld. Leave to appeal was also granted on the “or otherwise” ground in respect of the question whether it was right for the Judge to have determined which of the transactions were to be included or excluded in the computation of equitable compensation before according DBS Trustee and DHJ Management the opportunity to address the court on the causal connection (or lack thereof) between the breach of trust or fiduciary duty and the loss arising from the transactions. B. Did DBS Trustee and DHJ Management owe the supervisory duties found by the courts below? B.1 The Judge’s approach to the duties 41. The Judge found that the trustees had committed breaches of trust in that (i) they ought to have queried and not given approval for the further purchases by Wise Lords of USD83m worth of AUD from 25 July 2008 to 5 August 2008;[13] (ii) they wrongly approved the increase of Wise Lords’ credit facility to USD100m at the end of August 2008;[14] and (iii) they wrongly approved its purchase of the three decumulators on 18, 27 and 29 August 2008 respectively.[15] These were said to be breaches of the trustees’ “high level supervisory duty” over Wise Lords’ investments and also breaches of “their duty to act with due diligence, to act as would a prudent person, to act to the best of his ability and skill, and to act only in the interests of the beneficiaries”. In acting or omitting to act as they did, the trustees were found to have behaved in a “flagrantly negligent” manner. 42. Before exploring the legal basis for the Judge’s conclusions, it should be noted that the judgment does not clearly state what his Lordship considered to be the objectionable features of the three transactions so as to make their approvals grossly negligent breaches of trust. (a) The fact that DBS Bank was willing to increase Wise Lords’ credit facility to USD100m is not, on its face, a bad thing. It certainly does not constitute a loss to the trust fund. Previously, towards the end of May 2008, the facility had been increased to USD58m (HKD450m), approved by the trustees on 4 June 2008, but no complaint is made about the grant of that facility. While one might wish to ask how the increased facility is intended to be used, it is hard to see how approval of such an increase in itself merits description as a breach of trust, let alone a flagrantly negligent breach of trust. (b) Similarly, the Judge does not make it clear why he considered approval of Wise Lords’ purchase of an additional USD83m worth of AUD from 25 July 2008 to 5 August 2008 to be a grossly negligent breach of trust. One can of course see that such a purchase might be said to involve the risk of heavy concentration in a single currency against the background of a developing global financial crisis and a downward trend in the value of the AUD. But the Judge is not explicit in his criticism. Buying a foreign currency in a downward trend in anticipation of a rebound may be speculative, but it is hard to see how it constitutes flagrantly negligent conduct even leaving aside for now the anti-Bartlett provisions which authorise speculative and non-diversified investments. Ji had deliberately ignored many warnings about lack of diversification and her previous experience had led her to bet on the strength of the AUD against the USD. For instance, on 24 June 2008, a month before the start of the period of purchases complained of, Kenneth Cheung, vice-president of the investment advisory team of DBS:PB: “... told Ji that AUD was at a record high level and noted that Ji had purchased a lot of AUD. He asked Ji if she wanted to sell. Kenneth Cheung also suggested that Ji could sell half of her 43 million AUD holding and take profit, and alerted Ji to the risk of a possible USD rebound and a fall of the AUD. However, Ji said she would not do anything with the AUD.”[16] Thus, AUD purchases were evidently investments which could have been profitably realised at that stage. On 18 July 2008, Ji was pressing the Bank to increase her credit limit because she was anxious to purchase more AUD and EUR as they fell in the exchange rates.[17] (c) Similarly, it is hard to conclude that approving purchase of the three decumulators was a grossly negligent breach of trust. They were acquired in response to Ji’s unwillingness, in the face of repeated warnings by the Bank against over-concentration, to sell off any part of Wise Lords’ AUD holdings at anything less than breakeven point.[18] The decumulators were purchased as a possible means of liquidating part of Wise Lords’ AUD holdings at a more favourable rate of exchange against the USD. Whether that could be achieved depended on how the market would move in relation to the “knockout” rate agreed upon. The investment undoubtedly involved a risk since the AUD deposited for the purposes of the decumulators would be locked up for a year to be released only in weekly instalments with the possibility of losses due to further deterioration in the value of AUD. But at the same time, there was a possible gain if released instalments could be converted to USD at a higher than market rate. 43. Hints or indications of the grounds of the Judge’s disapproval can be found along the following lines. (a) In relation to the AUD83m purchases, his Lordship states that the trustees “ought to have queried and not given approval” and that approval “is not something which a trustee, complying with the duty to act prudently, could reasonably have done”.[19] In the earlier paragraphs referred to in support,[20] he describes the purchases of AUD as occurring “against a backdrop of substantial market nervousness created by the sub-prime crisis”.[21] The implication is therefore that the trustees ought to have disapproved of the purchases as “imprudent”, presumably because they should have been considered too speculative and entailed the risk of over-concentration in AUD in a “nervous” market. (b) The criticisms of the credit facility increase and the decumulator purchases are to similar effect. The charge that approvals of those transactions are “not something which a trustee, complying with the duty to act prudently, could reasonably have done” is repeated. While it is less easy to see why approving Wise Lords’ acceptance of an increased credit facility was imprudent, one may assume that the decumulators were disapproved of as presenting a risk of being “locked up” in AUD for a lengthy period and thus unable to react to a sharp depreciation in the currency’s value (at least, without paying the financial penalties eventually paid to unwind the transactions)[22]. 44. One may therefore proceed on the basis that the Judge held the approvals to be breaches of trust because they involved approving speculative and risky investments by Wise Lords constituting a breach of a duty to supervise (at a “high level”, whatever that might mean) Wise Lords’ investments, such breach being characterised as “flagrantly negligent”. It is, incidentally, noteworthy that the criticised approvals were in each case sought from and given by DBS Trustee after the transactions in question had already been effected on Wise Lords’ behalf at Ji’s instigation – a feature which somewhat colours the suggestion that such “approvals” were imprudent. 45. In any event, to conclude, as a matter of law, that the “high level supervisory duty” existed is in our view plainly inconsistent with the anti-Bartlett provisions. (a) The basic duty of the trustees was (subject to statutory requirements) to carry out and administer the trust in accordance with its terms.[23] It is difficult to see how a “high level supervisory duty” over Wise Lords’ investments, allegedly manifested as a duty to disapprove the three transactions in question, can have been imposed on the trustees given that in the Trust Deed: (i) paragraph 4(a)(i) of the First Schedule[24] expressly relieves them of any duty to exercise control or interfere or become involved with the management or conduct of the investment company’s business (whether qua shareholder or otherwise); and (ii) paragraph 4(a)(ii) thereof stipulates that the trustees “shall leave the administration management and conduct of the business and affairs of [Wise Lords] to the directors officers and other persons authorised to take part in the administration management or conduct thereof [i.e. Ji]”, going on expressly to relieve them of any duty to supervise such directors officers or other persons “so long as the Trustees do not have actual knowledge of any dishonesty relating to such business and affairs” on the part of those having conduct of the same. No such actual knowledge arises in this case. (b) The displacement of any supervisory duty is reinforced by paragraph 4(iii) of the First Schedule which requires the trustees to “assume at all times” that the business is being “carried on competently honestly diligently and in the best interests of the Trustees in their capacity as shareholders or howsoever they are interested therein until such time as they shall have actual knowledge to the contrary”, relieving them of any duty to take steps to see whether those assumptions are correct. Again, no such knowledge arises in the present case. (c) Furthermore, paragraph 5 of the First Schedule expressly relieves the trustees of any duty to seek or obtain information regarding management or conduct of the business and authorises them to assume that any information received is accurate unless they have actual knowledge to the contrary. (d) The Judge records without dissent that Matthews opined: “... that the duty of a trustee to obtain information about the affairs of a company in which the trust was interested as shareholder so as to enable it to make an informed decision whether to take any action for the protection of trust property in this case was modified or excluded by clauses 4 and 5 of the First Schedule. Clause 4 released the trustee from any obligation to interfere in the business of such a company, and clause 5 relieved it from any obligation to obtain information regarding such a company.”[25] (e) In the light of the express terms of the trust, endorsed as effective by both Jersey law experts, it is hard to see how there was any room for the existence of any “high level supervisory duty” which would require the trustees to query and disapprove of the transactions entered into by Wise Lords, thereby obviously interfering with Ji’s management of the company’s investment business which she had been duly authorised to conduct pursuant to the Investment Advisor Agreement dated 4 January 2005.[26] (f) Even if, upon examination of the relevant investments, one were to consider them speculative and over-concentrated in AUD, paragraph 2(b) of the First Schedule deems speculative investments to be authorised and relieves the trustees of any duty to diversify the trust fund’s investments. It also relieves them of any duty “to see that the value of the Trust Fund or any part thereof is preserved or enhanced in any way”, going on to exempt them from liability “for any failure in those respects whatsoever”. 46. In various parts of his judgment, the Judge points out that in practice, approval of certain transactions was routinely sought from and given by the trustees, generally after the event. This is a matter to which we return below.[27] But in the context of the Judge’s approach, it suffices to note that his Lordship does not provide any legal basis for contending that after-the-event approvals given routinely and formulaically as a matter of practice were transformed into a duty on the part of the trustees to exercise “high level supervision” over Wise Lords’ investments, contrary to the terms of the anti-Bartlett provisions. Nor does he provide a basis for finding that DHJ Management were subject to such a supervisory duty owed as a fiduciary duty to Wise Lords qua director. 47. It may be that the Judge thought that the high level supervisory duty represents a non-derogable duty under the 1984 Law, and in particular under Art 21 since he states[28] that the trustees were under a duty “to act honestly and in good faith, with due diligence, as would a prudent person, to the best of their ability and skill, only in the interests of the beneficiaries, and in accordance with the terms of the trust, and not in a grossly negligent manner” using language which rolls up and tracks in part Art 21(1) and (2). He uses similar language in finding a breach, holding that failure to discharge the high level supervisory duty “constituted a breach of their duty to act with due diligence, to act as would a prudent person, to act to the best of his ability and skill, and to act only in the interests of the beneficiaries”.[29] 48. We are unable to accept that Art 21 provides a viable basis for the “high level supervisory duty” purportedly imposed on the trustees. With respect, the Judge erroneously fails to note that Art 21(1) begins with the words: “A trustee shall in the execution of his or her duties and in the exercise of his or her powers and discretions [act with due diligence, etc]”. The point is that the requirement that trustees “act with due diligence, as would a prudent person, to the best of the trustee’s ability and skill; and observe the utmost good faith” operates to lay down the standards which trustees must adhere to in executing their duties or in exercising their powers. It does not create free standing duties to act prudently, etc. One must first identify the duty or power executed or exercised and then assess whether the trustees have executed or exercised the same in accordance with the specified standards or whether, on the other hand, their conduct was improper or so deficient as to constitute a breach of trust. Accordingly, if taking into account the effect of the anti-Bartlett provisions, any potentially relevant duty has been disapplied (so that, for instance, there is no duty to interfere or supervise management or to avoid speculative or non-diversified investments, or to preserve or enhance the value of the trust fund) the standards laid down by Art 21(1) do not come into play. 49. The Judge appears to have lost sight of this and based his finding of breach of trust on a postulated general duty to act prudently. Thus, he held that approving the three transactions was “not something which a trustee, complying with the duty to act prudently, could reasonably have done” and that “in approving the increased credit facility at the end of August 2008, DBS Trustee acted in a negligent manner and that their degree of negligence was a serious or flagrant degree of negligence”.[30] Unless DBS Trustee were under a duty as trustee to refuse approval in the execution of their duty or the exercise of their powers and discretions, the question of whether and to what extent granting approval (after the event) was negligent did not arise. 50. It is perhaps appropriate to add that the Judge may have been misled by the fact that, from the outset of the trial it was specifically conceded on behalf of the appellants that they had “a high-level, supervisory role in monitoring the overall performance of Wise Lords”[31], a rather imprecise and potentially misleading formulation of the functions of the trustees. The notion of a high-level supervisory role was adopted by the Judge in his judgment[32], but unfortunately he then went on to convert this “role” into a “high-level supervisory duty”[33]. He then held that this duty was breached by the appellants failing to challenge the purchase of the AUD and the decumulators[34], without apparently addressing the point that such a duty had been expressly excluded by the anti-Bartlett provisions, and would actually involve the trustees breaching the first part of paragraph 4(a)(ii) of the First Schedule to the Trust Deed[35]. 51. There is accordingly in our view no sustainable basis for the high level supervisory duties found by the Judge to have been imposed on DBS Trustee. 52. His Lordship also found DHJ Management to be in breach of its fiduciary duties as director of Wise Lords by approving the same three transactions, holding that such approvals constituted a failure to discharge their duties “to act in the best interests of the company and to exercise reasonable care, skill and diligence in the performance of their functions and their management of the company’s affairs”, and holding likewise “that DHJ Management acted in a negligent manner and that their degree of negligence was a serious or flagrant degree of negligence.”[36] It was on that basis that he held that Clauses 3(a) and 7 of the Services Agreement dated 13 September 2005 which provide for exemptions and indemnities except “in the case of gross negligence of the Nominees [including DHJ Management]” were of no avail.[37] We return to this issue in our discussion of breach in Section C below. B.2 The Court of Appeal’s approach to the duties 53. Cheung JA (who gave the main judgment) accepted that the trustees were not under any free-standing duty “not to act grossly negligently”. However, he upheld the Judge’s ruling that they were subject to a high level supervisory duty which had been breached when approvals were given to the three transactions.[38] 54. While Cheung JA also accepted that the anti-Bartlett provisions in paragraphs 4 and 5 of the First Schedule “are effective to exclude the obligations to which they refer”, he held that they do not exclude the “residual obligation” referred to by Matthews at paragraphs 87 and 88 of his expert report[39] and based the liability of the trustees on that “residual obligation” which his Lordship equated with the high level supervisory duty found by the Judge to exist.[40] 55. The relevant paragraphs in Matthews’ report state as follows: 87. The caselaw obligations of a sole trustee (such as the second Defendant was) of a Jersey law trust with more than one beneficiary (such as the Amsun trust is) include the duty to obtain sufficient information about the affairs of a company in which it is interested as shareholder so as to enable it to make an informed decision whether to take any action for the protection of the trust property. But in Jersey this duty is an aspect of the statutory duty contained in article 21(3) of the 1984 Law, considered above. And in practice such duty is commonly modified or excluded by express provision in the trust instrument, as indeed it was in this case, by clauses 4 and 5 of the First Schedule. Clause 4 released the trustee from any obligation to interfere in the business of such a company, and clause 5 relieved it from any obligation to obtain information regarding such a company. 88. In my opinion, given that the source of this caselaw duty is ultimately the statutory duty in article 21(3), and that duty is expressly made subject to the terms of the trust, these clauses are effective in Jersey law to exclude the obligations to which they refer. But there is a residual obligation cast on the trustee which these clauses do not exclude. The trustee as such trustee has in relation to the trust property all the powers of a natural person acting as the beneficial owner of such property. Although the trustee has no obligation to interfere in the business of the company, and no obligation to obtain information regarding the company, it still has a power to do so, because it is a member of the company. If circumstances were to arise where no reasonable trustee could lawfully refrain from exercising those powers, a failure to do so in such a case would amount to a breach of trust. It is true that clauses 4(d) and 5(d) purport to exclude liability for losses sustained in certain circumstances, but these clauses are of course subject to the statutory limits on trustee exemption clauses, set out in article 30(10) of the 1984 Law. This provision in effect renders an exemption clause ineffective to the extent that it purports to relieve a trustee for liability for gross negligence breach of trust, or for anything more serious.” 56. As Cheung JA noted,[41] there is a footnote to the phrase “If circumstances were to arise where no reasonable trustee could lawfully refrain from exercising those powers......” which states: “Such as where the trustee was informed by a credible source that the directors of the company were stealing its assets”. 57. Much depends on how one understands paragraphs 87 and 88 of the report and in particular, what Matthews meant by the “residual obligation” which he says is not excluded by the anti-Bartlett provisions in question. 58. Those provisions, contained in paragraph 4 of the First Schedule (“Trustees Not Bound to Interfere in Business of Company in which Settlement is Interested”) and in paragraph 5 thereof (“Trustees Not Bound to Obtain Information Regarding Company in which Settlement is Interested”), are set out in Section A.6 above. 59. In the aforesaid paragraphs of his report, Matthews endorses the effectiveness of those anti-Bartlett provisions, stating: “Clause 4 released the trustee from any obligation to interfere in the business of such a company, and clause 5 relieved it from any obligation to obtain information regarding such a company.” (paragraph 87) “In my opinion, given that the source of this caselaw duty is ultimately the statutory duty in article 21(3), and that duty is expressly made subject to the terms of the trust, these clauses are effective in Jersey law to exclude the obligations to which they refer.” (paragraph 88) 60. Accordingly, given that they are effective to relieve the trustees from any obligation to interfere with the business and any obligation to obtain information about the company, it would seem somewhat self-contradictory if Matthews were to be understood as saying simultaneously that the provisions are nonetheless ineffective should circumstances arise “where no reasonable trustee could lawfully refrain from exercising those powers [otherwise excluded]”. A legal basis has to be found to explain why any such “residual obligation” is not subject to the disapplication by the anti-Bartlett provisions. 61. One possible reading which avoids self-contradiction involves focussing on Matthews’ statement that “these clauses are effective in Jersey law to exclude the obligations to which they refer”, immediately followed by him adding: “But there is a residual obligation cast on the trustee which these clauses do not exclude”. Matthews could be understood simply to be recognising that paragraphs 4 and 5 do not entirely exclude the trustees’ obligation to interfere or to seek information. They are required to interfere in the conduct of the company’s business if they “have actual knowledge of any dishonesty relating to such business and affairs on the part of any of them”.[42] In other words, while paragraphs 4 and 5 are effective in excluding the obligations to which they refer, they preserve an obligation to interfere and seek information in cases where the trustees acquire actual knowledge of dishonesty. It is in that sense that such obligation is “residual”. This reading is supported by the content of the footnote which instances as a situation in which such a residual obligation arises (described as involving circumstances “where no reasonable trustee could lawfully refrain from exercising those powers”) a situation where the trustees are informed by a credible source that the directors are stealing the company’s assets. This is an instance involving dishonesty on the part of the directors relating to the company’s business and affairs. 62. Matthews goes on to explain that trustees do not lack power to interfere since they have all the powers of a natural person who beneficially owns the trust asset,[43] which would include power to interfere with the company’s management as a member of the company. On this reading, the power is translated into a residual obligation to act in cases involving actual knowledge of dishonesty not covered by the anti-Bartlett provisions. So understood, Matthews is not straying outside the four corners of the Trust Deed in asserting the existence of a “residual obligation”. And on this understanding, Matthews’s residual obligation does not provide any basis for the “high level supervisory duty” relied on by the respondents. 63. While the foregoing understanding of Matthews’ report is desirable, having the advantage of avoiding possible self-contradiction, it might well be said to be somewhat laboured. The comment may fairly be made that Matthews could simply have said that the anti-Bartlett provisions preserve an obligation to interfere where there is actual knowledge of dishonesty rather than referring to a power which has to be exercised in circumstances where “no reasonable trustee could lawfully refrain from exercising those powers”, which words do not appear to be confined to situations where the trustees gain actual knowledge of dishonesty. 64. But adopting the reading proposed by the Court of Appeal is even more difficult as it entails accepting that Matthews was intending to advocate the existence of a broad implied residual obligation arising outside of, and contradicting or overriding, the express anti-Bartlett provisions held by both experts to be valid under Jersey law. No authority is cited for such an obligation and it is difficult to see any principle which justifies its existence and, as mentioned, it would be inconsistent with Matthews’ earlier statement.[44] Anti-Bartlett provisions are generally incorporated in the Trust Deed in cases like the present because the parties wish to enable the settlor or the settlor’s nominee freely to exercise control and management of the underlying company, especially regarding matters such as its investment decisions, and to relieve the trustees of any management or supervisory duties in that regard (save where extreme situations such as those involving actual knowledge of dishonesty might arise). To postulate that the parties’ chosen scheme may be overridden by some implied, non-derogable external duty arising in circumstances “where no reasonable trustee could refrain from exercising otherwise excluded powers” would be to introduce an amorphous and ill-defined basis for undermining a legitimate arrangement consciously adopted by the parties, exposing the trustees to unanticipated risks of liability and sowing confusion as to the extent of their duties. 65. It is important to note that the residual obligation referred to by Matthews is not to be equated with the “irreducible core of obligations” which are “fundamental to the concept of a trust” recognised by Millett LJ in Armitage v Nurse.[45] As a matter of English law, those irreducible core obligations consist of “[t]he duty of the trustees to perform the trusts honestly and in good faith for the benefit of the beneficiaries” and do not include “the duties of skill and care, prudence and diligence”. They do not posit some broad duty to exercise available powers in circumstances “where no reasonable trustee could lawfully refrain from exercising those powers”. They do not operate to override express terms of a trust. They provide a touchstone for deciding whether the minimum requirements for constituting a trust have been met. 66. Despite the contrary arguments accepted by Cheung JA, we prefer a non-contradictory reading of the relevant paragraphs of Matthews’ report which accords with orthodox principles of trust law. In asserting the existence of a residual obligation, he was not seeking to go outside the terms of the Trust Deed but merely referring to the areas not covered by the anti-Bartlett provisions which are accepted by both experts to be effective. He was, in particular, not saying that there was some implied, peremptory, free-standing “high level supervisory duty” that came into being as an obligation that could not be excluded by the express terms of the Trust Deed. 67. Accordingly, with respect, we do not agree with Cheung JA’s analysis and hold that there is no basis for equating the “high level supervisory duty” found by the Judge with any purported residual obligation. 68. For completeness, we note that the latter part of Matthews’ paragraph 88 deals with a subsequent question, namely, whether given the apparently absolute exemptions from liability provided by Clauses 4(d) and 5(d) of the First Schedule, the trustees might in any event be relieved of the “residual obligation” under discussion. Matthews points out in this context that Art 30(10) of the 1984 Law “renders an exemption clause ineffective to the extent that it purports to relieve a trustee for liability for gross negligence breach of trust, or for anything more serious”, limiting the scope of the exemption clauses. No issue arises in the present appeal concerning the application of any exculpatory clause, whether in the First Schedule or the 1984 Law. 69. Madam Justice Yuen JA concurred with Cheung JA[46] and evidently equated the “high level supervisory role” accepted by the trustees with a failure to discharge the “residual obligation” referred to by Matthews.[47] Her Ladyship considered the aforesaid investments unsuitable for a discretionary trust with beneficiaries including minor children so that the trustees “could (and should) have called a halt”.[48] She did not, however, discuss the effect of the anti-Bartlett provisions in this context. 70. The Court of Appeal dealt with DHJ Management on the basis that they were in the same position as the trustees.[49] For the reasons already given in relation to DBS Trustees, there is no basis for treating DHJ Management as being subject to the same “high level supervisory duty” arising from the purported residual obligation. The only basis upon which DHJ Management could have been held liable was if it could be shown to have failed in its duties as a director of Wise Lords, and that that breach of duty amounted to gross negligence. B.3 The argument as to the duties advanced on this appeal 71. The argument advanced by Mr Barlow SC[50] to justify the Judge’s conclusion that the trustees had acted in breach of trust by approving the three transactions was quite distinct from the arguments accepted by the Judge and the Court of Appeal. 72. Counsel argued (i) that the trustees had all the powers of a natural person acting as the beneficial owner of the trust assets;[51] (ii) even though the anti-Bartlett provisions may effectively have disapplied any relevant duties, that if the trustees chose to exercise any of their powers, they came under a non-derogable “obligation” under Art 21(1), to act with due diligence, as would a prudent person, to the best of the trustee’s ability and skill; and to observe the utmost good faith; (iii) that the trustees had in fact chosen to exercise their power to supervise by approving individual investments and had assumed the role of “ultimate decision-makers” regarding Wise Lords’ investments; and (iv) that in so doing they were bound to act in accordance with Art 21(1) but had in fact acted with gross negligence in approving the three impugned transactions and so were liable for breach of trust. 73. At the centre of this argument is the proposition that by choosing to exercise the supervisory power of approving Wise Lords’ investments, the trustees subjected themselves to the non-derogable obligations laid down by Art 21(1) of the 1984 Law, providing the basis for their liability for breach of trust. As we have seen, Bharwaney J’s approach was different, his Lordship having treated the “high level supervisory duty” as arising pursuant to a free-standing obligation to act prudently and without gross negligence. The Court of Appeal also adopted a different approach premised on the existence of an external “residual obligation” identified by Matthews, requiring trustees to exercise available powers in circumstances where no reasonable trustee could refrain from acting. Mr Barlow’s argument was simply that if trustees in fact assumed a power to supervise Wise Lords’ investments, they had to exercise it in accordance with the standards laid down by Art 21(1) and were liable for breach of trust if they fell short. It was on this basis that he sought to uphold the existence of the “high level supervisory duty” found to exist by the courts below. Like the other versions of the duty, it was an unpleaded case and was never put to the Jersey law experts (who did not give oral evidence). For the reasons which follow, we do not accept the argument advanced. 74. The first objection to that argument is factual. As found by the Judge, DBS Trustee did not provide investment or portfolio management services.[52] Since the evidence makes it clear that in every case, the so-called “approvals” took place after the relevant transactions had been fully executed, it is untenable to allege, as Mr Barlow did, that the trustees had exercised a power to supervise and had in fact assumed the role of controllers and ultimate decision-makers regarding Wise Lords’ investments. That role and those decisions were taken by Ji as Wise Lords’ investment adviser and not by the trustees. Such “approvals” as the trustees gave did not amount to a meaningful form of supervision. The transactions were reported to them after the event and DBS Trustee’s so-called “approvals” represented merely a franking or acknowledgement of the information received. In no real sense can they be said to have assumed “control” of the conduct of Wise Lords’ investment business by exercising a power of supervision in respect thereof. Control over the possible undoing of any transaction after the event was also exercised by Ji and not the trustees. As we have seen, it was her decision to invest heavily in AUD and she consistently resisted all suggestions that Wise Lords’ AUD holdings be reduced given the risks identified. And it was Ji and not the trustees who eventually adopted the strategy of acquiring the decumulators as the means chosen for divesting Wise Lords of that currency, involving one of the very transactions complained of. 75. The second objection to Mr Barlow’s argument concerns the effect of the anti-Bartlett provisions. Counsel’s argument is premised on the trustees enjoying the very wide powers conferred by the beneficial owner clause in paragraph 1 of the First Schedule[53] which he then seeks to translate into a non-derogable duty. It is true that in general, such powers are potentially very wide. However, the powers conferred by paragraph 1 are expressly “[s]ubject always to any provisions or restrictions expressly contained in this Settlement” and it is clear that the parties to the Trust Deed intended that the trustees’ powers should be narrowly confined in one particular area, namely, in connection with the conduct of the investment business of the company in which the settlement is interested (i.e. Wise Lords). 76. Thus, paragraph 4(a) of the First Schedule (which we have examined above) not only relieves the trustees of a duty to interfere in the company’s business in the absence of actual knowledge of dishonesty, it stipulates in imperative terms that “... the Trustees shall leave the administration management and conduct of the business and affairs of such company to the directors officers and other persons authorised to take part in the administration management or conduct thereof ...”.[54] To emphasise the point, the trustees are then told[55] that they “shall assume at all times that the administration management and conduct of the business and affairs of such company are being carried on competently honestly diligently and in the best interests of the Trustees in their capacity as shareholders or howsoever they are interested therein until such time as they shall have actual knowledge to the contrary ...”. They are relieved of any duty “at any time to take any steps at all to ascertain whether or not the assumptions contained in this sub-clause are correct”. 77. As pointed out by Ms Warnock-Smith QC,[56] the trustees are thus consistently being told to keep their noses out of the company’s business and to leave those having conduct of the same free to manage it without interference. They are accordingly relieved of “any duty to obtain or to seek to obtain in any way whatsoever any information regarding the administration management or conduct of the business or affairs of any company in which this Settlement is or may be interested”[57] and are enjoined to “assume that such information as is supplied to them by any person relating to such company is accurate and truthful unless the Trustees have actual knowledge to the contrary”[58]. They are told that they “shall not be under any duty at any time to take any steps at all to ascertain whether or not the information is accurate and truthful”[59]. 78. Wise Lords’ business is of course that of investment. And, as we have seen, paragraph 2(b) and (g) of the First Schedule authorise the making of speculative and non-diversified investments. Of direct present relevance is paragraph (j) which states that: “In the exercise of the powers herein contained the Trustees shall not be under any duty to see that the value of the Trust Fund or any part thereof is preserved or enhanced in any way nor shall they be liable for any failure in those respects whatsoever.” 79. The effect of these provisions of the Trust Deed and its First Schedule is thus to restrict the power of the trustees to interfere in the conduct or management of Wise Lords’ investment business. The second objection to Mr Barlow’s argument is therefore that it fails to recognise that the powers which he contends have been exercised to interfere in the conduct of Wise Lords’ business were, on their true construction, unavailable to the trustees. Their conduct in receiving and “approving” the investments after they had been made did not constitute the exercise of a power of supervision whether factually or as a matter of law. 80. The consequence of the two abovementioned objections is that the respondents’ reliance on the non-derogable provisions of Art 21(1) is wholly undermined. Art 21(1) states as follows: “21 Duties of trustee (1) A trustee shall in the execution of his or her duties and in the exercise of his or her powers and discretions – (a) act – (i) with due diligence, (ii) as would a prudent person, (iii) to the best of the trustee’s ability and skill; and (b) observe the utmost good faith.” 81. As we have pointed out,[60] Art 21(1) lays down standards which trustees must adhere to in executing their duties or exercising their powers and does not create free-standing duties to act prudently, etc. Its operation therefore requires there first to be an applicable duty or power which is in fact executed or exercised by the trustees. Only then does one proceed to assess whether the trustees duly met those standards or whether, instead, they acted in bad faith or in a grossly negligently manner so as to constitute a breach of trust. The respondents are unable to establish a threshold case based on Art 21(1). 82. Mr Barlow’s argument relies on a power as opposed to a duty as the basis for a claim of breach of trust. The argument is aimed at circumventing the anti-Bartlett provisions in paragraphs 4 and 5 of the First Schedule which relieve the trustees, inter alia, of any duty to interfere in or supervise Wise Lords’ business. It aims to advance a claim for breach of trust without reference to any breach of “duty”. 83. However, as Art 1 of the 1984 Law stipulates, to constitute a “breach of trust” there must be “a breach of [a] duty imposed on a trustee by this Law or by the terms of the trust”. Thus, for the exercise of a power to constitute a breach of trust, such exercise must occur in such an improper or deficient manner as to amount to violation of a duty on the part of the trustee. What starts off as a deficient exercise of a power must attain the status of a breach of duty if it is to found the equitable claim. Thus, on a proper analysis, the anti-Bartlett provisions cannot be avoided simply by asserting that the complaint relates to a grossly negligent exercise of a power. It must necessarily be a claim for breach of a duty which in substance is one which is disapplied by the anti-Bartlett provisions. Art 21(1) is thus deprived of any impact since there is no applicable duty on which the Article’s non-derogable provisions can bite. 84. As for DHJ Management, Mr Barlow’s argument that they had a “high level supervisory duty” is untenable for the reason given in [74] above, namely that they did not exercise any powers of supervision over Wise Lords’ investments or the portfolio generally, and no other basis has been advanced to justify such a high level duty on the part of DHJ Management. Accordingly, it is unnecessary to address Question 5 any further. B.4 Conclusion as to the duties 85. For the foregoing reasons we conclude that there is no basis for the existence of the “high level supervisory duty” accepted in the courts below and advocated on this appeal. That is decisive of this appeal. However, since issues have been debated as to whether the alleged duties were breached and if so, what is the proper approach to equitable compensation, we shall touch briefly on those matters. C. Were there breaches of applicable duties? 86. It has throughout been correctly accepted by the respondents that to succeed, their claim based on breach by the trustees of the alleged high level supervisory duty by approving the three transactions had to involve proof of gross negligence on the trustees’ part. This is because exculpatory clauses such as Clauses 19 and 20 of the Trust Deed would protect the trustees from liability for any acts or omission falling short of fraud, wilful misconduct or gross negligence. Art 30(10) of the 1984 Law is to like effect. And in the case of DHJ Management, similar clauses in the Services Agreement dated 13 September 2005 provided for exemptions from liability and indemnities except “in the case of gross negligence” and, hence, the only basis on which it could have been held liable would have been if it had been grossly negligent in carrying out its duties as a director of Wise Lords. 87. As we pointed out in Section B.1 above, while the three transactions (involving an increase in Wise Lords’ credit facility to USD100m; the purchase of AUD83m and the purchase of the decumulators) can fairly be described as speculative and as carrying risks associated with heavy concentration in one currency and illiquidity, those are risks which the Trust Deed expressly authorises to be taken. And for the reasons there considered, we find it hard to see how approval of those transactions can be said to involve gross negligence on the trustees’ or nominee director’s part. 88. Bharwaney J did not explain in any detail why he considered gross negligence established. He defined “gross negligence” to mean “a serious or flagrant degree of negligence” and was largely content simply to assert that the negligence of the trustees was of a serious or flagrant degree.[61] 89. Cheung JA gave no reasons for finding gross negligence but appears to have held that breach of the high level supervisory duty ipso facto entailed gross negligence, stating: “Once DBS Trustee had been found in breach, it follows, like night following day, that the degree of breach was a serious and flagrant one. The nature of the breach speaks for itself. It had failed to perform the very task that it had assumed namely, high level supervision in order to assure that the trust asset is subject to appropriate control. The breach was not accidental but carried with it a serious and flagrant degree of negligence because the tasks to be performed by DBS Trustee clearly involved more than rubberstamping the transactions. It is a matter of common sense that, to properly discharge the task, DBS Trustee must have at least kept itself informed of the prevailing financial conditions. Although the intensity of the global financial crisis was unprecedented, it did not occur overnight and as the Judge had pointed out, the storm had been brewing since March 2007. In the circumstances, DBS Trustee plainly must exercise more caution before giving approval for the impugned transactions.”[62] 90. With respect, if contrary to our holding, the trustees were under a duty to supervise Wise Lords’ investments instigated by Ji as the company’s properly appointed investment adviser, no basis has been shown for concluding that approval of the three transactions constituted negligence to a “serious and flagrant degree”.[63] The claims against DBS Trustee would accordingly have failed since they would have been protected by the exculpatory clause covering acts or omissions short of fraud, misconduct or gross negligence. For the same reason, even assuming that they could otherwise have been liable for failing to exercise their duty as a director competently, DHJ Management cannot be liable as alleged by the Respondents because any failure on their part did not amount to gross negligence, and any liability is therefore excluded by a similar exculpatory clause in the Services Agreement. D. Concurrent findings of fact 91. A recurring theme running through the Respondents’ Case[64] is the assertion that the courts below made concurrent findings of fact and that, in keeping with this Court’s practice, those concurrent findings should not be overturned in the absence of exceptional and rare circumstances where an appellant can show some miscarriage of justice or violation of some principle of law or procedure to warrant such a review. 92. Relying on this practice, the respondents contend that there is no basis for overturning those concurrent findings of fact. In particular, it is contended that, “the Trust Deed was subject to independent requirements of Jersey law (both statutory and equitable) which the Appellants’ expert had explained and whose formulation ... both courts below concurrently found as fact.”[65] 93. The Court’s practice in respect of concurrent findings of fact is now both long-standing and firmly established. It was laid down in the judgment of Bokhary PJ in Sky Heart Ltd v Lee Hysan Co Ltd,[66] a case heard soon after the establishment of the Court. Save that the factor of unfamiliarity with local conditions is not relevant in this Court, the practice follows that of the Privy Council as explained by Lord Thankerton in Srimati Bibhabati Devi v Kumar Ramendra Narayan Roy.[67] It applies as much to primary findings of fact and findings of fact reached by drawing inferences from primary facts proved or admitted: Re Moulin Global Eyecare Holdings Ltd.[68] In Chinachem Charitable Foundation Ltd v Chan Chun Chuen,[69] it was unsuccessfully sought to challenge the practice and the invitation to abandon it was rejected by the Appeal Committee as not being reasonably arguable. 94. Nevertheless, the Court’s practice expressly admits of the possibility of exceptions to the general rule. The Court will not be inhibited by concurrent findings of fact where it can be shown that the findings are self-contradictory or there are facts pointing to the opposite conclusion: see, e.g., Kwan Siu Man v Yaacov Ozer[70] and Aktieselskabet Dansk Skibsfinansiering v Brothers.[71] In such a case, there will have been a “miscarriage of justice or violation of some principle of law or procedure” as described by Lord Thankerton in articulating the practice. 95. Foreign law is treated as a fact by the courts of Hong Kong and therefore ascertained by a process of pleading and proof involving expert witnesses, usually examined and cross-examined at trial.[72] However, although a question of fact, “foreign law ... is a question of fact of a peculiar kind” so that the considerations as to an appellate court’s power to interfere with ordinary findings of fact do not apply to the same extent.[73] As the UK Supreme Court held in Eli Lilly v Actavis UK Ltd: “[T]he notion that the resolution of a dispute as to foreign law involves a factual finding rather than a legal conclusion is somewhat artificial, and in any event, the Judge did not hear any oral evidence from the expert foreign law witnesses. We are therefore in as good a position as he was to analyse the effect of the evidence as to foreign law.” [74] 96. In particular, the court will examine the legal reasoning of a foreign law expert to determine the validity and reliability of the expert’s evidence as to the content of the foreign law in question, even where that evidence is uncontradicted.[75] A good summary of the court’s approach, which is well established, to assessing such evidence is provided in Dicey & Morris on the Conflict of Laws (15th Ed., 2012) at [9-016].[76] That paragraph (omitting footnotes) states: “If the evidence of the expert witness as to the effect of the sources quoted by him is uncontradicted, ‘it has been repeatedly said that the court should be reluctant to reject it,’ and it has been held that where each party's expert witness agrees on the meaning and effect of the foreign law, the court is not entitled to reject such agreed evidence, at least on the basis of its own research into foreign law. But while the court will normally accept such evidence it will not do so if it is ‘obviously false,’ ‘obscure,’ ‘extravagant,’ lacking in obvious ‘objectivity and impartiality’, or ‘patently absurd,’ or if ‘he never applied his mind to the real point of law’, or if ‘the matters stated by [the expert] did not support his conclusion according to any stated or implied process of reasoning’; or if the relevant foreign court would not employ the reasoning of the expert even if it agreed with the conclusion. In such cases the court may reject the evidence and examine the foreign sources to form its own conclusion as to their effect. Or, in other words, a court is not inhibited from ‘using its own intelligence as on any other question of evidence’. Similarly, the court may reject an expert’s opinion as to the meaning of a foreign statute if it is inconsistent with the text or the English translation and is not justified by reference to any special rule of construction of the foreign law. It should, however, be noted in this connection that quite simple words may well be terms of art in a foreign statute.” 97. As the above summary of the applicable principles shows: (a) Self-evidently, the Court’s practice of not entertaining challenges to concurrent findings of fact applies to findings which are findings of fact. Where it is sought to challenge conclusions of law based on the facts found concurrently by the courts below are involved, the Court will not be similarly constrained. (b) The Court’s practice is, in any event, not rigid and there are circumstances in which the Court will set aside findings of fact notwithstanding that they have been made by both a trial court and an intermediate appellate court. (c) Expert evidence of foreign law, although a matter of fact, will be treated differently to other, ordinary, findings of fact. (d) Even if uncontradicted, the court will examine the content of any expert opinion evidence as to foreign law and will reject it as unsatisfactory if the circumstances so warrant. 98. With these observations in mind, the respondents’ contentions as to the existence and effect of relevant concurrent findings of fact in the judgments below lose much of their force. 99. As the discussion in Sections B.1 and B.2 above demonstrates, the focus of the debate in this appeal is on the legal basis of the Judge’s and the Court of Appeal’s respective conclusions that DBS Trustee and DHJ Management owed high level supervisory duties to the respondents and, further, that in breaching those duties they had behaved in a “flagrantly negligent” manner. As we have observed above,[77] the conclusion that the high level supervisory duties existed is inconsistent, as a matter of law, with the anti-Bartlett provisions and is one without a sustainable basis in law. Likewise, the absence of a legal basis for transforming the giving of after-the-event approvals into a supervisory duty is similarly an error of law justifying appellate interference.[78] It follows that the findings regarding breach and flagrant negligence (the latter being, as we have seen in Section C above, unsupported by reasons) rest upon errors of law whether those findings are concurrent or otherwise. 100. The Court of Appeal’s approach to the duties owed was based on a reading of Matthews’ report at paragraphs 87 and 88 which we have concluded to be erroneous.[79] In circumstances in which neither Jersey law expert gave oral evidence before the Judge, and in which the more difficult and self-contradictory reading of Matthews’ report was never put to him in cross-examination,[80] this Court is in as good a position as the Judge to reach a conclusion as to the true import of Matthews’ evidence. 101. For these reasons, we conclude that the invocation of the Court’s practice on concurrent findings of fact does not preclude us from reaching contrary conclusions to the courts below on the issues of whether DBS Trustee and DHJ Management owed the duties found by the courts below and whether they had breached those duties in a “flagrantly negligent” manner. E. Equitable compensation 102. The conclusions reached above make it strictly unnecessary, in order to make an order dispositive of this appeal, to address the issue of relief and the “or otherwise” basis on which leave to appeal was granted. However, as with the issue of breach, we shall similarly touch on these matters briefly in view of their importance as a matter of law. E.1 The Judge’s conclusions as to the consequences of breach of duty 103. Having held DBS Trustee to have been in breach of trust[81] and DHJ Management to have been in breach of fiduciary duty,[82] the Judge went on to hold that Arboit and Sutton were entitled to “equitable restitution to the Trust for breach of trust in order to reconstitute the assets of the Trust so as to place the Trust in the position it would have occupied but for the said breaches”.[83] 104. The Judge then cited the principles discussed in this Court’s judgment in Libertarian Investments Limited v Thomas Alexej Hall (“Libertarian”)[84] in respect of the claim against DBS Trustee. At [424] of his judgment, the Judge concluded that the claim against DBS Trustee fell into the first category of cases identified in Libertarian, holding: “It is not in dispute that DBS Trustee was under a fiduciary duty to the Trust. I am satisfied that the breaches of duty by DBS Trustee, that I have found to have been established, have directly caused loss of the assets of the Trust in that they have directly led to diminution of the value of the assets held in Wise Lords’ portfolio. Our case also falls into the first category identified by Ribeiro PJ above, being a case where the breaches of duty by DBS Trustee have led directly to losses being suffered by Wise Lords’ portfolio, i.e. a case where ‘there are breaches leading directly to damage to or loss of the trust property’.” 105. This conclusion therefore formed the basis of his award of equitable compensation which he explained in his judgment at [425] as follows: “This is a case where it is appropriate to award equitable compensation against DBS Trustee in favour of Arboit and Sutton, that is to say, not compensation for loss but compensation that is restitutionary or restorative. I do not find any assistance from the plaintiffs’ submissions on the basis upon which I should award equitable compensation. It is not necessary to order an account as the assets in Wise Lords’ portfolio are well documented. I intend to adopt a robust approach to assess the equitable compensation by, firstly, attempting to determine the value of the assets in Wise Lords’ portfolio on the date of the issue of the Writ in these proceedings on 28 February 2011; and, secondly, by attempting to assess what that value might have been on 28 February 2011 if Wise Lords had not acquired US$83m worth of AUD from 24 July to 5 August 2008 and had not purchased the 3 Decumulators but had carried out the other transactions listed in Arboit’s 2nd Schedule that were unrelated to the purchases of US$83m worth of AUD from 24 July to 5 August 2008 and unrelated to the purchases of the 3 Decumulators; and, thirdly, by awarding the difference between the 2 values to Arboit and Sutton, being the trustees of the Trust, as equitable compensation. I need further assistance from the financial experts, Das and Malik, to enable me to do so and I give further directions below for a further joint report to be obtained from them.” 106. As regards DHJ Management, the Judge similarly held that it was appropriate to award equitable compensation against them. He concluded that DHJ Management’s duty to act bona fide in the best interests of Wise Lords was a fiducial obligation, the breach of which entitled Wise Lords to equitable compensation against DHJ Management.[85] He cited the principles in Libertarian and held (at [454]): “I am satisfied that the breaches of duty by DHJ Management, that I have found to have been established, have directly caused loss of the assets of Wise Lords in that they have directly led to diminution of the value of the assets held in Wise Lords’ portfolio. The present claim also falls into the first category identified by Ribeiro PJ above, being a case where the breaches of duty by DHJ Management have led directly to losses being suffered by Wise Lords’ portfolio, i.e. a case where ‘there are breaches leading directly to damage to or loss of the trust property’.”[86] 107. This conclusion similarly formed the basis of his award of equitable compensation against DHJ Management in favour of Wise Lords “that is to say, not compensation for loss but compensation that is restitutionary or restorative.”[87] Thus, he granted relief in favour of Wise Lords “for equitable restitution ... for breach of fiducial obligations in order to reconstitute the assets of Wise Lords so as to place Wise Lords in the position it would have occupied but for the said breaches.”[88] 108. The Judge recognised that his orders against both DBS Trustee and DHJ Management might result in a duplication of the relief awarded and so, in order to avoid double recovery, he ordered that if Arboit and Sutton obtained complete satisfaction from DBS Trustee, then Wise Lords could not seek equitable compensation from DHJ Management and vice versa. For the same reason, he ordered that partial satisfaction against one defendant would entitle Wise Lords to seek equitable compensation from the other defendant for the balance and vice versa.[89] 109. The Judge gave directions for the further conduct of the proceedings based on his judgment.[90] Those directions included the preparation of a joint expert report by the parties’ financial experts the purpose of which was to set out “the value of the assets in Wise Lords’ portfolio as at the date of the issue of the Writ in these proceedings” and to set out “their opinion on what that value might have been on 28 February 2011 if Wise Lords had not acquired US$83m worth of AUD from 24 July to 5 August 2008 and had not purchased the 3 decumulators but had carried out the other transactions listed in Arboit’s 2nd Schedule that were unrelated to the purchases of US$83m worth of AUD from 24 July to 5 August 2008 and unrelated to the purchases of the 3 decumulators.”[91] E.2 The Court of Appeal’s approach 110. The Court of Appeal upheld the Judge’s order for the assessment of equitable compensation. 111. Cheung JA rejected the appellants’ arguments that the Judge was wrong to have omitted to consider the question of causation in determining the availability of equitable compensation for the breaches held to have been established. He agreed with the Judge that the nature of the breach of trust in the present case was in the first category of breach discussed in Libertarian: “The present breach is the first category of breach discussed by the Court of Final Appeal in Libertarian Investments Ltd v Hall (2013) 16 HKCFAR 681 which approved the view of Tipping J in BNZ v NZ Guardian Trust Co Ltd [1999] 1 NZLR 664, namely, breach leading to damage or loss of the trust property.”[92] 112. Cheung JA applied the principles of causation applicable to that category of case as described in the judgment of Ribeiro PJ in Libertarian at [78] to [80], namely that “in cases within ... [the] first category, involving loss caused by the fiduciary to trust property, strict rules on causation apply.” Applying Libertarian at [93], Cheung JA held that the onus lay on the defaulting trustee or fiduciary and that the appellants “had not been able to disprove the apparent causal connection between the breach of duty and the loss apparently flowing therefrom.”[93] 113. Yuen JA addressed the appellant’s argument on causation more specifically. It had been argued that the Judge’s approach was flawed in that it “did not require an assumption to be made that the credit facilities were not increased” since (i) if there were no increase, it was likely that Ji would have terminated the banking relationship with DBS:PB and the trust relationship with DBS Trustee, after which she would have conducted the same investments elsewhere, and (ii) if there were an increase, it was likely that Ji would have utilised it for other investments which, in the market situation at the time, would have led to losses in any event.[94] 114. Her Ladyship rejected these arguments principally on the basis that, applying Libertarian at [93], the burden is on a defaulting trustee to disprove the apparent causal connection between the breach of duty and the loss apparently flowing therefrom. She noted that Ji had not apparently been cross-examined on the scenarios advanced by the appellants and “in light of the burden on the trustees, ... it cannot be assumed that if DBS Trustee had disapproved the AUD purchase and the 3 Decumulators ... that Ji would not have paused to consider their rationale for disapproval.”[95] E.3 Equitable compensation for breach of fiduciary duty 115. This Court examined the principles applicable to claims for compensation for breach of fiduciary duty in Libertarian. There, at [75], it was emphasised that the duties owed by fiduciaries are variable in nature and it may be important to ascertain the nature of the breach in question and the impact of that breach on any trust property. Referring to the judgment of Tipping J in Bank of New Zealand v New Zealand Guardian Trust Co Ltd,[96] three categories of breach were distinguished. These were: (a) the first category, being “breaches leading directly to damage to or loss of the trust property”; (b) the second category, being “breaches involving an element of infidelity or disloyalty which engage the conscience of the fiduciary”; and (c) the third category, being “breaches involving a lack of appropriate skill or care”. 116. The importance of distinguishing these categories of cases is that the rules of causation apply with varying degrees of strictness depending on which type of duty and breach is involved: Libertarian at [76]. 117. For first category cases, relating to misapplication or loss of trust property, the rules as to causation are strictest and are akin to those applicable to traditional trusts requiring the trustee to restore to the trust fund whatever loss he has caused as a result of his breach of trust. A strict “but for” causation test is applied and, if satisfied, the trustee is liable to make good the loss and the common law rules of remoteness and foreseeability do not apply. The claim is in the nature of a substitutive claim for restoration of the trust property either in specie or by value.[97] 118. Similarly, for second category cases, involving an element of infidelity or disloyalty which engage the conscience of the fiduciary, the rules on foreseeability and remoteness do not apply. Once the plaintiff has shown a loss arising out of a transaction to which the breach was material, the plaintiff is entitled to recover unless the defendant fiduciary, on whom the onus lies, shows that the loss or damage would have occurred in any event without any breach on his part.[98] 119. Thus, in both first and second category cases, it is irrelevant that the plaintiff’s loss was not foreseeable. Compensation will be assessed with the benefit of hindsight taking into account any post-breach events which affect the extent of the loss, e.g. market changes or currency fluctuations. Libertarian provides an example of a case in the first category. There, the defendant fiduciary, instead of using funds transferred to him for the specific purpose of acquiring shares in a listed company for his principal, used the funds for his own purposes. By the date of trial, the shares he should have bought had risen in value sharply because of an exceptionally high public offer price in a successful takeover bid. It was not relevant that the takeover bid and offer price were not foreseeable when the breach occurred. The defendant was ordered to pay equitable compensation on a wilful default basis reflecting the increased value in the shares. 120. Cases in Tipping J’s third category involving a lack of appropriate skill or care stand on a different footing to those in the first two categories. Liability in the third category of cases is imposed for the negligent execution of a fiduciary’s duties falling within the general duty to act with care imposed by law on those who take it upon themselves to act for or advise others. In such cases, the relationship of trustee or fiduciary is incidental and provides the context in which the breach of duty occurs. Equitable compensation for breach of the duty of skill and care is reparative and resembles common law damages awarded to a plaintiff for his loss and the common law rules of causation, remoteness of damage and measure of damages apply by analogy.[99] E.4 The categorisation of the present case 121. As we have seen, the courts below regarded this case as being in the first category of cases so that the award of equitable compensation was in the nature of a substitutive claim for restoration of the trust property. In consequence, in answer to the appellants’ causation arguments, the Court of Appeal held that the appellants had not discharged their onus of proving that the losses claimed were not caused by the breaches found. If the claim was properly to be regarded as in the first category of cases, their decisions would (subject to the observations in Section E.5 below) be correct. But was that categorisation correct? 122. With respect to the courts below, we do not think it was. On the contrary, the present case was, in our view, clearly not in the first category which applies to cases of misapplication or loss of the assets held on trust by the trustee or fiduciary, reflecting his obligation as custodian to account for his stewardship of the trust property and to restore to the trust estate the misapplied asset. However, at [424] and [454] of his judgment, cited at [104] and [106] above, the Judge (with whom the Court of Appeal agreed) treated this as a first category case on the basis that the breaches he found had “directly caused loss of the assets of the Trust in that they have directly led to diminution of the value of the assets held in Wise Lords’ portfolio” (emphasis added). 123. That was a mis-categorisation for two reasons. First, the only asset of the Trust, apart from the initial property of USD10 settled on it, was the sole share in Wise Lords. That share remained at all material times within the custody of the trustees. Even though the value of the share in Wise Lords had no doubt diminished, there was, properly analysed, no misapplication or loss of the trust asset that required substitution or restoration of trust property. 124. Secondly, even if the trust assets were regarded as consisting of the assets held by Wise Lords and even if there were a breach of duty in the three respects found by the Judge, that breach would have come within the third category of case, involving a lack of appropriate skill or care causing diminution in the value of the trust assets. This is clearly reflected in the language used by the courts below in describing the breaches in question. Negligence in the oversight of investments, even gross or flagrant, is by its nature a failure to exercise due skill or care. 125. Had we upheld the breaches of duty regarding the purchase of AUD, the increased credit facility and the approval of the purchase of the three decumulators we would have placed them in the third category. Consequently, equitable compensation would have been reparative as opposed to being substitutive or restorative (although these terms should be understood broadly since they can overlap to some extent) and, more importantly, the common law rules of causation, foreseeability and remoteness would therefore have applied to the assessment of any such compensation. It would accordingly have been inappropriate to place the onus of disproving the causal connection between the alleged breaches and the loss apparently flowing from them on the appellants and to order an assessment of compensation, as the courts below did, without giving the appellants the opportunity to raise the issue of causation. 126. In the course of argument, we were referred by Mr Barlow to the judgment of Brightman LJ (as he then was) in Bartlett v Barclays Trust Co. (No.2)[100] addressing the appropriate form of order for assessment of the compensation payable consequent upon his judgment on liability for breach of trust.[101] The claim had involved a trust which consisted of the shares in a company incorporated to take over and manage certain family properties. The directors of the company later pursued a policy of property investment through the incorporation of an investment company used as a vehicle to purchase and develop properties. The investment company pursued a speculative property venture during a property boom and, when the property boom ended, found itself insolvent. The plaintiffs claimed against the trustee to make good to the trust fund all loss accruing by reason of it having permitted the property company to engage in property development. The trustee was held liable for not exercising proper skill and care and for not intervening to prevent the investment company from having entered into the speculative property venture. 127. In relation to the form of order for compensation, Brightman LJ acceded to the plaintiffs’ invitation to direct an assessment of the compensation payable by the defaulting trustees by, first, determining the extent to which the assets of the company were wrongly depleted in consequence of the defendant’s breaches of trust and then, secondly, determining the additional proceeds of sale which would have accrued on the disposal of the shareholdings if the assets of the company had not been depleted.[102] 128. It was suggested by Mr Barlow that, in AIB Group (UK) plc v Mark Redler & Co Solicitors,[103] Lord Toulson JSC treated Bartlett as a first category type of case. However, it is not at all clear that he was intending to do so. He described it as “a case of breach of a trustee’s management stewardship duty” and went on to say: “In this type of case the order for payment by the trustee of the amount of loss is referred to by some as ‘reparative compensation’, to differentiate it from ‘substitutive compensation’, although in a practical sense both are reparative compensation.” That seems to us to be more consistent with Lord Toulson treating the case as being within the third category. 129. In any event, we do not consider that either Bartlett or Lord Toulson’s reference to that case in AIB provide a sound basis for categorising a trustee’s negligent breach of management of a company owned by a trust as a first category type of case, rather than a third category type of case. Neither Brightman LJ nor Lord Toulson addressed the character of the Bartlett claim in terms of the category of cases distinguished by Tipping J. As we have said above, the character of the claim is one involving an allegation of the failure to exercise appropriate skill or care. E.5 Other points not necessary to address 130. We were referred, by Ms Warnock-Smith, to a decision of the Jersey Court of Appeal in BNP Paribas Jersey Trust Corporation Limited and others v Cristiana Crociani and others,[104] a case concerning a claim against the trustees of a family trust, amongst others, in respect of the unauthorised payment out of the trust fund of substantially the whole of the artwork, investments and cash that had been accumulated by the trust over time. It was therefore a case which would fall into the first of Tipping J’s category of cases. However, it is to be noted that one of two sisters who were beneficiaries of the trust had dishonestly assisted in the breaches of trust. One of the issues on appeal was whether the liability of the trustees extended to reconstituting the whole of the trust or only that part held in trust for the other sister and her family. 131. The Jersey Court of Appeal, citing the UK Supreme Court’s decision in AIB, observed that “whilst the law in respect of equitable compensation for breach of trust is in a state of development – especially as regards issues of causation – the basic rule is that, where property has been misapplied and cannot be restored in its original form, the trustee must restore the trust fund to the position in which it would have been but for the breach.”[105] However, it noted that “[m]ore complex issues arise where the recipient of the misapplication is a member of one of the classes of beneficiaries”[106] and cited Lord Reed JSC’s statement in AIB that “the nature of the appropriate remedy for a breach of trust may well vary to reflect the terms of the trust in question and the breach in respect of which the complaint is made.”[107] 132. Significantly, the Jersey Court of Appeal held: “Separately, neither of the decisions in Target Holdings or AIB, or their underlying circumstances, assist where a beneficiary has been involved to some extent in the breach of trust which has taken place, whether as instigator, associate or recipient. For example, as Lord Browne-Wilkinson expressed matters in Target Holdings (at 436 C-D): “But the basic equitable principle applicable to breach of trust is that the beneficiary is entitled to be compensated for any loss he would not have suffered but for the breach.” Without hesitation we agree; but how that principle is applied where one or more beneficiaries are found to be instigator, associate or recipient in the breach is a question which requires to be resolved, on equitable principles, having regard to the individual facts of the particular case.”[108] 133. It is unnecessary for us to resolve the interesting, and difficult, question of whether, even if the case before us were in the first category of cases, it would nevertheless have been necessary to consider issues of causation, remoteness and foreseeability because Ji actively participated in making the decisions which led to the alleged losses suffered by the Trust. 134. Another point which it is not necessary to address in this judgment is the question of whether the Judge’s order for the assessment of equitable compensation would have permitted the recovery by the respondents of reflective loss since the basis on which it was argued the Trust had suffered loss was by reason of the diminution in the value of the sole share of Wise Lords. This point was sought to be raised by the appellants in the Court of Appeal but, as Yuen JA pointed out, there were a number of reasons why the Court of Appeal refused to permit this point to be argued for the first time on appeal.[109] Difficult questions arise in relation to the question of reflective loss, as manifested in the Judge’s attempt to manufacture an order to avoid double recovery,[110] due to the different identities of the plaintiffs and defendants in whose favour and against whom respectively equitable compensation was awarded. Since those questions were not before us, we do not propose to address them. However, we should not be understood as approving the Judge’s approach, which the Court of Appeal’s judgment left undisturbed. F. Conclusion 135. For the reasons set out above, if this case had not been settled, we would have allowed the appellants’ appeal and set aside the Judge’s order for the assessment of equitable compensation. Mr Justice Cheung PJ: 136. I agree with the joint judgment of Mr Justice Ribeiro PJ, Mr Justice Fok PJ and Lord Neuberger of Abbotsbury NPJ. Mr Justice Tang NPJ: 137. I agree with the joint judgment of Mr Justice Ribeiro PJ, Mr Justice Fok PJ and Lord Neuberger of Abbotsbury NPJ. Ms Shân Warnock-Smith QC, Mr Ashley Burns SC and Ms Bonnie Y.K. Cheng, instructed by Mayer Brown, for the 2nd & 4th Defendants (1st & 2nd Appellants) Mr Barrie Barlow SC and Mr Chan Pat Lun, instructed by MinterEllison LLP, for the 3rd & 4th Plaintiffs (1st & 2nd Respondents) [1] [2011] EWCA Civ 826; [2012] Bus LR 542. It has been applied in Greenwich Inc Ltd v Dowling [2014] EWHC 2451 (Ch) at [131] and [134] and in Re Dalnyaya Step LLC (In Liquidation) (No.2) [2017] EWHC 3153 (Ch); [2019] BCC 23; and has overtaken the earlier decision in Prudential Assurance Co Ltd v McBains Cooper (A Firm) & Ors [2000] 1 WLR 2000. [2] [2014] VSC 516 per Croft J at [23]. [3] [2004] 1 Qd R 212; [2003] QCA 252. [4] [2014] 1 NZLR 766 at [39]-[44]. [5] HCCL 2/2011, Judgment dated 13 April 2017 (“CFI Jmt”). [6] CACV 139/2017, [2018] HKCA 435 (Cheung, Yuen and Kwan JJA). [7] FAMV 117/2018, [2019] HKCFA 18 (Ribeiro Ag CJ, Fok PJ and Chan NPJ). [8] CACV 138/2017, [2018] HKCA 435 (Cheung, Yuen and Kwan JJA) (“CA Jmt”). [9] CACV 138/2017, [2018] HKCA 917 (Cheung, Yuen and Kwan JJA). [10] FAMV 126/2018, [2019] HKCFA 18 (Ribeiro Ag CJ, Fok PJ and Chan NPJ). [11] The full text of which is set out below at [80]. [12] [1980] 1 Ch 515. [13] CFI Jmt [407], referring back to CFI Jmt [303]-[305] and [309]-[319]. The date was corrected by Cheung JA; CA Jmt [4.3]. [14] CFI Jmt [408], referring back to CFI Jmt [303]-[315] and [317]-[319]. [15] CFI Jmt [409], referring back to CFI Jmt [324] and [362]-[381]. [16] CFI Jmt [297]. [17] CFI Jmt [305]. [18] CFI Jmt [324]-[326]. [19] CFI Jmt [407]. [20] CFI Jmt [303]-[305] and [314]-[315]. [21] CFI Jmt [304]. [22] See Section A.2 above. [23] Art 21(2) of the 1984 Law. [24] Set out in Section A.6 above. [25] CFI Jmt [104]. [26] CFI Jmt [83]. [27] Section B.3 below. [28] CFI Jmt [406]. [29] CFI Jmt [407]-[409]. [30] CFI Jmt [407]-[409]. [31] [152] of the Defendants’ opening submissions, quoted in CA Jmt [6.11]. [32] For instance, CFI Jmt [74], [83] and [129]. [33] For instance, CFI Jmt [407]-[409]. [34] Ibid. [35] See [45] above. [36] CFI Jmt [448]-[450]. [37] CFI Jmt [452]. [38] CA Jmt [6.18]-[6.20]. [39] CA Jmt [6.8] and [6.13]. [40] CA Jmt [6.10]. [41] CA Jmt [6.9]. [42] Paragraph 4(a)(ii) of the First Schedule, with words to similar effect in paragraphs 4(a)(iii) and 5(b). [43] Provided for by Art 24(1) of the 1984 Law and paragraph 1 of the First Schedule. [44] See [59] above. [45] [1998] Ch 241. [46] Kwan JA also concurred but dealt with other issues which do not presently arise. [47] CA Jmt [10] (emphasis added). [48] CA Jmt [11.5]. [49] CA Jmt [6.17]. [50] Appearing with Mr Chan Pat Lun for the respondents. [51] By virtue of paragraph 1 of the First Schedule. [52] CFI Jmt [34] and [161]. [53] Echoing Art 24(1) of the 1984 Law. [54] Paragraph 4(a)(ii) of the First Schedule. [55] Paragraph 4(a)(iii) of the First Schedule. [56] Appearing for the appellants with Mr Ashley Burns SC and Ms Bonnie Y K Cheng. [57] Paragraph 5(a) of the First Schedule. [58] Paragraph 5(b) of the First Schedule. [59] Ibid. [60] Section B.1 above. [61] CFI Jmt [406]-[409]. [62] CA Jmt [6.71]. [63] See, also, [42]-[44] above. [64] Respondents’ Case, passim; see, e.g., [5], [43]-[45] (Section E), [50] and [84]. [65] Ibid. at [5]; see, also, FN8 on p.6 which reads, referring to the 1984 Law, “Upon which, in all respects relevant to this appeal, both sides’ Jersey trust law experts (who were not cross-examined) were agreed.” [66] (1997-98) 1 HKCFAR 318. [67] [1946] AC 508 at pp.521-522. [68] (2009) 12 HKCFAR 621 at [37]. [69] (2011) 14 HKCFAR 798 at [40]-[44]; [45]-[47] and [59]. [70] (1997-98) 1 HKCFAR 343 at p.355F-H and p.364F. [71] (2000) 3 HKCFAR 70 at pp.90G-93C. [72] The Conflict of Laws in Hong Kong (3rd Ed.), Graeme Johnston and Paul Harris SC, at [2.060]. [73] Parkasho v Singh [1968] P 233 per Cairns J at p.250B-C. [74] [2017] UKSC 48, [2017] Bus LR 1731 per Lord Neuberger PSC at [93]. [75] Full Wisdom Holdings Ltd & Others v Traffic Stream Infrastructure Co Ltd & Others [2004] 2 HKLRD 1016 per Le Pichon JA at [23]. [76] As Ribeiro PJ, writing for the Appeal Committee, observed in Traffic Stream Infrastructure Co Ltd & Others v Full Wisdom Holdings Ltd & Others (2004) 7 HKCFAR 442 at [21] (citing the 13th edition of Dicey & Morris, virtually identical in this respect to the current 15th edition). [77] At [45] and [50]. [78] See [46] and [49] above. [79] See [63]-[67] above. [80] Neither expert gave oral testimony at trial. [81] CFI Jmt [407], [408] and [409]. [82] CFI Jmt [448], [449] and [450]. [83] CFI Jmt [413]. [84] (2013) 16 HKCFAR 681. [85] CFI Jmt [453]. [86] The different categories of claim and the legal consequences that flow from categorisation of a case as being within one rather than another category will be addressed further below. [87] CFI Jmt [455]. [88] CFI Jmt [456]. [89] CFI Jmt [457]. [90] CFI Jmt [479]-[480]. [91] CFI Jmt [479]. [92] CA Jmt [6.75]. [93] CA Jmt [6.78]. [94] CA Jmt [34]. [95] CA Jmt [35.2]. [96] [1999] 1 NZLR 664 at p.687. [97] See: Libertarian at [78]-[81]; Target Holdings Ltd v Redferns [1996] AC 421 at p.434; Bank of New Zealand v New Zealand Guardian Trust Co Ltd [1999] 1 NZLR 664 at p.687; and Equitable Compensation for Breach of Fiduciary Duty, Ribeiro PJ, Asia-Pacific Judicial Colloquium (Singapore, May 2019) at [63]-[64] and [66]-[67]. [98] See: Libertarian at [82] and [93]; Bank of New Zealand v New Zealand Guardian Trust Co Ltd [1999] 1 NZLR 664 at p.687; and Equitable Compensation for Breach of Fiduciary Duty, Ribeiro PJ, Asia-Pacific Judicial Colloquium (Singapore, May 2019) at [65]-[67]. [99] See: Bristol and West Building Society v Mothew [1998] Ch 1 at p.17; Bank of New Zealand v New Zealand Guardian Trust Co Ltd [1999] 1 NZLR 664 at p.687; and Equitable Compensation for Breach of Fiduciary Duty, Ribeiro PJ, Asia-Pacific Judicial Colloquium (Singapore, May 2019) at [68]-[70]. [100] [1980] 1 Ch 539. [101] That being his decision in Bartlett v Barclays Trust Co. (No.1) reported at [1980] 1 Ch 515. [102] [1980] 1 Ch 515 at p.542E-G. [103] [2015] AC 1503 at [52]-[54]. [104] Jersey Court of Appeal, unrep., 25 July 2018. [105] Ibid. at [28]. [106] Ibid. at [30]. [107] Ibid. at [32]. [108] Ibid. at [33]. [109] CA Jmt [27]-[28]. [110] See [108] above. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. In this appeal, which arises out of demonstrations mounted in a public gallery during sittings of a subcommittee of the Legislative Council (“LegCo”), the validity of two Administrative Instructions regulating admittance to and conduct within LegCo’s precincts is challenged on the basis that they infringe the right to freedom of expression. A. The relevant events 3. The facts are not in dispute. The appellant was a member of the Sai Kung District Council. Together with two of her assistants[1] and several others, she attended meetings of LegCo’s Public Works Subcommittee held on 7 and 13 May 2014 respectively when a project for extending the South-East New Territories landfill was discussed. It was a project which she and fellow residents strongly opposed. 4. On each occasion, she was admitted to the public gallery above Conference Room 1 in the LegCo complex where the meeting was being held. At the first meeting, she removed her jacket so that the characters 保衛將軍澳 (“Defend Tseung Kwan O”) were displayed on the T-shirt that she was wearing. She also handed to Cheung Mei-hung, one of her assistants, a paper poster depicting a Nazi Swastika with the characters 毒氣集中營 – 堆填區 (“Poison Gas Concentration Camp – Landfill”) which he displayed by holding it against the glass panel which walled off the public gallery above the conference room. This led to a commotion as security guards sought to seize the sign, leading to one guard sustaining bruising on her left wrist. The incident caused the Subcommittee’s meeting to be prematurely adjourned. 5. At the second meeting, the appellant and several others in the public gallery chanted slogans opposing the project, causing the Subcommittee’s chairman to warn them that they would be ejected if they did not stop. As the warnings were ignored and as the appellant and others linked arms to resist ejectment, the meeting was stood down and reconvened in a different conference room about an hour later with the public excluded from attendance.[2] B. The charges and the decisions below 6. Section 8 of the Legislative Council (Powers and Privileges) Ordinance (“LCPPO”)[3] regulates the admittance of persons to the LegCo complex. Section 8(3) provides that the President may: “... for the purpose of maintaining the security of the precincts of the Chamber, ensuring the proper behaviour and decorum of persons therein and for other administrative purposes, issue such administrative instructions as he may deem necessary or expedient for regulating the admittance of persons (other than members or officers of the Council) to, and the conduct of such persons within, the Chamber and the precincts of the Chamber.” 7. By LCPPO section 20(b), contravention of such an administrative instruction or any direction given thereunder is made an offence punishable by a fine of $2,000 and imprisonment for 3 months. 8. Pursuant to section 8(3), the President issued the Administrative Instructions for Regulating Admittance and Conduct of Person[4] (“Administrative Instructions”) which include the following provisions which are under challenge: “Section 11: Requirement for orderly behaviour (‘AI section 11’) Persons entering or within the precincts of the Chamber shall behave in an orderly manner and comply with any direction given by any officer of the Council for the purpose of keeping order. Section 12: Conduct in galleries (‘AI section 12’) (1) No person shall, in a press or public gallery, display any sign, message or banner. (2) No person shall, in a press or public gallery, display any sign or message on any item of clothing. (3) An officer of the Council may refuse admission to a press or public gallery to any person displaying any sign, message or banner, or to any person displaying any sign or message on any item of clothing, or to any person who, in the opinion of an officer of the Council, may so display any sign, message or banner, may so display any sign or message on any item of clothing or may otherwise behave in a disorderly manner. ...” 9. The appellant was charged jointly with Cheung Mei-hung with contravening AI section 12(1) at the 7 May 2014 meeting of the Subcommittee “in that they displayed sign and message in the public gallery”. They were also charged with contravening AI section 11 at the 13 May 2014 meeting “in that they failed to behave in an orderly manner within the precincts of the Chamber of the Legislative Council”. 10. The appellant was convicted after trial by the Magistrate[5] and fined $1,000 on each charge. Her appeal was dismissed by Wong J[6] who certified the following questions, reflecting the appellant’s constitutional challenge, as raising points of law of great and general importance, namely: (1) Whether the enactment of section 11 of the Administrative Instructions for Regulating Admittance and Conduct of Persons pursuant to section 20(b) of the Legislative Council (Powers and Privileges) Ordinance (Cap 382) is inconsistent with the principle of freedom of speech guaranteed by article 27 of the Basic Law and article 16 of the Bill of Rights, which rendered section 11 unconstitutional? (2) The same question ... in respect of section 12(1) of the Administrative Instructions for Regulating Admittance and Conduct of Persons. 11. Leave to appeal was granted by the Appeal Committee on the basis of the questions so certified.[7] C. The constitutional provisions 12. Article 27 of the Basic Law (“BL 27”),[8] on which the certified questions are based, states: Hong Kong residents shall have freedom of speech, of the press and of publication; freedom of association, of assembly, of procession and of demonstration; and the right and freedom to form and join trade unions, and to strike. 13. By BL 39, the Basic Law relevantly provides: (1) The provisions of the International Covenant on Civil and Political Rights ... as applied to Hong Kong shall remain in force and shall be implemented through the laws of the Hong Kong Special Administrative Region. (2) The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law. Such restrictions shall not contravene the provisions of the preceding paragraph of this Article.” 14. As is well-established,[9] the Hong Kong Bill of Rights Ordinance (“HKBORO”)[10] which enacts the Hong Kong Bill of Rights, implements the International Covenant on Civil and Political Rights (“ICCPR”) so that the Bill of Rights is given constitutional status by BL 39. Article 16 of the Bill of Rights (“BOR 16”) provides: (1) Everyone shall have the right to hold opinions without interference. (2) Everyone shall have the right to freedom of expression; this right shall include freedom to seek, receive and impart information and ideas of all kinds, regardless of frontiers, either orally, in writing or in print, in the form of art, or through any other media of his choice. (3) The exercise of the rights provided for in paragraph (2) of this article carries with it special duties and responsibilities. It may therefore be subject to certain restrictions, but these shall only be such as are provided by law and are necessary- (a) for respect of the rights or reputations of others; or (b) for the protection of national security or of public order (ordre public), or of public health or morals. 15. This Court has held[11] that there is no difference between the right of peaceful assembly guaranteed by BL 27 and that provided for in BOR 17.[12] The same applies to freedom of speech under BL 27 and freedom of expression under BOR 16. The Court has also noted[13] that the rights to freedom of expression, of public assembly and of procession and demonstration are closely related, making the case-law on these associated freedoms collectively relevant. 16. Accordingly, by the combined effect of BL 39 and BOR 16, if any purported restriction on the right of free expression is to be valid, it must have sufficient legal certainty to qualify as a restriction “prescribed by law”[14] and must be “necessary for respect of the rights or reputations of others; or for the protection of national security or of public order (ordre public), or of public health or morals.” It is established that the requirement of necessity involves the application of a proportionality test[15] and that the objectives listed in BOR 16 are exhaustive of purposes qualifying as legitimate aims to justify a purported restriction of the guaranteed right.[16] D. The issues in this appeal 17. The issues raised by Mr Hectar Pun SC[17] fall within a relatively narrow compass. 18. On the charge relating to the first incident, he submits that “the total and blanket prohibition imposed by section 12(1) of the Administrative Instructions on the display of any ‘sign, message or banner’ in a press or public gallery in the LegCo fails to satisfy the third stage of the proportionality test.” He argues that the only legitimate purpose said to be pursued by AI section 12 is the protection of public order within the LegCo building and that “by ... rendering all forms of display of any ‘sign, message or banner’ a criminal offence regardless of its purpose, nature, manner and its impact (if any) on the public order in the LegCo”, the section is unjustifiably wide.[18] He invites the Court to hold that AI section 12 is inconsistent with BL 27 and BOR 16 and thus unconstitutional. I shall call this the “blanket prohibition” argument. 19. As to the charge relating to the second incident, the appellant’s argument is that AI section 11 is invalid as a restriction on the guaranteed right because it is legally uncertain and does not constitute a restriction “prescribed by law”. Mr Pun SC submits that the section’s requirement for one to “ ‘behave in an orderly manner’ ... leaves a person entering or within the precincts of the Chamber of the LegCo unclear precisely what he must avoid doing in order to avoid committing a criminal offence under section 20(b) of the LCPPO”.[19] I shall call this the “prescribed by law” argument. Mr Pun SC contends that the Judge wrongly treated the relevant offence as having as one of its elements, a failure to comply with a direction given by an officer of the Council – an erroneous approach based on a misconstruction of AI section 11 and postulating an offence with which the appellant was never charged. 20. In response, Ms Anna Lai SC[20] begins by raising an important basic issue. She submits that members of the public, including the appellant, “do not have a right to exercise their freedom of expression in the public gallery” of LegCo so that “the constitutional challenges fail in limine.”[21] This involves the proposition, by analogy (so it is argued) with decisions relating to demonstrations sought to be mounted on private property, that there is “no freedom of forum” and that the right to freedom of expression does not apply if sought to be exercised on government premises to which the general public has not been given free access.[22] The contention is that on such properties, the guaranteed right simply does not arise so that there is no room for examining the aim, rationality or proportionality of the restriction. I shall call this the “rights not applicable” argument. 21. In the alternative, Ms Lai SC submits that if the right is engaged, the restrictions imposed by AI sections 11 and 12 are sufficiently certain and satisfy the test of proportionality, so that they are constitutionally valid. This responds to Mr Pun SC’s blanket prohibition argument and also to his prescribed by law argument. 22. Another issue raised by Ms Lai SC (although said to be “by no means determinative of the issue”[23]) involves the submission that regulation of the admittance and conduct of persons within the precincts of LegCo is an internal matter for the legislature so that, applying self-restraint in accordance with the non-intervention principle, the Court should refrain from interfering. I shall call this the “non-intervention” argument. E. The “rights not applicable” argument E.1 The nature and basis of the argument 23. Since it involves the submission that the constitutional challenge fails in limine, I shall deal first with the “rights not applicable” argument. 24. Ms Lai SC submits that the right of free expression is not engaged in the present case because members of the public are not given free access to the LegCo complex and because those that are allowed in only enter subject to restrictions on their conduct. Since LegCo, in common with the Government in relation to government-owned properties, is entitled to exercise its property rights so as to deny or restrict access to the LegCo complex, no one (so it is argued) can assert a right to be admitted and if admitted, no one can object to restrictions being placed on their conduct by invoking the right to freedom of expression. 25. This argument was accepted by Wong J who held: “Members of the public do not have the right to demonstrate in the Legislative Council Building, especially when meetings are in session. Having regard to the principle discussed in paragraphs 23 and 24 of this judgment,[24] I take the view that the rights the appellants mentioned have not been infringed. Neither the Basic Law nor the Bill of Rights gives members of the public the freedom to exercise such rights inside the Legislative Council Chamber, especially when they in fact do not have the absolute right to enter the Chamber.”[25] 26. A similar argument was made in the Canadian Supreme Court in Committee for the Commonwealth of Canada v Canada,[26] a case in which the respondents had been distributing leaflets and engaging passers-by in discussion of the respondents’ political aims in the state-owned passenger terminal of Montreal’s Dorval Airport. The question was whether the respondents were entitled to rely on the right of free expression guaranteed by section 2(b) of the Canadian Charter of Rights and Freedoms[27] so as to require the government to justify restricting their canvassing activities in accordance with section 1 of the Charter.[28] 27. One line of argument which is echoed by Wong J’s judgment and the respondent’s submissions, is what L'Heureux-Dubé J called the “hard line” taken by the Canadian Government in that case, “submitting that as owner of property, the government has the right to exclude whomever it wants, and to impose conditions on invitees to its property without limitation by the Charter”.[29] As McLachlin J explained, the Government’s position was that: “...there is no constitutional right to use any of its property for purposes of public expression. Only with its permission and where it considers it appropriate should individuals and groups be permitted to speak and demonstrate. The government submits that as the owner of all such property, it has the absolute right to exclude the use of the property for public expression if it chooses. It relies on the fact that the owners of property are generally entitled to control who enters on it and how it is used, a right which extends to the right to control expression on their property. The Crown, it contends, should be placed in no worse position than a private property owner.”[30] 28. In my view, for the following reasons, the “rights not applicable” argument cannot be accepted. E.2 Fundamental rights not trumped by property rights 29. The argument takes as its premise the existence of an entitlement on the part of the Government (or LegCo in this case) as property owner to exclude the public. It reasons from that premise to the conclusion that the guaranteed right of freedom of expression is excluded. It therefore subjugates a fundamental constitutional right to property interests, leaving the applicability or otherwise of that right to the unfettered choice of a Government agency regarding the grant of access. 30. Such an argument is, in my view, wrong in principle. As BL 8 makes clear,[31] an existing statutory or common law rule which comes into conflict with a constitutional requirement must give way. The “rights not applicable” argument inverts this principle. 31. For similar reasons, the Canadian Government’s “hard line” argument in the Montreal Airport case did not find favour with any of the judges. As Lamer CJ put it: “... the government’s right of ownership cannot of itself authorize an infringement of the freedom guaranteed by section 2(b) of the Charter.”[32] And as McLachlin J stated: “To accept the Crown's argument would be to restrict the freedom guaranteed by the Charter to limits much narrower than those with which it has traditionally been associated. Little would remain of the right. ... the state's property interest in a forum does not give it the absolute right to control expression on that forum.”[33] E.3 Other attempts at a priori exclusion of guaranteed rights 32. Examples can be found, including in Committee for the Commonwealth of Canada v Canada,[34] where judges have endeavoured to develop an intermediate doctrine for screening out the application of the right to free expression on some, but not all, state-owned properties by virtue of the state’s right as property owner to deny access. It is not, with respect, an attractive line to follow since it shares the flaws afflicting the “hard line” argument in respect of excluded properties and because major difficulties arise in trying to draw a definitional line between the two categories of properties. 33. Thus, Lamer CJ proposed a rule based on compatibility of the exercise of free expression with the function of the state-owned property in question: “In my opinion, the ‘freedom’ which an individual may have to communicate in a place owned by the government must necessarily be circumscribed by the interests of the latter and of the citizens as a whole: the individual will only be free to communicate in a place owned by the state if the form of expression he uses is compatible with the principal function or intended purpose of that place. ... Accordingly, it is only after the complainant has proved that his form of expression is compatible with the function of the place that the justifications which may be put forward under s 1 of the Charter can be analysed.”[35] 34. However, as McLachlin J pointed out, the appropriateness of this test may be questioned because of difficulties with the uncertain and relative concepts of “function” and “incompatibility”:[36] “... the concept of function presents difficulties. Does it mean normal function? Minimal or essential function? Optimum function? At what point does expression become incompatible with function? Presumably, only if the impairment of function were severe would s 2(b) be held inapplicable, with limitations relating to optimal (as opposed to minimal) function falling to be justified under s 1. Yet drawing this line may prove difficult in practice.” 35. The test which her Honour proposed for “defining what types of government property should prima facie be regarded as constitutionally available for forums for public expression”[37] was to require the claimant to: “... establish that the expression in question (including its time, place and manner) promote one of the purposes underlying the guarantee of free expression. These were defined in Irwin Toy[38] (at p 976) as: (1) the seeking and obtaining of truth; (2) participation in social and political decision-making; and (3) the encouragement of diversity in forms of individual self-fulfilment and human flourishing by cultivating a tolerant, welcoming environment for the conveyance and reception of ideas. Only if the claimant can establish a link between the use of the forum in question for public expression and at least one of these purposes is the claimant entitled to the protection of s 2(b) of the Charter.”[39] 36. With respect, I do not think that is a test that could or should be adopted in Hong Kong. BL 27 and BOR 16 guarantee the right to freedom of expression subject only to the specified permitted restrictions. It would not be appropriate for our courts to place hurdles in the way of a person claiming those rights where such rights are factually engaged, by requiring such person to show that the content, manner and form of the proposed expression promote the matters identified as the underlying purposes of free expression. As indicated in the discussion which follows, rather than imposing a burden on the claimant to prove that the proposed expression qualifies for constitutional protection, the burden is rightfully placed on the government to prove that its limitation of the guaranteed right is justified. 37. It also seems to me that an official trying to decide whether access to a venue should be granted or denied will find it hard to know beforehand whether the intended demonstration will meet the somewhat amorphous criteria articulated. Moreover, the criteria themselves appear debatable. It is well-established that, subject to permissible limitations, freedom of expression extends to views which may be “disagreeable, unpopular, distasteful or even offensive to others”.[40] It is unclear whether the suggested criteria accommodate the expression of such views. 38. What the two approaches just discussed have in common is a search for some a priori basis for dis-applying fundamental rights on state-owned properties, without ever proceeding to examine whether exclusion of the right is justified in terms of the legitimacy and proportionality of such restriction. I think that a more orthodox approach is preferable. E.4 Focussing on fundamental rights 39. In my view, the correct starting-point and the proper focus throughout is on the guaranteed right, adopting the assumption that it is universally applicable, subject to any constitutionally valid restriction. Thus, where the right to freedom of expression is invoked, one asks whether factually, that right is engaged. If so, the question becomes whether any restriction which purports to limit its exercise is valid, that is, whether it pursues a legitimate aim which falls within one of the permitted categories listed in BOR 16; and if so, whether it is rationally connected with accomplishing that aim; whether the restriction is no more than reasonably necessary for accomplishing that purpose; and whether a reasonable balance has been struck between the societal benefits of the encroaching measure on the one hand and the inroads made into the guaranteed right on the other.[41] 40. A person exercising the right to free expression or the related freedoms of peaceful assembly, demonstration and procession has choices. As BOR 16 recognizes, communication may be made “orally, in writing or in print, in the form of art, or through any other media of his choice”. If the expression is communicated using the postal service or by transmissions in the traditional media or by internet postings, no physical activity on the ground may be involved. But where the choice is to exercise the right through seeking, receiving or imparting information or ideas in some physical location, the exercise will have various physical and temporal dimensions. Choices will be made (amongst other matters) as to when, where, by whom, in what form and for what duration the exercise of the freedom will take place. 41. That the right inherently involves various dimensions in its exercise was recognized in a series of English cases involving demonstrations which took the form of tented encampments on prominent or symbolically important sites. One such case[42] involved the Aldermaston Women’s Peace Camp which had been going for some 23 years as a protest against nuclear weapons, with the women assembling on one weekend of each month on government land close to a weapons establishment. Laws LJ held that in some cases, including the case at hand: “... this ‘manner and form’ may constitute the actual nature and quality of the protest; it may have acquired a symbolic force inseparable from the protesters’ message; it may be the very witness of their beliefs. ... To them, and (it may fairly be assumed) to many who support them, and indeed to others who disapprove and oppose them, the ‘manner and form’ is the protest itself.”[43] 42. Another such case is Mayor of London v Hall,[44] which involved a demonstration by protesters who had, for over two months, set up a camp in Parliament Square Gardens bounded by the Palace of Westminster, Westminster Abbey and the United Kingdom Supreme Court. Lord Neuberger of Abbotsbury MR noted the importance of recognizing that the manner and location of the protest was part and parcel of the exercise of free expression: “The right to express views publicly, particularly on the important issues about which the defendants feel so strongly, and the right of the defendants to assemble for the purpose of expressing and discussing those views, extends to the manner in which the defendants wish to express their views and to the location where they wish to express and exchange their views. If it were otherwise, these fundamental human rights would be at risk of emasculation. Accordingly, the defendants' desire to express their views in Parliament Square, the open space opposite the main entrance to the Houses of Parliament, and to do so in the form of the Democracy Village, on the basis of relatively long-term occupation with tents and placards, are all, in my opinion, within the scope of articles 10 and 11.”[45] 43. His Lordship favoured the approach which is espoused in this judgment, namely, treating the rights as generally available and then subjecting the validity of particular restrictions to scrutiny. As Lord Neuberger MR put it: “... when freedom of assembly, and, even more, when freedom of expression, are in play, then, save possibly in very unusual and clear circumstances, article 11, and article 10, should be capable of being invoked to enable the merits of the particular case to be considered.”[46] 44. In my view, a coherent analysis of the issues arising where access to property is denied to would-be protesters is facilitated by recognizing that the right of free expression encompasses as one of its dimensions, the location of its exercise. If access to that place is denied, such denial is properly viewed as a restriction imposed on the exercise of the right so that the legitimacy, rationality and proportionality of that restriction fall to be considered. This involves applying established constitutional principles for determining when fundamental rights may validly be limited, without being diverted towards a doctrine involving an a priori or presumptive exclusion of such rights on the basis of a property owner’s interests in the proposed venue. 45. A rule presumptively excluding free expression at certain types of venues is too blunt an instrument. A proportionality analysis, on the other hand, enables the court to take into account the dimensions of any particular exercise of such rights and the exigencies of the intended venue. 46. A restriction may, for instance, be disproportionate in relation to a silent, orderly demonstration mounted by a limited number of protesters displaying placards for a few hours without unduly inconveniencing other members of the public. But the same restriction may be entirely justified in respect of a demonstration conducted in a manner or taking a form which is much more intrusive, affecting the rights of others or posing risks to public order, ordre public, or public health and morals, especially where more acceptable alternatives are open to the protesters. It was by application of a proportionality analysis, weighing up the manner, form and impact of each of the demonstrations at the sites concerned, that the English courts distinguished between the Aldermaston camp[47] (which was held entitled to continue) on the one hand and the much more invasive camps in Parliament Square[48] and around St Paul’s Cathedral[49] (which were held to have been justifiably curtailed) on the other. 47. As Lord Neuberger MR stated in The Mayor Commonalty and Citizens of London v Samede,[50] the case involving the camp around St Paul’s Cathedral: “[Establishing the limits of the right of peaceful assembly and protest on the highway] is inevitably fact-sensitive, and will normally depend on a number of factors. In our view, those factors include (but are not limited to) the extent to which the continuation of the protest would breach domestic law, the importance of the precise location to the protesters, the duration of the protest, the degree to which the protesters occupy the land, and the extent of the actual interference the protest causes to the rights of others, including the property rights of the owners of the land, and the rights of any members of the public.” E.5 “Obvious cases” for restricting rights 48. There are many locations in which, as a matter of common sense, one would think it obvious that demonstrations and similar exercises of free expression should be excluded. One might therefore be tempted to conclude that there ought to be some principle excluding in limine engagement of the right in such places. 49. Such views can be found in Committee for the Commonwealth of Canada v Canada,[51] where, for instance, L'Heureux-Dubé J stated: “… the Charter’s framers did not intend internal government offices, air traffic control towers, prison cells and Judges’ Chambers to be made available for leafletting or demonstrations.”[52] 50. Similarly, McLachlin J was of the view that: “It would be difficult to contend that [the purposes she had identified as underlying the guarantee of free expression referred to above] are served by ‘public’ expression in the sanctum of the Prime Minister's office, an airport control tower, a prison cell or a judge's private chambers, to return to examples where it seems self-evident that the guarantee of free expression has no place.”[53] 51. The examples given are all compelling and one can readily accept that in the vast majority of such cases, one would rapidly conclude that outsiders cannot validly assert a constitutional right to speak or demonstrate in such locations. However, the proper conclusion to draw in those cases is that the right to freedom of expression may validly be restricted and not that the Court should accept a presumptive rule excluding engagement of the right in limine,carrying with it the objections of principle and problems of definition that have been discussed above. A proportionality approach enables individualised and nuanced assessments and caters for the possibility that in exceptional circumstances such rights might be engaged even in such ostensibly improbable places. 52. Prisons are a good example. One might be inclined to think it obvious that for security reasons, demonstrations (say, against prison conditions) cannot be allowed to be mounted by outsiders within the precincts of a prison. However, other forms of exercising the right to free expression and other fundamental rights may have an important role to play. While prisoners necessarily forfeit their right to personal liberty as a consequence of their lawfully-imposed incarceration, the European Court of Human Rights (“ECtHR”) has held that generally, they otherwise continue to enjoy all the fundamental rights and freedoms subject to restrictions justified on principles of proportionality.[54] Absolute and non-derogable rights against torture and cruel, inhuman or degrading treatment or punishment which do not raise proportionality issues clearly remain applicable.[55] The Court of First Instance has held that the right to vote under BL 26 continues to apply and that a blanket exclusion of that right, regardless of the nature and gravity of the offence, the length and type of sentence, etc, is invalid as a disproportionate restriction of that right.[56] Similarly, the right to confidential legal advice under BL 35 may also be expected to persist, subject to proportionate limitations, within prison precincts.[57] 53. This approach is in principle applicable to the right of free expression. To take a hypothetical example, if the system of visits by Justices of the Peace,[58] traditionally designed to inspect prison conditions and to allow prisoners to air their grievances, were to be abolished and the JPs denied access, a challenge might well be made on the basis that such denial infringes the prisoners’ and the visiting JPs’ rights to freedom of expression so that the validity of the restriction would have to be assessed on the basis of proportionality. 54. Within the framework of restrictions permitted under BOR 16, the principle of proportionality presents a coherent basis for judging whether and to what extent guaranteed rights may be excluded or limited in prisons or similarly restrictive locations. In the vast majority of “obvious cases”, exclusions or limitations of rights, especially rights sought to be exercised in forms incompatible with the maintenance of public order or ordre public,will be upheld as necessary and proportionate. But the proportionality principle permits flexibility to be retained. A presumptive exclusionary rule is ill‑suited to the task. E.6 Private property 55. The discussion so far has concerned the exercise of free expression on government-owned property. Is the position different regarding private property? As this did not receive full argument, my remarks must be taken to be tentative and subject to possible adjustment in a future case. But some discussion is merited here because the respondent seeks to derive support for the “rights not applicable” argument from what is said to be the analogous position regarding private property rights. It is therefore relevant to consider whether any such analogy exists. 56. It has been suggested that a bright line can be drawn to exclude any right to exercise one’s freedom of expression on someone else’s private property. This is sometimes put in terms of there being “no freedom of forum” or of the Government having no positive duty to secure access to private property for use as a forum for expression. This represents another a priori argument in which the fact of private ownership presumptively precludes engagement of the right and application of the proportionality test. 57. Thus, for example, in Committee for the Commonwealth of Canada v Canada,[59] McLachlin J stated: “Freedom of expression does not automatically comport freedom of forum. For example, it has not historically conferred a right to use another's private property as a forum for expression. A proprietor has had the right to determine who uses his or her property and for what purpose. Moreover, the Charter does not extend to private actions. It is therefore clear that s 2(b) confers no right to use private property as a forum for expression.” 58. In my view, where the guaranteed right is engaged, the orthodox approach of ascertaining whether any restriction of access is proportionate and valid remains applicable but with the qualification that elements of particular significance regarding private property must be given special weight in the proportionality analysis. 59. Such considerations flow from the existence of constitutional protections relating to private property and the right to privacy. Thus, BL 6 requires the Government to protect the right of private ownership in accordance with law; and BL105 mandates such protection for the rights of individuals and legal persons to the acquisition and use, etc, of property. Particular protection is conferred on the “homes and other premises of Hong Kong residents”, BL 29 stating that they “shall be inviolable” and prohibiting arbitrary or unlawful intrusion. Similarly, BOR 14 prohibits “arbitrary or unlawful interference with [a person’s] privacy, family, home or correspondence”. 60. Consequently, where a limitation on the right to freedom of expression involves denial of access to private property, justification of the restriction as a measure necessary for the protection of the rights of others has to be given very substantial weight in the proportionality balance. 61. Such weight is further enhanced where the property involves a resident’s home or other premises, since such “rights of others” comprehend both property rights and the right to privacy and inviolability of the home. This was noted by the New Zealand Supreme Court in Brooker v Police,[60] in relation to section 14 of the New Zealand Bill of Rights Act 1990,[61] where Blanchard J stated: “The exercise of the s 14 right in the form of a protest is not confined to non-residential streets. However, what has to be borne by residents in an exclusively or predominantly residential area will be less than in areas where there is little or no residential character. This is because the common law has long recognised that men and women are entitled to feel secure in their homes, to enjoy residential tranquillity - an element of the right to privacy. They are justifiably entitled not to be subjected there to undue disturbance, anxiety or coercion.” 62. While “never say never” is “a wise judicial precept”,[62] applying the orthodox test, it is exceedingly difficult to envisage a case where refusal of access to other persons’ private homes (or their curtilages or common areas) to be used as a forum for free expression would be ruled a disproportionate limitation on the right. 63. In HKSAR v Au Kwok Kuen,[63] Andrew Cheung J[64] had to consider “... the limits of the right of assembly and of the right to freedom of expression when they are sought to be exercised in private residential property without the permission of its owners or occupiers.”[65] It involved 26 persons associated with a housing concern group forcing their way through a police cordon to enter and mount a demonstration in the common areas of a private residential development with the intention of delivering a petition to a person who resided in one of the flats. His Lordship noted that under our constitution, restrictions necessary for the protection of the rights of others are permitted[66] and that where one is concerned with a private residential development, it protects home and privacy rights.[67] He considered alternative possible locations for the demonstration and noted that ample opportunities exist to exercise the relevant rights in public places[68] and concluded that: “... the right of peaceful assembly and the right to freedom of expression stop, so far as physical or geographical limits are concerned, at the boundary of private residential property belonging to others, in the absence of any permission to enter.”[69] 64. While the passage just cited might be read as espousing a private property-based presumptive exclusion of the right to freedom of expression, the better view seems to me to be that the judgment as a whole proceeds on the orthodox basis that the denial of access, involving as it did a private residential development, was legitimate and valid, applying proportionality principles. 65. The position may, however, be a little less obvious when it comes to privately owned commercial properties such as shopping malls and similar premises, to which the public are freely admitted in furtherance of the owners’ business interests. While it will no doubt still be rare for a refusal of access amounting to a restriction on the right to freedom of expression to be held to be constitutionally invalid, such an outcome could be reached in an exceptional case. 66. This was acknowledged by the ECtHR in Appleby v United Kingdom,[70] where the applicants were prohibited from setting up stands at the entrance to a privately-owned shopping mall in the town centre when campaigning against the local authority’s decision to allow a playing field in the vicinity to be built upon, the complaint being that this would deprive their children of green areas for play. 67. The Court stated: “... notwithstanding the acknowledged importance of freedom of expression, [Article 10 of the ECHR] does not bestow any freedom of forum for the exercise of that right. While it is true that demographic, social, economic and technological developments are changing the ways in which people move around and come into contact with each other, the Court is not persuaded that this requires the automatic creation of rights of entry to private property, or even, necessarily, to all publicly owned property (Government offices and ministries, for instance). Where however the bar on access to property has the effect of preventing any effective exercise of freedom of expression or it can be said that the essence of the right has been destroyed, the Court would not exclude that a positive obligation could arise for the State to protect the enjoyment of Convention rights by regulating property rights. The corporate town, where the entire municipality was controlled by a private body, might be an example.”[71] 68. The Court indicated that the outcome in any case depends on the principle of proportionality and a balancing exercise: “In determining whether or not a positive obligation exists, regard must be had to the fair balance that has to be struck between the general interest of the community and the interests of the individual, the search for which is inherent throughout the Convention. The scope of this obligation will inevitably vary, having regard to the diversity of situations obtaining in Contracting States and the choices which must be made in terms of priorities and resources. Nor must such an obligation be interpreted in such a way as to impose an impossible or disproportionate burden on the authorities.”[72] 69. It follows from the foregoing discussion that there is only a limited and imperfect analogy between the approaches to freedom of expression as exercised in public versus private properties. In each case, assuming that the right is engaged, its limitation embodied by denial of access to the site is assessed on proportionality principles. But, where private property is concerned, special elements involving the constitutional protections of private property and privacy in the home enter the equation, weighing heavily in favour of validating restricted access although this may be subject to rare possible exceptions. E.7 Conclusion as to the “rights not applicable” argument 70. As indicated in the foregoing discussion, there are, in my opinion, two main reasons for rejecting the “rights not applicable” argument. First, whether in its “hard line” version of excluding the right to freedom of expression on all government-owned properties or in its intermediate form of excluding the right only on some of them, it impermissibly seeks to subjugate fundamental rights to property interests, inverting the usual principle. Secondly, it fails to recognize that the proposed location of a demonstration or other form of expression is an intrinsic dimension of the right so that exclusion from that location is properly analysed as a restriction of the right which requires to be justified on orthodox proportionality principles. In short, the proper approach is to focus on the right instead of on property interests, and to examine the validity of any restrictions imposed on its exercise by applying the principle of proportionality within the framework of the permitted restrictions listed in BOR 16. F. The “non-intervention” argument 71. Logically, the next argument to deal with is the respondent’s “non-intervention” argument which involves the suggestion that the Court should not interfere with LegCo’s decision to regulate conduct as set out in AI sections 11 and 12 because they represent aspects of the internal management of LegCo’s affairs. This can be dealt with briefly and must be rejected for two reasons. 72. The non-intervention principle was explained by this Court in Leung Kwok Hung v President of the Legislative Council (No 1),[73] as: “... the principle that the courts will recognise the exclusive authority of the legislature in managing its own internal processes in the conduct of its business, in particular its legislative processes. The corollary is the proposition that the courts will not intervene to rule on the regularity or irregularity of the internal processes of the legislature but will leave it to determine exclusively for itself matters of this kind ...” 73. Thus, the first answer to the respondent’s argument is that regulation of the admittance and conduct of strangers who wish to enter the precincts of the Chamber falls outside the category of managing LegCo’s internal processes so that the principle simply does not apply. 74. Secondly, as this Court held in the abovementioned case: “...it is important to recognise that the principle of non-intervention is necessarily subject to constitutional requirements. The provisions of a written constitution may make the validity of a law depend upon any fact, event or circumstance they identify, and if one so identified is a proceeding in, or compliance with, a procedure in the legislature the courts must take it under its cognizance in order to determine whether the supposed law is a valid law.”[74] 75. Accordingly, even if such regulation did fall within the internal management category, the non-intervention principle would have to give way since the Court is duty-bound to examine the validity of AI sections 11 and 12 in so far as they impose restrictions on the exercise of a constitutional right, all the more so when a criminal offence is involved. G. The “prescribed by law” argument 76. As we have seen,[75] BL 39 and BOR 16 require any purported limitation of the right of free expression to have sufficient legal certainty to qualify as a valid restriction “prescribed by law”.[76] In Mo Yuk Ping v HKSAR,[77] this principle was summarised as follows: “A criminal offence must be so clearly defined in law that it is accessible and formulated with sufficient precision to enable the citizen to foresee, if need be with appropriate advice, whether his course of conduct is lawful or unlawful. It is, however, accepted that absolute certainty is unattainable and would entail excessive rigidity. Hence it is recognised that a prescription by law inevitably may involve some degree of vagueness in the prescription which may require clarification by the courts.” 77. The appellant’s “prescribed by law” argument involves a challenge to the validity of AI section 11.[78] Mr Pun SC submits that penalising a person for failing to “behave in an orderly manner” leaves him, because of the vagueness of those words, “unclear precisely what he must avoid doing” and so constitutes an invalid restriction on the guaranteed right. G.1 The proper construction of AI section 11 78. I cannot accept that argument. To ascertain whether a provision is legally uncertain, it is necessary for it to be construed. And as the Court in HKSAR v Lam Kwong Wai[79] emphasised: “The modern approach to statutory interpretation insists that context and purpose be considered in the first instance, especially in the case of general words, and not merely at some later stage when ambiguity may be thought to arise.” 79. Mr Pun SC has not construed AI section 11. Instead, his argument merely focusses on the phrase “behave in an orderly manner” taken in isolation. When the section is properly construed in the light of its context and purpose, it is impossible to say that it lacks legal certainty or that it leaves a person “unclear precisely what he must avoid doing”. 80. In the first place, AI section 11 does not simply penalise a failure to “behave in an orderly manner” without more. The orderly behaviour is demanded only of persons who enter or are within precincts of the LegCo complex. The section also requires compliance with any directions given by an officer of the Council “for the purpose of keeping order”. The section is therefore self-evidently concerned with keeping order in those precincts. 81. As was pointed out in HKSAR v Chow Nok Hang,[80] citing Gleeson CJ in Coleman v Power,[81]“[concepts] of what is disorderly, or indecent, or offensive, vary with time and place, and may be affected by the circumstances in which the relevant conduct occurs.” It was held that: “Such concepts are best left to the trial judge to be applied in their ordinary meaning to the time, place and circumstances of the conduct in question.”[82] 82. In the present situation, the clear purpose of AI section 11 is to set a standard of orderly behaviour on the part of visitors congruent with LegCo’s institutional and social importance. Its context shows that it is part of a statutory framework aimed at creating a secure and dignified environment in the LegCo complex conducive to the legislature carrying out its constitutional functions at its sittings without disruption or disturbance, while permitting members of the public to observe the proceedings within the Chamber as an open legislative process. The section’s context is provided by its mother Ordinance, the LCPPO, and the other sections of the Administrative Instructions. 83. Thus, the LCPPO’s Long Title states that it aims (among other things): “... to secure freedom of speech in the Legislative Council; to make provision for regulating admittance to and conduct within the precincts of the Chamber of the Legislative Council ... and for offences in respect of such proceedings and related matters; and for purposes incidental to or connected therewith.”[83] 84. In LCPPO section 2, “Chamber” is defined to mean: “... the Chamber in which the proceedings of the Council are conducted, and any galleries and places therein provided for members of the public and representatives of the press, television and radio, and includes any lobbies, offices or precincts used exclusively in connexion with the proceedings of the Council.” And “Precincts of the Chamber” is defined to mean “... the Chamber and offices of the Council and any adjacent galleries and places provided for the use or accommodation of members of the public and representatives of the press, television and radio, and subject to any exceptions made by the President under subsection (2) includes, during the whole of any day the Council or a committee is sitting, the entire building in which the Chamber is situated and any forecourt, yard, garden, enclosure or open space adjoining or appertaining to such building and used or provided for the purposes of the Council.” 85. Protection of the core activity of free debate during proceedings in the Chamber is taken up in LCPPO section 3: “There shall be freedom of speech and debate in the Council or proceedings before a committee, and such freedom of speech and debate shall not be liable to be questioned in any court or place outside the Council.” 86. Section 5(b) immunises members from arrest “for any criminal offence whilst attending at a sitting of the Council or a committee” and section 6(1) forbids the service of civil proceedings or the process of civil execution “within the precincts of the Chamber while the Council is sitting”. 87. Section 8 is especially important. Section 8(1) provides: “Subject to this section, sittings of the Council shall be open to the public.” The statutory intention is thus to admit the public into LegCo’s precincts so that they can observe the sittings of the Council. Regulation of such attendance is obviously needed, so section 8(3) empowers the President to issue Administrative Instructions: “... for the purpose of maintaining the security of the precincts of the Chamber, ensuring the proper behaviour and decorum of persons therein and for other administrative purposes.” 88. Section 17(c) directly penalises interruptions of LegCo sittings: “Any person who ... creates or joins in any disturbance which interrupts or is likely to interrupt the proceedings of the Council or a committee while the Council or such committee is sitting, commits an offence and is liable to a fine of $10000 and to imprisonment for 12 months, and in the case of a continuing offence to a further fine of $2000 for each day on which the offence continues.” 89. And similarly, as we have seen,[84] section 20(b) penalises contravention of: “... administrative instructions issued under section 8(3), or any direction given thereunder, regulating the admittance of persons to or the conduct of persons within the Chamber or the precincts of the Chamber”. 90. AI section 2 provides specifically for admittance to public galleries during LegCo sittings: “Members of the public may be admitted to a public gallery whenever the Council or a committee is sitting in public in the Chamber or in a committee room.” 91. Taking into account the abovementioned provisions, there is nothing uncertain about AI section 11.[85] In the light of the context and purpose illuminated by those provisions, anyone reading the section, with appropriate advice if necessary, would know that creating a disturbance by demonstrating in the public gallery while a LegCo subcommittee was sitting constitutes a contravention. Indeed, that is a conclusion any layman would have no difficulty reaching, exercising common sense. G.2 The authorities relied on by the appellant 92. In support of his “prescribed by law” argument, Mr Pun SC relies on Steel v United Kingdom,[86] and Lau Wai Wo v HKSAR,[87] two cases concerning legal certainty regarding binding-over orders; as well as Brooker v Police,[88] a New Zealand case involving the summary offence[89] of behaving in an offensive or disorderly manner. These are clearly distinguishable authorities, involving as they do, an uncertainty argument relating to the requirement that someone “be of good behaviour” or the prohibition against “behaving in a disorderly manner” at large, without the contextual framework provided by a statutory scheme such as that applicable to AI section 11 just discussed. 93. Because of their open-endedness, binding-over orders have faced challenges on the ground that they lack legal certainty. Thus, in Steel v United Kingdom,[90] certain protesters were ordered to be bound over “to be of good behaviour” with imprisonment to follow if they refused to be bound over, such orders having been made after they were found to have committed a breach of the peace. The ECtHR regarded the requirement that a person “be of good behaviour” to be vague and imprecise, but it held that the prior finding of a breach of the peace provided a sufficient context to rescue the order from failing the “prescribed by law” test. The Court stated: “... the orders were expressed in rather vague and general terms; the expression ‘to be of good behaviour’ was particularly imprecise and offered little guidance to the person bound over as to the type of conduct which would amount to a breach of the order. However, in each applicant’s case the binding over order was imposed after a finding that she had committed a breach of the peace. Having considered all the circumstances, the Court is satisfied that, given the context, it was sufficiently clear that the applicants were being requested to agree to refrain from causing further, similar, breaches of the peace during the ensuing 12 months.”[91] 94. Hashman v United Kingdom,[92] is a contrasting case where fox hunting saboteurs disrupted a hunt and were bound over in the sum of £100 not to breach the peace and to be of good behaviour for 12 months, an order made solely on the basis of a finding that they had behaved in a manner contra bonos mores.[93] Since that concept (which had been defined in the English courts as behaviour which is “wrong rather than right in the judgment of the majority of contemporary fellow citizens”) offered no more certainty than the broad “good behaviour” requirement, the ECtHR was unable to adopt the Steel v UK technique and held that the order prospectively binding over the protesters to be of good behaviour at large was uncertain.[94] Distinguishing Steel, the Court stated: “Whilst in the case of Steel the applicants had been found to have breached the peace, and the Court found that it was apparent that the bind over related to similar behaviour, the present applicants did not breach of the peace, and given the lack of precision referred to above, it cannot be said that what they were being bound over not to do must have been apparent to them.”[95] 95. In Lau Wai Wo v HKSAR,[96] this Court was concerned with a case where the defendant was acquitted of the charge of common assault on his brother but was bound over by the Magistrate in the sum of $1,000 to be of good behaviour for one year and to keep the peace. In deciding that that order did not pass the “prescribed by law” test, the Court did not approve the Steel technique of discovering certainty on the basis of prior findings made. Lord Scott of Foscote NPJ, writing for the Court, held that a prospective requirement to be of good behaviour at large failed the test and that the order would only be valid if it was made specific by its own terms: “Although the traditional form of bind-over order is an order requiring the person concerned to keep the peace and be of good behaviour, without any greater precision, we do not think that an order simply in that general form should any longer be regarded as satisfactory. Nor do we regard an implied limitation by reference to the facts that prompted the making of the order to be satisfactory. We think the principle of legal certainty requires that the order spell out with precision, in the same way as would be expected of an injunction, what it is that the person must not do. For example, in the present case, if a bind-over order was to be made against the appellant, it should have been an order requiring him to keep the peace and be of good behaviour by abstaining from assaulting or threatening to assault his brother or by doing anything to give rise to a reasonable apprehension that he intended to assault his brother, or in some such terms. A guide to an acceptable formulation of a bind-over order would, in our opinion, be whether an injunction in those terms would be acceptable.”[97] 96. Brooker v Police,[98] posed a different problem. The appellant believed that a certain police constable had unfairly harassed him by obtaining a search warrant against him. He went to her home and knocked on her door, waking her up after a night shift, and then protested in the street outside, playing his guitar and singing songs critical of the constable. He was charged with a statutory offence of disorderly behaviour under section 4(1)(a) of the Summary Offences Act which provided: “Every person is liable to a fine not exceeding $1,000 who ... In or within view of any public place, behaves in an offensive or disorderly manner.” 97. It was thus an offence defined in minimalist terms. It had only two elements: (i) behaving in an offensive or disorderly manner; (ii) while being in or within view of any public place. It lacked any contextual framework to provide guidance as to the kind of behaviour which would qualify as “offensive or disorderly” and the Supreme Court of New Zealand was primarily engaged in construing the section to provide such guidance. The majority held that “disorderly” had to involve disruption of public order. As Elias CJ put it: “To constitute disorderly behaviour under s 4(1)(a) there must be an objective tendency to disrupt public order, by behaviour or because of the effect of words used. Whether behaviour is disorderly is not to be assessed against the sensibilities of individuals to whom the behaviour is directed or who are present to see and hear it, but against its tendency to disrupt public order.”[99] 98. The “prescribed by law” test was therefore not directly relevant. It was only obliquely touched on when one of the reasons given by Elias CJ for adopting “[a] narrower interpretation of ‘disorderly behaviour’, anchored in disruption of public order” was that this would be “more consistent with the fundamental principle that criminal law must be predictable.”[100] 99. These three authorities therefore had to grapple with issues that do not arise in connection with AI section 11 which has the benefit of a clear contextual focus and is not concerned with disorderly behaviour at large. Those authorities do not advance the appellant’s case and the “prescribed by law” argument fails. 100. Before leaving this argument, I should mention the criticism of the Judge’s construction of AI section 11 referred to in Section D above. With respect, I think that the Judge fell into error by eliding the two disjunctive strands in the Instruction and treating the offence as involving proof of both disorderly behaviour and a failure to comply with the direction of an officer of the Council.[101] His Lordship also erred in so far as he considered a “remedial interpretation” warranted.[102] The respondent does not seek to uphold the Judge’s approach on this matter. As the error does not affect the analysis or outcome, I will say no more about it. H. The “blanket prohibition” argument 101. It will be recalled that Mr Pun SC’s argument is that: “by ... rendering all forms of display of any ‘sign, message or banner’ a criminal offence regardless of its purpose, nature, manner and its impact (if any) on the public order in the LegCo”, [AI section 12] is unjustifiably wide.”[103] H.1 The proper construction of AI section 12 102. Just as the “prescribed by law” argument requires AI section 11 to be construed to decide whether it is legally uncertain, the “blanket prohibition” argument requires AI section 12 to be construed to see if it does indeed lay down a blanket prohibition. 103. It is once again important not to ignore material words in the section and to construe it in the light of its context and purpose. AI section 12 does not simply make “all forms of display of any ‘sign, message or banner’ a criminal offence”. It only does this with regard to displays “in a press or public gallery”, indicating its intended scope. 104. In considering AI section 12’s context and purpose, the exercise of construction undertaken regarding AI section 11 is equally applicable. For the reasons given in Section G.1 above, the purpose of the relevant rules in the LCPPO and the Administrative Instructions is to create a secure and dignified environment in the LegCo complex conducive to LegCo properly carrying out its constitutional functions at its sittings without disruption or disturbance. AI section 12 lays down certain prohibitions as a condition of allowing members of the public to observe legislative proceedings within the Chamber. 105. Thus, on a purposive construction, although couched in wide terms, the words of AI section 12 taken in context, do not render “all forms of display of any ‘sign, message or banner’ a criminal offence regardless of its purpose, nature, manner and its impact (if any) on the public order in LegCo” as the appellant alleges. On the contrary, the Administrative Instruction is issued very much with such impact in mind. The prohibitions are aimed at displays which entail the risk of disorder in public galleries and which may disturb LegCo sittings and the rights of others observing the proceedings. 106. This is a construction that receives support from AI section 12(3) which provides: “An officer of the Council may refuse admission to a press or public gallery to any person displaying any sign, message or banner, or to any person displaying any sign or message on any item of clothing, or to any person who, in the opinion of an officer of the Council, may so display any sign, message or banner, may so display any sign or message on any item of clothing or may otherwise behave in a disorderly manner.” 107. The words I have italicised strongly indicate that the prohibition of signs, messages or banners, including signs or messages on items of clothing in a public or press gallery, is aimed at conduct which amounts to behaving in a disorderly manner, thus contextually limiting the scope of subsections (1) and (2). A person in a public gallery wearing a T-shirt which happens to bear an innocuous message unconnected with the legislature’s proceedings, not brandished intrusively, is not intended to be caught. 108. This construction is reinforced if one keeps in mind the terms of section 8(3) of the LCPPO, which authorized the issue of AI section 12. Section 8(3) specifies that the President is to exercise this power “for the purpose of maintaining the security of the precincts of the Chamber, [and] ensuring the proper behaviour and decorum of persons therein ...”.[104] Thus, if he were to issue an Administrative Instruction laying down a blanket prohibition going far beyond that stated purpose, it would amount to an ultra vires exercise of the power and the rule of construction expressed in the maxim ut res magis valeat quam pereat would apply. That maxim is explained in Bennion on Statutory Interpretation[105] as follows: “... It is a rule of law that the legislator intends the interpreter of an enactment to observe the maxim ut res magis valeat quam pereat (it is better for a thing to have effect than to be made void); so that he must construe the enactment in such a way as to implement, rather than defeat, the legislative purpose.” 109. The contextual and purposive construction limiting the scope of AI section 12 discussed above, avoids an ultra vires outcome and preserves the validity of the rule. I arrive at this conclusion applying ordinary common law principles of construction. If it had been necessary, I would not have hesitated to arrive at the same conclusion applying a remedial interpretation in accordance with the principles explained in HKSAR v Lam Kwong Wai.[106] 110. I therefore conclude that properly construed, AI section 12 does not lay down the alleged “blanket prohibition” and the appellant’s argument advanced on that rejected basis fails. H.2 Does AI section 12 create a valid restriction on the right to freedom of expression? 111. AI section 12 plainly catches the conduct of the appellant and her co-defendants. Their demonstration was noisy and caused a commotion in the public gallery largely caused by the display of the Swastika sign and the slogan on their T-shirts opposing the landfill extension. It led on both occasions to the session being interrupted and, on the second occasion, to the exclusion of the public from observing proceedings in the substituted conference room. The question is whether, in limiting the right to freedom of expression by prohibiting such demonstrations, AI section 12 imposes a constitutionally valid restriction.[107] 112. This involves first examining whether the section operates as a restriction which is necessary for accomplishing one of the purposes listed in BOR 16. In my view, it clearly does. If not prohibited, demonstrations and interjections by persons in public galleries would inevitably interfere with debates and other proceedings on the floor of LegCo. A demonstration on a controversial issue is likely to be met with a counter-demonstration and a confrontation between the two camps in a public gallery would pose an obvious risk of conflict and public disorder. It would also interfere with the rights of persons who simply wish to observe the debate or other proceedings. 113. Given such prospects, the AI section 12 restrictions can be justified as necessary for respect of the rights of others and for the protection of public order or ordre public. The applicability of the categories of respect for the rights of others and the protection of public order is self-evident. However, the protection of ordre public as an aim may call for some elaboration. 114. Ordre public is a broad and flexible concept which, as Li CJ explained in HKSAR v Ng Kung Siu,[108] and as the majority reiterated in Leung Kwok Hung v HKSAR,[109]can be imprecise and elusive. It is much wider than “public order” in terms of the maintenance of law and order and “includes what is necessary for the protection of the general welfare or for the interests of the collectivity as a whole”being “a function of time, place and circumstances.” Various formulations of the meaning of ordre public were considered by the Court. The concept has been described as “a basis for restricting some specified rights and freedoms in the interests of the adequate functioning of the public institutions necessary to the collectivity...” Another formulation, taken from the Siracusa Principles[110] is that it “... may be defined as the sum of rules which ensure the functioning of society or the set of fundamental principles on which society is founded.” In my view, a restriction on the right to freedom of expression to safeguard the proper functioning of the legislature comes within these formulations of the protection of ordre public. 115. The next question is whether the restrictions are rationally connected with accomplishing the aforesaid aim, the answer to which is obviously “Yes”. They directly operate to restrict intrusive or disruptive conduct in the legislative Chamber. 116. Does the restriction do no more than reasonably necessary for accomplishing that purpose? In my view, it clearly does pass this test. AI section 12 has a limited scope, applying only to persons who are in a press or public gallery. It targets intrusive behaviour to protect good order during a LegCo meeting. The appellant was not prohibited from exercising her freedom of expression in opposition to the proposed landfill extension in other venues, including in designated areas of the LegCo complex. She was free to canvass support from fellow residents and to lobby elected legislative councillors who were members of the subcommittee concerned. She was free to campaign for public support using social media and other forms of public communication. What she was prohibited from doing was confined to her mounting a disruptive demonstration in the public gallery during the subcommittee’s sessions. 117. A reasonable balance has plainly been struck between the benefit to society of enabling LegCo properly to carry out its constitutional functions on the one hand and the limited restriction on the guaranteed right of freedom of expression on the other. AI section 12 in my view is clearly a proportionate and valid restriction on the right. I. Conclusion 118. Although the respondent’s arguments that the appeal should fail in limine have not been successful, the appellant’s arguments challenging the legal certainty of AI section 11 and the proportionality of AI section 12 must be rejected. I would accordingly dismiss this appeal. Mr Justice Fok PJ: 119. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Chan NPJ: 120. I agree with the judgment of Mr Justice Ribeiro PJ. Lord Neuberger NPJ: 121. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma: 122. The Court unanimously dismisses the appeal. Mr Hectar Pun, SC, Mr Harrison Cheung, and Mr Anson Wong Yu Yat, instructed by Y. S. Lau & Partners, assigned by the Director of Legal Aid, for the Appellant Ms Anna YK Lai, SC, DDPP and Mr Andrew Li, SPP of the Department of Justice, for the Respondent [1] Cheung Mei-hung and Fong Yu-ching. They were convicted as co-defendants at the trial but are not parties to this appeal. [2] Although a live video feed was arranged for the public. [3] Cap 382. [4] Cap 382A. [5] Mr Chu Chung Keung, ESCC 3792/2014 (7 October 2015). The other defendants were also convicted. [6] HCMA 666/2015 (19 May 2016). Wong J’s judgment was delivered in Chinese and references in this judgment are to the English translation. [7] Ribeiro, Tang and Fok PJJ, FAMC 29/2016 (8 February 2017). Enactment of AI sections 11 and 12 was of course pursuant to LCPPO section 8(3) rather than section 20(b) but the error is of no consequence. [8] In this judgment, Basic Law Articles are cited with the prefix “BL” followed by the number of the Article. [9] See eg, Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 at §53; and Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §19. [10] Cap 383. Articles of the Bill of Rights are cited with the prefix “BOR” followed by the number of the Article. [11] Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §20. [12] BOR 17: “The right of peaceful assembly shall be recognized. No restrictions may be placed on the exercise of this right other than those imposed in conformity with the law and which are necessary in a democratic society in the interests of national security or public safety, public order (ordre public), the protection of public health or morals or the protection of the rights and freedoms of others.” [13] Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §12; and HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837 at §31. [14] Under BL 39(2), or, to the same effect, as a restriction “provided by law” under BOR 16(3): see Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §18. [15] HKSAR v Ng Kung Siu (1999) 2 HKCFAR 442 at 461; and Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §33. [16] Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §35; HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837 at §36; and Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372 at §49. [17] Appearing for the appellant with Harrison Cheung and Anson Wong Yu Yat. [18] Appellant’s Written Case §§43-45. Italics in the original. [19] Appellant’s Written Case §57. Italics in the original. [20] Appearing for the respondent with Mr Andrew Li. [21] Respondent's Written Case §9(a). [22] Respondent's Written Case §§17(a) and (b), 18-20. [23] Respondent's Written Case §§26-31. [24] Those paragraphs refer to the need to accord fundamental rights a generous interpretation and to accommodate annoying or offensive views, citing Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229. [25] Judgment §35. [26] [1991] 1 SCR 139. [27] Section 2(b): “Everyone has the following fundamental freedoms: ... freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication; ...” [28] Section 1: “The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.” [29] Ibid at 192. [30] Ibid at 229. [31] BL 8: “The laws previously in force in Hong Kong, that is, the common law, rules of equity, ordinances, subordinate legislation and customary law shall be maintained, except for any that contravene this Law, and subject to any amendment by the legislature of the Hong Kong Special Administrative Region.” [32] Committee for the Commonwealth of Canada v Canada [1991] 1 SCR 139 at 155. [33] Ibid at 230. [34] [1991] 1 SCR 139. [35] Ibid at 156 and 158. [36] Ibid at 235. [37] Ibid at 237. [38] Irwin Toy Ltd v Quebec (Attorney General) [1989] 1 SCR 927. [39] Committee for the Commonwealth of Canada v Canada [1991] 1 SCR 139 at 238-239. [40] Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §2. [41] Applying the proportionality analysis discussed in Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372. [42] Tabernacle v The Secretary of State for Defence [2009] EWCA Civ 23. [43] Ibid at §37. [44] [2011] 1 WLR 504. [45] Ibid at §37. Articles 10 and 11 referred to are Articles in the European Convention for the Protection of Human Rights and Fundamental Freedoms (“ECHR”) which deal respectively with the rights to freedom of expression, to freedom of peaceful assembly and to freedom of association with others in terms very similar to the rights guaranteed by the Basic Law and HKBORO. See also R (Gallastegui) v Westminster City Council [2013] 1 WLR 2377 per Lord Dyson MR at §26: ““I accept that the importance of the right to express views publicly and to assemble for the purpose of expressing and discussing those views can extend to the manner in which it is wished to express the views and the location where they wish to express them...” [46] Ibid at §42. [47] Tabernacle v The Secretary of State for Defence [2009] EWCA Civ 23. [48] Mayor of London v Hall [2011] 1 WLR 504. [49] The Mayor Commonalty and Citizens of London v Samede [2012] EWCA Civ 160. [50] [2012] EWCA Civ 160 at §39. See also R (Gallastegui) v Westminster City Council [2013] 1 WLR 2377, for a proportionality analysis. [51] [1991] 1 SCR 139. [52] Ibid at 198. Although her Honour generally favoured the universal applicability of freedom of expression subject to justification under s 1 of the Charter. [53] Ibid at 241. [54] See Hirst v UK (No 2) [2005] ECHR 681 at §§69-71 and Velyo Velev v Bulgaria, Application No 16032/2007 (27 May 2014) at §30. Similar views have been expressed at common law: see eg, Raymond v Honey [1983] 1 AC 1 at 10 and R (Daly) v Secretary of State for the Home Department [2001] 2 AC 532 at §5. The question was addressed in HKSAR v Wan Thomas [2016] 5 HKLRD 656, but as an application for leave to appeal is pending before this Court, I say nothing about that decision. [55] See Lau Cheong v HKSAR (2002) 5 HKCFAR 415; Ubamaka v Secretary for Security (2012) 15 HKCFAR 743. [56] Andrew Cheung J in Chan Kim Sum v Secretary for Justice [2009] 2 HKLRD 166. [57] Cf Golder v United Kingdom (1979-80) 1 EHRR 524. See Prison Rules, Cap 234A, r 52 for the rules relating to such visits currently in force. [58] See Prisons Ordinance (Cap 234) section 23 and Prison Rules, Part III, especially rules 226-229. [59] [1991] 1 SCR 139 at 228. [60] [2007] 3 NZLR 91 at §60. [61] Section 14: “Everyone has the right to freedom of expression, including the freedom to seek, receive, and impart information and opinions of any kind in any form.” [62] Per Lord Nicholls of Birkenhead in a very different context: In re Spectrum Plus Ltd (In Liquidation) [2005] 2 AC 680 at §41. [63] [2010] 3 HKLRD 371. [64] As the Chief Judge of the High Court then was. [65] [2010] 3 HKLRD 371 at §1. [66] Ibid at §25. [67] Ibid at §26. [68] Ibid at §37. [69] Ibid at §53. [70] (2003) 37 EHRR 38. [71] Ibid at §47. [72] Ibid at §40. [73] (2014) 17 HKCFAR 689 at §28. [74] Ibid at §32. [75] Section C above. [76] Under BL 39(2) and “provided by law” under BOR 16(3): Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §18. [77] (2007) 10 HKCFAR 386 at §61, citing Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381. [78] Set out in Section B above. [79] (2006) 9 HKCFAR 574 at §63. [80] (2013) 16 HKCFAR 837 at §67. [81] (2004) 220 CLR 1 at §12. [82] (2013) 16 HKCFAR 837 at §68. [83] In the citations of legislative provisions in this Section of the judgment, the italics are supplied. [84] Section B above. [85] There being no suggestion that AI section 12 suffers from uncertainty. [86] (1999) 28 EHRR 603. [87] (2003) 6 HKCFAR 624. [88] [2007] 3 NZLR 91. [89] Under section 4(1)(a) of the Summary Offences Act. [90] (1999) 28 EHRR 603 [91] Ibid at §76. [92] (1999) 30 EHRR 241. [93] Ibid at §§25, 27 and 35. [94] Ibid at §§38-41. [95] Ibid at §40. [96] (2003) 6 HKCFAR 624. [97] Ibid at §49. [98] [2007] 3 NZLR 91. [99] Ibid at §41. [100] Ibid at §38. [101] Judgment §§81-82, 88 and 94. [102] Judgment §§83-84. [103] Appellant’s Written Case §§43-45. Italics in the original. [104] As well as for “other administrative purposes”. [105] Rule 198, Lexis Nexis On-line edition. [106] (2006) 9 HKCFAR 574. [107] The appellant does not suggest that AI section 11 is disproportionate. [108] (1999) 2 HKCFAR 442 at 457-460. [109] (2005) 8 HKCFAR 229 at §§69-74. [110] The Siracusa Principles on the Limitation and Derogation Provisions in the ICCPR, issued by the American Association for the International Commission of Jurists (1984) §22. 
Hon Poon JA (giving the Judgment of the Court) : 1. By a judgment handed down on 13 October 2017, Andrew Chan J found Mr Wong Ho Ming (“Mr Wong”),[1] together with 8 other respondents, guilty of criminal contempt (“Judgment”). Mr Wong now appeals.[2] 2. The background facts leading to the proceedings below are largely not in dispute.[3] They may be summarized as follows. A. BACKGROUND FACTS A1. The Amended Injunction 3. As part of the so-called “Occupy Movement”, large crowds of people had since September 2014 occupied significant portions of Nathan Road between Argyle Street and Dundas Street in Mongkok, Kowloon, which prevented the general public, to various extent, from using the public roads so occupied. Also affected were those taxi operators and taxi drivers who had to rely on normal vehicular traffic at the occupied areas for their livelihood. 4. On 20 October 2014, Lai Hoi Ping (suing on his own behalf and on the behalf of all other members of Hong Kong Taxi Association) and Tam Chun Hung (suing on his own behalf and on the behalf of all other members of Taxi Drivers and Operators Association) (“the Plaintiffs”) commenced proceedings in HCA 2104/2014. They made an ex parte application on notice for an injunction against “persons occupying portions of Nathan Road near to and between Argyle Street and Dundas Street to prevent or obstruct normal vehicular traffic from passing and repassing the occupied areas” (“the 1st Defendant”). After hearing counsel, Poon J (as he then was) granted the injunction sought until the return date on 24 October 2014. 5. At the same time, some other plaintiffs commenced HCA 2086/2014 and HCA 2094/2014, which respectively concerned other areas in Mongkok and Admiralty occupied by large crowds of demonstrators. Poon J also, upon their applications, granted similar injunctions until the same return date. 6. On 24 October 2014, several named individuals applied and were allowed to be joined as additional defendants in all the three actions. All the plaintiffs applied for continuation of the ex parte injunctions and for directions for enforcement of the same. Their applications were opposed. Au J heard arguments on 24 and 27 October 2014 and then reserved judgment, while continuing the ex parte injunctions in the meantime. On 10 November 2014, the learned judge handed down judgment ordering all the ex parte injunctions to be continued (“Injunctions Judgment”). The terms of the orders were subsequently finalized after a directions hearing on 13 November 2014 and amended by a judgment dated 21 November 2014. 7. For HCA 2104/2014, the amended injunction order (“Amended Injunction Order”) provided : “ (1) The Defendants, and each of them, whether by themselves or agent or servant or howsoever, be restrained until trial or further order of the Court, from doing, any of the following acts, namely :- (a) Occupying portions of Nathan Road between Argyle Street and Dundas Street (“the Area”) to prevent or obstruct vehicular traffic from passing and repassing the Area (“Para 1(a)”); (b) Erecting, building or otherwise set up tents, canopies, barriers, barricades or other structures obstacles or obstruction, or doing any other act, to prevent or obstruct vehicular traffic from passing and repassing the Area; or (c) Obstructing or interfering with, or doing any other act which deters, the Plaintiffs through their agents properly authorised in writing from or in dismantling or removing barriers and other obstacles and obstruction in or from the Area (“Para 1(c)”) (“Para 1”). (2) The bailiff does take all reasonable and necessary steps to assist the Plaintiffs and its agents to effect the clearance and removal of the obstructions as provided in the Injunction Order (“Para 2”). (3) The bailiff be authorized and directed to request the assistance of the Police where necessary (“Para 3”). (4) Any police officer be authorized to arrest and remove any person who the police officer reasonably believes or suspects to be obstructing or interfering any bailiff in carrying out his or her duties in enforcing the terms of the Injunction Order, provided that the person to be arrested has been informed of the gist of the terms of the Injunction Order and this Order and that his action is likely to constitute a breach of the Injunction Order and this Order and obstruction of the administration of justice, and that he may be arrested if he does not desist (“Para 4”). (5) Without prejudice to the provisions of the Police Force Ordinance (Cap 232) (including S 51 and S 52 thereof), any person so arrested by the police shall be brought before the Court, under lawful or legal processes, as soon as possible for further directions (“Para 5”).” 8. The amended injunction orders granted in HCA 2086/2014 and HCA 2094/2014 were similar in terms, including Para 4, save and except, for obvious reasons, the difference in the areas covered in Para 1 of the injunction. 9. Some of the defendants in the three actions applied for leave to appeal against the injunction orders. Au J refused them all. They then applied to the Court of Appeal. Their applications were likewise all dismissed by the Court of Appeal (Cheung CJHC and Lam VP).[4] In the judgment refusing the leave application by the 2nd Defendant in HCA 2086/2014 dated 21 November 2014 (“CA Judgment”), Lam VP clarified the requirements contained in the proviso of Para 4.[5] As explained by his Lordship, “(a) informing the person concerned as to the gist of the injunction; (b) warning him that his action is likely to constitute a breach of the injunction; and (c) warning him also the possibility of his being arrested if he does not desist … are steps to be taken by the bailiff in the execution of his duty under the order and the person concerned will be given a last opportunity to comply with the order voluntarily… [These] are steps a bailiff would have to take before a case of obstruction or interference with the carrying out of his duties can be made out. Whether a police officer would like to repeat any of these steps before he exercises his power of arrest is entirely a matter for him to decide.” 10. All the legal proceedings mentioned above including all the judgments handed down and orders made were widely covered by the local media, be it paper, electronic or internet. A2. Service of the Amended Injunction 11. On 24 November 2014, service of the Amended Injunction Order was effected on the 1st Defendant in the manner as provided for, namely, by way of advertisement in Sing Tao Daily and the Standard; and placing the same in a clear plastic envelope and attached to 11 different prominent locations at the portions of Nathan Road between Argyle Street and Dundas Street. 12. Service of the Amended Injunction Order was also widely reported in the local media. A3. Execution of the Amended Injunction 13. Prior to 26 November 2014, it was already widely reported in the local media that the Plaintiffs and the bailiff would take action to execute the Amended Injunction Order on that day. As a matter of fact, the bailiff had on 25 November 2014 executed the injunction order granted by Au J in HCA 2086/2014 in respect of Argyle Street between the junction of Tung Choi Street and Portland Street. Such enforcement action was also widely covered in the local media. 14. On 26 November 2014, the bailiff executed the Amended Injunction Order in the manner described below. A3.1 Preparatory steps 15. At about 8:00 am, Police Sergeant Li Kwong, (“Sergeant Li”) together with six other police constables, set up six loudspeakers and two metal stages at Nathan Road junction with Argyle Street (northbound and southbound carriageways), and two loudspeakers at Nathan Road junction with Dundas Street (northbound and southbound carriageways) (“the PA System”). The PA System worked properly throughout the entire operation on 26 November 2014. 16. Around 8:05 am, the Plaintiffs’ legal representatives and lawful agents, certain bailiff officers and police officers (“the Execution Team”) gathered at Argyle Street, preparing for the execution of the Amended Injunction Order. 17. Five rounds of announcements were then made at four different locations before actual steps were taken to execute the Amended Injunction Order. The manner in which the announcements were made was similar. A3.2 Announcements (1) First announcements (Argyle Street junction with Nathan Road) 18. At about 8:40 am, Assistant Chief Bailiff Chiu Shuk Man (“AC Bailiff Chiu) made an announcement at the junction of Argyle Street and Nathan Road (near HSBC at No 675, Nathan Road). Through the loudhailer, AC Bailiff Chiu explained to the crowd at the scene that the bailiff officers were there to enforce the Amended Injunction Order (“the Bailiff’s First Announcement”): “ 各位在場人士注意，我哋係司法機構執達事務組嘅執達主任。 根據高等法院案件編號2014年第2104號於2014年11月10日發出並於2014年11月21日修訂嘅禁制令，執達主任會協助原告人嘅合法代理人，清除及移走阻塞車輛在介乎亞皆老街與登打士街之間的彌敦道各部份正常交通運行嘅障礙物。請在上述範圍內嘅人士，立即收拾你哋嘅物品離開。 任何人士，如果作出任何阻礙執達主任執行職務嘅行為，將有機會觸犯藐視法庭罪，執達主任會要求警方協助作出拘捕行動。 以下時間會由原告人代表律師，宣讀禁制令嘅內容。” The Plaintiffs’ legal representatives followed suit by making an announcement of the contents of the Amended Injunction Order through the loudhailer in the following terms (“the Plaintiffs’ Announcement”) : “ 根據區慶祥法官於2014 年11 月10 日頒下並於2014 年11 月21 日修訂嘅法庭命令，要點為任何人均不得阻礙原告人嘅合法代理人清除及移走阻塞在介乎亞皆老街與登打士街之間的彌敦道各部份的正常交通運行嘅障礙物。 該法庭命令亦禁止任何人在上述範圍豎立或放置任何障礙物以堵塞該些入口及通道。 任何人不服從該法庭命令，有可能構成藐視法庭。 同時，法庭亦指示執達主任採取所有合理和所需嘅步驟協助原告人嘅合法代理人清除及移走該些障礙物。 在有需要時，執達主任可要求警方協助。 而當警方合理地認為或相信任何人正妨礙或阻礙執達主任履行本法庭命令，只要該等人士已獲悉 (1) 本法庭命令嘅要點； (2) 佢咁嘅行動可能構成違反法庭命令及妨礙司法工作； (3) 而如果佢仍然不停止咁嘅行動有可能會俾警方拘捕； 係咁嘅情況下警方有權拘捕或移走任何有關人士，並將被拘捕人士在切實可行範圍內儘快送到法庭上，作進一步嘅指示。” 19. Shortly before 8:44 am, AC Bailiff Chiu made another announcement to the effect that the lawful agents of the Plaintiffs would execute the Amended Injunction Order and clear the obstacles at the Area within 30 minutes; and that the crowd was advised to obey the Amended Injunction Order and to leave the Area immediately (“the Bailiff’s Second Announcement”). (2) Second announcements (Nelson Street junction with Nathan Road) 20. At about 8:45 am, Acting Assistant Chief Bailiff Yu Tak Shun (“AAC Bailiff Yu”) repeated the Bailiff’s First Announcement through a loudhailer near the junction of Nelson Street and Nathan Road outside Watson’s at No. 637 Nathan Road. The Plaintiffs’ legal representatives then repeated the Plaintiffs’ Announcement at about the same location. Shortly before 8:50 am, AAC Bailiff Yu repeated the Bailiff’s Second Announcement to the crowd. 21. After the above announcements, the Execution Team marched down Nathan Road towards Dundas Street, during which, a crowd of people (including Mr Wong Chi Fung and Mr Wong) kept asking about the terms of the Amended Injunction Order and the details of the enforcement action. In response, Chief Bailiff Kwan Sek Nam Sunny (“Chief Bailiff Kwan”) informed them that the bailiff officers would soon announce the details and invited them to listen carefully to such announcements. (3) Third announcements (Shantung Street junction with Nathan Road) 22. At about 8:51 am, AC Bailiff Chiu repeated the Bailiff’s First Announcement through a loudhailer at the junction of Shantung Street and Nathan Road. The Plaintiffs’ legal representatives then repeated the Plaintiffs’ Announcement at about the same location, followed by AC Bailiff Chiu’s repetition of the Bailiff’s Second Announcement shortly before 8:56 am. (4) Fourth announcements (Dundas Street junction with Nathan Road) 23. At about 9:05 am, AAC Bailiff Yu repeated the Bailiff’s First Announcement through a loudhailer at the junction of Dundas Street and Nathan Road. Following that, the Plaintiffs’ legal representatives also repeated the Plaintiffs’ Announcement at about the same location. AAC Bailiff Yu then repeated the Bailiff’s Second Announcement shortly before 9:10 am. (5) Final announcements (Argyle Street junction with Nathan Road) 24. At about 9:49 am, the Execution Team was back at the junction of Argyle Street and Nathan Road near the barricades set up there. At the time, ACC Bailiff Yu and AC Bailiff Chiu respectively announced, through the loudhailer, in Chinese and English, that the lawful agents of the Plaintiffs would immediately execute the Amended Injunction Order and clear the obstacles in the Area. Following the announcements, at about 9:53 am, AC Bailiff Chiu requested the Plaintiffs’ legal representatives and lawful agents to clear the barricades (which consisted of Mills barriers, planks and other objects) in front of them. A3.3 Clearance of the barricades 25. During the clearance of the barricades near HSBC (at the junction of Argyle Street and Nathan Road), Mr Wong Chi Fung and Mr Wong kept asking through loudhailer about the identity of the Plaintiffs’ lawful agents, and quarreled with the Plaintiffs’ lawful agents, who wore red-and-white T-shirts and red caps for easy identification on that date. The bailiff officers repeatedly requested both parties to keep calm. 26. At about 9:59 am, part of the barricades at the junction of Argyle Street and Nathan Road had already been cleared by the Plaintiffs’ lawful agents. At this juncture, there was a commotion in the crowd which cried out loudly and pushed against the Execution Team. Assistant Chief Bailiff Chim Fung Ling (“AC Bailiff Chim”) immediately advised the crowd (through the loudhailer) to keep calm and not to obstruct or impede the bailiff officers from carrying out their duties. It was the case of the Secretary for Justice that during the confusion, AC Bailiff Chim was pushed; AC Bailiff Chiu’s sunglasses fell off to the ground; and AAC Bailiff Yu’s right shank was hit by a Mills barrier. A3.4 Police assistance and action 27. AC Bailiff Chim immediately made the following warning : “ 根據區慶祥法官於2014 年11 月10 日頒下並於2014 年11 月21 日修訂嘅法庭命令，要點為任何人均不得阻礙原告人嘅合法代理人清除及移走阻塞車輛在介乎亞皆老街與登打士街之間的彌敦道各部份正常交通運行嘅障礙物。 你嘅行為係會阻礙執達主任執行職務。如再繼續此行為，有可能觸犯藐視法庭罪，請你立刻停止。否則，我哋會要求警方協助去拘捕你。” 28. Despite such warning, the crowd did not desist. Chief Bailiff Choi Tak Ming (“Chief Bailiff Choi”) hence formally requested the assistance from the police at about 10:00 am pursuant to Para 3 of the Amended Injunction Order. Acting on such request, the police started to take charge of controlling the crowd and clearing the barricades. 29. At about 10:05 am, Chief Inspector Lam Chi Yuen (“CIP Lam”) issued a verbal warning through the PA System at Nathan Road near Argyle Street to the crowd in Chinese and English in the following terms (“First Police Warning”) : “ 前面人群注意。這是警方發出的警告。高等法院的執達主任，正在此根據高等法院的命令，清除及移走路上的障礙物。高等法院命令的副本，已在附近張貼，並已在報章刊登。執達主任/代表原告人的律師亦已在這裡現場解釋法院命令的要點。任何人阻礙或干擾執達主任及協助執達主任的人員執行高等法院的命令，可能違反法院命令和阻礙司法工作。這些行為可能構成刑事藐視法庭和干犯香港法例第228 章《簡易程序治罪條例》第23 條的抗拒或阻礙公職人員罪，干犯此兩項的任何一項，都可被判監禁及罰款，此外，這些行為亦可能構成其他的刑事罪行。你們必須立即停止阻礙或干擾執達主任及協助執達主任的人員，否則你們可能會被拘捕及檢控。 This is a Police warning to the crowd in front of me. The bailiffs of the High Court are now clearing and removing the obstructions here in accordance with an order / orders of the High Court. Copies of the order(s) is / are displayed nearby and has/have been published in the newspapers. Bailiffs / solicitors acting for the plaintiff(s) have explained the gist of the court order(s) here. Any person who obstructs or interferes with any bailiff or person assisting him in enforcing the High Court order(s) may be in breach of court order(s) and obstructing administration of justice. Such conduct may constitute criminal contempt of court and an offence of Resisting or Obstructing a public officer under section 23 of the Summary Offences Ordinance (Cap. 228). Either of these may result in imprisonment and a fine. Furthermore, such conduct may also constitute other criminal offences. You must immediately stop your obstruction and interference with the bailiffs and those assisting them, otherwise you may be arrested and prosecuted.” 30. At about 10:12 am, CIP Lam made an announcement to the press at the scene requesting them to follow the instruction of the Police Media Liaison Team to retreat from the area of police operation. 31. At about 10:15 am, as the crowd ignored the verbal warning, CIP Lam gave a final verbal warning, again through the PA System, to the crowd at the same location in Chinese and English in the following terms (“Final Police Warning”) : “ 前面人群注意。這是警方發出的最後警告。你們必須立即停止阻礙或干擾高等法院執達主任及協助他們的人員清除及移走路上的障礙物。如你們不立即停止，警方會按涉嫌刑事藐視法庭，抗拒或阻礙公職人員罪，和你們已涉嫌干犯的其他罪行拘捕你們，如有必要會使用最低的武力執行拘捕，而不會再作任何警告。 This is the final warning made by the Police to the crowd in front of me. You must immediately stop obstructing or interfering with bailiffs of the High Court or those assisting them in clearing and removing the obstructions on the road. If you do not stop immediately, police officers will arrest you for suspected criminal contempt of court, resisting or obstructing a public officer, and other offences which you are suspected to have committed, and police officers will use minimum force for making arrests if necessary. No further warning of arrest will be issued before arrest action.” 32. Despite the repeated warnings, the crowd still refused to leave the Area. CIP Lam therefore instructed the police officers at the scene to arrest those people who caused obstruction in the Area and refused to leave. Upon CIP Lam’s instruction, the police check line (which was formed by police officers) progressed from the junction of Argyle Street and Nathan Road along Nathan Road towards the Tsim Sha Tsui direction. 33. Between 10:25 am and 12:00 pm, Chief Inspector Sean Lin (“CIP Lin”) and Senior Inspector Cheung Kin Pan (“SI Cheung”) took turn to make public announcements through the PA System and a microphone along Nathan Road from Argyle Street towards Dundas Street. Such public announcements included advice to the crowd, onlookers and press at scene to take care of their own safety; and that the police might raise the level of appropriate force, including the use of water jet pack of pepper-spray-based solution, if deemed necessary. CIP Lin and SI Cheung also advised the crowd to leave the Area via the Tsim Sha Tsui direction in an orderly manner and not to obstruct police officers in the execution of their duties. 34. At about 11:07 am, the police check line reached the junction of Shantung Street and Nathan Road. At the time, the crowd gathered again at the junction of Shantung Street and Nathan Road. Whilst the police officers at the scene continued to advise the crowd to leave the Area via the Tsim Sha Tsui direction, CIP Lam issued the Final Police Warning once again. Afterwards, police officers started arresting those people causing obstruction and refusing to leave. 35. The enforcement action of the police officers continued up to 12:40 pm when all the barricades on the portions of Nathan Road from Argyle Street to Dundas Street were cleared. 36. During the operation, several police video teams were deployed to capture the incident from different locations. A4. Arrest of Mr Wong 37. At about 10:18 am, after the first barricade near the junction of Argyle Street was removed, Mr Wong was standing in front of the police check line. Upon instructions, Detective Police Constable 4844 Chan Wang-pok intercepted Mr Wong and arrested him for suspected contempt of court and obstructing a public officer contrary to section 23 of the Summary Offences Ordinance.[6] 38. After arrest, Mr Wong was brought back to Kwai Chung Police Station for further enquiry. B. CONTEMPT PROCEEDINGS BELOW 39. Upon the application by the Secretary for Justice, Chow J on 30 July 2015 granted leave to commence the contempt proceedings against Wong and other respondents. On the Secretary’s case, the particulars of Wong’s acts which constituted criminal contempt were :[7] “45. As revealed from evidence, [Mr Wong] was found present in the Area for an extended period of time during the execution of [the Amended Injunction] and played a very active role in confronting/questioning the bailiffs :- 40. The Secretary pleaded :[8] “ 47. In the premises, [Mr Wong] : (1) was in the Area for an extended period of time and must have heard the repeated requests and warnings made by the Bailiff / Police Officers prior to the arrest; (2) had ample opportunities to leave the Area prior to the arrest but decided against doing so; and (3) the actions taken by [Mr Wong] constitute a breach of the Amended Injunction Order. By reason of the aforesaid, [Mr Wong] must have been aware of this and the breach was intentional. 48. The deliberate acts of [Mr Wong] identified in paragraph 45 above, coupled with his determination to continue flouting the Amended Injunction Order constituted interference and/or impediment to the due administration of justice by obstructing or resisting the Police and/or Bailiff Officers in executing the Amended Injunction Order, and therefore amounted to criminal contempt of court.” 41. The Secretary sought a committal order against Mr Wong that he be committed to prison and/or fined for criminal contempt of the court by interfering with and/or impeding the due execution of the Amended Injunction Order by refusing to leave the obstructed area covered by the Amended Injunction Order despite warnings. 42. Mr Wong opposed the Secretary’s application for committal. The hearing of his case, as well as other opposing respondents, before the Judge took 19 days to finish. A total of 29 witnesses gave oral evidence for the Secretary. The respondent in HCMP 796/2015, Ms Mak Ying Sheung, gave oral evidence and called 2 witnesses. None of the other respondents, including Mr Wong, elected to give evidence or call any witness. 43. Mr Lawrence Lok, SC, then representing Mr Wong and some other respondents, submitted that in order to succeed, the Secretary had to establish the necessary mens rea, namely that each of the respondents had the intention to interfere with the administration of justice. However, the Judge at [45]-[54] of the Judgment referred to Secretary for Justice v Ng Wai Bing [2012] 1 HKLRD 245, where Tang VP at [23] endorsed the judgment of Wright J in [2011] 5 HKLRD 620 at [53]-[58] : “ [53] … proof of a specific intent to interfere with the administration of justice would not be a prerequisite for liability for criminal contempt. … [57] In the circumstances I am satisfied that the applicant need only to show basic intent, that is that the respondents intended to perform the acts which constituted the actus reus of the contempt. The nature of the act [58] The act which is said to constitute contempt of court is conduct calculated to prejudice or interfere with the due administration of justice, is inherently likely to do so.” The Judge considered that the elements of criminal contempt had been settled by the Court of Appeal’s judgment in Ng Wai Bing. He therefore rejected Mr Lok’s submission on the requirement to prove specific intent. 44. Applying Ng Wai Bing, the Judge went on to say : “ 55. Translating that into the present case, the applicant in order to succeed, needed only to prove that each of the respondents with the intention to remain, did remain in the Area. Further, their conduct in remaining was inherently likely to prejudice or interfere with the due administration of justice given the circumstances that the Bailiffs and the Plaintiffs’ agents were about to execute or executing the Amended Injunction Order in effecting the clearance and removal of obstacles. Whether each of the respondents did, in fact, obstruct or interfere with those who were responsible for clearing up the obstacles was not part of any legal requirement. Of course, if the evidence shows that the respondent did in fact obstruct or interfere with the clearing up of the obstacles, a fortiori, it would amount to criminal contempt. 56. It was further argued that since the respondents were doing nothing more than remaining at the Area, ie given their mere presence, they were at most only liable for disobedience of Term (a) of the Amended Injunction Order, hence civil contempt. In any given date prior to the 26th, I accept that that may be the case. However, this argument in my view ignores (i) the factual matrix against which the injunction was granted in the first place and that people had continued to occupy the Area; (ii) that at each stage of the injunction proceedings, the subject matter of the hearings and the result had been widely reported and advertised in the Local Media; (iii) that the clearance of obstacles on the previous day at Argyle Street had received extensive media coverage; (iv) that the imminent operation on the 26 November 2014 was widely reported likewise; and (v) that the warnings given by the Bailiffs and the Plaintiffs’ solicitors, in particular the demand made by the Bailiffs to the crowd in the Area to immediately pack their belongings and leave. The Bailiff’s demand in Chinese was : “請在上述範圍內嘅人士，立即收拾你哋嘅物品離開。”. 57. Whether each of the respondents intentionally chose to remain and did remain at the Area, again adopting the words of Wright J, would be a matter of fact and degree in every instance, whether or not the alleged contempt was made out.” 45. After disposing of further legal issues which do not concern this appeal, the Judge then discussed the evidence against each of the respondents. Coming to Mr Wong at last, the Judge said : “ 116. Mr Wong was first seen in the footage at around 8:44 am. 117. Video footage showed that when the Bailiffs and the Plaintiffs’ solicitors were moving along Nathan Road and making announcements at different locations, Mr Wong had repeatedly asked questions on (i) whether people present in the Area amounted to obstacles; (ii) whether the Plaintiffs’ agents had been properly authorised; (iii) how to ensure the content of the Amended Injunction Order being communicated to people present at the Area; and (iv) demanding the Bailiffs to use a bigger loudhailer. His challenges to the Bailiffs and the Plaintiffs’ solicitors were broadcast to the crowd. Nevertheless, the ways and manners adopted by Mr Wong were, on any objective viewing, desired to a certain extent, ridicule the Plaintiffs’ solicitors as well as to incite the crowd to continue their defiance. 118. The terms of the Amended Injunction Order were clear on their face. There were extensive legal arguments before Au J and the Court of Appeal. The proper way to clarify any purported uncertainty on the meaning and scope of the injunction should be to apply to the court to seek a variation or clarification. The same should also be applicable to the issue of authority of the Plaintiffs’ agents. Given the impracticality, it was therefore not surprising to see that nowhere in the terms of the Amended Injunction Order was there a requirement to prove authority to the protestors. There were hundreds if not thousands of them. 119. After the fifth announcement made by the Bailiffs and the Plaintiffs’ solicitors, the Plaintiffs’ agents began to clear the barricades. Mr Wong continued to challenge the Plaintiffs’ agents’ identity and authority and demanded the sight of written authorization. His presence and challenges undoubtedly created enormous burden on the Bailiffs and Plaintiffs’ agents. The emotions on both sides were high. Accusations and counter-accusations had been exchanged and given the chaos at the time, there was no way the Bailiffs and the Plaintiffs’ agents on their own could have carried out the clearance operation. The only resort was for them to seek the police’s assistance as contemplated. 120. At 10:02 am, the Bailiffs sought the assistance of the police. At 10:03 am [CIP Lam] gave [the First Police Warning]. Mr Wong decided to remain in the Area. When the police moved in effecting his arrest, Mr Wong stood in front of the barricades. Video footage showed that Mr Wong seemingly invited the police to arrest him and even signalled to the police that he had no weapon on his body. The matters and observations mentioned from paragraphs 77 to 79[9] above are likewise applicable in Mr Wong’s case. His presence and challenges to the Bailiffs and the Plaintiffs’ agents were conducts not only inherently likely to but did factually cause prejudice to or interference with the clearance operation on that day hence the due administration of justice. Having considered all the evidence, I am satisfied that Mr Wong is guilty of criminal contempt.” C. GROUNDS OF APPEAL 46. Mr Martin Lee, SC (with Mr Hecter Pun, SC, Mr Earl Deng and Mr Anson Wong Yu Yat) raised a total of 11 grounds of appeal for Mr Wong. In his written submissions, Mr Lee re-organized them into 5 grounds. In gist : (1) Ground 1[10] complains that the Judge erred in law in holding that proof of specific intent to interfere with the administration of justice is not a prerequisite for liability for criminal contempt in the circumstances of this case. (2) Ground 2[11] is alternative to Ground 1. It complains that, if Ground 1 is not accepted, the Judge erred in law in holding that the Secretary in order to succeed, needed only to prove that Mr Wong, with the intention to remain, did remain in the Area, in that he should have held that in order to succeed, the Secretary had to prove basic intent, that is, that Mr Wong intended to perform the acts which constituted the actus reus of the contempt. (3) Ground 3[12] complains that the Judge made a host of errors in his findings of fact mainly relating to the events before the Final Police Warning. The thrust of the complaint is that the Judge had failed to have proper regard or at all to certain uncontroversial and undisputed evidence which was favourable to Mr Wong, and had further erred in law in convicting Mr Wong of criminal contempt when the evidence taken as a whole did not and could not support such conviction. (4) Ground 4 complains that the Judge erred in law when he used the First Police Warning given at about 10:03 am as the reference point in determining whether Mr Wong should have left the Area, in that he should but failed to decide whether Mr Wong had the intention to remain in the Area after the Police had finished giving the Final Police Warning at about 10:15 am; in which case he should have held that there was at least a reasonable doubt as to whether Mr Wong had the intention to remain in the Area at the material time because (according to the video evidence) Mr Wong and those around him were attempting to leave but were blocked by the police from leaving the Area at the time when the police was still giving the Final Police Warning. (5) Ground 5 complains that the Judge erred in holding that Mr Wong had, by remaining in the Area, intended to and did obstruct or impede the bailiff, as opposed to the police, from effecting clearance and removal of the obstructions as provided in the Amended Injunction Order. 47. As presently formulated, Grounds 1 and 2 concern the requirement of mens rea; Grounds 3 to 4, actus reus. Ground 5 questions if Mr Wong intended to and did obstruct the bailiff. It thus concerns both mens rea and actus reus. We will discuss each of the Grounds in turn below. D. GROUND 1 – MENS REA FOR CRIMINAL CONTEMPT D1. Overview of the parties’ submissions 48. In contending that the Judge erred in holding that specific intent to interfere with the administration of justice is not a prerequisite for liability of criminal contempt, Mr Lee took three main points. 49. First, Mr Lee argued that the starting point is that specific intent is required at common law. He cited Dallas v United Kingdom (2016) 63 EHRR 13, at [72]; Arlidge, Eady & Smith on Contempt, 5th Edition, 2017, §§3-273 to 3-275 and §3-45 in support. It would appear that Mr Lee had slightly revised his position in the course of his oral submissions in response to questions from the Court. He accepted that other than contempt in the face of the court or contempt akin to such form of contempt, which requires only basic intent, specific intent to interfere with the administration of justice is required for all other forms of criminal contempt. 50. Second, Mr Lee submitted that Ng Wai Bing is distinguishable in that it concerned covert recordings of conversations in a witness room during trial, which is treated as a form of “contempt in the face of the court” – a special category of criminal contempt which does not cover the present case. It is therefore not an authority that “proof of a specific intent to interfere with the administration of justice” is not a prerequisite for liability in all forms of criminal contempt. Indeed, the circumstances in which criminal contempts of court may arise are too varied for one mens rea to be applicable to all forms of contempt. Thus, instead of applying Ng Wai Bing, the Judge should have held that because the present form of criminal contempt is not the same as, but much more serious than, non-compliance with an injunction (as in civil contempt of court) or the summary offence under section 23 of the Summary Offences Ordinance, the Secretary had to prove specific intent in order to succeed against Mr Wong. 51. Third, Mr Lee submitted that in line with the common law position, Au J in the Injunctions Judgment at [123] held that “an intention and act to interfere with or impede the due administration of justice is a criminal contempt : Halsbury’s Laws of England, Vol 22, [5]”. The Judge ought to have followed and applied Au J’s approach to mens rea in the circumstances of the present case. 52. In response, Mr Jin Pao (with Mr Derek Chan[13]) submitted the assertion that the starting point is that specific intent is required under common law incorrectly conflates the common law of Hong Kong with that of England. The Hong Kong common law position is clearly stated in Ng Wai Bing, which is binding on the Judge as well as this Court. In Ng Wai Bing, both Wright J and the Court of Appeal followed the pre-1981 English cases which established that a specific intent to interfere with the proper administration of justice is not an essential ingredient of criminal contempt : R v Odhams Press Ltd, ex p Attorney-General [1957] 1 QB 73; Attorney-General v Butterworth [1963] 1 QB 696. As rightly observed by Wright J in Ng Wai Bing (CFI), at [45], the post-1981 English cases and views expressed by commentators after the Contempt of Court Act 1981 was introduced (“the 1981 Act”) on mens rea need to be approached with caution. Further, that no specific intent is required is also the common law positon in New Zealand and Australia : see Solicitor-General v Radio New Zealand Ltd [1994] 1 NZLR 48; Hinch v Attorney-General (Vic) (1987) 164 CLR 15 and R v Lupco Slaveski [2015] VSC 400. In the course of his oral submissions, Mr Pao clarified that for the purpose of this appeal, he was not advocating a broader proposition that for all forms of criminal contempt, only basic intent is required. He was limiting his case to the particular form of contempt arisen in this appeal which, he contended, only required basic intent. Finally, Mr Pao submitted that the reliance placed on Au J’s observation in the Injunctions Judgment, [123], is misplaced. 53. Regarding Wright J’s observation in Ng Wai Bing (CFI) that the post-1981 English authorities need to be approached with caution, Mr Lee submitted that what the 1981 Act did is to limit the strict liability rule, as defined by the Act,[14] to publication which creates a substantial risk that the course of justice in the proceedings in question will be seriously impeded or prejudiced. Mr Lee reasoned that section 6(c) of the 1981 Act, which provides that nothing in the foregoing provisions of the 1981 Act restricts liability for contempt of court in respect of conduct intended to impede or prejudice the administration of justice, makes it clear that the UK Parliament intended to retain the application of the common law of contempt generally. Thus its enactment only affects cases to which the strict liability rule applies and, apart from those cases, the common law requires proof of an intent to impede or prejudice the administration of justice : UK Archbold 2018, §28-35. D2. Discussion D2.1 Source of the common law 54. In Hong Kong, apart from some few forms of contempt in the face of the court which have been codified and are punishable by the appropriate courts, which are immaterial for present purposes, contempt has always been a common law creature. As such a creature, the law of contempt has been and continues to be developed and adapted to meet changing challenges to due administration of justice, an important facet of supremacy of the law. 55. Before 1997, the Hong Kong law on contempt derived its source from the comparable English common law. Mr Lee submitted that it remains the case after 1997. In support, he cited Hong Kong Civil Procedure 2018, Vol 1, §52/1/1 : “ The court’s jurisdiction in contempt is its inherent jurisdiction equivalent to the jurisdiction of the Queen’s Bench Division in England (High Court Ordinance (Cap.4) s.12(3)) …The substantive law of contempt is still the common law…” As an example, Mr Lee referred to the Injunctions Judgment where Au J (with the assistance of eminent senior counsel) at [123] only cited English authorities to explain the law of contempt as applicable to Hong Kong. With respect, Mr Lee’s submission on the source of common law on criminal contempt which Hong Kong courts may derive after 1997 is far too restrictive. And the Injunctions Judgment as an example is not conclusive on the matter. 56. After 1997, the source of common law available to Hong Kong is no longer limited to English common law. For article 84 of the Basic Law expressly provides that Hong Kong courts may refer to precedents of other common law jurisdictions, without limiting to any particular jurisdictions. Section 12(3) of the High Court Ordinance must now be read subject to article 84. Thus after 1997, the available source of Hong Kong common law is no longer confined to English common law only. It includes other common law jurisdictions, such as Australia and New Zealand. The benefits of tapping the jurisprudence of the common law world without limiting to any particular jurisdiction are self-evident. The primary imperative is to enable our courts, whenever necessary, to draw on the jurisprudential riches and experiences of the common law world, with a view to developing the Hong Kong common law pertaining to a given subject according to the local legal landscape and framework that best suits the needs of the local circumstances. When we consult other common law jurisdictions, we are of course keenly aware of the difference, if any, in terms of legal framework and societal circumstances, that may materially impact on how the subject should be approached or considered. When such differences exist, we will approach the overseas cases with caution. After giving due regard to the position in other common law jurisdictions, if necessary, it is ultimately a matter for our courts as to how we should develop our common law. 57. In light of the change introduced by article 84 of the Basic Law, the commentary in Hong Kong Civil Procedure 2018, Vol 1, §52/1/1 must now be read with circumspect. It cannot bear the meaning as contended by Mr Lee. Turning to the Injunctions Judgment, it is but one of the cases where the law of contempt was mentioned or discussed. The fact that Au J had only cited English authorities does not even begin to suggest that Hong Kong has since 1997 only followed English common law on contempt without regard to the comparable common law in other jurisdictions. A ready example in which Hong Kong courts did consult common law in jurisdictions other than England for contempt is of course Ng Wai Bing. Another example is Secretary for Justice v Oriental Press Group Ltd [1998] 2 HKLRD 123 where Chan CJHC and Keith J (as their Lordships then were) sitting as a divisional court applied New Zealand cases on contempt.[15] At the level of the Court of Appeal, Australian authorities on contempt were recently referred to in Secretary for Justice v Cheung Kai Yin [2016] 4 HKLRD 367. 58. For present purposes, we will consider not only the relevant English authorities but also those of Australia and New Zealand in our discussion below. As said, if the foreign legal regime is different from ours (eg the 1981 Act in England) and the difference has a material impact on the requirement of mens rea, we will read the authorities cautiously with that in mind. To the English authorities we first turn. D2.2 Mens rea under the English common law 59. In England, mens rea in the law of criminal contempt had been described as “something of a minefield” : Attorney-General v Newspaper Publishing Plc (CA) [1988] 1 Ch 333, per Sir John Donaldson MR at p 373H. His Lordship explained : “ The reason is that it is wholly the creature of the common law and has developed on a case by case basis, as no doubt it will continue to do. [The 1981 Act] did not seek to systematize the approach of the courts.” 60. As seen already, there is a line of pre-1981 authorities which established that specific intent to interfere with the administration of justice was not required for criminal contempt. R v Odhams Press Ltd, ex p Attorney-General and Attorney-General v Butterworth are often referred to as the leading authorities. However, it would appear that after 1981 the position has become less clear. To get a flavor of the uncertainty surrounding mens rea for criminal contempt, one only needs to refer to the discussions in Arlidge, Eady & Smith on Contempt at §5-120 onwards for publication contempt falling outside “the strict liability rule”; and §11-23 onwards for non-publication contempt. 61. The uncertainty remains despite some judicial observations which suggest that for criminal contempt an intent to impede or prejudice the administration of justice is required. For example : (1) In Attorney-General v Newspaper Publishing Plc (CA) (1988), a case involving a publication contempt, Sir John Donaldson MR said at p 374 G-H (and also p 383B) that what was contemplated by section 6(c) of the 1981 Act, and what was saved, was the power of the court to commit for contempt where the conduct complained of is specifically intended to impede or prejudice the administration of justice. Such an intent need not be expressly avowed or admitted, but can be inferred from all the circumstances, including foreseeability. However, in Connolly v Dale [1996] QB 120, a case involving a non-publication contempt, Balcombe LJ at pp 125H-126A only assumed, without deciding, that it was necessary to prove an intent to interfere with the course of justice, referring to Sir John Donaldson MR’s observation above.[16] The fact that Balcombe LJ chose only to assume without deciding on the question of specific intent must mean that according to his Lordship, the view of the Master of the Rolls was not conclusive. (2) In Dallas, which involved jury misconduct, the parties agreed that under the English common law of contempt, two elements should be present, namely, (1) there had to be an act which created a “real risk” of prejudice to the administration of justice; and (2) there had to be an intention to create the risk : see [72]. The Divisional Court held that in the circumstances of the case, if it was proved that the juror had deliberately disobeyed the judge’s direction prohibiting the obtaining of extraneous information relating to the trial, an intent to impede or prejudice the due administration of justice would also have been proved, which was upheld by the Strasbourg Court: see [73]-[74]. In Solicitor General v Cox & Another (Contempt of court: Illegal photography) [2016] EWHC 1241, Lord Thomas, LCJ at [66] observed that “the circumstances in which contempts of court arise are too varied for one mens rea to be applicable to all forms of contempt. Nor is that the law.” Thus for contempt in the face of the court or closely related to such contempt, a specific intent to interfere with the due administration of justice is not required. His Lordship at [73] held that where the act constitutes a contempt in the face of the court, or one closely akin to such a contempt, the deliberate breach of a court order of which he has notice will be sufficient. It is not necessary that the person additionally intended by his breach to interfere with the administration of justice although it will generally be inferred that such an intention is established. Relevantly for present purpose, his Lordship for the reasons that he gave at [76]-[78] considered Dallas was consistent with his view on mens rea on the facts of the case before him. 62. That being the present state of the common law in England, it is hardly surprising for the learned editors of Arlidge, Eady & Smith on Contempt to observe : (1) At §3-45 : “In the case of criminal contempts not falling within the strict liability rule, it would appear that an intention to interfere with the administration of justice is required, at least for publication contempts. For other types of criminal contempt, the mental element is less clear.”; (2) At §11-24 : “A question remains, however, as to the nature of the mens rea required for common law contempt outside the context of publication, and particularly with regard to acts of direct interference with the processes of justice.”; and (3) At §11-34 : “While no definitive answer can at present be given, it seems increasingly likely that except in the classes of cases considered by the Divisional Court in Solicitor General v Cox, the courts will insist as a necessary ingredient upon an intention to interfere, both in regard to publication contempts and other forms of interference with the administration of justice… It is necessary to enter the caveat that it is still possible that some categories of common law contempt will be treated as offences of strict liability.” 63. Given the present state of uncertainty, we do not consider the commentary in UK Archbold 2018, at §28-35 relied on by Mr Lee is definitive on the subject. It appears to us that in England, the common law position on mens rea for criminal contempt is still in a state of flux and hopefully clarity will emerge as the English courts continue to develop the law in the years to come. 64. Thus analyzed, contrary to Mr Lee’s argument, we cannot discern a starting point even under English common law that specific intent is required for all forms of criminal contempt, other than contempt in the face of the court or contempt akin to such contempt. D2.3 Mens rea under the common law of Australia and New Zealand 65. Contrast the present English position with that in both Australia and New Zealand. In the latter jurisdictions, the courts all follow the pre-1981 line of English authorities, mainly, R v Odhams Press Ltd, ex p Attorney-General; and Attorney-General v Butterworth and hold that for criminal contempt, no specific intent to interfere with administration of justice is required. See the Australian and New Zealand cases cited by Mr Pao at [52] above. 66. It would also appear that in Australia, no specific intent is required for all forms of non-publication contempt. For in R v Lupco Slaveski, King J of the Supreme Court of Victoria at [88] reiterated : “ There is a line of argument that what is required in non-media publication cases, is proof of an intent on the part of the respondent that he intended to commit a contempt of court. I agree … that it is not necessary to prove an intent to commit a contempt and that the mens rea required in respect of a contempt is proof that the person intended to do the acts which are capable of, or do, amount to contempt. In criminal terms it would be proof that the acts were conscious, voluntary and willed acts. The issue of intention is relevant to the seriousness of the charge, and the level of punishment that the contempt may attract.” 67. The approach adopted in Australia and New Zealand provides a much needed element of certainty and uniformity in this area of the law. But with respect, a universal requirement for mens rea in every form of criminal contempt seems to be too rigid : cf Lord Thomas’s remarks in Solicitor-General v Cox at [66], cited at [61(2)] above. For example, when the alleged contempt is committed by a third party to the underlying proceedings, a specific intent may well be necessary : see [72] below. D2.4 Mens rea decided in some Hong Kong cases 68. Coming back to Hong Kong, there are instances where the Court of First Instance had applied R v Odhams Press Ltd, ex p Attorney-General and Attorney-General v Butterworth to the particular forms of criminal contempt before them. For example, in Secretary for Justice v Oriental Press Group Ltd, Chan CJHC and Keith J at p 155 held that for (a) scandalizing the court by newspaper articles and (b) a sustained, physical, paparazzi-style pursuit of a judge who was the focus of the offending newspaper articles, specific intent to interfere with the proper administration of justice was not an essential ingredient for criminal contempt. In so holding, their Lordships respectively adopted Solicitor-General v Radio Avon Ltd (at p 153) and Attorney-General v Butterworth (at p 157) for each of the two forms of contempt in question. It is however important for present purposes to note that a careful reading of their joint judgment clearly showed that Chan CJHC and Keith J were searching for the proper mens rea for the particular forms of contempt before them. They did not hold that as a general proposition no specific intent was required for all forms of contempt. 69. Ng Wai Bing, which concerned covert recordings of conversations in witness room during trial, is another example. In Ng Wai Bing, Wright J at [40] rightly observed that post-1981, a divergence had arisen between the approach of the courts in England on the one hand and Australia and New Zealand on the other regarding mens rea for criminal contempt. After surveying the relevant authorities, Wright J preferred and adopted the approach of the latter jurisdictions which followed the pre-1981 English common law because like Australia and New Zealand, Hong Kong did not have any statutory intervention in this area like the 1981 Act. In the Court of Appeal, Tang VP simply endorsed Wright J’s judgment without giving his own views. It was in fact not necessary for him to do so as the Court of Appeal was only concerned with the sentence imposed by Wright J and not liability. However, although the Court of Appeal was only concerned with sentence, contrary to Mr Lee’s submission, it does not detract from Tang VP’s endorsement of the principles set out by him. 70. While both Wright J and Tang VP spoke in general terms regarding mens rea, we do not consider a general proposition such as that for every form of criminal contempt there is no requirement for specific intent, forms the ratio decidendi of their judgments. Such a general proposition is much wider than what was necessary to address mens rea for the particular form of contempt then before the court and hence unnecessary. It is at most a non-binding obiter dictum. 71. On the other hand, there is at least one instance where the Court of First Instance required proof of a specific intent for the forms of criminal contempt before the court. In Secretary for Justice v Yuen Oi Yee Lisa, HCMP 2390/2008, unreported, 25 October 2010, the respondent was found guilty of civil contempt by breaching certain court orders and her undertaking given to the court; as well as criminal contempt for interference with the due and proper administration of justice by engaging in various forms of misconduct. Relevantly, Chu J (as she then was) drew a dichotomy between civil contempt and criminal contempt in terms of mens rea. After referring at [18] of her judgment to Kao Lee &Yip (a firm) v Donald Koo Hoi Yan (2009) 12 HKCFAR 830, [43]-[53] for mens rea in civil contempt, she at [20] said that as for criminal contempt the mens rea required is an intent to interfere with the course of justice, but it needs not be the sole intent and it may be inferred. She cited Attorney-General v Punch Ltd [2003] 1 AC 1046, at [4], [66] and [87] and Attorney-General v Newspaper Publishing Plc (CA) (1988), at p 383B-C to support the proposition. 72. However, it should be noted that both Attorney-General v Punch Ltd and Attorney-General v Newspaper Publishing Plc (CA) (1988) concerned contempt by a third party to the underlying proceedings who did the acts prohibited by the court orders to which he was not a party. Thus the passages of the judgments referred to by Chu J were in fact made by the House of Lords and the English Court of Appeal in the context of a third party’s liability for criminal contempt. Insofar as English common law is concerned, it is well established that a specific intent to interfere with the due administration of justice is required in order to impose liability on a third party for criminal contempt : see Z Ltd v A-Z and AA-LL [1982] QB 558, per Eveleigh LJ at p 578; cited with approval in Attorney-General v Times Newspapers Ltd (HL) [1992] 1 AC 191, per Lord Oliver at pp 217H-218E. That being the case, we do not think Chu J’s judgment is capable of laying down a general proposition that specific intent is required for all forms of criminal contempt. The fact that her Ladyship required proof of specific intent for the particular forms of criminal contempt before her does not necessarily mean that as a matter of general proposition, the same is true for all or other forms of criminal contempt. D2.4 Mens rea for the present form of criminal contempt 73. Where does the brief survey of cases above lead us? As a matter of Hong Kong common law, how should mens rea be determined for criminal contempt generally and for the form of contempt involved in the present case specifically? In searching for an answer, we propose to go back to first principles. 74. The law of contempt is and must be founded entirely on public policy : Attorney-General v Times Newspaper Ltd (HL)[1974] AC 273, per Lord Reid at p 194 D. The relevant public policy considerations must guide the court in approaching and determining the content of mens rea for criminal contempt generally and for a particular form of contempt in a given case. As Eveleigh LJ in Z Ltd v A-Z and AA-LL, at p 579D-G, explained : “ In some kinds of contempt it is well established that no element of mens rea need to be shown… However, contempt of court may take a wide variety of forms and the fact that it is regarded as an absolute offence in one form does not necessarily require it to be so treated in another form. It is very much a matter of public policy. In Attorney-General v Times Newspaper Ltd [1974] AC 273, 308, Lord Diplock said : ‘ no sufficient public interest is served by punishing the offender if the only person for whose benefit the order was made chooses not to insist on its enforcement.’ I do not regard those words as saying that the court should ignore the fact that there has been a wilful disobedience of its order, but they emphasize the importance of the general public interest which exists in so many forms of contempt. It does not seem to me to be in the public interest that a person with no wrongful intent should be brought before the court, let alone be punished, unless there is some overriding public interest to the contrary.” 75. Although Eveleigh LJ was addressing a third party’s liability, two points arose from his judgment which are, in our view, of general applicability. First, mens rea depends on the form of criminal contempt in question. This echoes Lord Thomas’s remark in Solicitor General v Cox, at [66] that “the circumstances in which contempts of court arise are too varied for one mens rea to be applicable to all forms of contempt.” Second, the determination of the applicable mens mea for a given form of criminal contempt is guided by public policy considerations. In our view, they should be equally applicable in Hong Kong. In short, under Hong Kong common law, as a general approach, mens rea for criminal contempt should be determined by reference to the particular form of contempt before the court and the pertaining public policy considerations.[17] Unlike Australia or New Zealand, we do not apply basic intent to all forms of criminal contempt. Nor do we adopt specific intent as a starting point for all forms of criminal contempt as contended by Mr Lee. 76. In the instant case, the criminal contempt concerned a party who was bound by a court order but breached it by obstructing the bailiff in executing the same in the course of discharging his duty. It is so because : (1) The basis of the Secretary’s committal application against Mr Wong is that his deliberate acts, as identified in the Statement of Facts and Originating Summons, coupled with his determination to continue flouting the Amended Injunction Order constituted interference and/or impediment to the due administration of justice by obstructing or resisting the police and/or bailiff in executing the Amended Injunction Order : [40] of the Originating Summons. (2) Given the wide definition of “the 1st Defendant” as persons occupying the public roads in question and the express plea at [39(3)] of the Originating Summons that Mr Wong acted in breach of the Amended Injunction Order, Mr Wong was on the Secretary’s case a party to the underlying proceedings, HCA 2104/2014, and was, in that capacity, bound by but had breached the Amended Injunction Order by doing the prohibited acts as pleaded. 77. Regarding the form of criminal contempt that we need to deal with, the following are the pertinent public policy considerations and principles. 78. First, it is axiomatic that in Hong Kong where the rule of law reigns, the due administration of justice can in no way be interfered with. Those who strike at it strike at the very foundations of our society. As Salmon LJ observed in Jennison v Baker [1972] 2 QB 52 at p 61, quoted by Lord Oliver in Attorney-General v Times Newspaper Ltd (HL) (1992) at p 216C-D : “ The inherent power of the judges of the High Court to commit for contempt of court has existed from time immemorial.…The power exists to ensure that justice shall be done. And solely to this end, it prohibits acts and words tending to obstruct the administration of justice. The public at large no less than the individual litigant have an interest and a very real interest in justice being effectively administered. Unless it is so administered, the rights, and indeed the liberty, of the individual will perish.” In a like vein, Lord Oliver at p 216A-B spoke of the court’s power to commit for contempt thus : “ It is as essential as it is ancient, for unless litigants can be assured that the rights which it is the duty of the courts to protect can be fairly determined and effectively protected and enforced the system of justice necessarily ceases to command confidence and an essential foundation of the structure of civilized society is undermined.” 79. Second, there are many facets of administration of justice that need to be protected. So correspondingly, challenges to the administration of justice embrace various kinds of conduct. The result is that contempt, which is only a generic term, in reality takes many forms : See Attorney-General v Times Newspaper (HL) (1974), per Lord Diplock at p 307H. Despite its protean nature, contempt has been traditionally classified under two heads, namely, civil contempt and criminal contempt.[18] 80. Civil contempt involves a breach of a court order or undertaking given to the court by a party to the proceedings. As Lord Oliver in Attorney-General v Times Newspaper Ltd (HL) (1992) at p 217G-H explained : “ One particular form of contempt by a party to proceedings is that constituted by an intentional act which is in breach of the order of a competent court. Where this occurs as a result of the act of a party who is bound by the order or of others acting at his direction or on his instigation, it constitutes a civil contempt by him which is punishable by the court at the instance of the party for whose benefit the order was made…” For civil contempt, only basic intent in the sense that the contemnor’s conduct was intentional and that he knew of all the facts which made it a breach of the order, is required : Kao, Lee & Yip v Koo Hoi Yan, per Sir Gerard Brennan NPJ at [45]-[46] and [50]-[53]. 81. Criminal contempt involves other forms of interference with the due administration of justice, either in a particular case or more generally as a continuing process : cf Attorney-General v Newspaper Publishing Plc (CA), per Sir John Donaldson at p 362D. It may be committed by a party to the proceedings or a stranger to the proceedings. In case of a stranger, it may involve him doing the act prohibited by a court order although he is not a party to the proceedings. Again to quote Lord Oliver in Attorney-General v Times Newspaper Ltd (HL)(1992) at pp 217H-218A : “ When, however, the prohibited act is done not by the party bound himself but by a third party, a stranger to the litigation, that person may also be liable for contempt.” 82. Third, for civil contempt, it is ordinarily left to the party for whose benefit the order was made to pursue against the contemnor, that is, the party in breach of the order. The party pursing a civil contempt is essentially taking it as a form of execution to coerce the contemnor into compliance with the order that he had breached. For criminal contempt, it is a matter for the Secretary for Justice, in his capacity as the guardian of the public interest in the due administration of justice, to raise. The sanction imposed on the contemnor is punitive in nature : Attorney-General v Newspaper Publishing Plc (CA) (1988), per Sir John Donaldson at p 362D-F; Attorney-General v Times Newspaper Ltd (HL) (1992), per Lord Oliver at p 217G. Despite such a distinction, it does not mean that the public interest in due administration of justice in cases of civil contempt can be ignored. It is wholly inappropriate to regard an order of committal in civil contempt as no more than a form of execution. For the court still has a very substantial interest in seeing that its orders are upheld : see Arlidge, Eady & Smith on Contempt, §§3-73 to 3-74, and the authorities discussed there. 83. Fourth, there are occasions where the disobedience or breach of a court order is sufficiently flagrant to warrant the imposition of a penal rather than a merely coercive sanction. For example, the scale and nature of the breach of the order is such that the contempt threatens the due administration of justice as a whole. Or when the disobedience or breach of the order also gives rise to a particular form of criminal contempt. When such occasions arise, the civil contempt would also amount to a criminal contempt : see Report of the Committee on Contempt of Court 1974,[19] at §22.[20] 84. Fifth, execution of an order in a civil litigation is usually undertaken by the party in whose favour it was granted. He would ordinarily enlist the bailiff’s assistance in execution as a last resort when all other means have failed. Involving the bailiff is thus usually the tail end of the execution process. And the bailiff is there not solely for the benefit of a private litigant. In executing the order, the bailiff is also discharging his duty as an officer of the court. In so discharging his official duty, the bailiff’s conduct in executing the order is part and parcel of the due administration of justice. When a party in breach of the order obstructs the bailiff in executing the same, he is obstructing the due administration of justice, thereby committing a criminal contempt : Halsbury’s Law of England, Vol 22, §§17 and 49. As Sir Richard Scott VC in De Court Re [1997] TLR 601, explained : “ The administration of justice depends not simply on the judges or upon counsel in court, it depends upon court officials…discharging essential functions for the purpose of enabling cases to come to court and to be dealt with by judges and counsel in the way with which we are all familiar. In my judgment a physical interference with officers of the court while conducting their duty in furthering the administration of justice is indeed a contempt of court.” 85. In De Court Re, the contemnor physically assaulted a court officer while he was engaged in official business. It was held that he had the requisite mens rea if he intended to do what he had done and did it consciously. In other words, a specific intent to interfere with the due administration of justice was not required. In our view, there is no distinction in principle between an assault against a court officer in discharging his duty and other forms of interference with the latter’s performance of his duty. The same requirement for basic intent would also suffice for the latter. In cases of obstructing a bailiff in discharging his duty, such as executing a court order, the contemnor must know or be taken to know that in obstructing the bailiff, he is also at the same time obstructing the administration of justice. Any requirement for specific intent is therefore superfluous. In other words, no specific intent to interfere with the due administration of justice is required. 86. But not only that. When executing court orders, bailiffs act as an extended arm of the court in administering justice. As a matter of public policy, bailiffs must be fully protected from interference by parties when executing orders against them in the discharge of their duty. To require the proof of specific intent would open the door for arguments over what the alleged contemnor knew or did not know, intended or did not intend. It would not only introduce much unwanted uncertainty into the equation where certainty is most required, but also make the very often difficult job of bailiffs even more difficult. All this is bad for the due and effective administration of justice in Hong Kong. 87. Any sense of injustice to the alleged contemnor arising from not requiring the proof of specific intent is militated by the fact that, by definition, (1) the alleged contemnor was a party who had been served with and was bound by the court order in question; (2) the person whose discharge of duty the alleged contemnor is said to have interfered with was to his knowledge a bailiff; and (3) to his knowledge also, the duty the latter was discharging was the execution of the very court order that the alleged contemnor had been served with and was bound by. 88. Finally, if the order itself includes a provision to authorize the bailiff to execute the order and the party who is bound by the order obstructs the bailiff, he commits a breach of that aspect of the order, too. But what he has committed is more than a civil contempt. For the reasons stated at [84] above it also amounts to a criminal contempt. And for the reasons stated at [85] to [87] above, a basic intent of obstructing the bailiff’s execution in breach of the order is sufficient. No additional specific intent to interfere with the administration of justice is required. 89. Applying the above public policy considerations and principles, we hold that when a party to a court order obstructs or interferes with the bailiff in executing the order, whether or not the order contains a provision authorizing the bailiff to do so, he commits a criminal contempt. For such a criminal contempt, it is sufficient to prove as mens rea that he intended to do the acts which constituted the actus reus of the contempt. No specific intent to interfere with the administration of justice is required. In other words, so long as the acts of the alleged contemnor are inherently likely to obstruct or interfere with the bailiff when he is executing an order, and the contemnor intends to do the acts, that would be sufficient. There is no need to go further to prove an intention on the part of the contemnor to interfere or obstruct though very often such intent is self-evident or can be readily inferred. 90. Leaving these general principles and considerations and looking more closely at the facts of the present case, we have no doubt that this is a particular strong case for only requiring a basic intent for the criminal contempt committed by Mr Wong. (1) We at [76] above have already explained why on the Secretary’s case, Mr Wong was a party to the underlying proceedings bound by the Amended Injunction Order but had acted in breach of the same by conducting himself in the manner complained of. (2) The Amended Injunction Order is not a usual kind of interim injunction granted to a private litigant in an ordinary kind of civil litigation, which does not concern any persons not privy to the proceedings. It is an interim injunction, widely publicized through the local media, targeted at the sizeable crowd of occupiers, whose identity is unknown in the case of the 1st Defendant, who had been unlawfully occupying large portions of several main public roads in Kowloon for quite a long time, thereby creating serious obstruction and interference with normal vehicular access which had been seriously affecting the livelihood of the Plaintiffs and causing significant inconvenience and nuisance to the public at large. Its purpose is to forthwith restore the normal vehicular traffic to the public roads affected and remove the public nuisance occasioned pending the final determination of the disputes between the Plaintiffs and the Defendants by court process, which might well take some time to conclude. There are accordingly considerable public interests involved in the Defendants fully obeying the Amended Injunction in a timely manner. The public at large as well as the court itself have very substantial interests in seeing that is done. (3) Given the immense public interests involved in the due administration of justice by Mr Wong complying with the Amended Injunction Order, by thwarting the Amended Injunction Order in the manner complained of, Mr Wong had seriously (a) frustrated its very purpose as explained above; (b) undermined the public confidence in its effective enforcement; (c) diminished the respect for it as a court order and the court’s authority in granting it; and (d) defied the bailiff’s authority in enforcing it. His conduct had seriously interfered with the administration of justice by the established courts of law as a whole and constituted an affront to the rule of law. His conduct is more than a civil contempt. It plainly amounts to a criminal contempt which warrants the imposition of a punitive sanction. The nature of his breach of the Amended Injunction Order was sufficiently flagrant enough to render his conduct a criminal contempt. (4) Involvement of the bailiff for execution of the Amended Injunction Order became necessary when there had been a real and substantial risk that the due administration of justice and the respect for the authority of the court and therefore the rule of law would be seriously undermined. That was highlighted by Au J in the Injunctions Judgment, at [142]. At [143], he went on to allude to the fact that (a) the ex parte injunctions that Poon J granted had been openly disobeyed and flouted by the Defendants en masse and (b) various public figures (including some legally trained individuals) suggested (erroneously) to the public and the protestors and demonstrators en masse to the effect that ex parte injunctions need not be complied with until they had been determined after an inter partes hearing and that there was no challenge to the rule of law by merely disobeying civil orders, and that the rule of law was only threatened when there was disobedience of an actual order of committal for contempt of court. Although Au J’s observations were made in the context of explaining why the police authorization directions (Para 3 to Para 5 of the Amended Injunction Order) were required, they equally explained why the bailiff had to be appointed and authorized under Para 2 to execute the Amended Injunction Order. See also the CA Judgment, at [13]-[17]. Plainly, without the bailiff’s involvement, execution of the Amended Injunction Order in the then prevailing circumstances would be extremely difficult if not impossible. (5) Self-evidently, Au J appointed the bailiff to execute the Amended Injunction Order with a view to arresting the imminent and serious risk of interference with the administration of justice posed by the occupiers’ continuous disobedience and flouting of the injunctions granted by the court. Through the wide publication of the Injunctions Judgment in the local media, Mr Wong must be fully aware of the reason why the bailiff was so appointed. When Mr Wong breached the Amended Injunction Order in the manner complained of, he knew that his conduct was more than a breach. He knew that the very terms that he had breached, that is, those that specifically envisaged the involvement of the bailiff, were provided for in the Amended Injunction Order to prevent interference with the administration of justice as explained. He knew by committing the breach, he had interfered with the due administration of justice. In such circumstances, the questions whether his breach of the Amended Injunction Order was knowing and deliberate and whether it was intended to interfere with the course of justice really amount to the same question. The need for some specific intent over and above the deliberate and knowing breach of the Amended Injunction Order is simply not required. (6) Leaving aside the breach of the Amended Injunction Order, Mr Wong’s obstruction of the bailiff in executing the Amended Injunction Order is in itself an interference with the due administration of justice by obstructing a court officer in discharging of his duty. It is in itself a criminal contempt which only requires a basic intent. 91. To summarize the discussion so far, we reiterate that : (1) Under Hong Kong common law on mens rea for criminal contempt : (a) There is no starting point that for every form of criminal contempt, a specific intent to interfere with administration of justice is required. (b) Nor is there a rigid rule that for every form of criminal contempt, only a basic intent to perform the acts which constituted the actus reus of the contempt is required. (c) Whether a specific intent or a basic intent is required depends on the actual form of contempt before the court and the public policy considerations involved. (d) When a party to a court order obstructs the bailiff in executing the order, whether or not the order contains a provision authorizing the bailiff to do so, he commits a criminal contempt. For such a criminal contempt, no specific intent to interfere with the administration of justice is required. Proof of a basic intent would suffice. (2) In the present context, a basic intent on Mr Wong’s part to perform the acts which constituted the actus reus of the criminal contempt complained of is sufficient. A specific intent to interfere with the administration of justice is not required. 92. This dealt with the first two main points taken by Mr Lee at [49], [50] and [53] above. This leads us to his last main point at [51], which can be disposed of shortly. (1) We agree with Mr Pao that Mr Lee’s reliance on Au J’s observation at [123] of the Injunctions Judgment was, with respect, misplaced. There, Au J was discussing the arrest powers of the police in executing the Amended Injunction Order and other similar injunctions granted in the two other cases. It was in that context that he stated the general proposition at [123]. He did not examine and could not have examined the applicability of that general proposition in the context of the actual committal application before the Judge later. (2) In any event, for the reasons that we gave, the general proposition stated by Au J stated at [123] was not applicable in the present context. There is no substance in Mr Lee’s complaint that in the interest of certainty, the Judge ought to have followed that general proposition in Au J’s judgment, which is not apposite in the first place. (3) Mr Lee complained that it would be an abuse of process not to follow the general proposition in Au J’s judgment because the Secretary did appear before Au J and should not be allowed to “move the goal post” now. But it begs the question if the Secretary had ever submitted to Au J that such a proposition should be adopted. Mr Pao, who also appeared on behalf of the Secretary (as a non-party) before Au J, told us that he had not made any submission on mens rea before the learned judge. That is really the end of Mr Lee’s complaint. 93. To conclude, although we respectfully differ from the Judge in his reasoning, we hold that he was correct in ruling that no specific intent to interfere with administration of justice is required to establish liability against Mr Wong for the criminal contempt that he had committed. 94. Ground 1 is rejected. E. GROUND 2 – BASIC INTENT 95. As an alternative to Ground 1, Mr Lee submitted that the Judge erred in law in holding that “the Secretary in order to succeed, needed only to prove that each of the respondents with the intention to remain, did remain in the Area”. Mr Lee argued that the Judge should have at least held that in order to succeed, the Secretary had to prove “basic intent” in the sense that Mr Wong had “intended to perform the acts which constituted the actus reus of the contempt”, namely, to obstruct or resist the police and/or bailiff officers in executing the Amended Injunction Order, namely to effect the clearance and removal of the obstructions as provided in the Amended Injunction Order. Harking back to Au J’s observation at [123(2)] of the Injunctions Judgment, Mr Lee argued that this is clearly what Au J meant when he referred to the situation where a person “deliberately impedes the bailiff in the due execution of his duties”. Mr Lee relied on Hills v Ellis [1983] QB 680, at p 686B, and submitted that an intention means “doing deliberate actions with the intention of bringing about a state of affairs which, objectively regarded, amount to an obstruction, that is, making it more difficult for the police to carry out their duty”. He submitted in the present case, there was no evidence for the Judge to find that Mr Wong had any intention that his actions should result in making it more difficult for the bailiff and/or the police to carry out their duties under the Amended Injunction Order, that is, to remove the obstructions in the Area. 96. With respect, we disagree. Applying a basic intent, what is required to be proved is that Mr Wong intended to perform the acts which constituted the actus reus of the contempt. And it will be sufficient if his acts are inherently likely to interfere with the due administration of justice. Remaining in the Area in breach of the Amended Injunction Order is clearly inherently likely to interfere with the due administration of justice by impeding its effective enforcement by the bailiff and the police. In the circumstances under which the Amended Injunction Order was executed by the bailiff on 26 November 2014, when Mr Wong had the intention to remain, did remain in the Area, the requirement for basic intent is satisfied. If he did obstruct the clearance and removal of the obstacles, it would, a fortiori, amount to a criminal contempt. The Judge was entirely correct in his reasoning and holding at [55] of the Judgment. The same also applies to other acts constituting the actus reus found by the Judge against Mr Wong. For completeness, we wish to state that we also agree with Mr Pao’s submission that Mr Lee’s argument is an impermissible attempt to introduce specific intent to the present form of contempt by the back door. 97. We reject Ground 2. F. Ground 3 98. In contending that the evidence did not and could not support a finding of criminal contempt, Mr Lee complained that the Judge made a host of errors when he failed to consider or give proper regard to various aspects of the evidence. The alleged errors include matters such as (1) the actual cause of commotion at 9:59 am; in particular, whether the Plaintiffs’ agents also contributed to its cause: (2) whether AC Bailiff Chim was able to complete giving the warning; (3) whether the bailiffs warned specifically Mr Wong that his manner of asking questions was obstructing the clearance operations; (4) whether Mr Wong’s asking questions on the spot was a more efficient way to clarify the terms of the Amended Injunction Order; (5) whether Mr Wong was prevented from leaving when the Final Police Warning was being given. 99. We have carefully considered all the evidence including the video footages that the parties invited us to view with the benefit of their detailed submissions. For present purpose, it is unnecessary for us to discuss each and every matter relied on by Mr Lee. In our view, on an objective assessment of the evidence overall, how Mr Wong conducted himself at the material times showed, beyond reasonable doubt, that his acts were not only inherently likely to but did in fact impede and obstruct the bailiff officers in discharging their duty when executing the Amended Injunction Order. 100. By arraying a wholesale attack against the Judge’s evaluation of the evidence, Mr Lee is in effect asking the Court of Appeal to re-try the case. This is simply impermissible. For although the appeal is a rehearing, the Court of Appeal only reviews the findings made by the Judge by well-established principles. It does not re-try the matter : China Gold Finance Limited v CIL Holdings Limited, CACV 11/2015, unreported, 27 November 2015, per Lam VP at [11]-[25]; ZJW v SY, CACV 10/2017, unreported, 1 December 2017, per Poon JA at [25]. Mr Lee must persuade us that the findings made by the Judge were plainly wrong before we could interfere. But his attempt to nitpick holes and magnify minute imperfections in the Judge’s fact finding exercise does not get him anywhere. Contrary to Mr Lee’s argument, we are satisfied that the Judge’s finding that Mr Wong’s conduct did constitute impediment and obstruction against the bailiffs in discharge of their duty when executing the Amended Injunction Order is plainly correct. 101. Ground 3 is rejected. G. Ground 4 102. Mr Lee submitted that the Judge erred in law when he at [120] of the Judgment used the First Police Warning issued at about 10:03 am as the reference point in determining whether Mr Wong should have left the Area. Instead, he should have held that there was at least a reasonable doubt as to whether Mr Wong had the intention to remain in the Area at the material time because (according to the video evidence) Mr Wong and those around him were attempting to leave but were blocked by the Blue Team from leaving the Area at the time when CIP Lam was still giving the Final Police Warning. Mr Lee argued that the purpose of the Final Police Warning was to give people a reasonable time to leave the Area. The objective video evidence showed that Mr Wong had in fact attempted to leave the Area before CIP Lam finished reading out the Final Police Warning, but was prevented from doing so by the Blue Team. 103. Mr Lee’s argument is premised on the assumption that whatever Mr Wong’s conduct might be before the Final Police Warning had been given, it could not possibly constitute the actus reus of criminal contempt in question. He submitted that this is the effect of Para 4 of the Amended Injunction Order. With respect, Mr Lee’s reading of Para 4 is wholly misconceived. 104. To recap, Para 4 provides : “(4) Any police officer be authorized to arrest and remove any person who the police officer reasonably believes or suspects to be obstructing or interfering any bailiff in carrying out his or her duties in enforcing the terms of the Injunction Order, provided that the person to be arrested has been informed of the gist of the terms of the Injunction Order and this Order and that his action is likely to constitute a breach of the Injunction Order and this Order and obstruction of the administration of justice, and that he may be arrested if he does not desist.” 105. In the CA Judgment, Lam VP clarified the effect of Para 4 thus : “20. At one stage, we were concerned about whether the arrest provision places undue restrictions on the police by setting the following prerequisites: (a) informing the person concerned as to the gist of the injunction; (b) warning him that his action is likely to constitute a breach of the injunction; and (c) warning him also the possibility of his being arrested if he does not desist. These requirements were obviously laid down for the protection of the defendants and persons against whom the bailiff takes action pursuant to the order. On a proper reading of the order, it is clear to us that those are steps to be taken by the bailiff in the execution of his duty under the order and the person concerned will be given a last opportunity to comply with the order voluntarily. In our view, these are steps a bailiff would have to take before a case of obstruction or interference with the carrying out of his duties can be made out. Whether a police officer would like to repeat any of these steps before he exercises his power of arrest is entirely a matter for him to decide. Viewed thus, these requirements cannot be regarded as undue restrictions on the proper exercise of police powers.” (Emphasis supplied by Lam VP) As said, the CA Judgment was widely reported at the time when it was handed down, that is 21 November 2014, before the present criminal contempt arose. 106. The CA Judgment made it crystal clear that the question of whether Mr Wong and indeed any person had committed the actus reus of the present criminal contempt could not possibly depend on the issue of any police warning, let alone the Final Police Warning. 107. Mr Lee further submitted that pursuant to Para 4, the bailiff had to warn the person to be arrested that his action is likely to constitute (1) a breach of the Amended Injunction Order and (2) obstruction of the administration of justice before he could be said to have committed the actus reus of the contempt. With respect, Mr Lee’s submission is plainly wrong. (1) As a matter of law, the actus reus of the contempt in the form of obstructing a bailiff in discharging his duty does not depend on any prior warning given by the bailiff to the contemnor. And Para 4 could not possibly have the effect of altering the law. That is why at [20] of the CA Judgment, Lam VP in interpreting Para 4 did not include any warning by the bailiff to the contemnor that his action is likely to constitute an obstruction of justice as one of the prerequisites before a case of obstruction or interference with the carrying out of the bailiff’s duty can be made out. Para 4 must be understood according to the Court of Appeal’s interpretation. (2) As a matter of practicality, Mr Lee’s submission would render the bailiff’s task impossible. Take this example. A person did an act which was clearly an obstruction to the bailiff’s discharge of his duty when the latter was about to give the warning that the former’s act was likely to constitute an obstruction of the administration of justice under Para 4. But because the bailiff had yet to give the warning, his act could not have constituted the actus reus of the contempt. So that person could easily impede the bailiff in executing the Amended Injunction Order by continuing with his act while at the same time taking steps to prevent the bailiff from giving the warning. This makes any effort by the bailiff to effectively execute the Amended Injunction Order a futile exercise. Such absurdity could not have been intended by Para 4. 108. In any event, as a matter of fact, the bailiff officers had since 8:40 am given repeated warnings to the occupiers, including Mr Wong, to leave the Area lest they might have obstructed the bailiff and committed a criminal contempt. And when CIP Lam gave the First Police Warning at 10:03 am, he repeated the point that any person obstructing the bailiff might commit a criminal contempt. The evidence shows clearly (and Mr Lee has not otherwise submitted) that Mr Wong could have left the Area after the bailiff made the various announcements between 8:40 am and before CIP Lam gave the First Police Warning at 10:03 am. But he chose to remain throughout. By 10:03 am at the very latest, he had already committed the actus reus of the criminal contempt. In fact, the Judge could have used earlier moments, such as 8:44 am, that is, shortly after the first announcements by the bailiff officers, as the reference point in determining if Mr Wong should have left the Area. There is nothing wrong for the Judge to adopt the First Police Warning given at 10:03 am as the reference point, which is more favourable to Mr Wong. 109. Ground 4 is rejected. H. Ground 5 110. We finally come to Ground 5. 111. Mr Lee argued that the Judge erred in holding that Mr Wong had, by remaining in the Area, intended to and did obstruct or impede the bailiffs, as opposed to the police, from effecting clearance and removal of the obstructions as provided in the Amended Injunction Order. The Judge failed to take into account the uncontroversial evidence of Chief Bailiff Choi in cross-examination[21] that the police had decided, on its own, to take over the whole operation by replacing the bailiffs and taking it upon themselves to remove obstructions from the Area, even though the bailiffs had only invited the police to assist them in crowd control due to the commotion as opposed to clearing the obstructions in the Area.[22] After the police had automatically taken over and the bailiffs had left the scene, the police were no longer assisting the bailiffs to execute the Amended Injunction Order pursuant to any request from the bailiffs under Para 3 in removing obstructions from the Area, as the bailiffs had never requested the police to do so and the police were exercising their statutory power under section 10(f) of the Police Force Ordinance, Cap 232. Therefore, Mr Lee reasoned, the Judge failed to hold that in law Mr Wong could not have obstructed or impeded the bailiffs, as opposed to the Police, from clearing or removing the obstructions in the Area pursuant to the Amended Injunction Order, as opposed to the statutory power to remove obstructions from the Area. Thus, Mr Lee submitted that while Mr Wong might have been guilty of the offence of obstructing the police in the due execution of their duty, he could not have been guilty of criminal contempt in connection with the Amended Injunction Order. 112. Mr Lee’s submission ignored the fact that on the findings made by the Judge, even before police’s intervention, Mr Wong had already committed the criminal contempt by conducting himself in the manner complained of. And in any event, pursuant to Para 3, the bailiffs were specifically “authorized and directed to request the assistance of the police where necessary”. There is no limitation as to what assistance the bailiffs might wish to request from the police. The evidence clearly showed that the bailiffs were unsuccessful in removing the obstructions in the Area. It is for that reason that they requested the assistance of the police. The evidence of Chief Bailiff Choi must be understood in its totality. He did say that the bailiffs passed the responsibility of the clearance operation to the police in accordance with the Amended Injunction Order, and he understood that the police were permitted to execute the same.[23] In response to the Judge’s question as to the role of the police, Chief Bailiff Choi said the police were expected to carry out the Amended Injunction Order, which included clearing of obstructions.[24] When Chief Bailiff Choi’s evidence is understood properly, it is beyond doubt that the bailiffs had requested the police to assist in executing the Amended Injunction Order, and that the police did not take it upon themselves to remove the obstacles as contended by Mr Lee. The police were clearly assisting the bailiffs as officers of the court in the discharge of the latter’s duty, as expressly provided for by Para 3. The fact that the actual clearance was effected by the police and not the bailiffs and after the bailiffs had left the scene is wholly immaterial. Nor does it matter that Mr Wong’s current conduct might have amounted to obstructing the police, a separate criminal offence. For it is simply not a defence to his liability for criminal contempt. 113. We reject Ground 5. I. Dispositions 114. For the above reasons, none of the grounds of appeal succeeds. We accordingly dismiss Mr Wong’s appeal. 115. Costs should follow the event. We make an order nisi that Mr Wong do pay the Secretary costs of this appeal, to be taxed if not agreed, with a certificate for two counsel. Mr Wong’s own costs are to be taxed in accordance with the legal aid regulations. Mr Victor Dawes SC (by written submission only), Mr Jin Pao and Mr Derek CL Chan, instructed by the Department of Justice, for the Applicant Mr Martin Lee SC, Mr Hectar Pun SC, Mr Earl Deng and Mr Anson Wong Yu Yat, instructed by M/s JCC Cheung & Co, assigned by the Director of Legal Aid, for the Respondent [1] The proceedings against Mr Wong was HCMP 798/2015. The other proceedings were HCMPs 778/2015, 780/2015, 783/2015, 784/2015, 788/2015, 789/2015, 791/2015 and 796/2015. [2] On 17 January 2018, the Judge sentenced Mr Wong to a term of 4 months and 15 days’ imprisonment : [2018] HKCFI 77, [46] – [47]. Mr Wong has not lodged any appeal against sentence. [3] They are contained in the Amended Statement of Facts filed pursuant to Order 52, rule 2(2) of the Rules of the High Court, Cap 4A on 5 August 2015, which were adopted by the Judge : see Judgment, [4] – [43]. [4] For HCA 2104/2014, the Court of Appeal handed down the judgments refusing the 3rd Defendant’s and 2nd Defendant’s leave application in HCMP 2975/2014 and HCMP 3090/2014 on 15 and 27 November 2014 respectively. [5] HCMP 3028/2014, unreported, 21 November 2014, [20]. [6] Cap 228. The charge of obstructing public officer was withdrawn on 9 January 2015 after the Secretary for Justice decided to apply for leave to commit Mr Wong for criminal contempt on the same factual basis. [7] Amended Statement of Facts [45]; and Originating Summons filed on 5 August 2015 [38] . [8] Amended Statement of Facts [47]-[48]; and Originating Summons [39]-[40] . [9] [77]-[79] of the Judgment read : “ 77 As mentioned above, between the application for an injunction order in October and the Bailiffs and the Plaintiffs’ agents carrying out the operation in November, all the relevant matters had been widely reported in the Local Media for a month. On the 25th, there was a clearance operation in Argyle Street, a very short distance away, that too was widely reported. The Bailiff’s announcement was clear. It essentially informed the crowd to pack up their belongings and leave as the Bailiffs and the Plaintiffs’ agents would be soon carrying out an operation to clear up obstacles. 78. Given the much publicized event, the presence of strong contingent of police officers as well as local and international media and the announcements given by various parties at the time, it was highly unlikely that any citizen of Hong Kong would misapprehend the situation. In my view, those who had decided to remain and to stand in the middle of Nathan Road within the Area knew precisely what they were doing. 79. Given the length of time that [Ms Chu Pui Yan, the respondent in HCMP 778/2015] chose to remain at the Area, the acts and movements she performed and her knowledge of the clearance operation by the Bailiffs and police officers in the Area, I am satisfied that Ms Chu had no desire to leave the Area. The presence of individual protestors like Ms Chu also lent support to other fellow protestors too. As is common in mass protests, fellow protesters banded together to fight for their beliefs. Video footage clearly indicated that there was no way the Bailiffs and the Plaintiffs’ agents were able to carry out the clearance operation by themselves given the presence of the large crowd. Their presence made the job of clearing and removing obstacles along Nathan Road substantially more difficult. Police assistance was eventually sought by the Bailiffs. Their presence and conducts were, not only inherently likely, but factually calculated, to delay if not to frustrate the Bailiffs and the Plaintiffs’ agents’ efforts. I have no doubt that Ms Chu’s conducts on that day amounted to a serious interference with the administration of justice. [10] Original ground of appeal 1. [11] Original ground of appeal 2. [12] Consisting of original grounds of appeal 3 to 8 and 11. [13] The Secretary was also represented by Mr Victor Dawes, SC who had settled the written submissions. But because of a conflicting court commitment, Mr Dawes was unable to appear before the Court at the hearing on 5 March 2018. [14] As provided for in section 1 of the 1981 Act : “In this Act, ‘the strict liability rule’ means the rule of law whereby conduct may be treated as a contempt of court as tending to interfere with the course of justice in particular legal proceedings regardless of intent to do so.” And section 2(3) provides that “the strict liability rule applies to a publication only if the proceedings in question are active within the meaning of this section at the time of the publication.” [15] See [68] below. [16] Balcombe LJ was actually one of the three members of the Court of Appeal sitting in Attorney-General v Newspaper Publishing Plc (CA) (1988). [17] How the actual mens rea was determined in the Hong Kong cases that we have discussed is entirely consistent with this general approach. Thus, our judgment should not be read as having the effect of casting any doubt on those cases. [18] The distinction has been invariably described as “misleading”, “unhelpful” or “largely meaningless” given the substantial similarities between the two heads of contempt in terms of procedure, procedural safeguards and standard of proof (beyond reasonable doubt). Two practical differences were identified in the course of arguments. First, a conviction for criminal contempt will result in a criminal record for the contemnor. Second, the Secretary for Justice has the locus to bring criminal contempt in respect of obstruction to the administration of justice to safeguard the public interest. Contrary to the submission of Mr Lee, the concept of purging a contempt can be applicable in some form of criminal contempt : see Re Mahesh J Roy (No 2) [2017] 5 HKLRD 830, per Lam VP at [35] and [47]. [19] Commonly referred to as the Phillimore Report. [20] This proposition is accepted by Mr Lee. In fact, in the course of his oral submissions, Mr Lee gave a vivid example of how a serious and defiant breach of §1(a) of Amended Injunction Order may also amount to a criminal contempt in the context of the present case. [21] Day 10, transcript 10/11/14-15. [22] Day 10, transcript 10/12/3. [23] Day 10, transcript 10/11/16-17. [24] Day 10, transcript 10/13/17-26. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Tang PJ: 3. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: 4. The question of law for which leave to appeal was granted in this case raises this short point of criminal procedure: “May a single Justice of Appeal who refused leave to appeal to the Court of Appeal, lawfully sit as a member of the panel if the applicant seeks to have his application determined by the Court of Appeal?” A. The relevant statutory provisions 5. The statutory provisions governing criminal appeals to the Court of Appeal are contained in the High Court Ordinance (Cap.4) (“HCO”) and the Criminal Procedure Ordinance (Cap.221) (“CPO”). In determining the question arising on this appeal, it will be necessary to consider the proper construction of those relevant statutory provisions. 6. By section 13(3) of the HCO, it is provided that: “The criminal jurisdiction of the Court of Appeal shall consist of – (a) appeals from the Court of First Instance or District Court under Part IV of the Criminal Procedure Ordinance (Cap.221); ...” 7. Part IV of the CPO concerns “appeals, questions of law reserved and referred and review”. Section 82(1) provides that a person convicted of an offence on indictment may appeal to the Court of Appeal against his conviction.[1] Unless the ground of appeal involves a question of law alone, or unless the trial judge grants a certificate that the case is fit for appeal, leave of the Court of Appeal is required: see section 82(2).[2] 8. Section 34 of the HCO concerns sittings of the Court of Appeal in the exercise of its criminal jurisdiction. Sub-sections (2) and (3) of that section relevantly provide: “(2) The Court shall be duly constituted for the purpose of determining any appeal – ... (b) under Part IV of the Criminal Procedure Ordinance (Cap.221), if it consists of an uneven number of Justices of Appeal not less than 3. ... (3) No judge shall sit as a member of the Court of Appeal on the hearing of, or shall determine any application in proceedings incidental or preliminary to – (a) an appeal from a judgment or order made by him; or (b) an appeal against a conviction before him or a sentence passed by him.” 9. There are provisions of both the HCO and the CPO which deal with the powers of a single Justice of Appeal of the Court of Appeal in its criminal jurisdiction. Thus, section 34A of the HCO provides: “(1) Subject to subsection (2), in pursuance of the criminal jurisdiction of the Court of Appeal, any power of the Court of Appeal not involving the determination of the appeal may be exercised by a single Justice of Appeal in the same manner as it may be exercised by the Court of Appeal and subject to the same provisions. (2) Where a single Justice of Appeal refuses an application to exercise any power of the Court of Appeal in pursuance of subsection (1), the applicant shall be entitled to have the application determined by the Court of Appeal.” 10. These powers are reflected in section 83Y of the CPO which relevantly provides: “(1) The powers of the Court of Appeal under this Part which are specified in subsection (2) and the power to give directions under section 156(5) of the Crimes Ordinance (Cap.200) may be exercised by a single judge in the same manner as they may be exercised by the Court of Appeal and subject to the same provisions. (2) The said powers are the following – (a) to give leave to appeal; ... (3) If the single judge refuses an application on the part of an appellant or applicant to exercise in his favour any of the powers above specified, the appellant or applicant shall be entitled to have the application determined by the Court of Appeal. (4) The references in this section to a single judge are to a single judge of the Court of Appeal or of the Court of First Instance.” 11. Although the single judge could be a judge of the Court of First Instance or a Justice of Appeal, Practice Direction 4.2 (Criminal Appeals to the Court of Appeal) directs (at paragraph 6A(1)) that applications for leave to appeal against conviction or against sentence involving less than 7 years’ imprisonment (or in respect of which a sentence other than one of imprisonment has been imposed) will be heard and determined by a single Justice of Appeal. As this Court has held, in Chau Ching Kay v HKSAR,[3] the single judge procedure under section 83Y of the CPO serves two useful purposes: first, it aims to relieve the full Court of Appeal of some of its workload; and secondly, it also serves as a screening process for unmeritorious applications. B. The factual background and the appellant’s trial 12. The appellant was tried in the District Court on a charge of burglary and convicted on 19 September 2014.[4] Although the underlying facts of the burglary charge are not directly material to this appeal, it is clear that the prosecution case was very strong and the defence case correspondingly weak. The appellant was seen by a police officer in Flat 34C of a building in Tai Kok Tsui apparently trying to break out from the inside through the locked security gate. He was subsequently discovered hiding in another unlocked flat on the same floor of the building. Flat 34C was found to have been ransacked and a sum of RMB600 stolen from it. The appellant’s fingerprints were found on a drawer in Flat 34C and RMB600 was found in his trouser pocket. The defence case was that he had been framed for the burglary, so the principal issue at trial was the credibility of the prosecution evidence. The Judge accepted the prosecution evidence and was satisfied that it had proved its case beyond reasonable doubt. She described the evidence as “overwhelming”. 13. Upon conviction, the appellant was sentenced by the Judge to three years’ imprisonment. C. The procedural history after trial 14. On 29 September 2014, the appellant applied for leave to appeal against his conviction and sentence. This leave application was heard by Macrae JA sitting as a single Justice of Appeal under sections 83Y(1) and 83Y(2)(a) of the CPO and, on 2 June 2015, he dismissed it.[5] In his judgment, Macrae JA reviewed the prosecution and defence cases and the reasons for verdict and then considered the grounds of appeal sought to be raised, which included medical reports purporting to support the appellant’s case that he had been beaten by the police in the course of being framed for the burglary. Macrae JA did not detect any error of law or in the Judge’s fact-finding process and did not consider that the medical evidence sought to be relied on would have made any difference to the verdict. He commented (at [15]) that, had the medical reports been produced, he was: “doubtful they would have had any significant impact on what is on the face of it an overwhelming case.” 15. Accordingly, Macrae JA refused leave to appeal both in relation to conviction and sentence. He concluded his judgment (at [20]) by saying: “These applications are hereby refused and the applicant is duly advised of his right to renew either or both applications to the Court of Appeal, and the consequences of so doing, which may include a direction for the loss of any time spent in custody pending his appeal, if the Court comes to the conclusion that there was no justification for the renewal of the application.” 16. So advised and warned, the appellant duly elected to renew his applications for leave to appeal against conviction and sentence under section 83Y(3) of the CPO. The renewed application for leave to appeal was heard, and dismissed, on 2 December 2015 by the Court of Appeal consisting of Macrae, McWalters and Pang JJA.[6] Pang JA gave the written reasons for judgment of the court, in which he noted (at [19]) that the appellant’s grounds of appeal “concern nothing that the judge and/or Single Judge had not covered”, save for one matter which he concluded were “of no help” to the appellant. He concluded, by way of disposition of the application, as follows (at [22]): “The evidence against the applicant was overwhelming. His renewed applications were, on the other hand, completely lacking in merit. That being the case, these applications must be refused with the order that there be a loss of time of 6 weeks pursuant to section 83W of the Criminal Procedure Ordinance, Cap 221.” 17. The appellant then applied to the Court of Appeal (differently constituted) for leave to appeal to this Court, which dismissed the application on the basis that the questions proposed to be raised on appeal were not reasonably arguable.[7] On 17 August 2016, the Appeal Committee granted leave to appeal on the single question set out in paragraph 4 above.[8] D. The appellant’s contentions 18. In support of the appeal, Mr Gerard McCoy SC, counsel for the appellant,[9] contended that section 83Y(3) of the CPO and sections 34(3) and 34A(2) of the HCO must be construed purposively and in a manner consistent with the rights guaranteed under relevant constitutional instruments (which are addressed below). He submitted that, so construed, it followed that a single Justice of Appeal who has previously refused leave to appeal cannot sit as a member of the Court of Appeal hearing the renewed application for leave to appeal. On that basis, the Court of Appeal in the present case, consisting of Macrae, McWalters and Pang JJA, was not properly constituted because it did not consist of “an uneven number of Justices of Appeal not less than 3” as required.[10] 19. It was also contended on behalf of the appellant that, if a single Justice of Appeal who has heard the initial application for leave to appeal were to sit on the Court of Appeal hearing the renewed application, a fair-minded observer would perceive that he was not independent and impartial. On the basis of this perceived bias, it was the appellant’s case that, under section 83Y(3) of the CPO, the Court of Appeal hearing a renewed application for leave can never, as a rule of general application, consist of a single Justice of Appeal who has previously refused leave to appeal in that case. 20. Finally, it was contended for the appellant that, in this particular case, Macrae JA’s sitting in the Court of Appeal on the renewed application gave rise to the appearance of bias so that the judgment of the Court of Appeal refusing leave to appeal against conviction and sentence and the order for loss of time made by it should be set aside. E. The construction of sections 34(3) and 34A(2) of the HCO and section 83Y(3) of the CPO 21. In a system governed by the rule of law, it is axiomatic that adjudicative processes must be fair and impartial and observably so: justice must not only be done, it must be seen to be done.[11] This is particularly important in respect of those charged with, or convicted of, criminal offences since personal liberty will often be at stake. These principles are firmly embedded in the laws of Hong Kong at a constitutional level and also in its domestic law: rights under the International Covenant on Civil and Political Rights (“ICCPR”) are given constitutional protection under Article 39 of the Basic Law and are applied in Hong Kong through the Hong Kong Bill of Rights Ordinance (Cap.383). 22. These rights necessarily form part of the context in which the provisions of the HCO and CPO at the heart of this appeal are to be construed. Of particular relevance are the rights contained in Articles 10 and 11(4) of the Hong Kong Bill of Rights[12] (“BOR 10 and BOR 11(4)”) which respectively provide: “10. All persons shall be equal before the courts and tribunals. In the determination of any criminal charge against him, or of his rights and obligations in a suit at law, everyone shall be entitled to a fair and public hearing by a competent, independent and impartial tribunal established by law. ... 11. ... (4) Everyone convicted of a crime shall have the right to his conviction and sentence being reviewed by a higher tribunal according to law.” 23. The effect of the statutory provisions governing appeals in criminal cases, as noted above, is that a convicted person has the right to appeal to the Court of Appeal against his conviction and sentence, but that leave is required unless he relies on a ground of appeal which involves a question of law alone or unless a certificate is granted by the trial judge. Such leave can be granted by a single Justice of Appeal or by the Court of Appeal (consisting of at least three judges of that court). 24. It is accepted in the Appellant’s Printed Case that the leave requirement imposed by section 82(2) of the CPO is lawful and does not infringe BOR 11(4).[13] This is clearly correct since the leave threshold for a full criminal appeal cannot be regarded as constituting an impermissible dilution of the right of review of a criminal conviction or sentence by a higher tribunal “according to law” as that expression is used in BOR 11(4).[14] Nevertheless, it is the appellant’s case on this appeal that, properly construed in the context of that constitutionally protected right, the references in section 34(2) of the HCO and section 83Y(3) of the CPO to the determination by the Court of Appeal of a renewed application for leave to appeal after a single Justice of Appeal’s refusal of the initial application requires a different decision-maker and one which does not include the same single judge who refused the initial application. It is contended that a Court of Appeal hearing a renewed application for leave to appeal which includes a Justice of Appeal who previously refused the application is not an independent and impartial tribunal and therefore infringes his rights under the BOR. 25. The wording of sections 34(3) and 34A(2) of the HCO and section 83Y(3) of the CPO do not, in my view, support the appellant’s contentions in this appeal, even when those provisions are purposively construed in their context including the rights under BOR 10 and BOR 11(4). 26. Section 34A(2) of the HCO, as reflected in section 83Y(3) of the CPO, plainly refers to the same application – in this case an application for leave to appeal but it could be any application not involving the determination of the appeal – which has been made to a single Justice of Appeal being determined by the full Court of Appeal consisting of at least three judges. This is clear from the statutory language referring to “an application” refused by a single judge and the entitlement to have “the application” (emphasis added) determined by the full court. At the time when the single judge refuses the original application, it is not finally determined until after its review by the full court or after the applicant elects not to renew the application. 27. For this reason, the appellant’s argument that the renewed application is an appeal from the single judge’s decision and therefore one on which he is prohibited from sitting by reason of section 34(3) of the HCO is unconvincing. It is not an appeal from the single judge’s decision but instead a further hearing of the same application which will eventually lead to a final determination of the application. Section 34(3) of the HCO is therefore not engaged in the present situation. Instead, that section operates to exclude a judge from hearing either a substantive appeal, or an application for leave to appeal, from his own judgment or order issued in a court inferior to the Court of Appeal. 28. It is wholly plausible that the legislature would have intended that a single judge should be eligible to be part of the three-member full court to re-hear the leave application. On the original application, which as a matter of practicality would be able to be heard sooner than would be the case if a full three-judge court had to be assembled, the applicant would have the benefit of the single judge’s view of the prospects of success of the application. If leave to appeal were granted by the single judge, the application for leave would go no further and a substantive appeal hearing would then follow. But if the single judge were minded to refuse the application, the applicant would have the benefit of the judge’s reasons for doing so before exercising his right to have that application for leave to appeal finally and fully determined by a three-judge court. That full Court of Appeal would be constituted by a majority of the judges other than the single judge and would have the power and practical ability to overrule the single judge who initially refused leave to appeal. The single judge may well change his mind on the rehearing and is not required to adhere to his original refusal. This process would be a genuine power of review by the Court of Appeal and is similar to other statutory review processes. For example, the legislature has seen fit to provide for a magistrate, sitting alone, to review his own decision.[15] 29. Nor do the BOR rights relied upon suggest a different construction of those provisions of the HCO and CPO. There is no question of the right of review under BOR 11(4) being contravened by participation of the single judge on the renewal of the leave application before the full court. As already indicated, he is free to change his mind and is not bound by his previous refusal. In any event, the other two members of the court form a majority which could overrule the single judge. In the present case, there is no question that the appellant has not enjoyed the right to have his conviction and sentence reviewed by a tribunal higher than the District Court where he was tried, convicted and sentenced. The powers of that higher tribunal, the Court of Appeal, could be exercised by a single Justice of Appeal and, even if the matter had rested with his refusal of leave to appeal, there would have been no denial of the appellant’s BOR 11(4) right. The appellant’s right to renew the application after an initial refusal by a single judge confers an additional protection of his BOR 11(4) right of review by a higher tribunal. 30. All judges are subject to the rules of disqualification for bias, either actual or apparent.[16] Since those rules will apply in respect of any single Justice of Appeal who may be listed to sit as part of a three-judge Court of Appeal hearing a renewed application for leave to appeal, I do not think that the Court of Appeal re-hearing such an application pursuant to its statutory power to do so loses its independence or impartiality – as a “competent, independent and impartial tribunal” within BOR 10 – simply because one of its members, who would be bound by the majority, has previously sat on the initial leave application and reached what, at that stage, will simply be a provisional view. Instead, it is by the application of the rules concerning disqualification for bias that the existence of an independent and impartial tribunal for any particular case is ensured. 31. This conclusion is consistent with UK and Strasbourg jurisprudence as to the content of the right to an independent and impartial tribunal under Article 6(1) of the European Convention on Human Rights.[17] It has been authoritatively held that there is no difference between the common law test for bias and the requirement under ECHR 6(1) of an independent and impartial tribunal: see Lawal v Northern Spirit Ltd [2003] ICR 856 per Lord Steyn at [14], R v Abdroikov [2007] 1 WLR 2679 per Lord Bingham at [14] and O’Neill v HM Advocate (No.2) [2013] 1 WLR 1992 per Lord Hope of Craighead DPSC at [47]. In my view, the same conclusion follows in Hong Kong in respect of the right to an independent and impartial tribunal under BOR 10. 32. Reference was made to the provisions governing applications to the Court of Appeal for leave to appeal in civil matters. Under Order 59 rule 2A(5) of the Rules of the High Court (Cap.4A), applications for leave to appeal may be determined on the papers. Under Order 59 rule 2A(7), where a party is aggrieved by that determination, it may ask the Court of Appeal to reconsider it at an oral hearing and, by Order 59 rule 2A(9), it is expressly provided that the court may include a Justice of Appeal who has previously determined the application on the papers. These provisions were contrasted with the provisions concerning criminal appeals and it was contended that, had the legislature intended that a single Justice of Appeal would be eligible to sit in the Court of Appeal hearing a renewed application for leave to appeal, the legislation would have said so. However, the fact that the Rules of the High Court provide expressly for what, in my view, is clearly inherent in the proper construction of sections 34(3) and 34A(2) of the HCO and section 83Y(3) of the CPO does not militate against that construction. 33. Reference was also made to:- (1) Practice Direction SL4 (Crime: Sentence: Loss of Time), paragraph 11 of which reads: “In the case of applicants who are unrepresented and whose application first comes before a single judge, the single judge has himself the power to order loss of time. Where an application has been refused by a single judge and the application is nonetheless renewed and the court before which the renewed application is made also takes the view that the application is wholly without merit, an order for loss of time will ordinarily be made.” (Underlining added) (2) Practice Direction 4.2 (Criminal Appeals to the Court of Appeal), paragraph 6A(7) of which reads: “...Where an application devoid of merit has been refused by the single judge he may indicate that the Full Court should consider making a direction for loss of time on renewal of the application. However the Full Court may make such directions whether or not such an indication has been given by the single judge.” (Underlining added) 34. It was contended that the underlined portions of the above two practice directions support the appellant’s construction of the relevant provisions of the HCO and CPO since they suggest that the full Court of Appeal will consist of judges other than the single judge who has initially heard the leave application. However, the proper construction of sections 34(3) and 34A(2) of the HCO and section 83Y(3) of the CPO falls to be determined on the basis of their own wording read in the light of their statutory context and purpose and not by reference to practice directions which, having no statutory force, are subsidiary to, and cannot override, the proper construction of the governing legislation. F. Whether apparent bias because of pre-determination and structural lack of impartiality 35. The appellant’s second contention is that the participation of any Justice of Appeal who has previously refused leave to appeal in the full Court of Appeal’s consideration of the renewed application for leave to appeal gives the appearance that the single judge will have pre-determined that application and therefore of a structural lack of impartiality in the constitution of the Court of Appeal. It is no part of the appellant’s case that there is actual bias on the part of Macrae JA in the present case, simply that the system of having the single judge who has initially refused the application sit in the Court of Appeal on the renewed application gives rise to the perception of a real possibility of bias (i.e. apparent bias). 36. The test for apparent bias in Hong Kong is the reasonable apprehension of bias test applied by the Appeal Committee in its determination in Deacons v White & Case Ltd Liability Partnership & Ors[18]at [18] to [21], namely: “The court must first ascertain all the circumstances which have a bearing on the suggestion that the judge was biased. It must then ask whether those circumstances would lead a fair-minded and informed observer to conclude that there was a real possibility ... that the tribunal was biased.” Although that was said in the context of a civil case, the position of a civil judge is in principle the same as that of a criminal judge in this regard and there is no reason why the same test should not apply where it is said that there is apparent bias on the part of a judge in a criminal case: see O’Neill v HM Advocate (No.2) (supra.) at [51]. 37. The test of apparent bias is an objective one and the issue is considered from the viewpoint of an observer who is both fair-minded and informed. The qualities of such an observer were described by Lord Hope in Helow v Secretary of State for the Home Department and another:[19] “2. The observer who is fair-minded is the sort of person who always reserves judgment on every point until she has seen and fully understood both sides of the argument. She is not unduly sensitive or suspicious, as Kirby J observed in Johnson v Johnson (2000) 201 CLR 488 , 509, para 53. Her approach must not be confused with that of the person who has brought the complaint. The “real possibility” test ensures that there is this measure of detachment. The assumptions that the complainer makes are not to be attributed to the observer unless they can be justified objectively. But she is not complacent either. She knows that fairness requires that a judge must be, and must be seen to be, unbiased. She knows that judges, like anybody else, have their weaknesses. She will not shrink from the conclusion, if it can be justified objectively, that things that they have said or done or associations that they have formed may make it difficult for them to judge the case before them impartially. 3. Then there is the attribute that the observer is ‘informed’. It makes the point that, before she takes a balanced approach to any information she is given, she will take the trouble to inform herself on all matters that are relevant. She is the sort of person who takes the trouble to read the text of an article as well as the headlines. She is able to put whatever she has read or seen into its overall social, political or geographical context. She is fair-minded, so she will appreciate that the context forms an important part of the material which she must consider before passing judgment.” 38. In the present case, the appellant argues that the fair-minded observer will perceive the mind to have been closed if it appears that the determining judge had determined the same issue on a previous occasion. 39. A significant flaw in the appellant’s argument is that, contrary to Mr McCoy’s submission, the single judge who refuses leave to appeal is not finally determining the application for leave. That is because, pursuant to section 83Y of the CPO, the applicant is entitled to renew the application before the Court of Appeal consisting of three judges. That hearing, as explained above, is not an appeal from the single judge but a rehearing of the same application for leave to appeal. Whilst it is true that the issue before the single judge is the same issue as that on the renewed application before the full court, namely whether to grant leave to appeal, the issue on the renewed application is a reconsideration of the initial, and necessarily provisional, refusal of leave by the single judge and may involve additional grounds and, if admissible, new evidence. The single judge is not bound by any views previously expressed and can change his mind as to the outcome. The decision by the Court of Appeal on the rehearing is a fresh and final determination of the leave application. 40. It is also contended by the appellant that the participation of the same single judge in the Court of Appeal will give the appearance of a closed mind, when it should instead have been open to reconsideration. 41. This contention, however, ignores the important fact that, in applying the test for apparent bias, the consideration by the fair-minded observer is necessarily informed by objective appreciation of the qualities of a judge and the essential characteristics of the judicial function and process. As the High Court of Australia said in Johnson v Johnson:[20] “12. ... At the same time, two things need to be remembered: the observer is taken to be reasonable; and the person being observed is ‘a professional judge whose training, tradition and oath or affirmation require [the judge] to discard the irrelevant, the immaterial and the prejudicial’. 13. Whilst the fictional observer, by reference to whom the test is formulated, is not to be assumed to have a detailed knowledge of the law, or of the character or ability of a particular judge, the reasonableness of any suggested apprehension of bias is to be considered in the context of ordinary judicial practice. ...” 42. The significance of the judicial oath, which in Hong Kong (like in other jurisdictions) imposes on judges a duty to “safeguard the law and administer justice without fear or favour, self-interest or deceit”,[21] is not to be overlooked: it is an important bulwark of judicial impartiality. It expresses the “general rule of the common law ... that judges must apply the law as they understand it to the facts of individual cases as they find them without fear or favour, affection or ill-will, that is, without partiality or prejudice.”[22] The fair-minded and informed observer will be aware of the oath and that judges will generally “try to live up to the high standard which it imposes”: see Davidson v Scottish Ministers [2004] HRLR 34 at [18]. 43. That the participation of the same judge in a subsequent hearing involving the same parties and subject matter does not necessarily give rise to an appearance of bias is illustrated by the case of Sengupta v Holmes [2002] EWCA Civ 1104. That case involved a single judge (Laws LJ) who had refused leave to appeal on the papers and who then sat on the Court of Appeal hearing the substantive appeal after a renewed application for leave to appeal was successful. It was suggested that Laws LJ should recuse himself from sitting in the substantive appeal, having previously refused leave to appeal. The English Court of Appeal rejected this contention and held that there was no apparent bias in the circumstances. Laws LJ considered what would be known to the fair-minded and informed observer, which included the following features. He regarded it as material that, in the ordinary case, a single judge having initially refused leave to appeal will have done so “in the knowledge that, at the option of the applicant, his view may be reconsidered at an oral hearing” and that there was, in the circumstances, no reasonable basis for supposing he might not bring an open mind to bear on the substantive appeal if he subsequently sat on the full court.[23] He regarded a readiness to change one’s mind on an issue to be “a commonplace of all the professions, indeed of the experience of all thinking men and women” and the fact that judges do change their minds under the influence of oral argument to be a central feature of the adversarial system.[24] Another important feature of the English legal culture which Laws LJ considered important was the expectation that a judge would be positively expected to alter an earlier expressed view if objectively persuaded by his colleagues that it should be altered.[25] 44. In my view, all these features of the English common law system described by Laws LJ apply equally to Hong Kong and inform the view of the fair-minded observer in this jurisdiction. So informed, that observer will expect that a single judge who has previously refused leave to appeal will hear the renewed application for leave under section 83Y(3) of the CPO with an open mind and with a readiness to grant leave to appeal if persuaded by the arguments advanced to the full court and the deliberations of his judicial colleagues. Although Sengupta v Holmes concerned an initial refusal of leave on the papers before a substantive appeal hearing, I do not consider there to be any material distinction between that and the circumstances of the present case involving an initial refusal of leave at an oral hearing followed by a renewed application of that same application before the full court. 45. However, that is not to say that there might never be circumstances in which a single judge who has refused leave to appeal would be in a position in which the fair-minded and informed observer would consider there to be a real possibility that the judge might be biased on a subsequent hearing of the underlying merits of the case. In Sengupta v Holmes, Laws LJ gave as examples: (i) where a judge of first instance has committed himself to a view of the facts which he had the responsibility to decide (at [32]); (ii) where he had decided that a party or witness was a crook or a rogue (at [33]); and (iii) where he had expressed himself in such vituperative language that any reasonable person would think him unable to take a fair view of the case if called on to revisit it (at [34]).[26] Since cases of apparent bias are necessarily fact sensitive,[27] this list is not closed and there may be other particular instances in which apparent bias may be shown. Other than in these types of case, though, there is no reason to think that, in the ordinary case of judicial reconsideration of an issue such as leave to appeal, the fair-minded and informed observer would harbour any misgivings as to the ability of the judge to bring a fair and open mind to bear on the issue before him. 46. This conclusion in respect of section 83Y of the CPO is not limited to that particular procedure and other analogous circumstances can be called to mind. For example, a single permanent judge of this Court, may have to express a view on the merits of an underlying application for leave to appeal when faced with a bail application, but, absent unusual circumstances, there would normally not be any question of apparent bias on his part if he were to sit as a member of the Appeal Committee hearing the subsequent application for leave. Similarly, it would be an unusual case where a single permanent judge of this Court, having refused leave to appeal when sitting as part of the Appeal Committee,[28] would be required to recuse himself from sitting on the substantive appeal to the Court on the grounds of apparent bias arising from his prior refusal of leave to appeal. 47. The appellant’s reliance on R v Taito [2003] 3 NZLR 577 in support of the contention that there was structural impartiality in the present case was misplaced. There, on an appeal from the Court of Appeal of New Zealand, the Privy Council concluded that the practice of the Court of Appeal in sitting to determine substantive appeals with a member of the court who had previously sat as part of the court refusing legal aid on the merits of the prospective appeal gave rise to a breach of the New Zealand Bill of Rights Act 1990 section 25.[29] The material part of the Privy Council’s judgment, delivered by Lord Steyn, at [14] reads as follows: “... A delivery Judge was always in an impossible position: he either had no knowledge of the dossier of available information or he was parti pris because he had advised against the granting of legal aid. In Nicholls Tipping J, at p 443, with the agreement of Smellie J at p 461, expressed doubts about this last feature of the practice. In the circumstances the participation of a Judge who had concluded that legal aid should not be granted would have suggested to a fair-minded and informed observer that the Judge was not independent: Porter v Magill [2002] 2 WLR 37 at pp 83 - 84, per Lord Hope of Craighead. The requirement that a judgment of the Court of Appeal must be ‘in accordance with the opinion of the Judges present’ was incapable of fulfilment: s 59 of the Judicature Act 1908. Relying on the fact that three Judges of the Court of Appeal had earlier concluded that legal aid should not be granted, the ex parte decisions were purely formalistic or mechanical acts involving no exercise of judicial judgment. It was the phenomenon of tabulated legalism against which Lord Wilberforce had warned in Fisher. Moreover, the system of ex parte decisions was not authorised by the legislation. It follows that the dismissal of all the appeals under consideration pursuant to the ex parte procedure was of no force or effect.” 48. The position in Taito is distinguishable from the circumstances of the present case in a number of respects. First, the determination of the legal aid application there was, unlike the initial application to a single judge here, a conclusive one and not one subject to review by the full court consisting of three judges who would revisit the same decision. Secondly, the subsequent appeal in Taito was simply a “formalistic or mechanical” act, unlike the full hearing of the renewed application for leave here. Thirdly, the subsequent appeal process in Taito was in any event not authorised by the domestic legislation, unlike the position here (on the basis of the proper construction of the relevant provisions of the HCO and CPO set out in Section E above). G. Whether apparent bias for Macrae JA to sit in renewed application for leave in this case 49. In the CA Leave Judgment, the Court of Appeal said (at [31]): “In our view, there is no risk that a reasonable, fair-minded and well-informed observer would conclude that there is a real possibility, or a real danger that the single Justice of Appeal who has previously determined the leave application would be biased.” With respect, this does not state the law accurately insofar as it may suggest that a single Justice of Appeal could never be disqualified by reason of apparent bias from sitting in the Court of Appeal on a renewed application for leave. On the contrary, despite the conclusion (in Section F above) that there is no structural lack of impartiality in a single Justice of Appeal who has refused an application for leave to appeal under section 83Y of the CPO sitting on the Court of Appeal hearing the renewed application for leave, it remains essential that the application is determined by an independent and impartial tribunal. A judge who may give the appearance of being biased is not such a tribunal and may not sit on the Court of Appeal on the renewed application for leave. 50. It is therefore necessary to consider the appellant’s fallback contention that the circumstances of the present case are such that a fair-minded and informed observer would conclude that there was a real possibility that Macrae JA was biased when sitting on the full Court of Appeal hearing the renewed application for leave. The appellant’s contention is a specific allegation of apparent bias related to the facts of this particular case and, again, Mr McCoy emphasised that it is no part of the appellant’s case that Macrae JA was actually biased. 51. Applying the reasonable apprehension of bias test to the specific facts of this case, I have no doubt that the fair-minded and informed observer would not think that there was any risk that Macrae JA was biased when he came to sit in the Court of Appeal hearing the appellant’s renewed application for leave to appeal. 52. A review of the Single JA Judgment shows that Macrae JA expressed himself in a measured and objective manner and did not use pejorative language in refusing leave to appeal. Contrary to the submission in the Appellant’s Printed Case,[30] Macrae JA did not characterise the applications as “hopeless” and “devoid of any merit”, nor was he required to do so in order to refuse the initial application for leave. He did not make a loss of time order himself (although it should not be thought that it would automatically follow that, had he done so, he could not have sat on the renewed application for leave since the Court of Appeal would have power to revisit his loss of time order and to set it aside). In giving the loss of time warning, he did not indicate that the Court of Appeal should make such an order but merely gave a neutral warning that it might impose such an order if it were to come to the conclusion there was no justification for renewing the application. The context of the decision as a provisional refusal of leave was clear to the single judge since he advised the appellant of his right to renew the application to the Court of Appeal. 53. When the application was renewed before the Court of Appeal, although Macrae JA was, by dint of seniority, the presiding judge, his was but one equal voice amongst three on the court. The CA Judgment was delivered by Pang JA speaking for the court, rather than Macrae JA, indicating that the court’s views were collective. Mr McCoy criticised Pang JA for quoting Macrae JA’s conclusions on the new medical evidence sought to be introduced by the appellant and his conclusion on the application for leave to appeal. This, however, does not indicate a lack of impartiality on the part of Macrae JA or a lack of independence on the part of McWalters and Pang JJA. The fact that McWalters and Pang JJA both agreed with those views simply means that nothing said on the hearing of the renewed application persuaded any of the court that Macrae JA’s initial views were wrong. Indeed, given the strength of the case against the appellant and the weakness of his grounds of appeal, this is hardly surprising. It is important to note that the CA Judgment contained separate discussion of the merits of the application for leave (at [19] and [21]) and a separate and independent conclusion with its order of loss of time (at [22]). There is no suggestion this was simply an adoption of Macrae JA’s earlier views. Pang JA’s comment (at [19]) that the grounds of appeal concerned “nothing that the judge and/or Single Judge had not covered” does not mean that the Court of Appeal did not collectively reconsider afresh those grounds to determine if there was anything new or different to say about them. The brevity of the Court of Appeal’s treatment of the appellant’s grounds of appeal in the CA Judgment was appropriate in view of the weakness of those grounds. 54. For all these reasons, there is no basis for suggesting that the fair-minded and informed observer would think that there was a risk Macrae JA was biased and no reason to interfere with the CA Judgment. H. Answer to certified question of law and disposition of appeal 55. Accordingly, I would answer the certified question of law as follows. Subject to the rules concerning disqualification of a judge from sitting by reason of bias, whether actual or apparent, a single Justice of Appeal who has refused leave to appeal to the Court of Appeal under section 83Y of the CPO may lawfully sit as a member of the panel if the applicant seeks to have his application determined by the Court of Appeal. 56. There being no apparent bias on the part of Macrae JA in sitting on the renewed application for leave, I would dismiss this appeal. Lord Hoffmann NPJ: 57. I agree with the judgment of Mr Justice Fok PJ. Mr Gerard McCoy SC, Mr Albert N B Wong and Mr Kim J McCoy, instructed by Ho Tse Wai & Partners, for the Appellant Mr William Tam SC, DDPP and Ms Janice Cheuk, SPP (Ag.), of the Department of Justice, for the Respondent [1] CPO s.83G provides a right of appeal to the Court of Appeal against sentence. [2] CPO s.82(2) reads: “The appeal may be – (a) on any ground which involves a question of law alone; and (b) with the leave of the Court of Appeal, on any ground which involves a question of fact alone, or a question of mixed law and fact, or on any other ground which appears to the Court of Appeal to be a sufficient ground of appeal; but if the judge of the court of trial grants a certificate that the case is fit for appeal on a ground which involves a question of fact, or a question of mixed law and fact, an appeal lies under this section without the leave of the Court of Appeal.” [3] (2002) 5 HKCFAR 540 at [59]. [4] In DCCC 492/2014, before HH Judge Woodcock. [5] CACC 330/2014, Judgment dated 2 June 2015 (“the Single JA Judgment”). [6] CACC 330/2014, Reasons for Judgment dated 10 December 2015 (“the CA Judgment”). [7] CACC 330/2014 (Yeung, Lam and Lunn VPP), Judgment dated 5 May 2016 (“the CA Leave Judgment”), reported at [2016] 5 HKLRD 79. [8] FAMC 25/2016 (Ribeiro, Tang and Fok PJJ), Determination dated 17 August 2016. [9] Appearing with Mr Albert N.B. Wong and Mr Kim J. McCoy. [10] HCO section 34(2). [11] See Hobbs v Tinling (C.T.) and Company Limited [1929] 2 KB 1 at p.48 per Sankey LJ; and R v Abdroikov [2007] 1 WLR 2679 per Lord Bingham at [14] and Davidson v Scottish Ministers [2004] HRLR 34 per Lord Bingham at [7]. [12] In the same terms as ICCPR Articles 14(1) and 14(5). [13] At [33], citing Lumley v Jamaica, Communication No.662/1995 of the United Nations Human Rights Committee (“UNHRC”) (adopted on 31 March 1999) at [7.3]. [14] See UNHRC General Comment No.32 (CCPR/C/GC/32), 23 August 2007, at [45]. [15] Magistrates Ordinance (Cap.227), s. 104. [16] See, generally, the Guide to Judicial Conduct (October 2004), Part D. [17] (“ECHR 6(1)”); which is in similar terms to BOR 10 and ICCPR 14(1). [18] (2003) 6 HKCFAR 322. [19] [2008] 1 WLR 2416. [20] (2000) 201 CLR 488 per Gleeson CJ, Gaudron, McHugh, Gummow and Hayne JJ at [12]-[13] (footnotes omitted). [21] The full text of the judicial oath (set out in the Oaths and Declarations Ordinance (Cap.11), Schedule 2, Part V) reads: “I swear that, in the Office of a Judge of the Judiciary of the Hong Kong Special Administrative Region of the People’s Republic of China, I will uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China, serve the Hong Kong Special Administrative Region conscientiously, dutifully, in full accordance with the law, honestly and with integrity, safeguard the law and administer justice without fear or favour, self-interest or deceit.” [22] Bolkiah & Ors v The State of Brunei Darussalam & Anor [2007] UKPC 62 per Lord Bingham at [14] citing Locabail (UK) Ltd v Bayfield Properties Ltd [2000] QB 451 at [2]. [23] [2002] EWCA Civ 1104 at [35]. [24] Ibid. at [36] and [38]; see also, his reference at [14] to the judgment of Peter Gibson LJ in Mahomed (CHANI 1999/1003/B3, 3 February 2000). [25] Ibid. at [39]. [26] Falcon Private Bank Ltd v Borry Bernard Edouard Charles Ltd (2014) 17 HKCFAR 281 is an example of situation (iii). [27] See Davidson v Scottish Ministers (supra.) at [17]; JSC BTA Bank v Ablyazov and others (No.9) [2013] 1 WLR 1845 at [65]; and Falcon Private Bank Ltd v Borry Bernard Edouard Charles Ltd (supra.) at [29]. [28] As a matter of the Appeal Committee’s practice, this would be an exceptional case. [29] Equivalent to BOR 10 and BOR 11(4). [30] At [67]. 
The Court: 1. The appellant, who at the time of his alleged offences was Chief Executive of the Hong Kong Special Administrative Region, and President of the Executive Council (“ExCo”), was charged with an offence of bribery, and a further offence of misconduct in public office, arising out of his dealings with Wong Cho-bau, a mainland businessman, in connection with the refurbishment and re-decoration of a residential property in Shenzhen which the appellant proposed to occupy following his retirement from office. The property was owned by a company controlled by Mr Wong. The appellant was tried before Andrew Chan J and a jury. The jury could not agree on the first charge. The appellant was convicted on the second charge, and sentenced to a term of imprisonment. An appeal against his conviction was dismissed by the Court of Appeal. By leave, he now appeals to this Court. There was a third charge against the appellant, arising out of one aspect of the property dealings, of which the appellant was acquitted by the jury. It is presently of marginal significance. There was a re-trial of the first charge, but, again, the jury could not agree. 2. The essence of the charge of bribery was that the arrangements concerning refurbishment and re-decoration of the residential property were a bribe, being an inducement or reward in connection with the appellant’s handling of certain applications in relation to broadcasting licences made by Wave Media Limited (“WML”), later re-named Digital Broadcasting Corporation Hong Kong Limited (“DBC”), a company in which Mr Wong held shares. 3. The essence of the charge of misconduct in public office was failing to disclose to, or concealing from, the ExCo, the dealings and negotiations concerning the residential property when the ExCo was considering the applications. 4. Although the particulars of the second count were framed in the alternative as “by failing to declare or disclose…or by concealing”, the primary case for the prosecution was one of deliberate concealment, the object of which was to hide the bribe. There was an alternative case. The jury’s failure to agree on the bribery charge requires the nature of the alternative case to be analysed. 5. According to the defence, the dealings concerning refurbishment and re-decoration of the subject premises were arm’s-length, and commercial. The appellant and his wife expected to occupy the premises under a lease for some three years following his retirement in mid-2012, after they were renovated in accordance with their requirements. The proposed rental was to be the market rental, having regard to the renovations. The proposals conferred no improper benefit on the appellant. They were unconnected with the broadcasting applications which, for their part, were straightforward, even if commercially important, matters that were dealt with in accordance with proper administrative process, and which had an outcome to which no one could take exception. 6. If the jury had convicted the appellant on the first charge, conviction on the second was almost inevitable. Acquittal on the first count, however, would not require acquittal on the second count. It was still open to the jury to conclude that the dealings and negotiations in respect of the property should have been disclosed, and that their non-disclosure amounted to the offence of misconduct in public office. That would require consideration of a number of issues which, although they could be resolved easily if the dealings amounted to bribery, could not be resolved so easily in the absence of such a finding. 7. The trial judge told the jury that they must consider counts 1 and 2 separately. The appellant complains that he did not properly explain to them how to do that. His summing up, the appellant argues, did not address, or did not address adequately, the legal principles and factual issues which arose if the second count fell to be considered on the basis that the allegation of bribery was not made out, and was to be regarded, not as the corollary of the first count, but as a true alternative. In consequence, it is said, there was a failure of process at the trial. 8. The resolution of the appeal requires attention to the way the case was conducted, and to the elements of the offence of misconduct in public office. One general principle, however, is clear. As the charges against the appellant were framed, the jury were required to consider both the first and second counts separately. They were so directed. They might have come to consider the second count with a view on the first count that was at least partly favourable to the appellant. Since the case was not left to the jury on the basis that the two counts must stand or fall together, the judge was required to explain the elements of the second charge in a way that related them to a possible view of the facts that did not involve bribery, and to identify the issues to be determined on that hypothesis in a manner that fairly reflected the defence case.[1] The background facts[2] 9. An investigation into the matter followed media publicity, in February 2012, concerning allegedly improper associations between the appellant and prominent business people. [3] The appellant responded to those allegations in media interviews at the time, including interviews on the “Talkabout” programme on 22 February 2012, and the “Beautiful Sunday” programme on 26 February 2012. The appellant did not give evidence at his trial but transcripts of the interviews were tendered, and were treated as his account, or explanation, of his conduct, including his conduct in relation to the Shenzhen property. 10. In late 2009, the Government of the Hong Kong Special Administrative Region (“HKSAR”) invited applications for sound broadcasting licences. In April 2010, four such applications were made, one of them by WML. Ultimately three of the applications, including that of WML, were successful. The fourth was withdrawn. Evidently the applications were not in competition with one another. Between 2010 and 2012 Mr Wong held 20% of the shares in WML. [4] Other shareholders were prominent Hong Kong identities, including Mr Arculli, a member of the ExCo [5]. In late 2010, WML also applied to surrender an earlier licence it held. Between July and November 2011, a further application was made to allow one of the WML (by then DBC) shareholders, Mr Arthur Li, to be a director and Chairman notwithstanding that he was disqualified by a provision of the relevant regulations. The two latter applications were also successful. 11. There was no evidence to suggest anything irregular about the applications, or the administrative process by which they were handled. The licence applications were dealt with by the ExCo after and in accordance with favourable reports by the Broadcasting Authority and the Commerce and Economic Development Bureau.[6] There was no evidence or argument to suggest any reasonable basis upon which the ExCo could have refused them. The appellant participated in the deliberations of the ExCo. His participation was not alleged itself to amount to misconduct for the purposes of the second count. 12. The appellant made no declaration of interest to the ExCo. Mr Arculli did, and took no part in the ExCo’s deliberations. Mr Arculli, as a shareholder in WML, had a personal financial interest in the applications. On many other occasions during his term of office the appellant had made declarations of interest. He was obviously aware of the importance of declaring interests where necessary. 13. In early 2010, the appellant had entered into discussions with Mr Wong concerning the Shenzhen property. The property was in need of renovation. In the Beautiful Sunday interview the appellant said, in substance, that the proposal was that he or his wife would take a lease of the property for three years commencing on 1 July 2012, following his retirement, that the rent would be at market rental value, which was later agreed at RMB800,000 per annum, and that the owner of the property (a company controlled by Mr Wong) would renovate the premises at its expense and in accordance with his and Mrs Tsang’s requirements.[7] A well-known interior designer, Mr Ho, was later engaged by the company for that purpose. (The appellant’s recommendation of Mr Ho for an award in the Hong Kong system of honours was the subject of the third charge, of which the appellant was acquitted). 14. It was the agreement to refurbish the premises (which previously consisted of a multi-level recreational facility at the top of an apartment building) at the expense of the owner which was said to constitute the advantage that was an inducement or reward in connection with the broadcasting applications, and that amounted to a bribe. The cost of the refurbishment was to be around HK$3.5 million. Mr Ho’s fees were to be HK$350,000. There was no evidence at trial to show that RMB800,000 per annum was not a proper market rental having regard to the anticipated state of the property at the time the refurbishment was completed. 15. On 2 November 2010, the first of the above three applications was approved in principle. 16. On 17 November 2010, the sum of RMB800,000 was paid by Mrs Tsang to a company related to the company that owned the Shenzhen property. The prosecution case was that this payment was never properly explained. 17. On 24 March 2011, the first and second of the above applications were finally approved. 18. On 20 January 2012, the third application was approved. 19. The refurbishment of the property began in late 2011. 20. On 20, 21 and 22 February 2012 the media reports referred to earlier surfaced. The media responses of the appellant were a form of damage control. On 26 February 2012 he told the media that he had rented the property.[8] He referred to the term and the annual rent, as did a press release of the same day.[9] (The Court of Appeal erroneously said “the lease agreements and their contents were not spoken to on 26 February 2012”.[10]) An investigation by the Independent Commission Against Corruption commenced. Ultimately, in September 2013, the solicitors for the appellant produced a lease of the property dated 21 February 2012, said to support the explanation the appellant had given to the media. The prosecution did not accept the genuineness of the document. The prosecution case was that the purported lease was part of an attempt by the appellant to put an innocent complexion on dealings, including payments and transfers of money, that were corrupt, although the true nature of the arrangements concerning the residential property remained obscure. As far as the bribery charge was concerned, however, it was the refurbishment and re-decoration of the property at the expense of the owner of the property that was said to constitute the bribe. It follows that, even if the lease was a sham, there must have been some arrangements concerning future occupation of the property by the appellant and his wife. The indictment 21. The first and second counts in the indictment, as amended, were as follows: First Count STATEMENT OF OFFENCE Chief Executive accepting an advantage, contrary to sections 4(2B)(a) and 12 of the Prevention of Bribery Ordinance, Cap. 201. PARTICULARS OF OFFENCE TSANG Yam-kuen, Donald ("Donald Tsang"), being the Chief Executive of the Hong Kong Special Administrative Region ("Chief Executive") and President of the Executive Council ("ExCo"), between the 1st day of January 2010 and the 30th day of June 2012, without lawful authority or reasonable excuse, accepted an advantage, namely the refurbishment and re-decoration of a three-storey residential property situated at East Pacific Garden, Futian, Shenzhen as an inducement to or reward for or otherwise on account of Donald Tsang's performing or abstaining from performing or having performed or abstained from performing acts in his capacity as the Chief Executive and President of ExCo, namely, considering and making decisions in relation to applications made by Wave Media Limited ("WML") (subsequently renamed Digital Broadcasting Corporation Hong Kong Limited ("DBC")), in particular: (a) The application of WML for a sound broadcasting licence for the provision of digital audio broadcasting services submitted to the Commerce and Economic Development Bureau ("CEDB") in April 2010 and approved in principle and formally granted by Donald Tsang acting in his capacity as the Chief Executive-in-Council at meetings of ExCo held on the 2nd day of November 2010 and the 22nd day of March 2011 respectively; (b) The application of WML to surrender a sound broadcasting licence for the provision of Amplitude Modulation radio services submitted to the CEDB in September 2010 and approved in principle and formally granted by Donald Tsang acting in his capacity as the Chief Executive-in-Council at meetings of ExCo held on the 2nd day of November 2010 and the 22nd day of March 2011 respectively; (c) The applications made by DBC between July and November 2011 for LI Kwok-cheung, Arthur to exercise control of DBC as a director and chairman of the board of the company and approved by Donald Tsang acting in his capacity as the Chief Executive-in-Council at a meeting of ExCo held on the 20th day of January 2012. Second Count STATEMENT OF OFFENCE Misconduct in public office, contrary to Common Law and punishable under section 101I(1) of the Criminal Procedure Ordinance, Cap. 221. PARTICULARS OF OFFENCE TSANG Yam-kuen, Donald, being the holder of a public office, namely the Chief Executive of the Hong Kong Special Administrative Region and President of the Executive Council ("ExCo"), between the 1st day of January 2010 and the 30th day of June 2012, in the course of or in relation to his public office, without reasonable excuse or justification, wilfully misconducted himself by failing to declare or disclose to, or by concealing from, the ExCo his dealings and negotiations with Wong Cho-bau, the major shareholder of Wave Media Limited ("WML") in respect of a three-storey residential property situated at East Pacific Garden, Futian, Shenzhen when he, in his capacity as the Chief Executive and President of ExCo, was involved in decision making in relation to applications made by WML (subsequently renamed Digital Broadcasting Corporation Hong Kong Limited ("DBC")) for: (a) a sound broadcasting licence for the provision of digital audio broadcasting services submitted to the Commerce and Economic Development Bureau ("CEDB") in April 2010; (b) the surrender of a sound broadcasting licence for the provision of Amplitude Modulation radio services submitted to the CEDB in September 2010; (c) permission for LI Kwok-cheung, Arthur to exercise control of DBC as a director and chairman of the board of the company. Conviction, sentence and appeal 22. On 17 February 2017, the appellant was convicted on count 2. He was sentenced to 20 months’ imprisonment. He appealed to the Court of Appeal against his conviction and sentence. On 20 July 2018, the Court of Appeal delivered its judgment dismissing the appeal against conviction, but allowing the appeal against sentence and reducing the term of imprisonment to 12 months (a term that has now been served).[11] 23. On 20 December 2018, the Appeal Committee of this Court granted leave to appeal on the basis of two points of law expressed in the following questions. (1) In respect of the mental element of the offence of misconduct in public office contrary to common law: (a) What is the proper direction to the jury on the element of wilful misconduct? To what extent is it necessary for the jury to be directed that the official must know that his conduct was unlawful (and/or disregarded the risk of such) in addition to a direction that the act itself must be “deliberate”? (b) In circumstances where the misconduct in question is premised upon an allegation of failure to declare or disclose, or concealing, a conflict of interest, to what extent is the trial judge required to give directions or assistance to the jury on how they are to resolve the issue of whether the accused was under a duty to disclose? If there are two possible failures (one corrupt, one non-corrupt) to what extent should the duty direction be modified or explained? (2) In respect of the element of seriousness in the offence of misconduct in public office contrary to common law, to what extent is the trial judge required to provide assistance to the jury on how the element of seriousness is to be assessed? Is the Court of Appeal entitled to draw any conclusions where a jury has failed to agree that the defendant acted corruptly? 24. Leave was also granted on the basis of a possible substantial and grave injustice, the issue being expressed as follows: In the context of Count 2 standing on its own, and in circumstances of a hung jury on Count 1, whether the trial judge’s directions on the element of wilful misconduct were flawed by a failure to direct the jury on the need to find that the [appellant] knew that his conduct was unlawful in accordance with Sin Kam Wah v HKSAR (2005) 8 HKCFAR 192 at §42, thus leaving it open for the jury to convict on the basis that the [appellant] simply made a deliberate decision not to disclose his dealings with Mr Wong Cho-bau in respect of the property that was the subject of the charge, as opposed to the [appellant] deliberately concealing those dealings with the requisite knowledge. Misconduct in public office and non-disclosure 25. For the purposes of the law of Hong Kong, the elements of the common law offence of misconduct in public office are as follows:[12] the offence is committed where: (1) a public official; (2) in the course of or in relation to his public office; (3) wilfully misconducts himself; by act or omission, for example, by wilfully neglecting or failing to perform his duty; (4) without reasonable excuse or justification; and (5) where such misconduct was serious, not trivial, having regard to the responsibilities of the office and the officeholder, the importance of the public objects which they serve and the nature and extent of the departure from those responsibilities. It is (3) and (5) that are of present relevance. Element (4) is an additional requirement that may become relevant where a wilful failure to meet standards has occurred but is sought to be excused or justified.[13] 26. The history of the offence was considered by this Court in Shum Kwok Sher v HKSAR.[14] The Court rejected an argument that the offence was impermissibly vague and could not survive scrutiny under modern human rights jurisprudence. Sir Anthony Mason NPJ[15] cited with approval what was said by PD Finn in an article entitled “Official Misconduct”:[16] “…official misconduct is not concerned primarily with the abuse of official position for pecuniary gain, with corruption in the popular sense. Its object is simply to ensure that an official does not, by any wilful act or omission, act contrary to the duties of his office, and does not abuse intentionally the trust reposed in him.” 27. Shum Kwok Sher v HKSAR was a case concerning awarding of public contracts to a company without the necessary qualifications by a public official who had a family relationship with the controllers of the company. The particulars of the misconduct asserted three cumulative elements: failing to disclose the relationship; failing to abstain from the decision making process; and displaying partiality by awarding the contract to an unqualified tenderer. 28. The relevant misconduct may or may not involve an act proscribed by a statute such as, in Hong Kong, the Crimes Ordinance,[17] or the Prevention of Bribery Ordinance.[18] In HKSAR v Hui Rafael Junior[19] this Court examined the relationship between the common law offence of misconduct in public office and the statutory offence of bribery. Bribery is an example of a particular form of misconduct that became the subject of special legislative treatment in common law jurisdictions in the nineteenth and twentieth centuries. 29. The common law offence does not necessarily involve venality. For example, A-G’s Reference (No 3 of 2003)[20] concerned an alleged failure by police officers to take proper care for the safety of a person in custody. HKSAR v Ho Hung Kwan Michael[21] concerned a medical officer who gave favourable treatment to members of his family. Neglect of duty or abuse of power may amount to criminal misconduct even though the accused gains no personal benefit from it. 30. In Shum Kwok Sher v HKSAR[22] Sir Anthony Mason NPJ pointed out that the misconduct may take various forms, ranging from fraudulent conduct, through nonfeasance of a duty, misfeasance in the performance of a duty, an exercise of a power with a dishonest, corrupt or malicious motive, acting in excess of power or authority with a similar motive, or oppression. In that case the appellant’s duties in relation to conflicts of interest and the obligation not to favour relatives or friends were the subject of written guidelines. The appellant was found to have set out to favour the company controlled by his relatives. His motive for concealment of his relationship was his desire to advance the interests of that company. His conduct was dishonest, as he must have realised.[23] The concealment of his relationship with the company was an integral part of his scheme to benefit the company by awarding it contracts notwithstanding its lack of proper qualification. 31. Just as the relevant conduct may or may not in itself be illegal, so also it may or may not be contrary to rules or guidelines, and attract possible disciplinary sanctions. It is the capacity of the offence, which has sometimes been described as “misbehaviour in a public office”,[24] to criminalise acts or omissions that would otherwise be no more than departures from civil standards of behaviour that has led to complaints about the uncertainty of its scope, and to repeated emphasis on the importance of the elements of wilfulness and seriousness. The offence strikes at abuse of powers or duties, not at errors of judgment. “A mistake, even a serious one, will not suffice”.[25] As to seriousness, the English Court of Appeal said, in R v Dytham,[26] (a case concerning a police constable’s failure to intervene in a fight in which a man was beaten to death): “The allegation made was not of mere non-feasance but of deliberate failure and wilful neglect. This involves an element of culpability which is not restricted to corruption or dishonesty but which must be of such a degree that the misconduct impugned is calculated to injure the public interest so as to call for condemnation and punishment. Whether such a situation is revealed by the evidence is a matter that a jury has to decide.” 32. In a case where the exercise of a power, such as the grant of a licence, is not itself alleged to be, or to be part of, the relevant misconduct, but the misconduct is said to consist of not making disclosure in the course of the exercise of the power, a characterisation of the conduct as “deliberate” may involve an ambiguity that has a bearing on the elements of wilfulness and seriousness. A considered decision not to disclose information may be deliberate in the sense that it is not inadvertent, but it may at the same time result from an error of judgment. To describe a decision not to disclose something as “deliberate concealment” adds a pejorative element. Where, as in the present case, the particulars of the alleged misconduct are failing to declare or disclose or concealing, there may be a need for care in distinguishing between the alternative possibilities. Concealment implies dishonesty. Failure to disclose, even if deliberate, may be the result of an error of judgment. 33. In HKSAR v Ho Hung Kwan Michael[27] Chan ACJ, with whom the other members of the Court agreed, said:[28] “(29) In cases where corruption, dishonesty or other illegal practices are involved, it is not necessary to specifically consider the consequences of the misconduct in deciding whether it is serious enough as to constitute the offence of misconduct in public office. The misconduct speaks for itself: the seriousness of the consequences of such corrupt, dishonest or illegal practices will be obvious. (30) In other cases, where corruption, dishonesty or other illegal practices are not involved, the consequences of the misconduct may not be obvious.” In a case of non-disclosure which falls into the latter category, leaving a jury to its own devices in assessing seriousness may be dangerous. It has been said by the English Court of Appeal, with reference to other forms of misconduct, that it is not sufficient simply to tell the jury that the conduct must be so serious as to amount to an abuse of the public’s trust in the office holder without giving them assistance as to how to assess seriousness and harm by putting the conduct into its proper factual context.[29] The same applies to non-disclosure. Deliberate non-disclosure, as distinct from deliberate concealment, may fall into the second of the two categories identified by Chan ACJ. 34. In the present matter, the primary prosecution case was that the dealings between the appellant and Mr Wong in respect of the renovations to the Shenzhen property were corrupt, the form of corruption being that alleged in Count 1, and that they were deliberately concealed in order to hide that corruption. If that case had been accepted, there would have been a conviction on Count 1, and the elements of wilfulness and seriousness in respect of Count 2 would have presented little difficulty. That case, however, was not accepted by some members of the jury. If the dealings concerning the renovations were not corrupt, then issues as to the wilfulness and seriousness of their non-disclosure would become prominent. The central question in this appeal is whether the jurors who were not prepared to convict on Count 1, but who convicted on Count 2, were given appropriate guidance on how they were to approach those issues. Disclosure obligations 35. There was evidence from a senior official as to Civil Service Bureau Circulars and other published requirements concerning conflicts of interest, and disclosure of interests. The primary obligation on senior officers was to avoid a situation giving rise to real or potential conflicts of interest. 36. In the context of proceedings in the ExCo, disclosures of actual or possible conflicts were made to the Council, not to the public. Even so, one of the objects served by disclosure is to give the person or persons to whom the disclosure is made the opportunity to seek further information. Disclosure may be an iterative process. 37. The appellant relied on the fact that there was no obligation to declare rented property. This was in the context of the annual declaration of registrable interests required to be made by members of ExCo. While properties owned by members should be entered on the register, rental properties need not.[30] It does not follow, however, that current negotiations about future rental arrangements need never be disclosed. 38. It is not easy to understand how, on the prosecution’s primary case, the appellant could have made a full disclosure about his dealings and negotiations with Mr Wong in respect of the Shenzhen property without revealing facts which showed that a bribe was involved. Anything less would be misleading. On the primary case, the appellant was being prosecuted both for taking a bribe and, additionally, for not disclosing that he had taken a bribe. There were frequent references in argument to the appellant being “hopelessly compromised”. That would certainly be correct if this were a case of bribery. It seems superfluous, however, to suggest that the appellant’s duty was to disclose that he was hopelessly compromised because he had been bribed. On that basis it might be thought that the substantive misconduct consisted of taking a bribe and then participating in the decision-making process. That reinforces the view that, for the second count to be regarded as a true alternative to the first count, the consequences of accepting (at least as a possibility) a non-corrupt explanation of the appellant’s dealings and negotiations with Mr Wong required careful analysis. 39. In the Respondent’s Case in this Court it was said: “7.4 Count 2 was an allegation of misconduct in public office. It alleged that the Appellant abused his position as CE by failing deliberately to declare or to disclose to the ExCo his negotiations with Wong in respect of the Shenzhen property. 7.5 In the course of ExCo meetings, the Appellant appeared to be acting impartially in the public interest. In fact, he was involved in furthering his own private interests.” 40. The reference to the appellant being “involved in furthering his own private interests” is another example of ambiguity. If it means that he was pursuing a corrupt benefit, that is one thing. If it means no more than that he was making commercial arrangements for his accommodation after retirement, that is something quite different. Whether the latter required disclosure, what the form of such disclosure might be, and what degree of seriousness attached to non-disclosure, were matters that would appear in a different light. Perhaps there were possibilities other than the two just mentioned, but, if so, they were never spelled out. The prosecution case on non-disclosure 41. It was the prosecution’s primary case (corruption) that was emphasised in opening and closing address. Relatively little was said about Count 2 as a true alternative, and what was said reflected the ambiguities noted above. 42. In opening, and closing, counsel said that Counts 1 and 2 were linked. “[T]he counts on the indictment all fit together, and they all link into the Shenzhen property”.[31] The essence of Count 2 was said to be that the appellant was concealing his links (to Mr Wong and the property) even though he was hopelessly conflicted and compromised.[32] The appellant knew that if he made a disclosure to ExCo it would lead to the property, which was secret, and “the reason why it was secret was because it was a corrupt bargain, part of a corrupt bargain”.[33] The information “was concealed deliberately and for an obvious reason: it would have led to a declaration of the true nature of his relationship with Wong”.[34] It was not a good faith transaction. It was not honest. “The prosecution case is that it was clearly corrupt; highly valuable, secretive…and it created an obvious and flagrant conflict of interest”.[35] 43. In closing, counsel for the prosecution referred only briefly to the alternative. He said[36] “Whatever the position, whatever you make of count 1 of the indictment, it was still an obligation of disclosure on the defendant when he was presiding before ExCo.... Just ask yourselves this: what would have been on the defendant’s mind throughout this period…So he had all this period of time to think about the negotiations, the redecoration, the redesign, and it was clearly a conscious and deliberate decision not to reveal what the true positon was, even though it was a classic conflict of interest situation.” 44. Assuming the expression “whatever you make of count 1” means, or includes, “even if you do not find [the appellant] guilty under count 1”, then the “true position” and the “classic conflict of interest situation” must mean, or include, dealings and negotiations without the element of corruption which was the essence of the primary case. There was no form of corruption alleged other than bribery. The circumstances might well have been regarded as savouring of cronyism, and transfers of money revealed by the evidence, and the late production of the lease agreement, were suspicious, but a sense that something “untoward” (to take up a term used by the prosecution in argument in this Court) was going on would not be a substitute for the findings of fact necessary to justify a conclusion of criminal misconduct. The defence case on non-disclosure 45. In his radio interview on 26 February 2012 (the Beautiful Sunday interview) the appellant said that he made it clear to Mr Wong that the lease had to be at the market rate and when the agreement was actually signed he again sought and received confirmation that the rental was at the market rate. He said he had wanted the matter to be handled “in low profile”. The lease was only short-term. As to his state of mind, he said:[37] “Look. Let me tell you this first. I mean, I had, at that time, in examining, at the time of discussing this issue at the Executive Council, I did not make any declaration. However, let me explain it to you. As regards this, in issuing a broadcasting licence, the Hong Kong SAR Government has a very strict system. It is not for the Executive Council itself to be, er, er, to be, er- er- er- er- to be the judge, as well as to do the research, as well as to approve--er, to, everything in relation to considering whether or not it has reached a pass and how it would be operated in the market, we have the Broadcasting Authority to do--the Broadcasting Authority to handle this. They will submit a report to the Executive Council. This time, the report was, they had a unanimous consent for approval. Therefore, the involvement of the Executive Council was very little. In case we had to work on it, if we were to overturn (their recommendation), there must be a very special ground. At that time, there was one thing that I considered. Indeed and truly, I didn't give any thought at all that I, in Shenzhen, not in Hong Kong, had an intention, at that time, it was not yet rented, the tenancy agreement was not yet entered, there was the intention (to do it), wanting to rent a place, which was in his, being a property of a shareholder of the company. In addition, the tenancy was to rent for a, a short period of time. At that time, indeed and truly, it did not occur to me that a declaration was required. If a declaration had to be made, in fact, it was pretty farfetched. Well, the decision would not be influenced. However, I felt, let me tell you this, at that time, I did not declare it. There was a good, a good reason, a good reasoning, for not making declaration. Indeed and truly, I personally did not realize there was any conflict of interest re-re-re-re—regarding commercial conflicts. Most important of all, (I was) aware that, at that time, we were, were handling the matter at the market rate. Moreover, at that time, no tenancy agreement was entered yet. I only had an intention to rent it. Well, now, it influences, that is, not a, not, re-regarding the comments, the recommendation, it was entirely brought up by the Broadcasting Authority unanimously. The Executive Council did not indicate that, huh, it had to overturn or to amend its recommendation. Hm.” 46. The appellant moved between stating objective facts, recalling his state of mind, and commenting on his obligations. It would hardly be fair, however, to say that his explanation for not making a disclosure was that he simply overlooked the issue. 47. The trial judge directed the jury[38] that if the account given by the appellant was or may be true, then he must be acquitted. He also told the jury, conventionally, that they could accept part and reject part of any evidence.[39] Some parts of what the appellant said were capable of being accepted or rejected as a whole. What he said about his state of mind in relation to non-disclosure could not be dealt with so simply. For example, the jury might have rejected any suggestion that he did not think about disclosure (if that is what he was intending to suggest), but could have had more difficulty with his contention that he regarded an obligation of disclosure, in the circumstances as he described them, as “farfetched”. The former might have been regarded as obviously implausible, but the latter, on his version of events, is not. 48. The possibility of an error of judgment about disclosure was raised in defence submissions. Counsel said:[40] “He may think there’s no conflict; you may disagree. But even if you think he’s got it wrong, that is not a crime. It is only a crime if he deliberately, knowing that he ought to disclose, fails to because he is dishonest and corrupt.” 49. At trial, defence counsel did not accept that there was an obligation of disclosure, the dealings being at arm’s-length, and commercial, and having no connection with the broadcasting applications, which were themselves uncontroversial. That is consistent with what the appellant said in the Beautiful Sunday interview. Even if there were an obligation of disclosure, it was argued that the non-disclosure was not wilful misconduct, and was not of a degree of seriousness to attract criminal liability. Counsel sought from the trial judge[41] a direction on seriousness in accordance with the decision in R v Chapman.[42] In response, counsel for the prosecution[43] referred to the judgment of Chan ACJ cited at [33] above, referring to cases where corruption, and dishonestly failing to make declarations of serious conflicts of interest, is comfortably through the threshold of seriousness. The summing up at trial 50. The trial judge, on the issue of wilfulness, said:[44] “‘Wilfully’ in this context means deliberately. Deliberately, rather than by accident and inadvertence or oversight. So members of the jury, you have to be satisfied that the act or omission was deliberately done and it was not merely overlooked by the defendant. So again, if the account put forward by the defendant in Beautiful Sunday is true or may be true, then you should acquit the defendant, because overlooking is not enough for conviction. It has to be deliberately alone.” 51. On the issue of seriousness, the trial judge said:[45] “So it has to be serious, not trivial, and you judge whether it is serious enough to warrant his conviction of the offence by considering the defendant’s responsibilities as well as the responsibilities of the office and the importance of the public objects which he, as the Chief Executive and President of the Executive Council, served, and the nature and extent of his departure from those responsibilities.” 52. On the topic of separate consideration of Counts 1 and 2, the trial judge pointed out that the particulars of Count 2 did not allege bribery or corruption and said:[46] “So if, at the end of your deliberation in relation to Count 1, you come to the conclusion that there exists no corrupt practice, you still have to consider the above, the specific allegations that I have just mentioned. You still have to go back to Count 2, go back to the particulars of the offence. You still have to consider the specific allegations, whether they were deliberately made, not by accident, not by inadvertence, not by oversight, without reasonable excuse or justification. And you have to decide whether the defendant, as a public official, was placing himself in a serious conflict of interest where his duties as a public official were hopelessly compromised.” 53. The judge then took the jury on a comprehensive review of the evidence, reminding them of what each witness said. Towards the conclusion of that review[47] he reminded the jury, after referring to evidence about Mr Barrie Ho and his engagement in respect of the Shenzhen property, that in respect of Count 2, counsel for the appellant said the arrangements concerning the lease and the renovations were bona fide and commercial and gave the appellant no interest in the WML applications, which themselves were uncontroversial. Accordingly, it had been argued there was no need for a declaration of interest. 54. It is submitted for the appellant that what is set out above was inadequate to deal with the way Count 2 was to be approached if the prosecution’s primary case, of concealment of corruption in the form of bribery, was not accepted. The adequacy of the directions on Count 2 55. Underlying the complaints about the way Count 2 was left to the jury is the ambivalence of the prosecution’s alternative case, which fell to be considered in the absence of a finding of bribery. This is exemplified in the Respondent’s Case in this Court, para 18.1: “While neither a prosecutor nor the trial court is required to identify an ‘alternative case theory’, the basis of the case was obvious. The Appellant had deliberately concealed the truth knowing that his relationship with Wong would call his integrity into question. Even in the absence of a conviction on Count 1, the relationship between the Appellant and Wong was suggestive of impropriety and, the Appellant was in a position of seriously divided loyalties. The misconduct was a serious abuse of office.” 56. The primary prosecution case on Count 2, as well as its case on Count 1, was that the reason why the appellant was “hopelessly compromised”, and the reason why his dealings with Mr Wong were kept secret was that those dealings were corrupt, and that the appellant had taken a bribe. That was the alleged “true nature of his relationship with Wong”. 57. It is understandable that, for forensic reasons, counsel might not have wanted to be seen to retreat from an allegation of corruption to a weaker contention that there was a relationship suggestive of some unspecified and unknown impropriety, giving rise to divided loyalties. However, if the jury were to be invited to convict on that basis, then such an approach to the case required explanation. References in argument, and in the summing-up, to an obvious conflict of interest, and to the appellant being hopelessly compromised, and to deliberate concealment, took their colour from what was alleged to be the “true nature of the relationship” that required disclosure. The only alleged relationship specified was that of the giver and taker of a bribe. No more anodyne version, based on “divided loyalties”, ever surfaced. 58. The trial judge, as appears from [52] above, referred briefly to the possibility that the jury might come to the conclusion that there existed no corrupt practice. He did not, however, elaborate on the possibility of some non-corrupt “impropriety”. He told the jury to go back to the particulars of Count 2. (If they did that, they would find an allegation of failing to declare or disclose, or concealing, the dealings and negotiations about the Shenzhen property, unaccompanied by any allegation of impropriety in those dealings.) He then told them to consider whether the failure to declare or disclose, or the concealment, was deliberate, which he contrasted with “not by accident, not by inadvertence, not by oversight [and] without reasonable excuse or justification”. 59. That direction was consistent with what the judge had said earlier (see [50] above) as to the meaning of “wilfully” in explaining the elements of the offence charged in Count 2. He contrasted that with “overlooking”, which he said was the account put forward in the Beautiful Sunday interview. The reason he gave for the direction that the jury should acquit if the Beautiful Sunday account might be true was: “… because overlooking is not enough for conviction.” The plain implication was that “wilful” meant deliberate, in the sense that the question of disclosure was not overlooked. 60. Part of the defence case, also put forward on the basis of what was said in the Beautiful Sunday interview, was that the appellant regarded a requirement for disclosure of his dealings with Mr Wong as far-fetched and, further, that, even if the jury disagreed with that, what was involved was an error of judgment. That was consistent with a deliberate decision not to disclose, but one that was not wilful. There was no reference to that in the directions on wilfulness. (If the judge’s direction meant that such a defence could only succeed if the jury regarded the decision as reasonable, that would itself have involved an error of law, but it does not appear to have been taken that way by counsel.) 61. In a note on directions handed up by the prosecution prior to the summing-up, with which the defence expressed no disagreement, it was submitted: “‘Wilfully’ in this context means deliberately, rather than by accident or inadvertence or oversight, in the sense that the Defendant either knew his conduct was unlawful or deliberately disregarded the risk that his conduct was unlawful”.[48] In closing submissions, the defence made the argument in para [48] above. After the summing-up the defence invited the judge to direct on knowing unlawfulness.[49] 62. A failure by a decision-maker to disclose an interest in the subject matter of the decision may be deliberate in the sense that the decision-maker thought about disclosure and decided against it, but not wilful because the decision-maker did not know, or believe, there was an obligation to disclose in the circumstances of the case, and did not disregard the risk of there being such an obligation. 63. The Court of Appeal, consistently with the argument of the respondent which was later repeated in this Court, answered the criticism of the directions on wilfulness in two ways. 64. First, it was said, on a proper analysis of the conduct of the trial, and the summing-up, the jury, as shown by their verdicts, must have concluded that the appellant had deliberately concealed the true nature of his relationship with Mr Wong and therefore the jury must have rejected the factual hypothesis upon which the supposed need for further direction arose. The prosecution case, from beginning to end, was one of deliberate and dishonest concealment and that must have been accepted by the jury.[50] 65. Furthermore, it was said, the judge directed the jury that if they accepted that the Beautiful Sunday account of his conduct was, or might be, true, they should acquit.[51] That reinforces the first point. 66. Since the jury were unable to agree on Count 1, and therefore on the primary case for the prosecution on Count 2, an interpretation of their verdict on Count 2 can hardly be made with confidence if the alternative prosecution case on Count 2 was never clearly explained, and the elements of the offence, and the possible defences, were never related to that alternative case. 67. A consistent theme in the prosecution case was that what was being concealed was a bribe, and that was why the appellant was “hopelessly compromised”. Evidently, that was not accepted by some jurors. What the trial judge described as an approach to Count 2 following “the conclusion that there exists no corrupt practice”, by reference to the particulars alleged in the indictment, was never made clear. Consistently with the prosecution’s deliberate concealment argument, it may have been based upon a suspicion of some non-corrupt impropriety, whatever that might be. 68. As to the direction to acquit if what was said in the Beautiful Sunday programme was, or might, be true, that compendious account of the defence case required closer examination. The defence case did not depend upon the proposition that disclosure was simply “overlooked”. The appellant advanced a reason why he regarded an obligation to disclose as far-fetched. Moreover, a rejection of the bribery theory of the case suggests that some jurors were prepared to accept that some part of what the appellant said might be true. The prosecution’s alternative case on Count 2 was so nebulous, and the directions were so confined, that it is difficult to attribute any particular process of reasoning to the jurors who did not accept the primary case. 69. The Court of Appeal rightly referred to the need for it to avoid falling into the trap of assuming that there was something criminal about the applicant’s dealings with Mr Wong. Macrae VP said:[52] “To make assumptions, other than in the clearest circumstances, would be akin to convicting the applicant of an uncharged act, when the jury were unable to resolve the issue of a criminal (in this case corrupt) transaction. However unwise and imprudent it may have been for the applicant to have entered into private, and apparently secret, negotiations with someone who at the same time had official business with the Government of which the applicant as Chief Executive was the head, and however ambiguous the circumstances of the transaction may or may not seem, there has been no determination that the circumstances of their dealings were corrupt.” To those observations may be added the comment that a prosecution case based on suspicion of some unspecified form of impropriety in the dealings concerning the Shenzhen property would go close to reversing the onus of proof by requiring the appellant to provide a reasonable explanation of his conduct so as to dispel the suspicion. 70. In a case where the allegedly wilful misconduct consists of a failure to comply with an obligation to disclose information, and there is a viable issue as to whether disclosure was, and was regarded as, necessary, a direction which treats a conscious decision not to disclose as the equivalent of deliberate failure to disclose, or, even worse, concealment, is dangerously ambiguous. 71. As the present case was argued, and left to the jury, references to the true nature of the relationship between the appellant and Mr Wong, and a serious conflict of interest, and to the appellant’s being hopelessly compromised, were put in the context of the prosecution’s primary case, on Count 2 as well as Count 1, of corruption. In that context, there was no viable issue of the kind referred to above. However, on an approach to Count 2 without the element of corruption, those matters fell to be considered in a different light. There was a viable issue on the element of wilfulness, and it was not explained to the jury. 72. It is in the nature of the offence of misconduct in public office that a jury is required to make an assessment of whether the alleged misconduct is so serious as to involve an element of culpability which is of such a degree that the misconduct is calculated to cause injury to the public interest so as to call for condemnation and punishment.[53] This does not involve a quasi-legislative process. Jurors are not required to give reasons for their decision, but they are expected to have them. This expectation is meant to be satisfied by a trial process that involves reasoned argument by counsel, and judicial directions appropriate to the case. 73. In a passage in Shum Kwok Sher v HKSAR[54] which was cited with approval in A-G’s Reference (No 3 of 2003)[55] Sir Anthony Mason NPJ said: “Whether it is serious misconduct in this context is to be determined having regard to the responsibilities of the office and the officeholder, the importance of the public objects which they serve and the nature and extent of the departure from those responsibilities.” To that, the English Court of Appeal added,[56] “the seriousness of the consequences which may follow from an act or omission.” 74. Here again, on the prosecution’s primary case of corruption as the motive for concealment, the element of seriousness required little elaboration, and the directions of the trial judge were adequate. It is argued by the respondent that the trial judge’s directions were tailored. So they were, but the pattern to which they were tailored was that of the respondent’s primary case. When corruption was taken out of the equation, then an evaluation of the nature and extent of the appellant’s departure from his responsibilities, and the seriousness of the consequences which might follow from his omission required consideration of the motives behind his omission, what it was the appellant was required to disclose, and the consequences of non-disclosure. Without proper analysis of the competing possibilities, there was a danger the jury would fall into the very trap against which the Court of Appeal warned itself. Conclusion 75. The trial judge’s directions on wilfulness and seriousness were inadequate. That inadequacy would have been immaterial if the jury had convicted on both Count 1 and Count 2. Failure to secure a conviction on Count 1 exposed features of the prosecution’s alternative case on Count 2 that required more examination than they received. Disposition 76. The appeal should be allowed and the appellant’s conviction and sentence should be quashed. 77. It has not been argued, and could not be argued, that it was not open to a jury, properly instructed, to convict the appellant on Count 2 of the indictment. However, it is submitted that there should be no order for a new trial because the appellant has already served the whole of the sentence imposed on him. 78. Section 83E of the Criminal Procedure Ordinance[57] provides that where an appeal against conviction is allowed and it appears to the Court that the interests of justice so require it may order the appellant to be re-tried. The principles according to which the discretion should be exercised were considered in the judgement of Lord Woolf NPJ, with whom Li CJ, Bokhary, Chan and Ribeiro PJJ agreed, in Ting James Henry v HKSAR.[58] 79. The appellant was granted bail pending his appeal to the Court of Appeal, but, after the decision of that Court, went into custody and served his sentence. 80. Because of the high office held by the appellant there was a public interest in a definitive resolution of the charges against him. That has already proved impossible in the case of the bribery charge.[59] It is unproductive to speculate about the shape of a prosecution case on Count 2 in the absence of a charge of bribery. The admissible evidence in such a case may be different, and the particularisation of the alleged “impropriety” would come into sharper focus. However, the appellant has already suffered what the Court of Appeal considered a just punishment for the offence in respect of which he would be re-tried. That weighs heavily in favour of a conclusion that the interests of justice do not require a new trial. There should not be such an order. The orders of the Court are: 81. For the above reasons, the appeal is allowed. It follows we further order that the appellant’s conviction and sentence in relation to the second count in the indictment be quashed. Ms Clare Montgomery QC, Mr Derek Chan SC and Ms Betty Chiu, instructed by King & Wood Mallesons, for the Appellant Mr David Perry QC and Ms Maggie Wong SC, Counsel on fiat, Ms Alice Chan SADPP and Ms Irene Fan ADPP (Ag), of the Department of Justice, for the Respondent [1] Mraz v The Queen (1955) 93 CLR 493 at 514. [2] The facts are more extensively set out in the reasons of Macrae VP in the Court of Appeal, CACC 55/2017, [2018] HKCA 425, [2018] 3 HKLRD 564 (Yeung VP, Macrae VP and Pang JA), judgment dated 20 July 2018 (“CA Judgment”). These facts are largely taken from [9] to [13], [17], [24] to [26] and [33] to [36] with some additions and comments. [3] Part B of the Record (“Part B”) pp54-55. [4] Part B p20. [5] Part B p21. [6] Summing up of trial judge (“Summing Up”) p95. [7] Part B pp201-203; Summing up pp64-68. [8] Part B pp201-204, pp208-209. [9] Part B pp436-438. [10] CA Judgment [83]. [11] [2018] HKCA 425, [2018] 3 HKLRD 564. [12] Sin Kam Wah v HKSAR (2005) 8 HKCFAR 192 at 210-211 per Sir Anthony Mason NPJ. [13] A-G’s Reference (No 3 of 2003) [2005] QB 73 at 86, 91. The relationship between elements (3) and (4) has not been fully explored in the authorities, but it is clear that (4) is additional to (3). In some cases the potential for excuse or justification may be built into the standard itself. [14] (2002) 5 HKCFAR 381 at 404-408. [15] (2002) 5 HKCFAR 381 at 408. [16] (1978) 2 Crim LJ 307 at 308. [17] Cap 200. [18] Cap 201. [19] (2017) 20 HKCFAR 264 at 281-288. [20] [2005] QB 73. [21] (2013) 16 HKCFAR 525. [22] (2002) 5 HKCFAR 381 at 408, 410. [23] (2002) 5 HKCFAR 381 at 398, 414. [24] E.g. R v Llewellyn-Jones [1968] 1 QB 429. [25] A-G’s Reference (No. 3 of 2003) [2005] QB 73 at 90. [26] [1979] QB 722 at 727-728. [27] (2013) 16 HKCFAR 525. [28] (2013) 16 HKCFAR 525 at 534. [29] R v Chapman [2015] QB 883 at 895, R v France (Anthony) [2016] 4 WLR 175 [27], [28]. [30] CA Judgment [37]. [31] Transcript (“T”) Day 7 p52. [32] T Day 5 pp73-74. [33] T Day 22 pp15-16. [34] T Day 23 p75. [35] T Day 23 p73. [36] T Day 22 p16. [37] Part B pp208-209 (The interview was in Chinese. What appears in the citation is the certified translation.). [38] Summing up p17. [39] Summing up p8. [40] T Day 24 p80. [41] T Day 21 pp22-23. [42] [2015] QB 883. [43] T Day 21 pp30-31. [44] Summing up p35. [45] Summing up p37. [46] Summing up p37. [47] Summing up p129. [48] Prosecution Note dated 6 February 2017. [49] Summing up pp140, 151. [50] CA Judgment [87], [111], [121]. [51] CA Judgment [107], [111]. [52] CA Judgment [85]. [53] R v Dytham [1979] QB 722 at 727-728. [54] (2002) 5 HKCFAR 381 at 409-110. [55] [2005] QB 73 at 87-88. [56] [2005] QB 73 at 88. [57] Cap 221. [58] (2007) 10 HKCFAR 632 at 651-653. See also Reid v The Queen [1980] AC 343 at 348-350. [59] CA Judgment [5]. 
Hon Poon CJHC: 1. Article 46(1) of the Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region (“NSL”) provides: 「對高等法院原訟法庭進行的就危害國家安全犯罪案件提起的刑事檢控程序，律政司長可基於保護國家秘密、案件具有涉外因素或者保障陪審員及其家人的人身安全等理由，發出證書指示相關訴訟毋須在有陪審團的情況下進行審理。凡律政司長發出上述證書，高等法院原訟法庭應當在沒有陪審團的情況下進行審理，並由三名法官組成審判庭。」 Its English translation reads: “In criminal proceedings in the Court of First Instance of the High Court concerning offences endangering national security, the Secretary for Justice may issue a certificate directing that the case shall be tried without a jury on the grounds of, among others, the protection of State secrets, involvement of foreign factors in the case, and the protection of personal safety of jurors and their family members. Where the Secretary for Justice has issued the certificate, the case shall be tried in the Court of First Instance without a jury by a panel of three judges.” 2. The core issue raised in this appeal is whether a decision by the Secretary for Justice to issue a certificate under NSL 46(1) is amenable to conventional judicial review challenge such as the principle of legality and procedural safeguards. A. Proceedings below A1. Prosecution case 3. The applicant is the defendant in HCCC 280/2020. He is facing a count of incitement to secession, contrary to NSL 20 and NSL 21; and another count of terrorist activities, contrary to NSL 24, and an alternative count of causing grievous bodily harm by dangerous driving, contrary to section 36A of the Road Traffic Ordinance[1]. The prosecution alleged that on 1 July 2020, after NSL was promulgated, a large number of protestors took part in riots in Wanchai and Causeway Bay. Police were deployed to maintain public order there. In the afternoon, the applicant rode his motorcycle at speed in Wanchai, flying a black flag emblazoned with words in white “光復香港時代革命” and “LIBERATE HONG KONG REVOLUTION OF OUR TIMES”. The public actively responded to his conduct. The police tried to stop him at several locations but failed. Eventually, the applicant rammed into the police officers at the checkline at the junction of O’Brien Road and Jaffe Road, after ignoring the police’s warning, and injured three police officers seriously. He himself fell off his motorcycle and fractured his ankle. 4. On 16 November 2020, the applicant was ordered by a magistrate to be committed to the Court of First Instance for trial. The indictment was preferred on 23 November 2020. A2. Issue of the certificate under NSL 46(1) 5. On 5 February 2021, the Secretary for Justice pursuant to NSL 46(1) issued a certificate directing that the proceedings in HCCC 280/2020 be tried in the Court of First Instance without a jury (“the Certificate”). The Certificate is written in Chinese and it reads: “本人現行使在《2020年全國性法律公布》(2020年第136號法律公告)附表中的《中華人民共和國香港特別行政區維護國家安全法》第四十六條第一款賦予的權力,對高等法院原訟法庭進行的就危害國家安全犯罪案件提起的上述刑事檢控程序,發出證書指示相關訴訟毋須在有陪審團的情況下進行審理。證書是經顧及並考慮所有相關的情況及資料,為有效防範、制止和懲治危害國家安全犯罪,基於以下理由而發出： (一) 保障陪審員及其家人的人身安全;及/或 (二) 若審訊在有陪審團的情況下進行,有可能會妨礙司法公義妥爲執行的實際風險。 日期: 2021年2月5日 Its English translation is: “ In exercise of the power vested in me by Article 46(1) of the Law of the People's Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region in Schedule to the Promulgation of National Law 2020 (L.N. 136 of 2020), in the above criminal proceedings in the Court of First Instance of the High Court concerning offences endangering national security, I hereby issue a certificate directing that the case shall be tried without a jury. For the effective prevention,suppression andpunishmentofoffencesendangeringnationalsecurity,the certificate is issued on the following ground(s) having taken into accountand considered all the relevant circumstances andinformation: (1) Protection of personal safety of jurors and their family members;and/or (2) If the trial is to be conducted with a jury, there is a real risk that the due administration of justice might beimpaired. Dated this 5th day of February, 2021 As a result, the case is listed for trial in the Court of First Instance before a panel of three judges. It will commence on 23 June 2021 with 15 days reserved. A3. The Judge’s judgment 6. On 7 April 2021, the applicant applied before Alex Lee J (“the Judge”) for leave to judicial review of the decision to issue the Certificate on the ground that it engaged the principle of legality and procedural safeguards which the Secretary for Justice had failed to observe. After a rolled-up hearing on 10 May 2021, the Judge handed down his judgment on 20 May 2021,[2] refusing to grant leave to the applicant to apply for judicial review. 7. The Judge rejected the applicant’s primary contention that, since an indictment has been preferred against him, he has a constitutional right to a jury trial in the Court of First Instance; and that this right is protected by general principles of legality and other ordinary procedural safeguards. While recognising that the practice in Hong Kong has been for jury trials to be held in criminal proceedings in the Court of First Instance, the Judge held that it does not mean that the accused has a constitutional right to jury trial. Even if there had been any previous right to a jury trial, such right would have been abrogated by NSL 46(1) and NSL 62 in respect of trial for national security offences[3]. After the enactment of NSL, there are two modes of trial in the Court of First Instance for these offences. The traditional mode is trial by jury, and the new mode is trial by a panel of three judges without jury. This new mode is engaged when a certificate is issued by the Secretary for Justice under NSL 46(1). 8. The Judge observed that the direction in the NSL 46(1) certificate is mandatory, and that the NSL does not provide that the Secretary has a duty to hear or notify an accused before she issues the certificate. The Judge further observed that NSL 46(1) sets out non-exhaustively the grounds for issuing a certificate. These grounds involved matters which the Secretary would reasonably be expected not to engage in discussion with an accused before trial, namely, protection of state secrets, involvement of foreign factors and jury protection. 9. The Judge further held that issuing the certificate directing a trial without jury is a prosecutorial decision, which is to be free from any interference under the Basic Law (BL 63). Thus, the certificate cannot be reviewed by the court on ordinary judicial review grounds. It is only reviewable on the limited grounds such as (i) acting in obedience to political instruction, (ii) bad faith, and (iii) rigid fettering of prosecutorial discretion: see Re Leung Lai Fun [2018] 1 HKLRD 523. 10. Referring to Re Hutchings’ Application for Judicial Review [2020] NI 801, the Judge held that there was no requirement to hear from or to inform the applicant before issuing the certificate. He therefore rejected the applicant’s challenge on the ground of procedural impropriety or unfairness based on the lack of notice and opportunity to make representation. 11. Following his ruling that issuing the Certificate is a prosecutorial decision, the Judge held that the mere absence of detailed reasons is not sufficient to meet the very high evidential threshold for reviewing such a decision. Absent any allegation of bad faith or dishonesty, there was no basis for the court to interfere. The challenge on the ground of illegality, with the applicant contending that the lack of reasons meant that there was no rational basis for the Certificate (so that the Secretary must have misinterpreted or misapplied the law), was rejected. 12. As for the challenge based on Wednesbury unreasonableness, the Judge held that there was nothing inherently unreasonable in directing a trial by a panel of three judges without jury, when, on the face of the Certificate, there is a perceived risk in relation to the safety of jurors or their family, or that due administration of justice might be impaired. 13. Finally, the Judge held that, since the applicant has no right to a trial by jury, and his right to a fair trial is not engaged (as a fair trial can be conducted without a jury), the Certificate does not constitute a restriction of any of his rights. Therefore, the proportionality test as set out in Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372 is not engaged. B. Appeal B1. The applicant’s grounds of appeal and submissions 14. In the notice of appeal, Mr Philip Dykes SC for the applicant[4] put forward numerous grounds of appeal, which can be categorized as two main complaints: (1) the Judge erred in ruling that there was no constitutional right to a jury trial in the Court of First Instance; and (2) he erred in ruling that issuing a NSL 46(1) certificate was a prosecutorial decision protected under BL 63 from interference. 15. On constitutional right to jury trial, Mr Dykes submitted that, before the enactment of NSL, an accused was entitled to a verdict from a jury if he were to be tried in the Court of First Instance. Trial before a jury is therefore a right. It was further submitted that it is a right protected under the Basic Law, as the principle of trial by jury is to be maintained under BL 86; and that this right was covered by principles previously applied to criminal proceedings and was a right previously enjoyed in Hong Kong under BL 87. Mr Dykes further pointed out that under section 65F of the Criminal Procedure Ordinance (“CPO”)[5], a case can be transferred from the Court of First Instance to the District Court or a Magistrates’ Court to be tried without a jury. He submitted that this tended to show that there was a right to jury trial, as the prosecution had to make an application for and justify such a transfer. 16. As the constitutional right to jury trial cannot be lightly curtailed or limited, NSL 46(1) should be construed as requiring procedural fairness to be accorded to the applicant before a certificate is issued, counsel reasoned. 17. Regarding the Judge’s observation that the Secretary for Justice would not be expected to engage in discussion with an accused before trial on matters set out in NSL 46(1), Mr Dykes submitted that, while standards of procedural fairness may have to be adjusted in respect of matters involving public interest immunity, this does not mean right to jury trial should be affected. He submitted that, in the instant case, the grounds relied on by the Secretary, namely, (i) the safety of jurors and family or (ii) unspecified grounds relating to administration of justice, did not appear to give rise to public interest immunity precluding discussion before trial. 18. Based upon the argument that there is a right to a jury trial, Mr Dykes submitted that the proportionality test was applicable when restrictions are to be imposed on this right. 19. As to procedural safeguards, on the basis that there is a right to jury trial, Mr Dykes submitted that the Judge erred in holding that the issue of the Certificate was a prosecutorial decision protected from interference by BL 63. He further pointed out that, unlike other prosecutorial decisions such as a decision to prosecute and a decision on trial venue, there is no public policy or guidelines on issuing a NSL 46(1) certificate. Seeking to distinguish Re Hutchings’ Application for Judicial Review, Mr Dykes submitted that ordinary procedural fairness requires that the Secretary should hear from the applicant before issuing a certificate and to give reasons for doing so. B2. The putative respondent’s submissions 20. Mr Jenkin Suen SC for the putative respondent[6] submitted that there is no constitutional right in Hong Kong to a jury trial, whether generally or in the limited sense suggested by the applicant. Trial by jury in the Court of First Instance was a practice, not a right. In the context of the NSL, there are competing concerns pertaining to the mode of trial, so that trial by jury cannot be characterized as a right of the accused. In any event, it was submitted that the Judge was correct that any previous right to jury trial would have been abrogated by the NSL. The mechanism under section 65F of the CPO for transfer of cases from the Court of First Instance to lower courts could not assist the applicant. 21. Mr Suen submitted that issuing a NSL 46(1) certificate was a prosecutorial decision, which could not be reviewed on ordinary judicial review grounds. The applicant was wrong to contend that BL 63 does not apply simply because a trial without jury is directed. He also submitted that, on the face of the Certificate, there was sufficient basis to invoke NSL 46(1), as it referred to (1) a risk of personal safety of the jurors and their family, or that (ii) the due administration of justice might be impaired. 22. Given that no right has been infringed and that the issue of the Certificate was a prosecutorial decision protected from interference, Mr Suen submitted that the applicant’s challenge on ordinary judicial review grounds (procedural impropriety, illegality and proportionality) could not succeed. B3. Respondent’s notice 23. By a respondent’s notice, it is contended that the application for leave amounted to an inappropriate collateral attack on an underlying criminal proceedings, and that if the applicant were to take issue with the fairness of those proceedings, he should do so within those proceedings, and not through other satellite litigation. It is further contended that the leave application was premature as the Certificate is an intermediate or procedural step dealing with case management, not the substantive or conclusive outcome of the criminal proceedings. If the applicant were concerned that the Certificate would adversely affect the outcome of the criminal trial, it is a matter to be raised in an appeal against conviction. 24. In reply, Mr Dykes submitted that nothing in the NSL excludes or limits the applicant’s right to apply for judicial review. The leave application was not an inappropriate collateral attack because there were no suitable alternative remedies to challenge the certificate. The NSL does not provide for an appeal mechanism in respect of the Certificate. The panel of three judges in the criminal trial does not have jurisdiction to entertain an application for a jury trial. Lack of a jury trial is not a ground of appeal against conviction under section 83 of the CPO. Referring to Re Hutchings’ Application for Judicial Review, supra, he submitted that judicial review is a proper way to challenge the Certificate. C. The applicant’s real complaint 25. It is important to bear in mind the following incontestable facts to remind oneself of what this appeal is really about. 26. In HKSAR v Lai Chee Ying (2021) 24 HKCFAR 33, the Court of Final Appeal at [37] held that in light of Ng Ka Ling v Director of Immigration (No. 2) (1999) 2 HKCFAR 141, at p.142, the NSL is not subject to review on the basis of any alleged incompatibility as between the NSL and the Basic Law or the ICCPR as applied to Hong Kong. It means that NSL 46(1) cannot be reviewed in courts. Here, Mr Dykes did not mount a constitutional challenge against NSL 46(1). Indeed, any disguised attempt, if raised, to challenge it based on an alleged incompatibility with any article of the Basic Law or ICCPR as applied to Hong Kong would have been firmly rejected. 27. NSL 45 states that: “Unless otherwise provided by this Law, magistrates’ courts, the District Court, the High Court and the Court of Final Appeal shall handle proceedings in relation to the prosecution for offences endangering national security in accordance with the laws of the Hong Kong Special Administrative Region.” (Emphasis supplied) The local laws referred to in NSL 45 plainly cover those concern jury trial. By virtue of the introductory words in NSL 45 (italicized), NSL 46(1), if invoked, will allow a non-jury trial to take precedence. Thus, even assuming that there is a right to jury trial entrenched in BL 86, Mr Dykes accepted, as he must, that it is not absolute and may be abrogated by NSL 46(1). Nor did Mr Dykes contend that a non-jury trial directed by NSL 46(1) case is per se unconstitutional. 28. Acknowledging that he could not constitutionally challenge NSL 46(1) which may direct a non-jury trial, Mr Dykes accepted that a trial before a panel of three judges under NSL 46(1) does not impede on the applicant’s constitutional right to a fair trial. In other words, Mr Dykes accepted that the applicant will still have a fair trial before the panel of three judges even though it is not a jury trial. 29. Finally, it is not the applicant’s case that the Secretary for Justice was motivated by bad faith, dishonesty or other ulterior motives in issuing the Certificate. 30. Properly understood, the applicant’s complaint boils down to his assertion that he has a constitutional right to a jury trial under BL 86 and that when the Secretary for Justice issued the Certificate, her decision had the effect of depriving him such a right, therefore engaging the principle of legality and procedural safeguards. As said at the outset, it is eminently a matter of construction of NSL 46(1). D. Construction of NSL 46(1) D1. Purposive approach 31. As to the approach to be adopted by Hong Kong courts in construing the NSL, the judgment of the Court of Final appeal in Lai Chee Ying, supra, is instructive. 32. There, the respondent was charged with one count of collusion with a foreign country or with external elements to endanger national security, contrary to NSL 29(4). The Chief Magistrate refused bail and remanded him in custody. Upon his application, the Judge granted him bail with special conditions, which was later set aside by the Court of Final Appeal. 33. In analyzing the issues raised in the appeal, the Court of Final Appeal undertook an interpretative exercise of NSL 42(2), a specific provision on bail in the NSL. The Court at [8] pointed out that determination of the meaning and effect of NSL 42(2) requires that provision to be examined in the light of the context and purpose of the NSL as a whole. In the outline of the legislative process through which the Standing Committee of the National People’s Congress enacted the NSL, the Court referred to the Decision of the National People’s Congress on Establishing and Improving the Legal System and Enforcement Mechanisms for Safeguarding National Security in the Hong Kong Special Administrative Region dated 28 May 2020 and the Explanation of a Draft Decision (which was eventually adopted to be the above Decision) presented to the NPC on 22 May 2020. The Court at [11] emphasized that given the special status of the NSL as a national law applied under Article 18 of the Basic Law, and given the express reference in NSL 1 to that process, regard may properly be had to the Explanations and Decisions made in proceedings of the NPC and the NPCSC regarding promulgation of the NSL as a law of the HKSAR as extrinsic materials relevant to consideration of the context and purpose of the NSL. Turning specifically to NSL 42(2), the Court at [42] and [45] highlighted the importance of examining the matrix in which NSL 42(2) exists, consisting of the applicable human rights and rule of law principles, referred to in NSL 4 and NSL 5, the rules regarding bail under HKSAR law and the provisions of the NSL read as a coherent whole, with a view to ascertaining how NSL 42(2) is intended to operate in that context. 34. What the Court of Final Appeal adopted in interpreting NSL 42(2) is evidently the well-established common law technique of purposive and contextual construction. The same must apply to the construction of NSL 46(1) here. D2. Construing NSL 46(1) 35. Pursuant to the purposive approach, I now examine the context in which NSL 46(1) operates, consisting of the legislative process of the NSL, various articles in the NSL and the relevant articles of the Basic Law and the Hong Kong Bill of Rights (“BOR”) that shed light on context and purpose, and the corpus of law concerning jury trial and prosecutorial decisions protected by BL 63. I will distil the interpretative factors that inform the construction exercise as appropriate and construe the provision accordingly. 36. NSL 46(1) has already been set out in full at [1] above. There is a considerable measure of agreement between the parties regarding the following basic features of NSL 46(1): (1) NSL 46(1) is a specific provision applicable only to criminal proceedings in the Court of First Instance concerning offences endangering national security. (2) The discretion vested solely in the Secretary for Justice. (3) The discretion is open-ended as the stated grounds are non-exhaustive. (4) The direction of the Secretary for Justice for trial without a jury is mandatory. (5) The decision by the Secretary for Justice to issue the certificate is not a judicial function. (6) There is no express reference to an accused’s right to jury trial. Nor is there any express provision for an accused to object or to be consulted before the Secretary for Justice decides to issue a non-jury trial certificate. These features set the scene for the construction exercise. D2.1 Giving effect to the primary purpose of the NSL 37. NSL 46(1) has first to be examined in the light of the general context and purpose of the NSL as a whole, taking into account the constitutional basis upon which the NSL is applied to Hong Kong: Lai Chee Ying, supra, at [8]. In this regard, I would respectfully adopt the outline drawn by the Court of Final Appeal at Part B on the genesis and legislative process of the NSL, cumulating in the promulgation by the Chief Executive on 30 June 2020. As pointed out by the Court of Final Appeal, the NSL was enacted to fill the vacuum in the laws of the HKSAR regarding safeguarding national security arising from the failure to enact local legislation under BL 23 for the past 23 years. As a national law applied to the HKSAR, the NSL has a special constitutional status focusing specifically on safeguarding national security and preventing and suppressing acts endangering national security in the Region. 38. NSL 46(1) seeks to give full effect to that primary purpose of the NSL: (1) It applies to criminal proceedings in the Court of First Instance concerning offences endangering national security and no more. (2) Of the three stated grounds upon which the Secretary for Justice may issue a non-jury trial certificate, protection of State secrets and involvement of foreign factors plainly arise from the special nature and needs of the offences of endangering national security. If and when it is necessary to issue a certificate on either of those two grounds, it clearly serves the NSL’s primary purpose. The third stated ground concerning personal safety of jurors and their family members has to be understood with fairness of a criminal trial in mind. D2.2 Ensuring a fair trial 39. NSL 4 mandates respect for and protection of rights and freedoms which the residents of the HKSAR enjoy under the Basic Law and the ICCPR as applied to Hong Kong. Of immediate significance is BL 87, which states: “In criminal or civil proceedings in the Hong Kong Special Administrative Region, the principles previously applied in Hong Kong and the rights previously enjoyed by parties to proceedings shall be maintained. Anyone who is lawfully arrested shall have the right to a fair trial by the judicial organs without delay and shall be presumed innocent until convicted by the judicial organs.” 40. The right to fair trial is also contained in BOR 10 in these terms: “In the determination of any criminal charge against him… everyone shall be entitled to a fair and public hearing by a competent, independent and impartial tribunal established by law.” 41. Further, NSL 5 requires the adherence to the rule of law principles in preventing, suppressing, and imposing punishment for offences endangering national security. It expressly provides that a person who has committed an act which constitutes an offence under the law shall be convicted and punished and no one shall be convicted and punished for an act which does not constitute an offence under the law. This is the cardinal principle of conviction and punishment of crimes as prescribed by law. NSL 5 goes on to stipulate: “A person is presumed innocent until convicted by a judicial body. The right to defend himself or herself and other rights in judicial proceedings that a criminal suspect, defendant, and other parties in judicial proceedings are entitled to under the law shall be protected. No one shall be liable to be tried or punished again for an offence for which he or she has already been finally convicted or acquitted in judicial proceedings.” These principles of presumption of innocence, protection of parties’ rights in fair trial and protection against double jeopardy tie in with BL 87 and mirror the similar provisions in BOR 11. 42. NSL 46(1) has to be read together with NSL 4 and NSL 5, and BL 87 and BOR 10 and BOR 11, to ensure that the defendant’s constitutional right to a fair trial as embodied in those provisions is not compromised. The prosecution also has a legitimate interest in maintaining the fairness of the trial. 43. The third of the stated grounds for issuing a non-jury trial certificate in NSL 46(1) concerns the protection of personal safety of jurors and their family members. When the personal safety of jurors or their family members is under threat, it will seriously undermine the integrity of the criminal process. This is where the paramount importance of a fair trial comes into play. Granted jury trial is the conventional mode of trial in the Court of First Instance, it should not be assumed that it is the only means of achieving fairness in the criminal process. Neither BL 87 nor BOR 10 specifies trial by jury as an indispensable element of a fair trial in the determination of a criminal charge. When there is a real risk that the goal of a fair trial by jury will be put in peril by reason of the circumstances mentioned in the third ground, the only assured means for achieving a fair trial is a non-jury trial, one conducted by a panel of three judges as mandated by NSL 46(1).[7] Such a mode of trial serves the prosecution’s legitimate interest in maintaining a fair trial and safeguards the accused’s constitutional right to a fair trial. 44. In the present case, the Certificate specified two reasons. The first is self-evidently based on the third stated ground. The second concerns the risk of impairment to due administration of justice. Both reasons are amply supported by the considerations of fair trial discussed above. The trial by a panel of three judges now mandated by NSL 46 (1) is to ensure that the applicant will have a fair trial. It is therefore not surprising that the applicant did not say that he would be deprived of a fair trial even if it is not a jury trial. D2.3 Reading with BL 63 and BL 86 coherently 45. NSL 1 states that the NSL is enacted in accordance with the Constitution and the Basic Law. Thus, there can be no inconsistency between the NSL and the Basic Law. It follows that there can be no inconsistency or incompatibility between NSL 46(1) with BL 63 or BL 86. This requires the court to read NSL 46(1), BL 63 and BL 86 as a coherent whole. 46. BL 63 provides: “The Department of Justice of the Hong Kong Special Administrative Region shall control criminal prosecutions, free from any interference.” BL 86 stipulates that: “The principle of trial by jury previously practised in Hong Kong shall be maintained.” The parties hotly disputed about the scope of BL 63 and BL 86. 47. To recap, Mr Dykes argued that the right to jury trial in the Court of First Instance is entrenched in BL 86. Since a certificate issued under NSL 46(1) removes that right, the Secretary for Justice’s decision, unlike ordinary prosecutorial decisions protected by BL 63, engages the principle of legality, and because NSL 46 does not abrogate common law protections, her decision attracts procedural safeguards. Mr Suen argued that no such right to jury trial is entrenched in BL 86. The decision by the Secretary for Justice made under NSL 46(1) is protected by BL 63 and is not open to judicial review on the grounds advanced by the applicant. 48. Taking BL 86 first, counsel spent considerable efforts in making good their respective contention. I only need to deal with their submissions briefly as the answer to this question does not in my view impact on the construction issue in any substantial way. 49. In support of his argument, Mr Dykes relied on numerous overseas authorities, including Newell v R [1936] 55 CLR 707, per Latham CJ at pp.710-712; Kingswell v R [1985] 159 CLR 264, per Deane J at pp.299-300, 303; AK v State of Western Australia (2008) 232 CLR 438, per Heydon J at [93]-[98]; and R v Mirza [2004] 1 AC 1118, per Lord Steyn at [7] and Lord Hobhouse at [144]. Those authorities described the immense value and unique features of jury trial in the administration of criminal justice in the common law system and invariably referred to it as a right. But even as a right, the authorities recognized that it is not absolute and may be amended or circumscribed by express legislation: R v Twomey [2010] 1 WLR 630, per Lord Judge CJ at [10];[8] Re Hutchings’ Application, supra, per Lord Kerr at [37].[9] 50. In Hong Kong, it would appear that older cases did not speak unanimously on the matter. For example, in Re David Lam Shu-tsang [1977] HKLR 393, the Full Bench at pp.393-400 expressed some doubt if there was a common law right to elect trial by jury. When the case reached the Court of Appeal (CACV 42 & 43/1977, 7 November 1977, unreported), Pickering JA (at pp.6-7), Li J (at p.14) and Cons J (at pp.19-20) all regarded trial by jury as a common law right. Both the Full Bench and the Court of Appeal confirmed that such a right could be abrogated by the District Court Ordinance.[10] 51. However, modern appellate authorities relied on by Mr Suen suggest that there is no free-standing right to jury trial in Hong Kong. In Chiang Lily v Secretary for Justice [2009] 6 HKC 234, Ma CJHC (as he then was) at [24] disavowed Pickering JA’s observation in Re David Lam Shu-tsang, ibid, and stated in unambiguous terms that no right to jury trial exists in Hong Kong. In a similar vein, Li CJ in Chiang Lily v Secretary for Justice (2010) 13 HKCFAR 208, at [9], reiterated that there is no right to trial by jury in Hong Kong. In HKSAR v Chan Huandai [2016] 2 HKLRD 384, in the context of addressing jury irregularities, this Court, in referring to jury trial as a tradition deeply rooted in the common law, observed that it is an integral and indispensable feature of the criminal justice system and has been entrenched in BL 86. There was however no reference to trial by jury as a right as such. 52. Mr Dykes accepted that there is no free-standing right for jury trial. He further accepted that any common law right to jury trial could be abrogated by statute and had in fact been abrogated by the District Court Ordinance back in 1953. However, Mr Dykes argued that once an indictment is preferred and the case has been committed to the Court of First Instance for trial, the accused is entitled, as his right, to a trial by jury until and unless the Secretary for Justice offers no evidence or otherwise discontinues the prosecution. Mr Dykes relied on the combined effect of section 14A(2) and section 41 of the CPO that a trial on indictment in the Court of First Instance is by way of jury trial. He augmented his contention by relying on the transfer mechanism in section 65F of the same Ordinance. That section provides for the mechanism whereby the Secretary for Justice may, on grounds shown, apply for a transfer of the proceedings on indictment pending in the Court of First Instance to the District Court and the judge may allow the application having regard to the interests of justice. 53. However, what appears to contradict Mr Dykes’s submission is Ma CJHC’s observation in Chiang Lily, supra, at [42(1)] that there is no absolute right to a jury trial in Hong Kong although where the venue chosen in the Court of First Instance, there is in that instance a requirement for jury trial. A requirement for jury trial refers to the mode of the criminal trial that must be adopted in the Court of First Instance. Whether it entails a right the accused enjoys as such is open to debate. 54. In short, Mr Dykes’s contention that there is a right to jury trial in the Court of First Instance does not appear to sit well with the modern local appellate authorities on the subject. However, as said, it is not necessary for me to come to a definitive view on the matter. The reason is this. Even assuming that BL 86 has entrenched a right to jury trial in the Court of First Instance, it does not encompass the principle of legality or procedural safeguards as contended. It is because, as elaborated below, a decision made by the Secretary for Justice which results in a non-jury trial under the relevant enactment is a prosecutorial decision protected by BL 63. 55. Section 88 of the Magistrates Ordinance[11] mandates the magistrate, upon an application by the Secretary for Justice, to transfer the charge or complaint relating to indictable offences an accused is facing to the District Court. In Chiang Lily, supra, the defendant applied for judicial review of the decision of the Secretary for Justice to transfer several charges against her to the District Court. She wished to have a trial by jury and made no constitutional challenge to section 88. The judge dismissed her application and a magistrate made an order under section 88. The defendant then mounted a second judicial review of the magistrate’s decision, arguing that section 88 was unconstitutional as its effect was to confer the judicial power to select the venue for a criminal trial exclusively on the executive. That application was also dismissed by the lower courts. The defendant sought leave to appeal to the Court of Final Appeal, which was dismissed by the Appeal Committee. Li CJ first held that the question of constitutionality of section 88 could and should have been raised in the first application. His Lordship in any event held that the defendant’s contention was not reasonably arguable. He at [15]-[18] held that choice of venue for a prosecution is clearly a matter covered by BL 63 which gives control of prosecutions to the Secretary for Justice, having regard to the context and basis of such decision. 56. It follows from the Appeal Committee’s judgment in Chiang Lily that the Secretary for Justice’s decision on venue, which results in a non-jury trial by the operation of a statutory provision, is a prosecutorial decision protected under BL 63 and is not reviewable on conventional judicial review grounds. Likewise, issuing a certificate under NSL 46(1) is undeniably a prosecutorial decision made by the Secretary for Justice in the criminal process. NSL 46(1) then mandates a non-jury trial. Applying Chiang Lily, BL 63 shields the decision to issue a NSL 46(1) certificate from any conventional judicial review challenge. 57. Support for the conclusion that the decision of the Secretary for Justice to issue a NSL 46(1) certificate is a prosecutorial decision may also be derived from the comparable case law in England. As already alluded to, although jury trial is regarded as a right there, it is not absolute and can be circumscribed by express legislation. There are cases in which statutory provisions enable the Attorney General to make a decision, including issuing a certificate, directing or resulting in a trial by a judge without a jury. For present purpose, it will be sufficient to look at three appellate judgments. 58. In Shuker’s Application [2004] NIQB 20, each of the applicants was charged with an offence that is “scheduled”, that is of a type specified in Schedule 9 of the Terrorism Act 2000. By virtue of section 75(1) of the 2000 Act, such offences are tried by a judge sitting without a jury. Certain offences are stated to be subject to Note 1 of the Schedule. Those offences shall not be scheduled offences when the Attorney General certifies that they are not to be treated as such. The offences with which the applicants were charged are subject to Note 1 and may therefore be certified, or “de-scheduled”, resulting in a jury trial. The applicants applied for judicial review to challenge the decision of the Attorney General not to exercise his power to certify the offences with which they have been charged. One of the issues is whether the Attorney General’s decision was justiciable. 59. In determining this issue, Kerr LCJ drew on the well-established limitations on the review of prosecutorial decisions. He then drew an analogy between the Attorney General’s decision not to “de-schedule” and the decision whether to prosecute in these terms: “[25] …What the various cases that deal with the reviewability of prosecutors’ decisions have in common is an approach to the question that is firmly based on the practical implications of permitting judicial review of the decision, whether it is a decision not to prosecute or a decision to withhold reasons. Ultimately, therefore, the question whether the Attorney General should be subject to judicial review in respect of decisions about de-scheduling must be answered in a way that takes account of the particular features of this process of decision-making. We have concluded that it is not a process which is suitable for the full panoply of judicial review superintendence. In particular, we do not consider that the decision is amenable to review on the basis that it failed to comply with the requirements of procedural fairness. [26] The exercise involved in deciding whether offences should be de-scheduled is in some respects akin to the decision whether to prosecute. It involves the evaluation of material that will frequently be of a sensitive nature and the assessment of recommendations made by or on behalf of the Director of Public Prosecutions based on his appraisal of matters that may not be admissible in evidence or whose disclosure would be against the public interest. This is par excellence a procedure on which the courts should be reluctant to intrude. It is, moreover, a task that has been entrusted by Parliament to the Attorney General and while this will not in all circumstances render judicial review impermissible, it signifies a further reason for reticence.” 60. In Arthurs’ Application [2010] NIQB 75, the applicants sought judicial review of a decision made by the Director of Public Prosecutions whereby he certified that their trial be conducted without a jury under the Justice and Security (Northern Ireland) Act 2007 (“2007 Act”). After rejecting the applicants’ argument that the right to fair trial under Article 6 of the European Convention on Human Rights, Girvan LJ adopted Kerr LCJ’s reasons in Shuker’s Application and rejected the procedural arguments relied on by the applicants. Referring to section 7 of the 2007 Act, which limits the grounds of challenge to the DPP’s decision to dishonesty, bad faith and other exceptional circumstances, he at [25] observed that the language is inspired by the principle of exceptionality applicable in the context of prosecutorial decisions and section 7 gives statutory recognition to the common law reticence in the scrutiny of decisions made in the field of prosecutorial decision-making. 61. Similarly, in Re Hutchings’ Application, supra, the UK Supreme Court had to grapple with a challenge against the certificate issued by the Attorney General under section 1 of the 2007 Act that the trial on indictment of the applicant was to be conducted without a jury on the ground that he was satisfied that there was a risk that the administration of justice might be impaired if the trial were to be conducted with a jury. 62. One of the arguments raised by the applicant was procedural. He argued that he should have been provided with the reasons that the DPP had been minded to issue a certificate and with the material on which his consideration of that question was based, and that the applicant should have been given the opportunity to make representations on whether the certificate should be issued, in advance of any decision, and furthermore that effective representations could not be made in the absence of that information. The applicant also argued that he had a right to jury trial and because of that, his case fell within the exceptionality clause of section 7 of the 2007 Act. 63. In rejecting those arguments, Lord Kerr disposed of the applicant’s reliance on section 7 by observing at [54] that a curtailment of the full spectrum of judicial review challenge was obviously intended by section 7; and ruling at [55] and [56] that the right to jury trial alone was not enough to shift the applicant’s case into a condition of exceptionality; and that something more was required beyond a claim that there is right to a jury trial to qualify for exceptionality clause. Relevantly, he went on to say: “[57] Quite apart from the statutory imperative requiring that there be exceptional circumstances in the absence of bad faith or dishonesty, the decision whether to issue a certificate is obviously one which should not be subject to the full spectrum of conventional judicial review challenge. Unlike most decisions taken in the public law arena, it is not founded exclusively on the evaluation and weighing of hard evidence. It will usually be motivated by sensitive information which cannot be disclosed. It is a decision which the Director of Public Prosecutions must take according to his personal reaction to the material with which he has been presented and his own estimation of the matters at stake. In sum, a decision to issue a certificate does not readily admit of scrutiny of the reasoning underlying it because it will usually be of the impressionistic and instinctual variety, for the reasons earlier explained.” Lord Kerr then referred to the parallels drawn in the cases including Shuker’s Application, supra, and Arthurs’ Application, supra between this species of decisions and decisions whether to prosecute and endorsed it by observing at [60] that the parallels between them are obvious. 64. Plainly, when one considers the context and process in which a decision to issue a non-jury trial certificate under NSL 46(1) is made, the same considerations and evaluative assessment undertaken by the Attorney General as described in the English authorities are involved. The decision-making process undertaken by the Secretary for Justice under NSL 46(1) may involve classified information such as State secrets, confidential intelligence concerning involvement of foreign factors, sensitive materials on risks of personal safety of jurors or their family members or threats to due administration of justice. The information or materials are ordinarily of such a nature that it would not be in the public interest to disclose. Or for the Secretary for Justice to reveal to or discuss with the accused before trial. The Secretary for Justice has to take into account all the relevant circumstances in assessing all the materials available to her, some of which may not be admissible in evidence, and make a judgment call. It is usually, as aptly described by Lord Kerr, of the impressionistic and instinctual variety. And NSL 46(1) entrusts the Secretary for Justice alone with this enormous task. The reasons articulated in the English authorities as to why the Attorney General’s decision or certificate is a prosecutorial decision not amenable to conventional judicial review challenge are equally apposite to the decision by the Secretary for Justice to issue a certificate under NSL 46(1). 65. In his attempt to distinguish the English authorities, Mr Dykes took two points. 66. He first submitted that they were decided against their own statutory scheme, which is different from ours. But such differences do not detract from the cogent reasoning in the English authorities which clearly have general application in terms of common law, and should be adopted for present purpose. 67. Mr Dykes next submitted that in England the right to jury trial is not constitutionally entrenched, unlike BL 86. But in England, the right to jury trial is jealously guarded and has been regarded as a hallowed principle of the administration of the criminal justice. In the circumstances, whether it is entrenched is quite irrelevant. It does not detract from the cogency and weight of the reasoning in the English cases and their applicability to the issues in this appeal. 68. In sum, reading NSL 46(1), BL 86 and BL 63 as a coherent whole with the relevant case law in mind, even if there is a right to jury trial in the Court of First Instance entrenched in BL 86, the decision by the Secretary for Justice to issue a non-jury trial certificate under NSL 46(1) is a prosecutorial decision protected by BL 63. Like other prosecutorial decisions, it is not amenable to conventional judicial review challenge. The contrary submissions by Mr Dykes are all rejected. D2.4 Timely disposal of NSL cases 69. Finally, NSL 42(1) directs the law enforcement and judicial authorities of the Region to ensure that, when applying the laws concerning matters such as the detention and time limit for trial, cases concerning offence endangering national security are handled in a fair and timely manner so as to effectively prevent, suppress and impose punishment for such offence. This direction for timely disposal of the trial strongly militates against the applicant’s contention that the decision by the Secretary for Justice to issue a certificate is amenable to conventional judicial review challenge. It is because such a challenge will definitely breed elaborate and protracted satellite proceedings, thereby frustrating the directive of NSL 42(1) by delaying if not derailing the criminal process. 70. NSL 46(1) contains no express provision for an accused to mount a conventional judicial review against the issue of a certificate. Allowing such a challenge by way of construction would defeat the purpose of NSL 42(1). It cannot possibly be the intent of NSL 46(1) read with NSL 42(1). D2.5 Conclusion 71. For the reasons I have given, on a proper construction, NSL 46(1) does not admit of a conventional judicial review as contended by the applicant. The decision of the Secretary for Justice to issue a certificate under NSL 46(1) is a prosecutorial decision protected by BL 63. It is only amenable to judicial review on the limited grounds of dishonesty, bad faith and exceptional circumstances as explained in the case law. I would therefore answer the core issue identified at [2] with a “No”. 72. This concludes the discussion on construction. E. The applicant’s challenge must fail on facts 73. Under the common law, a prosecutorial decision is amenable only on limited grounds under the rubric of dishonesty, bad faith or other exceptional circumstances: R v DPP (Ex p. Kebilene) [2000] 2 AC 326, per Lord Hope at p.376B-C; see also Re Leung Lai Fun, supra, at [10]. Mr Suen accepted, and rightly so in my view, that a decision by the Secretary for Justice to issue a certificate under NSL 46(1) is subject to these limited grounds of review. By definition and as confirmed in the case law, these challenges are rare: see Re Hutchings’ Application, supra, at [58]. 74. Applying the principle here, as said, the applicant did not allege dishonest or bad faith. Mr Dykes availed himself of the exceptionality ground by harking back to his contention that the applicant has a constitutional right to jury trial in the Court of First Instance under BL 86. However, a similar argument was rejected in Re Hutchings’ Application, supra, at [55]-[56] in the context of section 7 of the 2007 Act. Since section 7 reflects the common law exceptionality principle applicable to the scrutiny of prosecutorial decisions (Arthurs’ Application, ibid), the applicant’s asserted right to jury trial alone is not sufficient to render his circumstances exceptional for the purpose of challenging the Certificate. It follows that the applicant’s case must fail on the facts. F. Disposition 75. In consequence, I agree with the Judge that the applicant’s case is not reasonably arguable and he is right in refusing to grant the applicant leave to apply for judicial review. I would dismiss the appeal. Hon Yeung VP: 76. I have the advantage of reading the judgment of the Chief Judge in draft and I agree with it entirely. I, too, would dismiss the appeal. Hon Lam VP: 77. I also agree respectfully with the judgment of the Chief Judge and for the reasons given by him the appeal should be dismissed. 78. In addition, I am unable to accept Mr Dykes’ submission that a right of being consulted or heard before the Secretary for Justice issued her certificate under NSL 46(1) can be derived from BL 86. If such contention were correct, similar argument could be mounted with regard to the choice of venue by the prosecution for a criminal trial as between the District Court and the High Court. This line of argument must be rejected in light of the Appeal Committee’s holding in Chiang Lily, supra that choice of venue is a matter covered by BL 63. 79. Mr Dykes tried to overcome such difficulty by arguing that since the prosecution is proceeded by way of indictment, it gives rise to a right of a defendant to be tried by jury. In this connection, counsel referred us to section 41(2) of the CPO. 80. In my judgment, this argument fails to give full effect to the provisions in the NSL. Though NSL 41(3) prescribes that prosecution of NSL offences shall be brought by way of indictment, NSL 46(1) and (2) clearly provides that apart from trial before a judge and jury, another possible mode of trial in the Court of First Instance is a trial before three judges without a jury. Construing these provisions together in a coherent manner, it is not possible to imply from the way in which prosecution is brought under NSL 41(3) by indictment that the trial can only be conducted before a judge and a jury. 81. It is provided in NSL 45 and NSL 62 that in respect of the procedure for trials of NSL offences the provisions in NSL shall prevail over other laws in Hong Kong. Hence, insofar as there is conflict between section 41(2) of CPO and NSL 46, the latter shall prevail. 82. In light of this analysis, even assuming that there is any right to jury trial for prosecution brought by way of indictment, such right had been curtailed by NSL 46. I do not accept Mr Dykes’ submission that such right was not curtailed by NSL 46 but by the decision of the Secretary for Justice. In this connection, the position is not materially different from that Lord Kerr alluded to in Re Hutchings’ Application for Judicial Review, supra at [64]. 83. Thus, properly understood, the decision of the Secretary for Justice to issue the certificate under NSL 46(1) is no different in nature from a decision on the venue of trial. Just as BL 86 cannot be relied upon to mount a conventional judicial review against the decision on venue, neither can it be relied upon to mount such challenge against the decision to issue a certificate under NSL 46(1). The policy considerations leading to the restricted ambit of judicial review are essentially the same. Hon Poon CJHC: 84. The appeal is accordingly dismissed. The parties are directed to file written submissions (limited to no more than 3 pages) on costs within 14 days. We will then dispose of it on paper. Mr Philip J Dykes SC and Ms Tina Mok, instructed by Bond Ng Solicitors, for the applicant Mr Jenkin Suen SC and Mr Michael Lok, instructed by the Department of Justice and Ms Leona Cheung PGC (Ag), of the Department of Justice, for the putative respondent [1] Cap 374. This alternative charge was introduced with leave of the court on 7 June 2021. [2] [2021] HKCFI 1397 (“Judgment”). [3] NSL 62 provides that the NSL shall prevail where provisions of the local laws of the Hong Kong Special Administrative Region are inconsistent with it. [4] Together with Ms Tina Mok (Mr Dykes also appeared for the applicant before the Judge below). [5] Cap 221. [6] Together with Ms Leona Cheung and Mr Michael Lok (all of whom also appeared before the Judge below). [7] Cf: In England, in the case of jury tampering, sections 44 and 46 of the Criminal Justice Act 2003 provides for a trial before a single judge in lieu of a jury trial. [8] In that case, the legislation in question is the 2003 Act which imposes restrictions on jury trial in case of, among others, jury tampering. [9] There, the legislation in question is the Justice and Security (Northern Ireland) Act 2007. [10] Cap 336. [11] Cap 227. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Tang PJ: 3. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: 4. This appeal concerns the regime under the Prison Rules[1] for visits to prisoners on remand awaiting trial (“prisoners awaiting trial”). In particular, it raises the questions of (i) whether under those Rules visits to such prisoners are limited to their relatives and friends, and (ii) if so, whether under the visiting regime a prisoner’s friends are limited only to those persons who are personal acquaintances and known to him so that, absent special authority, visits by strangers or persons he has not previously met are excluded in all cases and regardless of the purpose of the visit. The proceedings below 5. The appellants, with others,[2] were charged with conspiracy to defraud[3] in that, between 27 August 2011 and 19 August 2012, they conspired with other unknown persons to defraud officers of the Correctional Services Department (“CSD”) by dishonestly and falsely representing to those officers that each of them was a friend of an inmate remanded at Lai Chi Kok Reception Centre (“LCKRC”), thereby inducing the officers to act contrary to their public duty, namely to grant them permission to visit the relevant inmates at LCKRC which the officers would not otherwise have granted. 6. On 13 September 2013, the appellants[4] were convicted after trial before the magistrate,[5] who imposed community service orders on them.[6] The appellants appealed against their convictions. The appeal was ordered to be transferred from the Court of First Instance to the Court of Appeal[7] and, on 17 October 2016, the Court of Appeal[8] gave judgment dismissing the appellants’ appeals.[9] 7. The Court of Appeal having refused to certify that questions of law of great and general importance were involved in the appeal,[10] the Appeal Committee granted the appellants leave to appeal in respect of the questions of law set out at [8] below and also on the substantial and grave injustice ground on the footing that it was reasonably arguable that the evidence was incapable of sustaining a conviction for conspiracy to defraud.[11] 8. The questions of law in respect of which leave to appeal was granted are: “(a) On a proper construction, what is the meaning of ‘visitors’ in the context of the Prison Rules, Cap 234A? (b) If in the context of Rule 203, ‘visitors’ bears the same meanings under Rule 48 and ‘friends’ as a category of visitors means ‘personal friends’: (i) Is Rule 203 compatible with article 6(2)(a) of the Hong Kong Bill of Rights? (ii) Is Rule 203 compatible with article 14 of the Hong Kong Bill of Rights?” The facts 9. The facts were substantially undisputed at trial. The 1st appellant set up a company, IPS-Care Company, which offered what were called “representative visiting services” to the family and friends of prisoners on remand who were detained in LCKRC. Those visiting services included visiting prisoners awaiting trial on behalf of their family and friends, procuring items for delivery to prisoners such as newspapers, books and magazines, as well as ordering and delivering meals and conveying messages to and from prisoners. The 2nd appellant was employed by the 1st appellant to provide the visiting services. IPS-Care Company had a website advertising its services and its employees also distributed leaflets for the same purpose outside LCKRC. There was evidence that staff wore green “polo” shirts when carrying out the visits. The visiting services were charged at a rate of HK$84 per visit, later increased to HK$120 per visit, and there was a minimum total charge of HK$300. Between August and November 2011, the two appellants visited a total of 31 prisoners awaiting trial who were in custody in LCKRC. 10. As regards the procedure for visiting prisoners at LCKRC, the prosecution evidence was that visits were processed by the Visitor Registration Unit and all visitors were required to complete a Visit Request Slip on which a visitor must specify his or her name, identity card number, address and relationship with the prisoner. CSD staff would input the information from the Visit Request Slip into a computer for verification purposes. In respect of the relationship of the visitor to the prisoner, the computer system only had two categories, namely “relative” and “friend”. There was a single code for “friend”, whereas for “relative” there were various codes corresponding to different family members. 11. After verification in this way, a computer-generated Visit Arrangement Slip was generated and given to the visitor, on the strength of which the visitor would be admitted to visit the prisoner. So far as the relationship between the visitor and the prisoner was concerned, CSD staff were not able to ascertain or verify that relationship. The prosecution evidence was that if a person was not a relative and did not fill in “friend” in the Visit Request Slip, or if the CSD knew that a claimed relationship was false, the matter would be referred to a senior officer for determination.[12] 12. There was no distinction in the procedure between prisoners awaiting trial and convicted prisoners, although convicted prisoners would, on admission to the institution in which they were detained, be asked to provide a Declared Visitors List setting out the names and relationships of all prospective visitors which would then be input into the CSD’s Penal Record Information System. It would appear that prisoners awaiting trial at LCKRC might not have completed a Declared Visitors List before they received visits and this was the case in respect of each of the prisoners awaiting trial who were the subject of visits by the appellants. 13. In the present case, in respect of each visit made to a prisoner by the appellants, each appellant indicated in the relevant Visit Request Slip that he or she was a friend of the prisoner being visited. As noted above at [5], that claimed relationship was alleged to be the misrepresentation of fact on which the conspiracy in the charge was based. The Prison Rules and the questions of construction arising on this appeal 14. The Prison Rules consist of various rules and regulations for the government and management of prisons, prison staff and prisoners authorised by the rule-making power under section 25 of the Prisons Ordinance.[13] Although other rules featured in the course of argument to which reference will be made, three rules are of particular relevance in this appeal, namely Rules 2, 48 and 203. 15. Rule 2 is in Part I of the Prison Rules entitled “General Rules for the Government of Prisons” and provides: “2. Application of rules The rules in this Part shall apply to all classes of prisoners except in so far as they may be inconsistent with the rules made to govern any particular class or classes of prisoners.” 16. Rule 48, under Subdivision 7 of Division 3 of Part I of the Prison Rules entitled “Communications and Visits”, provides: “48. General provisions as to visits No persons, other than the relatives and friends of a prisoner, shall be allowed to visit him except by special authority. Such visits by relatives and friends shall, subject to such restrictions as may be imposed for the maintenance of discipline and order in the prison and for the prevention of crime, be allowed in the manner following – (a) they shall be allowed to visit a prisoner twice a month and no more than 3 persons shall be allowed at one time; (b) the visits of the relatives and friends of a prisoner shall be recorded in a book kept for that purpose and the visits shall be limited to 30 minutes on each occasion; (c) a prisoner shall be visited in the presence of an officer of the Correctional Services Department; (d) the Superintendent shall fix the days and times for visits which shall be publicly notified at the gates of the prison; (e) (Repealed L.N. 65 of 1969) (f) visitors shall not be admitted until they have recorded their names and addresses, their relationship to or connexion with the prisoner they wish to visit; (fa) visitors shall not be admitted unless they have satisfied an officer of the Correctional Services Department, if so required, as to their identity; (g) the Superintendent may, in special cases, extend the duration of a visit; (h) the Superintendent may permit any convicted prisoner to see his relatives or friends for the purpose of making arrangements respecting his property or for any other special reason; (i) the Superintendent may allow a prisoner who is entitled to a visit to write a letter instead of receiving such visit.” 17. In Part II of the Prison Rules entitled “Special Rules for Particular Class of Prisoners”, Rule 203, under Subdivision 7 of Division 1 entitled “Visits and Communications”, provides: “203. Rule as to visitors (1) Every prisoner awaiting trial shall, subject to the order of the Superintendent, be permitted to be visited by one visitor, or if circumstances permit, by two at the same time, for a quarter of an hour on any week day, during such hour as may from time to time be appointed. (2) The Superintendent may, in special cases, permit the visit to be prolonged, and allow more than 2 visitors to visit such prisoner at one time.” 18. As will be seen, Rule 203 grants to prisoners awaiting trial an entitlement, subject to the order of the Superintendent of the prison, to be visited by one or more “visitors”. The first question that arises in this appeal is whether, as contended by the appellants, that entitlement extends the class of persons who may visit prisoners awaiting trial to any persons or whether, as contended by the prosecution, “visitors” in Rule 203 are confined to the “relatives and friends” of such prisoners. Both the magistrate and Court of Appeal considered that the prosecution’s construction of Rule 203 was correct.[14] 19. If the prosecution’s construction is correct, as was held below, the next question that arises is whether the category of visitors who are “friends” of a prisoner is limited to those persons who are in a pre-existing personal relationship with the prisoner and are known to him and cannot include strangers or those whom the prisoner has never met. This was the prosecution’s contention below and the magistrate and the Court of Appeal both accepted that contention[15] and rejected the appellants’ invitation to give the word “friends” a wider construction. On this basis, since the prisoners they had visited were strangers whom they had not met before, the courts below held that the appellants had made misrepresentations to the CSD staff when completing the Visit Request Slips for their visits to prisoners in LCKRC in the period covered by the charge. 20. The questions in issue being ones of statutory construction, it is necessary for the Court to construe the language used in the Prison Rules having regard to its context and purpose. The context of a particular rule will include the other provisions of the Prison Rules and this may include the history of those rules. The words used must be construed consistently with their purpose. That purpose is ascertained using a flexible and open-minded approach. For these uncontroversial and well-settled principles, see the recent judgment of Ma CJ in Town Planning Board v Town Planning Appeal Board (2017) 20 HKCFAR 196 at [29] and the references there cited. The proper construction of Rule 203 (a) Do “visitors” mean “relatives and friends”? 21. The rival constructions of Rule 203 respectively advanced by the parties to this appeal differ as regards the extent to which Rule 48 applies to prisoners awaiting trial. 22. In this respect, the appellants emphasise that the regime provided for prisoners awaiting trial in Rule 203 is, and has always been, a particular self-contained regime for such prisoners separate and distinct from that applying to convicted prisoners. This, they say, is inherent in the presumption of innocence and is also reflected in Article 6(2)(a) of the Bill of Rights. The appellants also rely on the fact that the Legislature has used the words “visitor” and “visitors” in Rule 203 rather than the words “relatives and friends” as in Rule 48. Relying on Rule 2, the appellants contend that, where there is an inconsistency between that particular regime applying to prisoners awaiting trial and the regime applying generally to prisoners under Rule 48, the general regime in Rule 48 yields to the particular regime in Rule 203. Thus, since Rule 203 says that prisoners awaiting trial may be permitted to be visited by one or more “visitors”, this is not restricted to “relatives and friends” which is the category of persons generally able to visit prisoners under Rule 48. 23. For the following reasons, I would reject the appellants’ argument as regards the construction of Rule 203 and hold that the word “visitors” in that rule means “relatives and friends”. 24. To construe Rule 203 in context, it is necessary to have regard to the other rules in the Prison Rules. This requirement is expressly reinforced in the Prison Rules by Rule 2 which specifies that the rules in Part I “shall apply to all classes of prisoners except in so far as they may be inconsistent with the rules made to govern any particular class or classes of prisoners”. Rule 48 is a rule within Part I and so, unless it is inconsistent with the rules relating to prisoners awaiting trial, it must apply. 25. Rule 203 relating to prisoners awaiting trial is inconsistent with Rule 48 in respect of the maximum number of persons who may visit and also as to the total duration and frequency of visits. Thus, under Rule 203, the maximum number of persons who may visit may be more than two. In contrast, under Rule 48, the maximum may never exceed three. Under Rule 203, prisoners awaiting trial may receive daily weekday visits up to an aggregate of approximately 300 minutes per month, whereas, under Rule 48, the maximum number of visits is two per month for an aggregate of 60 minutes per month. To that extent, in relation to the maximum number of persons who may visit, the total duration and frequency of visits, Rule 48 must yield to the specific provisions of Rule 203 applying to prisoners awaiting trial. 26. Rule 48 must, however, otherwise apply to prisoners awaiting trial. (1) First, the provisions of Rule 48 sub-paragraphs (c), (f) and (fa) prescribe particular procedures to be complied with in relation to visits to prisoners. These must apply to visits to prisoners awaiting trial since Rule 203 is silent as regards those procedures and it cannot be the case that such visits are unregulated by any procedure. (2) Secondly, Rule 48 sub-paragraph (h) contains a particular provision concerning visits to “any convicted prisoner”. This distinction drawn in sub-paragraph (h) points clearly to the provisions elsewhere in Rule 48 applying to other classes of prisoners and thus to prisoners awaiting trial. 27. I do not discern any inconsistency between Rule 48 and Rule 203 as to the identity of the persons who may visit and there is no reason in principle why there should be a difference in respect of the identity of visitors. Although the words “visitor” and “visitors” are used in Rule 203, there is no indication that they are intended to mean something different to the meaning they bear elsewhere in the Prison Rules. Generally speaking, where the same word is used in different parts of the same piece of legislation, it should be given the same meaning unless the context otherwise clearly indicates a different meaning is intended. As Lord Walker of Gestingthorpe noted, in R v Islam [2009] 1 AC 1076, at [23]: “It is as true today as it was in 1869 that ‘it is a sound rule of construction to give the same meaning to the same words occurring in different parts of an Act of Parliament’ …”.[16] 28. With that rule in mind, it is important to note that the word “visitors” is also used in sub-paragraphs (f) and (fa) of Rule 48. On a reading of Rule 48 as a whole, “visitors” must be read as meaning “relatives and friends” because the second sentence of the introductory part of Rule 48 provides that “Such visits by relatives and friends shall … be allowed in the manner following – …” (emphasis added).[17] The word “visitors” having been used in sub-paragraphs (f) and (fa) of Rule 48, it would be very surprising if the same word used in Rule 203 was intended to have a different meaning and I would hesitate to reach that conclusion in the absence of clear legislative intent. 29. It is true that prisoners awaiting trial are generally treated differently to and more liberally than convicted prisoners under the Prison Rules. For example, prisoners awaiting trial must be kept apart from convicted prisoners (Rule 190); they may receive their own food rather than the prison diet (Rule 192, cf. Rule 31); they may receive malt liquor (Rule 192, cf. Rule 25); they may wear private clothes (Rule 196, cf. Rule 26); they may but are 30. Other provisions in the same subdivision of the Prison Rules support the conclusion that reading “visitors” in Rule 203 literally and to convey a different meaning as to the category of persons who may visit prisoners generally under Rule 48 is unlikely to have been intended. Thus: (1) In Rule 205, which is headed “Right to see visitors for the purpose of finding bail” (emphasis added), it is provided that: “Every prisoner awaiting trial who is in prison in default of bail shall be permitted to see any of his relatives or friends, on any week day, at any reasonable hour, for the bona fide purpose of providing bail.” If the visiting regime under Rule 203 extends to persons other than “relatives and friends” it would be odd that a prisoner cannot have resort to that wider category of persons to assist him to find bail. Rule 205 also shows the word “visitors” (albeit appearing in a heading to a legislative provision[18]) being equated with “relatives or friends”. (2) In Rule 204, relating to visits from a private medical adviser for the purpose of his defence, it is provided that the choice of such adviser may be “by him or by his friends or legal adviser” and in Rule 206(1), concerning written communications, prisoners awaiting trial are entitled to be provided with a reasonable amount of paper and other writing materials “for [the] purpose of communicating with his friends or for preparing his defence”. It seems unlikely that the use of “friends” in Rules 204 and 206(1) was intended so as to exclude, for example, “relatives” of those prisoners awaiting trial and this suggests a broader purposive, rather than literal, approach to the construction of the term “friends”. 31. Although the Court was taken through the legislative history of the Prison Rules, that exercise does not materially assist. In each of the precursors to the current version of the Prison Rules,[19] there has consistently been a provision that prisoners awaiting trial are subject to the general rules applicable to other prisoners except insofar as they are inconsistent with specific rules relating to them. The precursor Prison Rules have had their equivalent rules to the current Rules 48 and 203 but I do not consider that the wording of those precursors provides a clear answer to the question of construction of those current rules now under consideration. It appears from Hansard that the 1954 version of the Prison Rules were intended to be a complete revision of the old rules and to bring them up-to-date and so resort to previous versions of the rules prior to 1954 is of little assistance.[20] 32. Similarly, the United Nations’ Standard Minimum Rules for the Treatment of Prisoners[21] and the Nelson Mandela Rules[22] do not materially assist in the construction of Rule 203. Those international standards are, as their name suggests, merely prescribed minimum rules. Moreover, the visitation entitlements contained in those rules are framed in terms of visits by “family and friends” and so do not imply the necessity of a wider category of visitors.[23] (b) What is the meaning of “friends” and were the appellants within that meaning in relation to the prisoners they visited? 33. The conclusion that “visitors” in Rule 203 is to be construed to mean “relatives and friends” of the prisoners awaiting trial leads to the next question as to who constitute the “friends” of such a prisoner and whether the appellants could properly be said to be within that category of visitor. As noted at [19] above, the magistrate and Court of Appeal both concluded that under Rule 48 a “friend” of a prisoner must be a personal acquaintance of the prisoner and could not mean a stranger. In this appeal, the prosecution invited this Court to uphold that conclusion. 34. The reasoning leading to the Court of Appeal’s narrower definition of the word “friends” in the Prison Rules is contained in the CA Judgment at [58] which reads as follows: “As a piece of subsidiary legislation dealing with penal establishments, one of the primary concerns of the Prison Rules is the maintenance of discipline and order in the prison and the prevention of crime. These are the considerations that underpin the restrictions that the Superintendent may impose under Rule 48 and Rule 203. On the other hand, due regard must also be given to the benefits that social visits may confer on the prisoners. As will be seen shortly, it is universally accepted in modern civilized societies that family and social support is crucial for the prisoners’ adaption to prison life and overcoming adjustment problems and for their rehabilitation and facilitation of their future re-integration into society. With these considerations in mind, ‘friends’ in the context of the Prison Rules cannot possibly mean someone who is a stranger of the prisoner. He must be a personal acquaintance of the prisoner. He and the prisoner must know each other in order to give the desired effects and benefits to the social visits. The same definition of ‘friends’ applies to the entire statutory regime of visits in the Prison Rules, including Rule 48 and Rule 203.” 35. With respect, the reasoning in this paragraph provides neither a sound basis nor compelling policy for construing the word “friends” so restrictively. The fact that one of the primary concerns of the Prison Rules is the maintenance of discipline and order and the prevention of crime in prisons does not provide any logical basis for restricting the meaning of “friends”. There is no reason to think that giving that word a wider meaning to include the appellants would have any detrimental effect on discipline and order and there is certainly no question of the appellants’ activities being contrary to good order and discipline in LCKRC or otherwise for an unlawful purpose. In linking family and social support to prisoners’ “rehabilitation and facilitation of their future re-integration into society”, the Court of Appeal seems to have misapprehended the position of prisoners awaiting trial. Unlike convicted prisoners, prisoners awaiting trial have no need for rehabilitation since they are presumed innocent. In any event, this is not a good reason for holding that, for the purposes of providing a prisoner awaiting trial with the desired effects and benefits of social visits, the “friend” who performs the function of providing the moral and material support to which the prisoner is entitled must necessarily be his personal acquaintance and that the two must be known to one another. 36. For the reasons which follow, I would respectfully disagree with the courts below as regards the meaning of “friends”. I consider that a wider definition should be given to the word “friends” to include persons in the position of the appellants. Before articulating what I consider to be a comprehensive practical definition of “friends” in Rule 48, I shall first explain my reasons for concluding that, for the purposes of the Prison Rules, the appellants in this case should clearly be considered to be “friends” of the prisoners awaiting trial that they visited. 37. The dictionary definition of the word “friend” includes a wide range of meanings including “a person joined by affection and intimacy to another”, “a near relation”, “a person who is not hostile or an enemy to another; one who is on the same side”, “a person who wishes another, a cause, etc., well”, and “an acquaintance, an associate; a stranger whom one comes across or has occasion to mention again”.[24] 38. This wide range of meanings is reflected in In re Barlow’s Will Trusts,[25] where Browne-Wilkinson J (as he then was) noted, in relation to the meaning of the word “friends” in a will, that: “The word has a great range of meanings; indeed, its exact meaning probably varies slightly from person to person. Some would include only those with whom they had been on intimate terms over a long period; others would include acquaintances whom they liked. Some would include people with whom their relationship was primarily one of business; others would not. Indeed, many people, if asked to draw up a complete list of their friends, would probably have some difficulty in deciding whether certain of the people they knew were really ‘friends’ as opposed to ‘acquaintances.’” 39. The difficulties in defining the relationship of friend as requiring personal acquaintance is further complicated by the general use of the term “friend” to describe a number of what may be relatively impersonal relationships. Thus, for example, the law recognises a “McKenzie friend” in the context of litigation who is a person who may sit with a litigant in person to give him advice and help him with the presentation of his case.[26] Similarly, under RHC Order 80 rule 2(1), a person under disability may not bring, or make a claim, in any proceedings except by his “next friend”, which capacity does not require a relationship having any particular degree of intimacy. To take a further example, a “friend of the court” (amicus curiae) does not imply personal friendship between the advocate and the judge. Nor, of course, is such personal friendship implied by the professional courtesy of barristers referring to each other in court as “my learned friend”. 40. Apart from the wide meaning of the word itself, in the context of the Prison Rules, there are practical difficulties in evaluating the quality of relationship that should qualify a person as a “friend” of a prisoner. Does a friend have to be on intimate or close terms to the prisoner? Is a Facebook friend, whom the prisoner may never have met, sufficient? What about other social media contacts or “pen pals” or friends of friends? As the procedure for visiting prisoners at LCKRC described above shows, there is no practical way in which CSD staff can verify the degree of any claimed friendship. A prisoner is left to determine who he is prepared to identify as his relatives and friends when he completes the Declared Visitors List and it would not seem realistic for CSD staff to vet this list to ascertain the quality of an asserted friendship. 41. A principled approach to determining the meaning of “friends” in Rule 48 in the present case is to examine the word by reference to the purposes for which visits by friends are made to prisoners awaiting trial. Those will be similar to the purposes for which visits are made to convicted prisoners subject to differences arising from the important distinction that the imprisonment of convicted prisoners is in part rehabilitative in nature. The most important purpose of any prison visit to prisoners awaiting trial is contact with persons outside the prison and the provision of moral and material support. But the purpose of visits to prisoners awaiting trial is additionally to enable them to take advantage of the more liberal regime to which they are subject. The entitlement of a prisoner awaiting trial to procure for himself food and malt liquor (Rule 192), his own clothes (Rule 196), newspapers and other means of occupation (Rule 202(2)) are of no practical benefit if these things cannot actually be supplied to him. 42. In practice, it is easy to envisage situations in which relatives and close personal friends will be unable to visit a prisoner awaiting trial. They may be working or otherwise unable to attend during visiting hours or they may be incapacitated through ill health or other indisposition. Yet such persons may have a strong desire to visit a prisoner awaiting trial and the prisoner himself may have a corresponding desire to be visited by them for the purposes contemplated by the Prison Rules. For prisoners awaiting trial who are overseas nationals with no family or social network in Hong Kong, the practical difficulties are all the more obvious. 43. In those circumstances, if (say) the disabled grandmother of a prisoner awaiting trial wanted to deliver some food that the prisoner particularly liked or to convey a personal message of support to him, for example, it would seem odd if the grandmother could not send her own friend to deliver that food or convey that message. It would do no violence to the language of Rule 48 construed purposively to describe that friend of the grandmother as a friend of the prisoner. Objectively, looked at from the point of view of the prisoner, the grandmother’s friend would almost certainly be regarded by him as a friend lending support or performing a service for his benefit and be willingly received as a visitor. 44. In the present case, the appellants were contracted by relatives of the prisoners awaiting trial whom they visited and they were requested to perform the functions that would otherwise have been performed by family members or personal friends (see the description at [9] above). The representative visiting services which they offered consisted of providing to prisoners awaiting trial the support they might receive as contemplated under the Prison Rules. In those circumstances, it does not stretch the meaning of the word “friend” to describe the appellants as friends of the prisoners awaiting trial whom they visited. The appellants would, in these circumstances, properly be described from the point of view of the prisoner as his friends. Similarly, this would be the case even if the services were contracted directly by the prisoner himself. For these reasons, I would conclude that the appellants were “friends” of the prisoners awaiting trial whom they visited. 45. Having rejected the Court of Appeal’s definition of “friends”, I turn to consider how the word “friends” in Rule 48 should be defined. For the reasons set out above, I consider that, as well as personal acquaintances, “friends” of such a prisoner can also include a person: (a) who has been requested to visit the prisoner, either directly by the prisoner himself or indirectly through a relative or personal acquaintance of the prisoner; (b) who wishes to visit the prisoner in order to provide him with some moral or material benefit consistent with the statutory purposes of visits to that category of prisoner; and (c) by whom the prisoner is willing to be visited. If those conditions are satisfied, the visitor should, objectively, be regarded as a “friend” of the prisoner. 46. The magistrate was concerned that a wider definition of “friends” which could include persons not known to each other would extend its meaning too far and frustrate the Prison Rules. In the SoF at [27], she held: “Given the special nature of the CSD and the need to restrict the source of visitors, if people who do not know each other can be called “friends”, its meaning would extend boundlessly, and all people could be broadly divided into two categories, namely “relatives” and “friends”. As such, the rules would be completely ineffective and unenforceable.” 47. I do not agree. The definition proposed above (at [45]), so far as it relates to persons not previously personally acquainted, has specific limits and is not “boundless”. It does not, with respect, render the Prison Rules ineffective or unenforceable. On the contrary, it seeks to formulate the concept of friendship in Rule 48 in the context of the Prison Rules as a whole and for the purposes of those rules. Obviously, the putative “friend” of a prisoner must have a legitimate reason for visiting him and must be seeking to do so in good faith and not for some improper ulterior purpose. 48. I would add that the above definition of the word “friends” does not make redundant any residual category of visitors to whom special authority might be granted under Rule 48. A person not fulfilling the three conditions set out in the definition might still wish to apply for special authority to visit a prisoner. Representatives of religious groups or other voluntary organisations are examples of persons who might fall into this category. It is also to be noted that the Prison Rules themselves provide for visits by visiting justices (Part III, Division 1) and prison visitors (Part III, Division 2). The special authority category in Rule 48 is consistent with the existence of these separate special classes of visitors who may be strangers to the prisoners they visit. 49. In the course of argument, the concept of agency was advanced as a basis for holding that representative visitors like the appellants were friends of the visited prisoners since they were agents acting on behalf of the prisoners’ relatives and friends. However, although agency has a superficial attraction in the case of relatives and friends who engage the appellants’ services, I do not consider that it provides an appropriate or comprehensive principle on which to define friends in the Prison Rules. Whilst relatives and friends may appoint an agent who may be said to visit a prisoner on their behalf, the agency concept cannot work for a prisoner who himself contracts for the representative visiting services offered by the appellants, since the prisoner’s agent represents the prisoner himself and a prisoner cannot visit himself under the Prison Rules. The constitutional questions 50. The conclusions reached above as to the proper construction of the Prison Rules mean that the appellants’ statements on the Visit Request Slips that they were friends of the prisoners whom they visited were true and not misrepresentations. There could accordingly be no question of the CSD officers being induced to act contrary to their public duty to admit persons who were not entitled to visit the prisoners concerned. On this basis, the appellants’ appeals must be allowed. 51. It therefore follows that it is unnecessary to consider the constitutional questions raised (set out at [8] above) concerning the compatibility of Rule 203 with Articles 6(2)(a) and 14 of the Hong Kong Bill of Rights and the circumstances of this case do not involve any interference with or restriction of rights raising any question of compatibility with the Hong Kong Bill of Rights Ordinance.[27] In keeping with the Court’s general practice, where a constitutional issue is raised but it is not necessary to determine that issue for the purpose of determining the appeal before it, the Court usually will not address that constitutional issue: see, for example, Fateh Muhammad v Commissioner of Registration & Anor (2001) 4 HKCFAR 278 per Bokhary PJ at 287C-E; and GA v Director of Immigration (2014) 17 HKCFAR 60 per Ma CJ at [42]. Substantial and grave injustice 52. The above conclusions as to the proper construction of the Prison Rules also mean that it is strictly unnecessary to consider the appellants’ submissions that substantial and grave injustice was done to them in that the evidence was incapable of sustaining a conviction of conspiracy to defraud. Nevertheless, the charges against the appellants being serious and involving dishonesty, it is appropriate to address them briefly since, in my view, there are two respects in which the evidence fell short of establishing necessary elements of the charge against the appellants. 53. The first respect relates to the element of dishonesty, which was fundamental to the alleged conspiracy to defraud. Although the appellants did not give evidence as to what they understood by the word “friend”, the meaning of that word in the Prison Rules was by no means clear and settled. The word “friend” being inherently broad (see above), it does not follow automatically, as the magistrate and Court of Appeal seem to have thought, that the appellants must have known that the correct definition of the word was that found by the courts below. In determining if a defendant has made a dishonest misrepresentation, it is necessary to consider whether he honestly believed the representation to be true in the sense in which he understood it. As Lord Jenkins, giving the opinion of the Privy Council in Akerhielm v De Mare, observed: “The question is not whether the defendant in any given case honestly believed the representation to be true in the sense assigned to it by the court on an objective consideration of its truth or falsity, but whether he honestly believed the representation to be true in the sense in which he understood it albeit erroneously when it was made. This general proposition is no doubt subject to limitations. For instance, the meaning placed by the defendant on the representation made may be so far removed from the sense in which it would be understood by any reasonable person as to make it impossible to hold that the defendant honestly understood the representation to bear the meaning claimed by him and honestly believed it in that sense to be true. But that is not this case.”[28] 54. Here, there were no guidelines from the CSD as to the meaning of the word, which is undefined in the Prison Rules. When the representative visiting services conducted by the appellants first came to the attention of the CSD in August 2011, legal advice was sought by the CSD from the Department of Justice concerning that service and it was not until August 2012 that the appellants were arrested.[29] There was evidence that CSD staff might differ over the meaning of the word. It is therefore entirely possible that the appellants might have believed that they were “friends” of the prisoners awaiting trial whom they visited within the meaning of the Prison Rules and it does not follow, merely because they did not give evidence, that the prosecution case that they dishonestly misrepresented their status as friends must be accepted. 55. The second respect in which the evidence was insufficient to support the charge against the appellants related to the evidence of the existence of a conspiracy agreement. The alleged conspiracy was an agreement to carry out the representative visiting services by dishonest means, namely by falsely representing that they were friends of the prisoners awaiting trial to be visited. Although the appellants were from the same company and there was therefore necessarily a uniformity in their actions, there were only two categories of relationship on the Visit Request Slips for them to complete. If they were not a “relative” of the prisoners, the only other possible status was “friend”. The mere fact that the appellants were all from the same company does not support the conclusion that the only irresistible inference here was that they all agreed together that they would select the category “friends” to insert in the Visit Request Slips in order to induce the CSD staff to admit them as visitors. 56. In the course of argument, Lord Collins of Mapesbury NPJ observed that the prosecution of the appellants for conspiracy to defraud might be described as “heavy handed”. I would respectfully agree. The present case involved a company providing visiting services to prisoners awaiting trial for relatives and friends of the prisoners unable to attend in person and their activities were known to the authorities for some time. The company’s services (described at [9] above) provided at a modest cost were entirely within the purposes for which visits to such prisoners are permitted under the Prison Rules. Staff of the company openly distributed leaflets outside LCKRC and wore green shirts as a uniform to identify themselves as representatives of the company. There was no clear consensus as to what the word “friend” in the Prison Rules meant. In these circumstances, it is a matter of some surprise that the appellants were charged as they were. Conclusion 57. For the above reasons, I would allow the appellants’ appeals and quash their convictions. Lord Collins of Mapesbury NPJ: 58. I agree with the judgment of Mr Justice Fok PJ. (Joseph Fok)				Permanent Judge (Lord Collins of Mapesbury)				Non-Permanent Judge Mr Johannes Chan SC and Mr Douglas Kwok, instructed by Tang, Wong & Chow, assigned by the Director of Legal Aid, for D1/the Appellant in FACC 6/2017 Mr Eric TM Cheung, Solicitor Advocate, instructed by ONC Lawyers,	assigned by the Director of Legal Aid, for D2/the Appellant in FACC 7/2017 Mr David Leung SC, DPP, Ms Audrey Parwani, SPP and Ms Gladys Chan, PP, of the Department of Justice, for the Respondent in FACC 6 & 7/2017 [1] (Cap.234A). The current version of the Prison Rules is dated 2015 and, save where otherwise indicated, references in this judgment to the Prison Rules will be to that version. [2] There were a total of nine defendants charged. D4, D5 and D6 pleaded guilty before trial and the trial proceeded against the appellants (who were D1 and D2 respectively) and D3, D7, D8 and D9. [3] Contrary to common law and section 159C(6) of the Crimes Ordinance (Cap.200). [4] Together with the other defendants who had pleaded not guilty (i.e. D3, D7, D8 and D9). [5] Ms Kennis Tai Chiu-ki, sitting as a Deputy Magistrate, in KTCC 2097/2013. [6] D1 and D2 were respectively sentenced to perform 240 hours and 120 hours of community service. [7] Pursuant to section 118(1)(d) of the Magistrates Ordinance (Cap.227). [8] Yeung VP, Poon & Pang JJA. [9] HCMA 700/2013, Judgment dated 17 October 2016 (“CA Judgment”). [10] HCMA 700/2013, Judgment dated 24 January 2017. [11] FAMC Nos. 10/2017 & 17/2017 (Ribeiro, Tang & Fok PJJ), Determination dated 27 October 2017. [12] Visits by volunteer groups were handled differently from the above procedure applying to relatives and friends: KTCC 2097/2013, Statement of Findings (“SoF”) at [9(b)]. [13] (Cap.234). [14] SoF at [19] & [21]; CA Judgment at [55] & [56]. [15] SoF at [26]-[28]; CA Judgment at [57]-[58]. [16] Courtauld v Legh (1869) LR 4 Exch 126 per Cleasby B at 130; see also Bennion on Statutory Interpretation (6th ed., 2013), at p.1034; and R v Kansal (No 2) [2002] 2 AC 69 per Lord Hutton at 117G-H. [17] To this extent, I respectfully disagree with the CA Judgment at [54] which holds that “visitors” in sub-paragraphs (f) and (fa) of Rule 48 must also apply to persons who are given special authority to visit under Rule 48. This construction is contrary to the plain language of Rule 48. Instead, it would appear that special authority visitors will be subject to the discretion of the Commissioner of the CSD or his duly authorised representative as regards the procedures for their visits to prisoners. [18] And so subject to s.18(3) of the Interpretation and General Clauses Ordinance (Cap.1). [19] Prison Rules 1885 (No.18 of 1885); Prison Rules 1925 (originally No.4 of 1899); and the Prison Rules 1954 (No.17 of 1954). [20] Hong Kong Hansard, Session 1954, in the speech of the Attorney-General at p.158; and see also para.[10] of the Objects and Reasons for the Bill at p.160. [21] (“SMR”); Adopted by the First United Nations Congress on the Prevention of Crime and the Treatment of Offenders, held at Geneva in 1955, and approved by the Economic and Social Council by its resolutions 663C (XXIV) of 31 July 1957 and 2076 (LXII) of 13 May 1977. [22] Resolution adopted by the General Assembly on 17 December 2015, A/RES/70/175. [23] See, SMR Rules [37], [87] and [92]; and Nelson Mandela Rules, Rule 58(1). [24] Shorter Oxford English Dictionary (6th Edition), Vol.1, p.1042. [25] [1979] 1 WLR 278 at 281B-D. [26] The term comes from the case of McKenzie v McKenzie [1971] P 33; see also Lobo v Kripalani [1998] 2 HKLRD 325 per Godfrey JA (as he then was) at 328F-H. [27] (Cap.383). [28] [1959] AC 789 at 805. [29] SoF at [9(g)]; CA Judgment at [22] & [27]. 
The Court: A. Introduction 1. This appeal concerns equality under the law and involves an application of the legal principles identified and applied in this Court’s recent decision in QT v Director of Immigration.[1]It arises in the context of a claim to entitlement to spousal medical and dental benefits under the Civil Service Regulations (“CSRs”) and to opt for joint assessment of salaries tax under the Inland Revenue Ordinance.[2] As will be seen, the appellant claims he has been unlawfully discriminated against on the basis of his sexual orientation. A.1 The parties 2. The appellant is a Hong Kong permanent resident of Chinese nationality. He commenced employment as an immigration officer with the Government of the Hong Kong Special Administrative Region in 2003 and, as such, is subject to the CSRs. He is homosexual and, in 2005, met his partner, Mr Scott Adams. The couple began cohabiting in 2013 and, on 18 April 2014, they were married in New Zealand. Same-sex marriage is legal in New Zealand and, upon their marriage there, the couple were issued with a New Zealand Marriage Certificate formally acknowledging their marriage and its registration by the Registrar of Births, Deaths and Marriages of New Zealand. 3. The 1st respondent is the Secretary for the Civil Service (“the Secretary”), whose responsibilities include the administration of the regulations which apply to civil servants, including the appellant. The 2nd respondent is the Commissioner of Inland Revenue (“the Commissioner”), who is responsible for administering the scheme of taxation in Hong Kong under the IRO. 4. The International Commission of Jurists (“ICJ”) appeared as intervener in this appeal and served written submissions in support of the appeal.[3] A.2 The challenged decisions 5. Under the CSRs,[4]a civil servant is entitled to the provision by the Government of various medical and dental benefits. These benefits are extended to a civil servant’s family, as defined in CSR 900(2), including his “spouse”. CSR 513 requires a civil servant: “to inform his Department immediately of … any change in his marital status, including marriage, divorce, or the death of his wife …”. In anticipation of his marriage to Mr Adams, the appellant wrote to the Civil Service Bureau to inquire if he was required to update his marital status pursuant to CSR 513, having regard to the fact that same-sex marriage is not recognised in Hong Kong. He was informed that his intended marriage to Mr Adams would not constitute a change in marital status for those purposes. 6. Following his marriage, the appellant wrote to the Secretary to complain that he had been denied the right to update his marital status and that his spouse was denied access to the spousal medical and dental benefits under the CSRs. The Secretary’s reply dated 17 December 2014 maintained that the appellant’s same-sex marriage with Mr Adams was not a marriage within the meaning of Hong Kong law, so that Mr Adams was not the appellant’s spouse for the purposes of the CSRs and that accordingly, Mr Adams was not entitled to the spousal benefits. This was the decision of the Secretary referred to in the courts below as the “Benefits Decision”. 7. Under section 10 of the IRO, the salaries tax of spouses is to be paid separately unless they elect to be jointly assessed.[5]In May 2015, the appellant sought to file his income tax return for the year of assessment 2014/2015 online, using the e-filing system of the Inland Revenue Department (“IRD”), but was unable to enter Mr Adams’ name, in order to make an election for joint tax assessment, as it had the same prefix as his own name. The appellant raised this issue with the IRD by email and referred to the IRD’s guideline stating that “spouse” meant “lawful husband or wife under a valid marriage recognized by Hong Kong law or other legal marriage recognized by the law of the place where it was entered into”. He claimed that, since he and Mr Adams were legally married in New Zealand, he was validly married to Mr Adams. 8. On 9 June 2015, the Commissioner replied to the appellant stating that a same-sex marriage was not regarded as valid for the purposes of the IRO because: “Although the definition of ‘marriage’ in section 2(1) [of the IRO] does not expressly oust one between persons of the same sex, it does make reference to a marriage between a ‘man’ and any ‘wife’. Under section 2, ‘husband’ means a married man and ‘wife’ means a married woman. ‘Spouse’ is defined under the same section as a husband or wife. Marriage in the context of the [IRO] is thus intended to refer to a heterosexual marriage between a man and a woman. Parties in a same-sex marriage cannot be ‘husband/wife’ and they would be incapable of having a ‘spouse’.” Accordingly, the Commissioner maintained that the appellant was not entitled to elect for joint assessment with Mr Adams because their same-sex marriage did not fall within the meaning of section 2(1) of the IRO. This was the decision of the Commissioner referred to in the courts below as the “Tax Decision”. 9. On 14 September 2015, the appellant filed a paper tax return for the year of assessment in question, in which he sought to elect for joint assessment with Mr Adams. This election was refused by the Commissioner on the ground that the appellant and Mr Adams were not husband and wife for the purposes of the IRO and the appellant’s income was assessed on an individual basis.[6] A.3 The proceedings below 10. The appellant challenged the Benefits Decision and the Tax Decision by way of judicial review proceedings. It was his contention that the decisions unlawfully discriminated against him on the ground of his sexual orientation. 11. In the Court of First Instance, by his judgment dated 28 April 2017, Chow J allowed the appellant’s application for judicial review of the Benefits Decision but dismissed his application for judicial review of the Tax Decision.[7]The Judge held that the Benefits Decision constituted differential treatment of the appellant on the basis of his sexual orientation.[8] As such, it required to be but was not, he held, justified by the 1st respondent and so amounted to unlawful discrimination.[9] In respect of the Tax Decision, the Judge held that, as a matter of statutory construction of the IRO, the appellant’s right to equality was not engaged and that his marriage was not a “marriage” for the purposes of the IRO.[10] 12. Against that judgment, the 1st respondent appealed and the appellant cross-appealed to the Court of Appeal. By its judgment dated 1 June 2018, the Court of Appeal allowed the 1st respondent’s appeal and dismissed the appellant’s cross-appeal.[11]The Court of Appeal allowed the 1st respondent’s appeal, holding that, although the Benefits Decision might constitute indirect discrimination against same-sex married couples on the ground of sexual orientation, it was justified as being no more than reasonably necessary to achieve the legitimate aim of protecting and not undermining the status of marriage as understood in Hong Kong.[12] 13. The Court of Appeal dismissed the appellant’s cross-appeal against the Tax Decision. It upheld Chow J’s decision as to the statutory construction of the IRO, namely that “marriage” for the purposes of that ordinance meant only an opposite-sex marriage and not a same-sex marriage. The Court of Appeal nevertheless also held that the Tax Decision might constitute indirect discrimination but that the construction of section 10 of the IRO confining the availability of election to joint assessment to heterosexual married couples was justified for similar reasons to the Benefits Decision to achieve the legitimate aim of protecting and not undermining the status of marriage as understood in Hong Kong.[13] A.4 Leave to appeal to this Court 14. The Court of Appeal granted the appellant leave to appeal[14]to this Court in respect of the following questions of great general or public importance: (1) “Question 1: (a) Is the legitimate aim of protecting and/or not undermining the concept and/or institution of marriage, being the voluntary union for life of one man and one woman to the exclusion of all others, as understood in and under the laws of Hong Kong, rationally connected to the difference in treatment, between a person who is a party to such a marriage and a person who is a party to a same-sex marriage entered into outside Hong Kong according to the law of the place in which it was entered, for the purpose of conferral of spousal benefits under the Civil Service Regulations; (b) Are the local legal landscape and societal circumstances including prevailing socio-moral values of society on marriage relevant to the issue of proportionality and/or justification; and (c) Has the First Respondent justified the difference in treatment?” (2) “Question 2: (a) Is the legitimate aim of protecting and/or not undermining the concept and/or institution of marriage, being the voluntary union for life of one man and one woman to the exclusion of all others, as understood in and under the laws of Hong Kong rationally connected to the difference in treatment, between a person who is a party to such a marriage and a person who is a party to a same-sex marriage entered into outside Hong Kong according to the law of the place in which it was entered, for eligibility for joint assessment under section 10 of the Inland Revenue Ordinance (Cap.112) (“IRO”); (b) Are the local legal landscape and societal circumstances including prevailing socio-moral values of society on marriage relevant to the issue of proportionality and/or justification; and (c) Has the Second Respondent justified the difference in treatment?” B. The applicable principles 15. It is unnecessary to set out the legal principles which apply in the present case at length because they were fully addressed in this Court’s judgment in QT. There was no dispute between the parties to this appeal as to those principles or that this appeal fell to be resolved by applying them to the facts of this case. In the circumstances, it is sufficient to summarise those principles briefly. B.1 The nature of discrimination 16. It is a cardinal principle of a system governed by the rule of law that all persons are equal before the law and that principle is enshrined in the Basic Law of the Hong Kong Special Administrative Region and the Hong Kong Bill of Rights: (1) Art. 25 of the Basic Law provides: “All Hong Kong residents shall be equal before the law.” (2) Art. 1(1) of the Hong Kong Bill of Rights provides: “The rights recognized in this Bill of Rights shall be enjoyed without distinction of any kind, such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.” (3) Art. 22 of the Hong Kong Bill of Rights provides: “All persons are equal before the law and are entitled without any discrimination to the equal protection of the law. In this respect, the law shall prohibit any discrimination and guarantee to all persons equal and effective protection against discrimination on any ground such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.” 17. As this Court made clear in its judgment in QT, unlawful discrimination is “fundamentally unacceptable” (QT at [27]-[28]). However, the law has to draw distinctions between different situations or types of conduct, to which different legal consequences may attach. Principles have therefore been established “for determining when distinctions drawn by legal or administrative measures are rational and fair and when such distinctions constitute unlawful discrimination” (QT at [29]). 18. It is now acknowledged that there are three forms of differential treatment, which may be described as discriminatory. In summary, these are: (i) direct discrimination where like cases are not treated alike; (ii) direct discrimination where unlike cases are treated in the same way; and (iii) indirect discrimination where an ostensibly neutral criterion is applied which operates to the significant prejudice of a particular group (QT at [31]-[33]). B.2 Differential treatment and the justification test 19. In every alleged case of discrimination, the correct approach is, first, to determine whether there is differential treatment on a prohibited ground and, only if this can be demonstrated, then, to examine whether it can be justified. Differential treatment which is justified does not constitute unlawful discrimination. However, where differential treatment is not justified it is unlawful discrimination (QT at [81]-[83]). 20. The initial step of determining whether there is differential treatment on a prohibited ground essentially involves a comparison exercise. As was said in QT (at [38]): “A person complaining about discrimination generally has in mind one or more comparators. The question asked is: Why is the complainant being treated less favourably than individuals in a relevant comparator group? Here, QT asks: ‘Why am I denied a dependant visa which would be granted to a married spouse of a sponsor?’ The Director's answer is essentially simply to state: ‘Because she is married and you are not’. As we have seen, the Director recognises someone as married only if he or she is a party to a marriage which, if celebrated here, would be valid under Hong Kong law, in other words, a party to a monogamous and heterosexual marriage, wherever it might have been contracted.” Accordingly, the initial step must be for the complainant to demonstrate that he or she has been treated differently to a person in a comparable position and that the reason for this difference in treatment can be identified as a prohibited ground, such as race, religion or sexual orientation. Only after this is demonstrated does it then become necessary to consider whether such differential treatment is lawful. If the treatment is held to be unlawful, then the complainant will be entitled to remedies. 21. In order to determine whether differential treatment is unlawful, the courts apply the same test used to determine if incursions into constitutionally protected rights are lawful (QT at [84]-[86]). When applied in the context of an analysis of constitutionality, that test is usually referred to as the “proportionality” test. When applied in the context of determining whether differential treatment is unlawful, that test is usually referred to as the “justification” test. 22. The justification test consists of four steps or elements: (i) does the differential treatment pursue a legitimate aim; (ii) is the differential treatment rationally connected to that legitimate aim; (iii) is the differential treatment no more than necessary to accomplish the legitimate aim; and (iv) has a reasonable balance been struck between the societal benefits arising from the application of differential treatment and the interference with the individual’s equality rights (QT at [86]-[87]). C. The parties’ respective cases 23. It is the appellant’s case that the Benefits Decision and the Tax Decision constitute unlawful discrimination against him on the ground of his sexual orientation. That is because the marriage criterion that is applied by the 1st respondent in respect of the conferment of the spousal benefits and by the 2nd respondent in respect of the availability of joint tax assessment for a married couple is limited to heterosexual married couples and excludes same-sex married couples. As a gay man, he is never going to be able to qualify within the marriage criterion as applied by the respondents. 24. The appellant contends that he is validly married to Mr Adams under New Zealand law and, as such, should be regarded as falling within the marriage criterion by which it is determined whether civil service spousal benefits and joint tax assessment are available. He contends that the Benefits Decision constitutes unlawful discrimination and is irrational and unreasonable and that section 10 of the IRO, as construed by the Commissioner as the basis for the Tax Decision, is unconstitutional as it unlawfully discriminates against same-sex couples contrary to the right to equality protected under Art. 25 of the Basic Law and/or Arts. 1(1) or 22 of the Hong Kong Bill of Rights. 25. For their part, the respondents contend that their respective decisions do not amount to unlawful discrimination against the appellant. It is, however, accepted (and this concession, see Section D.2 below, will be examined in closer detail in Section D.3 below) that there is differential treatment on the ground of sexual orientation amounting to indirect discrimination. It is therefore also accepted that, for the respondents to succeed in this appeal, they must justify that differential treatment applying the justification test. This, they contend, they can do (and this contention will also be examined in closer detail in Section E below). 26. It was common ground between the parties that the challenges to the Benefits Decision and the Tax Decision both fall to be considered applying the same principles and that, in effect, they stand or fall together. 27. Although the appellant asserts that his claim for unlawful discrimination on the basis of his same-sex marriage places him in a relevantly comparable position to a man married to a woman, this appeal does not concern the question of whether same-sex couples have a right to marry under Hong Kong law. In W v Registrar of Marriages,[15]which concerned the right of a post-operative transsexual woman to marry a man in her new gender as a woman, it was “common ground that a marriage for constitutional as for common law purposes is the voluntary union for life of one man and one woman to the exclusion of all others”.[16] As was the case in QT, it was not argued in this appeal that the constitutional freedom to marry and raise a family[17]makes marriage available to same-sex couples.[18] D. Whether differential treatment on the basis of sexual orientation D.1 The appellant’s case 28. In the appellant’s printed case, it is contended: “The Appellant’s married relationship with his husband is in substance completely indistinguishable from that of a heterosexual married relationship. The Appellant’s marriage is characterised by life-long commitment, monogamy, sexual intimacy and interdependence, exactly like a heterosexual (monogamous) marriage.”[19] 29. By letter from the Registrar to the parties prior to the hearing of the appeal, the Court invited written submissions from the parties on the following two questions: (1) “Whether the same sex marriage contracted by the applicant in New Zealand can, as a matter of Hong Kong law, be regarded as valid, in particular whether he had the capacity to enter into the marriage?” (2) “If not valid, what are the consequences for the purposes of this appeal?” 30. In the appellant’s answer to those questions, it is contended: “The Applicant’s case is that the fact of his relationship with Mr Adams, the nature and substance of which is evidenced inter alia by that marriage, puts him and Mr Adams in an analogous position to persons who are married in accordance with the marriage law of Hong Kong for the purposes of Civil Service benefits and taxation. This mirrors the position as it was in Director of Immigration v QT [2018] HKCFA 28, [2018] 4 HKC 403 for the purposes of immigration control.”[20] 31. As already noted at [2] above, the appellant and his husband began cohabiting in 2013 and, on 18 April 2014, they were married in New Zealand where same-sex marriage is legal. Importantly, upon their marriage in New Zealand, the couple were issued with a New Zealand Marriage Certificate formally acknowledging their marriage and its registration by the Registrar of Births, Deaths and Marriages of New Zealand. 32. The appellant’s case is that, by the respondents’ application of an exclusively heterosexual marriage criterion in reaching the Benefits Decision and the Tax Decision, he is being discriminated against on the ground of his sexual orientation. D.2 The respondents’ concession of differential treatment 33. The respondents accept, for the purposes of this appeal, that “a same-sex married couple is in an analogous position to that of an opposite-sex married couple”.[21]This is an important part of the respondents’ concession and constitutes a crucial aspect in the circumstances of this case. 34. It was also accepted by the respondents, for the purposes of this appeal, that the denial of spousal benefits to a same-sex married couple and their inability to elect for joint tax assessment “constitutes indirect discrimination against same-sex couples on the ground of their sexual orientation if not justified, as they cannot get married lawfully under the prevailing marriage laws of Hong Kong”.[22] 35. The respondents maintained this stance in their responses to the Court’s questions referred to at [29] above.[23]They did so notwithstanding that they contended that the appellant’s same-sex marriage contracted in New Zealand could not be regarded as valid as a matter of Hong Kong law. That invalidity flowed, it was contended, from the fact that, under the conflict of laws rules in Hong Kong, capacity to enter into a marriage is a matter of essential validity and is determined by reference to the law of each party’s place of ante-nuptial domicile.[24] 36. It is relevant to add, however, that the respondents did not contend, in response to the Court’s questions, that the appellants’ same-sex marriage was also invalid as a matter of New Zealand law. This argument had been alluded to in written submissions made on behalf of the respondents below in the Court of Appeal but the Court of Appeal noted that this point had been abandoned.[25]For that reason, and because there was no evidence of New Zealand law to suggest the contrary, the appeal before this Court proceeded on the basis that, as a matter of New Zealand law, the appellant and Mr Adams are lawfully married in that jurisdiction. D.3 The appellant is subject to differential treatment requiring justification 37. The need to justify the differential treatment of the appellant being conceded by the respondents, it might be thought unnecessary to discuss the issue of whether there is relevantly differential treatment in this case. As the Court’s questions (identified at [29] above) demonstrate, though, there might well have been some scope for debate as to the true boundaries of the analogous situations existing between same-sex couples and opposite-sex couples. Not all couples are in a truly analogous situation and the use of the status of marriage itself as a distinguishing characteristic may or may not be determinative of the question of whether there is relevant differential treatment in any given case. As the Court held in QT: “This is not to suggest that a person’s marital status is irrelevant as a condition for the allocation of rights and privileges. Such status may in some circumstances be highly important or even decisive. The point we make is that the relevance and weight to be attributed to that status is taken into account in considering whether a particular difference in treatment is justified as fair and rational, and that a person’s marital condition cannot determine presumptively that discrimination does not exist.”[26] 38. The question of whether treatment is relevantly different such as to require justification is always a matter that is context dependent. The Court observed in QT: “44. The second major objection to the Director’s first argument is that the identification of comparators does not of itself permit a proper conclusion to be reached as to whether a given difference in treatment is or is not discriminatory. As Lord Walker pointed out in R (Carson) v Secretary of State for Work and Pensions, the real issue in the case at hand was: ‘… why the complainant had been treated as she had been treated. Until that question was answered, it was impossible to focus properly on the question of comparators.’ 45. The notion of whether the comparators are analogous or relevantly similar is elastic both linguistically and conceptually. As his Lordship pointed out in the same judgment: ‘Some analogies are close, others are more distant’. It is therefore generally unprofitable to debate in the abstract whether a given comparator is or is not sufficiently analogous to require like treatment. The context of the question is crucial.”[27] 39. In QT, the relevant context was immigration control and the issue of dependency visas to persons from overseas who were dependents of those who had been admitted to live and work in Hong Kong. QT and SS had entered into a same-sex civil partnership in England under the UK’s Civil Partnership Act 2004. In the context of that case, the relevant comparison was between QT and her same-sex partner SS and a heterosexual married couple from overseas since it was the differential application of the dependency visa policy to both couples that gave rise to the differential treatment of QT on the basis of her sexual orientation. The Court held that, in the immigration context for the purposes of the dependency visa policy, there was no obvious difference between homosexual civil partners and heterosexual married couples: each status was recognised under UK law, and both homosexual and heterosexual couples were capable of having equivalent interdependent and interpersonal relationships.[28] 40. The present case is concerned with the conferment of financial benefits on spouses in the contexts of employment and taxation. Those benefits are conferred on the basis of marriage. The nature of the relationship between the appellant and Mr Adams is one of same-sex marriage valid under the law of the place where it was entered into. It is a relationship which has the same characteristics of publicity and exclusivity which distinguish a heterosexual marriage. 41. The characteristic of publicity is established by the formality of the marriage entered into by the couple under the laws of New Zealand and by the New Zealand Marriage Certificate issued to them under the Births, Deaths, Marriages, and Relationships Registration Act 1995. That certificate, which bears a formal registration number, states: the full particulars of the first and given names and surnames of the parties to the marriage; their sex, age, date and place of birth; their usual occupation, relationship status and usual residential address; and also full particulars of the first and given names and surnames of the couple’s parents. The date and place of the marriage are also recorded and the certificate states that it is a true copy of particulars recorded by a Registrar. 42. Similarly, in Hong Kong, there are detailed provisions for publicity of a marriage prescribed in the Marriage Ordinance.[29]Notice of an intended marriage must be given to the Registrar of Marriages either directly or through a civil celebrant.[30] The Registrar is required to file and exhibit every notice of intended marriage in his office.[31] A marriage may be celebrated in a licensed place of worship or before a civil celebrant[32]and there are prescribed forms of marriage certificate that must be delivered to the parties in both such ceremonies.[33] The Registrar is required to register all certificates of marriage[34]and such certificates of marriage are admissible as evidence of the marriage to which they relate.[35] Searches of the register of marriages may be made and certified copies of marriage certificates obtained.[36] 43. The characteristic of exclusivity of the appellant’s and Mr Adams’ same-sex marriage is constituted by the consequence of their marriage under New Zealand law. Unless and until that marriage is legally dissolved in New Zealand, it must be taken to be valid and subsisting under that law. The appellant will not be free to marry someone else under New Zealand law whilst his marriage to Mr Adams subsists. 44. Exclusivity is also an essential characteristic of a marriage under Hong Kong law. Under the Marriage Ordinance, the form of marriages under that ordinance is limited to “Christian marriage or the civil equivalent of a Christian marriage”[37]and such a marriage “implies a formal ceremony recognized by the law as involving the voluntary union for life of one man and one woman to the exclusion of all others”.[38] The fact that one or other of the parties to an intended marriage was already lawfully married at the time of the marriage renders such a marriage void and is a ground for a decree of nullity under the Matrimonial Causes Ordinance.[39] Again, unless the marriage is dissolved in court proceedings, it must be taken as valid and subsisting. 45. In her submissions on behalf of the appellant, Ms Karon Monaghan QC,[40]pointed to the appellant’s New Zealand marriage certificate as evidence of his intimate relationship bearing family incidents with Mr Adams. That it may well be, as a matter of fact, but the material fact that the marriage certificate establishes is that the appellant and Mr Adams are in a valid same-sex marriage under the laws of New Zealand, where such marriages are lawfully recognised. It is not just the relationship between the appellant and Mr Adams that is important for the purposes of determining if there is differential treatment in respect of the Benefits Decision and the Tax Decision. A mere relationship (whether opposite-sex or same-sex) will not have the same readily identifiable characteristics of publicity and exclusivity described above that positively identify a same-sex married couple as being in materially the same position as an opposite-sex married couple. It is to be noted that the relationship between QT and SS in QT, having entered into a civil partnership, similarly had these characteristics of publicity and exclusivity. 46. For these reasons, the respondents’ concession of differential treatment requiring justification was properly made. In the context of the present case, concerned with financial spousal benefits, a same-sex married couple and an opposite-sex married couple are relevantly analogous and the appellant was treated differently to a heterosexual married man in respect of the two challenged decisions on the ground of his sexual orientation. E. Whether differential treatment justified 47. There being differential treatment of the appellant on the basis of sexual orientation, it was common ground that the respondents bear the burden of justifying that treatment or else it will constitute unlawful discrimination. As noted at [22] above, the justification test is a four-step analysis and the respondents must show that the differential treatment satisfies each of the four steps in question. E.1 The legitimate aim identified by the respondents 48. The legitimate aim relied upon by the respondents in this case has been articulated in various, slightly different, ways. 49. First, as identified in the questions for which leave to appeal to this Court was given, the legitimate aim is said to be that “of protecting and/or not undermining the concept and/or institution of marriage, being the voluntary union for life of one man and one woman to the exclusion of all others, as understood in and under the laws of Hong Kong”. This “not undermining the institution of marriage” aim reflects the way in which the legitimate aim was expressed in the evidence filed on behalf of the respondents in opposition to the appellant’s judicial review application.[41] 50. Secondly, the legitimate aim is said to be that “of protecting and not undermining the institution and unique status of marriage as understood and recognised in Hong Kong”.[42]This “protection of the unique status of marriage” aim is consistent with the way the legitimate aim was identified by each member of the Court of Appeal: (1) Cheung CJHC (as Cheung PJ then was) referred to the fact that “by allowing same-sex couples to share in benefits or privileges that up till now have long been exclusively associated with or enjoyed by married couples, one is undermining the uniqueness of the status of marriage in society”;[43] (2) Lam VP said: “The unique status of marriage, as the laws in Hong Kong presently stand, is only confined to heterosexual marriages. It is important to bear this in mind in the application of the proportionality test in the present case where protection of marriage is put forward as the justification for the differential treatments of the [appellant].”[44] (3) Poon JA held: “Heterosexual marriage, as the only form of marriage recognized in Hong Kong and deeply imbedded in our legal system, must for obvious reason deserve full protection of the law.”[45] 51. Thirdly, Lord Pannick QC, on behalf of the respondents,[46]expressed the legitimate aim in terms of protecting the coherence of the law of Hong Kong on the subject of marriage. It was contended that there is a legitimate aim in avoiding the incoherence at the level of the laws of Hong Kong that would arise if the respondents were required to extend marital benefits to the appellant even though, as a matter of Hong Kong law, he is not married. This third variation of the legitimate aim argument may be described as the “coherence” aim. 52. Each of these ways of expressing the legitimate aim relied upon by the respondents is a variation of a similar rationale that was sought to be raised on behalf of the Director of Immigration at the hearing of the appeal in this Court in QT but which the Court did not permit the Director to argue as it had formed no part of the analysis of the courts below in that case and had not been addressed in evidence or submission by the parties.[47] E.2 Illegitimate or irrelevant aims 53. Before addressing the three expressions of the legitimate aim identified at [49] to [51] above, it is convenient to identify two rationales for the differential treatment of the appellant that are either illegitimate or irrelevant. These rationales are alluded to in the main judgment in the Court of Appeal given by Poon JA (with which the other two members of the court agreed). 54. The first such argument is based on the concept of core rights and benefits unique to marriage. This was referred to by Poon JA in Section G of the CA Judgment. Although he rejected the argument that differential treatment of same-sex couples in respect of spousal benefits and the right to elect joint assessment did not require to be justified, he examined them in terms of whether they were core rights.[48]The “core rights” argument had its genesis in the Court of Appeal’s judgment in QT. In this Court’s judgment in QT (which was handed down after the judgment of the Court of Appeal in this case), the Court rejected the Court of Appeal’s approach of recognising that there were certain core rights pertaining to marriage and that differential treatment based on those core rights could not be regarded as discriminatory.[49] It is therefore illegitimate to speak in terms of core marriage rights or to regard differences of treatment as not being discriminatory because they relate to such rights. 55. The other argument that can be shortly disposed of is the suggestion that prevailing views of the community on marriage are relevant to identifying a legitimate aim and justification of differential treatment. In the Court of Appeal, Poon JA said “protecting and not undermining the status of marriage in light of the prevailing views of the community on marriage as mentioned in Part G1 above is plainly a legitimate aim.”[50]In Part G1 of the CA Judgment, Poon JA referred to the fact that: “As demonstrated by the results of recent surveys, public opinion on same-sex marriage remains divided with the majority firmly against it. The corollary is that the majority of the community still regards heterosexual marriage as the only acceptable form of marriage in the local social circumstances.”[51] 56. In Section F.9 of their joint judgment in W v Registrar of Marriages, Ma CJ and Ribeiro PJ rejected the absence of a majority consensus as a reason for rejecting a minority’s claim as being inimical in principle to fundamental rights. They quoted with approval the extra-judicial comments of the Chief Justice of Ireland, Murray CJ, in the following terms: “… The use of consensus as an interpretive tool is inherently problematic, not only because of any perceived inconsistency in the application of the doctrine by the [ECtHR], but fundamentally because the very application of a doctrine of consensus by a court required to adjudicate on fundamental rights begs important questions of legitimacy. How can resort to the will of the majority dictate the decisions of a court whose role is to interpret universal and indivisible human rights, especially minority rights?”[52] In his concurring judgment, Bokhary NPJ similarly rejected societal consensus as a relevant consideration.[53] 57. It follows therefore that the “prevailing views of the community on marriage” as identified by Poon JA, even if this can confidently be gauged in the first place, are simply not relevant to a consideration of the justification exercise. To the extent that Questions (1)(b) and (2)(b) (set out at [14] above) refer to the “prevailing socio-moral values of society on marriage”, they are to be answered in the negative. E.3 Is the aim identified legitimate? 58. The appellant accepted that it has been held “that in principle the protection of the ‘traditional family’ as constituted by heterosexual marriage may constitute a legitimate aim”[54]and the respondents relied on the appellant’s acceptance of this legitimate aim.[55] To this extent, the question of whether the aim of the differential treatment is legitimate was common ground between the parties. 59. The respondents did not suggest that there was any difference in substance between the three ways in which the legitimate aim relied upon was articulated (see Section E.1 above) or that any of them differed in substance from what, as noted above, the appellant accepted to be a legitimate aim. They are materially the same as the argument which the Director of Immigration was not permitted to raise in QT (see [52] above) and the slightly different ways the legitimate aim has been put were, as Lord Pannick QC described them in his oral submissions, complementary and “variations on a theme”. 60. The proposition that the protection of the traditional family constituted by heterosexual marriage is a legitimate aim is supported by a number of authorities, including the House of Lord’s decision in Ghaidan v Godin-Mendoza[56]: see also, Mata Estevez v Spain,[57]Karner v Austria,[58]In re G (Adoption: Unmarried Couple)[59]and Kozak v Poland.[60] In Şerife Yiğit v Turkey, the European Court of Human Rights stated: “With regard to art.12 of the Convention, the Court has already ruled that marriage is widely accepted as conferring a particular status and particular rights on those who enter it. The protection of marriage constitutes, in principle, an important and legitimate reason which may justify a difference in treatment between married and unmarried couples. Marriage is characterised by a corpus of rights and obligations that differentiate it markedly from the situation of a man and woman who cohabit. Thus, states have a certain margin of appreciation to treat differently married and unmarried couples, particularly in matters falling within the realm of social and fiscal policy such as taxation, pensions and social security.”[61] 61. There can be no doubt, therefore, that the protection of the institution of marriage in Hong Kong, being the voluntary union for life of one man and one woman to the exclusion of all others, is a legitimate aim and that differential treatment directed to that aim may be justified if the other elements of the justification test are satisfied. To this extent, in answer to Questions (1)(b) and 2(b) of the certified questions of law, the protection of the institution of marriage as defined under, and in its context within, the laws of Hong Kong is part of “the local legal landscape and societal circumstances” for the purposes of the issue of proportionality and/or justification. 62. The real contest between the parties in this appeal was the question of whether the differential treatment of the appellant was rationally connected to that legitimate aim of the protection of the traditional family in the circumstances of the present case and it is that to which we now turn. E.4 Is the differential treatment rationally connected to the legitimate aim? 63. In addressing this question, it is necessary to remember that one has reached it having already concluded that a same-sex married couple such as the appellant and Mr Adams are in a relevantly analogous position to an opposite-sex married couple and that there has been differential treatment that requires justification (see Section D.3 above). 64. It is also important to keep in mind the context of the differential treatment when addressing this question, just as context is relevant in determining whether one is considering an analogous situation giving rise to differential treatment (see the citation at [38] above). 65. Here, as we have already noted, the relevant context is the conferment of financial benefits on spouses in the contexts of employment and taxation. Traditionally, those benefits were not conferred in order to protect the institution of marriage or even to encourage people to marry one other. Instead, they were provided to acknowledge the economic reality of the family unit with one member of a couple, usually the male, being the principle breadwinner for the family and, in the case of employment within the civil service, to encourage the recruitment and retention of staff. Medical and dental benefits were therefore extended to a civil servant’s spouse and dependent children as a perquisite of employment. Joint tax assessment helped to lessen the overall tax burden on a couple living together and meeting expenses traditionally from one source of earned income. It was (and is) no part of the Secretary’s or Commissioner’s functions that they were responsible for protecting (much less promoting) the institution of marriage. The Secretary’s principal responsibility was (and is) the efficient administration of government and that of the Commissioner was (and is) the raising of revenue through the taxation system. 66. In these circumstances, having concluded that the appellant has been subject to differential treatment because he is in a same-sex marriage rather than an opposite-sex marriage, one looks to see how denying the appellant spousal employment benefits (the Benefits Decision) and the right to elect for joint assessment (the Tax Decision) is rationally connected to the legitimate aim of protecting or not undermining the institution of marriage in Hong Kong. 67. It is here that the respondents’ case faces great difficulty. How is it said that allowing Mr Adams medical and dental benefits weakens the institution of marriage in Hong Kong? Similarly, how does permitting the appellant to elect for joint assessment of his income tax liability under the IRO impinge on the institution of marriage in Hong Kong? It cannot logically be argued that any person is encouraged to enter into an opposite-sex marriage in Hong Kong because a same-sex spouse is denied those benefits or to joint assessment to taxation. 68. As Lady Hale said, in Rodriguez v Minister of Housing (a Privy Council appeal from Gibraltar): “Privileging marriage can of course have the legitimate aim of encouraging opposite-sex couples to enter into the status which the state considers to be the most appropriate and beneficial legal framework within which to conduct their common lives. Privileging civil partnership could have the same legitimate aim for same-sex couples. But, to paraphrase Buxton LJ in the Court of Appeal’s decision in Ghaidan v Mendoza [2002] EWCA Civ 1533, [2002] 4 All ER 1162 at [21], it is difficult to see how heterosexuals will be encouraged to marry by the knowledge that some associated benefit is being denied to homosexuals. They will not be saying to one another ‘let’s get married because we will get this benefit and our gay friends won’t’.”[62] 69. Similarly, in her speech in the House of Lords in Ghaidan v Godin-Mendoza, Baroness Hale observed: “The traditional family is not protected by granting it a benefit which is denied to people who cannot or will not become a traditional family. What is really meant by the ‘protection’ of the traditional family is the encouragement of people to form traditional families and the discouragement of people from forming others. There are many reasons why it might be legitimate to encourage people to marry and to discourage them from living together without marrying. … But, as Buxton LJ [2003] Ch 380, 391, para 21 pointed out, it is difficult to see how heterosexuals will be encouraged to form and maintain such marriage-like relationships by the knowledge that the equivalent benefit is being denied to homosexuals. The distinction between heterosexual and homosexual couples might be aimed at discouraging homosexual relationships generally. But that cannot now be regarded as a legitimate aim. It is inconsistent with the right to respect for private life accorded to ‘everyone’, including homosexuals, by article 8 since Dudgeon v United Kingdom (1981) 4 EHRR 149.”[63] 70. Lord Pannick QC submitted that the Benefits Decision and the Tax Decision were rationally connected to the aim of protecting and not undermining the institution and unique status of marriage as understood and recognised in Hong Kong and invited the Court to uphold the Court of Appeal’s analysis concluding that the differential treatment was rationally connected to that legitimate aim. That analysis is variously addressed in the judgments of the Court of Appeal but in particular: (1) In the judgment of Cheung CJHC at CA Judgment [12]: “… by allowing same-sex couples to share in benefits or privileges that up till now have long been exclusively associated with or enjoyed by married couples, one is undermining the uniqueness of the status of marriage in society; one is in fact equating relationships which society does not recognise as a marriage relationship and is not prepared to accord the status of marriage to, with what society regards as constituting a marriage relationship and is prepared to accord the status of marriage to. To do so, in the eyes of those holding the objection, is to offend, challenge, question, confuse, or subtly change society’s established understanding and concept of marriage, which, as explained, is rooted in its traditional, historical, social, moral or religious background and values, as embedded in article 37 of the Basic Law. To do so, in other words, is to dilute or diminish the unique status of marriage in society. …” (2) And in the judgment of Poon JA at CA Judgment [90] and [126]: “90. … The protection must be full in the sense that the law must safeguard the special status of marriage from any impermissible inroads or encroachments lest the very institution of marriage, guaranteed by the Basic Law, would be shaken or would even collapse. … … 126. … In the local context, if Spousal Benefits and joint assessment, which have been long associated closely and exclusively with marriage, were made available to homosexual couples, it would per se undermine, or be perceived by many to undermine, the status of marriage. Thus using marital status as the benchmark is plainly rationally connected to the aim of protecting marriage in the societal context of Hong Kong.” 71. With great respect, we cannot agree with the Court of Appeal’s analysis. Restricting these financial benefits to opposite-sex married couples on the ground that heterosexual marriage is the only form of marriage recognised in Hong Kong law is circular and therefore proceeds on the fallacious basis rejected by the Court in QT at [42]. It amounts to the application of a self-justifying reasoning process and denies equality to persons of different sexual orientation who are accepted to be in a relevantly analogous position. Ultimately, a line is merely drawn without any further attempt to justify it. 72. In any event, we are unable to accept the proposition that heterosexual marriage would be undermined by the extension of the employment and tax benefits to same-sex married couples. Whilst the Court recognised in QT (at [76]) that a person’s marital status might well be relevant to the allocation of rights and privileges and that “the relevance and weight to be attributed to that status is taken into account in considering whether a particular difference in treatment is justified as fair and rational”, we are satisfied that this is not such a case. Heterosexual marriage is not promoted by the differential treatment in question. 73. The suggested rational connection between the differential treatment and the legitimate aim is all the more illogical in respect of the Benefits Decision when one takes into account the Government’s published policy as an equal opportunities employer: (1) The appellant’s evidence was that, when he applied to join the Government as a civil servant, the advertisement of vacancies was accompanied by a note similar to that which is presently appended to Government employment advertisements. That note states: “As an Equal Opportunities Employer, the Government is committed to eliminating discrimination in employment. The vacancy advertised is open to all applicants meeting the basic entry requirement irrespective of their disability, sex, marital status, pregnancy, age, family status, sexual orientation and race.”[64] (2) Furthermore, the Secretary has adopted a “Code of Practice against Discrimination in Employment on the Ground of Sexual Orientation”, which includes the following statement: “5.1 Government is committed to the principle of equal pay for equal work and encourages all employers to share that commitment. … [This] means that – in principle – all employees are entitled to the terms and conditions of employment or access to employment or access to benefits, facilities or services commensurate with their rank, duties, seniority and experience and any other special circumstances of their employment, irrespective of their sexual orientation.”[65] 74. It is difficult to see how the Secretary can adhere to the published employment policies dedicated to the elimination of discrimination on the ground of sexual orientation by denying to a married same-sex couple the same employment benefits that are available to a married opposite-sex couple. 75. The contention that there is a rational connection between the Tax Decision and the legitimate aim of protecting the institution of marriage as understood under Hong Kong law (i.e. heterosexual and monogamous) is, similarly, further undermined by the fact that section 2(1) of the IRO recognises a polygamous marriage in that it extends the definition of “marriage” to that between a man and his principal wife. The IRO simply does not serve the purpose of promoting traditional heterosexual monogamous marriage. 76. Nor is it necessary to restrict the spousal employment and tax benefits to those in an opposite-sex marriage as recognised under Hong Kong law in order to draw a “bright line” in order to achieve administrative workability. As explained in Section D.3 above, the appellant in this case can demonstrate without any difficulty that he and Mr Adams are parties to a same-sex marriage having the characteristics of publicity as a formal marriage and exclusivity that distinguish it from a mere relationship. There is therefore no administrative difficulty posed by the appellant’s case and the “bright line” argument provides no rational justification for upholding the Benefits Decision or the Tax Decision. 77. For these reasons, we conclude that the respondents are unable to justify the differential treatment in the present case in respect of the Benefits Decision and the Tax Decision. E.5 The third and fourth steps of the justification test 78. Having concluded that the two challenged decisions cannot be justified because the restriction of the spousal employment and tax benefits to opposite-sex married couples is not rationally connected to the legitimate aim of protecting the institution of marriage under Hong Kong law, it is not necessary to go on to consider the third and fourth steps of the justification test (set out at [22] above). 79. It was similarly unnecessary to go into those questions in the case of QT (at [100]), although the Court discussed the standard of review briefly in view of the full arguments received. Had it been necessary to go on to consider the third and fourth steps of the justification test, we would have followed the same reasoning as set out in QT at [103] to [109] and, albeit that the standard of review is case specific (QT at [110(c)]), would have applied the reasonable necessity standard in assessing the justification exercise in this case. 80. Realistically, though, both Ms Monaghan QC and Lord Pannick QC recognised in argument that their respective cases in this appeal stood or fell depending on the answer to the question of whether the differential treatment was rationally connected to the legitimate aim in question. Thus, in view of the Court’s conclusion in Section E.4 above, the answers to the third and fourth steps in the justification test would likely have been that the differential treatment was not proportionate and was not a fair balance in that it resulted in an unacceptably harsh burden on the appellant. But, as we have said, it is not necessary to express a concluded view on those issues. F. Conclusion 81. We therefore answer the certified questions of law as follows: (1) As to Question 1: (a) No, see Section E.4 above. (b) The local legal landscape and societal circumstances are relevant to the issue of proportionality and/or justification but not the prevailing socio-moral values of society on marriage, see [57] and [61] above. (c) No. (2) As to Question 2: (a) No, see Section E.4 above. (b) The local legal landscape and societal circumstances are relevant to the issue of proportionality and/or justification but not the prevailing socio-moral values of society on marriage, see [57] and [61] above. (c) No. 82. For the reasons set out above, we would allow the appellant’s appeal. 83. Both parties requested that the Court invite further written submissions on the form of relief that should follow in the event that we were to allow the appeal. It would also be appropriate for the parties to address the issue of the costs of the appeal in those written submissions. Accordingly, we direct that:- (1) The appellant file written submissions as to the form of relief and the costs of the appeal within 14 days of the date of handing down of this judgment; (2) The respondents file written submissions in response thereto within 14 days thereafter; (3) The appellant file any written submissions in reply to those of the respondents within 7 days thereafter; (4) Save with the leave of the Court, the appellant’s and respondents’ respective written submissions should not, in aggregate, exceed 5,000 words in length. 84. Unless otherwise directed, after receipt of the parties’ written submissions pursuant to the above directions, the Court will proceed to determine the relief to be granted and the disposition of the costs of the appeal without any further oral hearing. Ms Karon Monaghan QC, Mr Nigel Kat SC and Mr Azan Marwah, instructed by Daly & Associates, assigned by the Director of Legal Aid, for the Appellant Lord Pannick QC, Mr Stewart Wong SC and Mr Johnny Ma, instructed by the Department of Justice, for the 1st and 2nd Respondents Hogan Lovells, for the Intervener (by written submissions only) [1] (2018) 21 HKCFAR 324 (“QT”). [2] (Cap.112) (“IRO”). [3] The Court of Appeal gave leave to the ICJ to intervene by way of written submissions below. [4] Specifically, CSRs 900 to 925 and 950 to 954. [5] IRO, section 10 provides: “(1) In the case of a husband and wife, unless an election is made under subsection (2), salaries tax shall be payable on the net chargeable income of each spouse ascertained under this Part by the spouse to whom the income has accrued. (2) Where in any year of assessment a husband and wife, not being a wife living apart from her husband, both have assessable income and – (a) either the husband or wife is entitled to deductions under Part 4A and allowances under Part 5 which, in aggregate, are in excess of his or her net assessable income; or (b) both also have a net chargeable income and the aggregate of the salaries tax which would be payable by them if subsection (1) applies exceeds the salaries tax which would be payable if an election is made under this subsection, an election may be made by them, subject to section 11, to be assessed to salaries tax in the manner specified in subsection (3). (3) Where an election is made by a husband and wife under subsection (2) salaries tax shall be payable on their aggregated net chargeable income as ascertained under section 12B(2) and in the case of an election – (a) under subsection (2)(a), the spouse who would have been chargeable to salaries tax in the absence of such an election; (b) under subsection (2)(b), the spouse who is nominated by them, shall be chargeable to salaries tax in respect of such aggregated net chargeable income. (4) Where a husband or wife is deceased an executor shall have the same right to make an election under subsection (2) as the deceased would have had if the deceased had not died. (5) For the purposes of subsection (3), where an election is made under subsection (2) by a husband and wife who married one another in the year of assessment to which the election relates, they shall be deemed, for the purposes of ascertaining their aggregated net chargeable income for that year, to have married at the commencement of that year.” [6] In fact, the appellant was not adversely affected by this basis of assessment since, even if he and Mr Adams had been assessed for salaries tax by way of joint assessment as a married couple, they would not have obtained any reduction of their total tax liability. [7] HCAL 258/2015, Judgment dated 28 April 2017 (“CFI Judgment”). [8] CFI Judgment at [46]-[60]. [9] CFI Judgment at [64]-[66] and [76]-[78]. [10] CFI Judgment at [87]-[90]. [11] CACV 126/2107, [2018] HKCA 318 (Cheung CJHC, Lam VP and Poon JA), Judgment dated 1 June 2018 (“CA Judgment”). [12] CA Judgment at [12]-[17] (per Cheung CJHC), [24] and [33] (per Lam VP), and [125]-[130] (per Poon JA). [13] Ibid. [14] Under s.22(1)(b) of the Hong Kong Court of Final Appeal Ordinance (Cap.484), in CACV 126/2017, [2018] HKCA 638, Judgment dated 24 September 2018. [15] (2013) 16 HKCFAR 112. [16] Ibid. at [63]. [17] Under Art. 37 of the Basic Law which provides: “The freedom of marriage of Hong Kong residents and their right to raise a family freely shall be protected by law.” [18] (2018) 21 HKCFAR 324 at [25]-[26]. [19] Case for the Appellant at [9] (footnote omitted). [20] Appellant’s Supplemental Case, 23 April 2019, at [4]. [21] Case for the Respondents at [38]. [22] Case for the Respondents at [26] and, similarly, [64] (italics in original). [23] Supplemental Case for the Respondents at [11] and [15]. [24] Ibid. at [2]-[10]. [25] CA Judgment at [62(1)], footnote 16, which reads: “In her written submissions, Ms Carss-Frisk took the point that because of lack of capacity as a matter of the law of his domicile, which is Hong Kong law, the applicant could not be regarded as lawfully married to Mr Adams even under New Zealand law. But in her oral submissions, she effectively abandoned this point because, as she readily accepted, it did not help inform the proper analysis of the issues before the Court.” [26] (2018) 21 HKCFAR 324 at [76]. [27] Ibid. at [44]-[45] (italics in original; footnotes omitted). [28] Ibid. at [46]-[52]. [29] (Cap.181). [30] Ibid., section 6(1). [31] Ibid., section 7. [32] Ibid., sections 19(1) and 21(1) respectively. [33] Ibid., sections 20(3) and 21(6) respectively. [34] Ibid., section 23. [35] Ibid., section 24. [36] Ibid., section 26. [37] Ibid., section 40(1). [38] Ibid., section 40(2). [39] (Cap.179), per section 20(1)(c). [40] Appearing with Mr Nigel Kat SC and Mr Azan Marwah. [41] Affidavit of Tse Wing Yee Winnie on behalf of the 1st respondent dated 20 June 2016 at [65]; Affidavit of Kung Chun Fai Frederick on behalf of the 2nd respondent dated 20 June 2016 at [51]. [42] Case for the Respondents at [27] and [38]. [43] CA Judgment at [12]. [44] Ibid. at [23]. [45] Ibid. at [89]. [46] Appearing with Mr Stewart Wong SC and Mr Johnny Ma. [47] (2018) 21 HKCFAR 324 at [89]. [48] CA Judgment at [103]-[116]. [49] (2018) 21 HKCFAR 324 at [62]-[66]. [50] CA Judgment at [125]. [51] Ibid. at [109(2)]. [52] (2013) 16 HKCFAR 112 at [116]. [53] Ibid. at [219]-[220]. [54] Case for the Appellant at [59] (italics in original). [55] Case for the Respondents at [19] and [27]. [56] [2004] 2 AC 557 per Lady Hale at [138]. [57] (Application No. 56501/00, 10 May 2001), ECHR 2001-VI. [58] (2004) 38 EHRR 24 at [40]. [59] [2009] 1 AC 173 at [108]. [60] (2010) 51 EHRR 16 at [98]. [61] (2011) 53 EHRR 25 at [72] (footnotes omitted, italics added). [62] [2009] UKPC 52, [2010] 3 LRC 653 at [26]. [63] [2004] 2 AC 557 at [143] (italics in original). [64] Affirmation of Leung Chun Kwong, 24 December 2015, at [9]. [65] Ibid. at [28]-[29], referring to http://www.cmab.gov.hk/en/issues/full_code_of_practice.htm (21 December 2015). 
Chief Justice Ma: 1. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Tang PJ: 3. The appellant, MGET was the principal operating subsidiary of Moulin Global Eyecare Holdings Ltd (Holdings), a listed company in Hong Kong since 1993, which had a majority of public shareholders. MGET’s case is that its profits had been fraudulently inflated over the tax years 1998/1999 to 2003/2004 by its then management.[1] MGET’s tax returns were prepared based on balance sheets which had included such inflated profits. Assessments were made based on such returns and the tax[2] paid as a result were excessive. Two directors of MGET and other members of its management have been convicted of a number of fraud charges involving false accounting in respect of Holdings accounts. I do not believe it is seriously challenged that some of the directors of MGET were innocent. But it does not matter, because for the present purpose, I shall proceed on the basis that there were innocent directors. 4. MGET was ordered to be wound up on 5 June 2006. This appeal arose out of the liquidators’ applications under s 64(1)(a) and s 70A of the Inland Revenue Ordinance Cap 112. 5. Under s 64(1), a taxpayer has one month after the date of the assessment to give notice of objection to the assessment with the possibility of an extension: “(a) if the Commissioner is satisfied that owing to absence from Hong Kong, sickness or other reasonable cause, the person objecting to the assessment was prevented from giving such notice within such period …” 6. The Liquidators applied under s 64(1)(a) for extensions of time to object to the assessments made during those tax years … on the ground that the fraud of its previous management (which was not discovered until after the provisional liquidators were appointed in June 2005) had prevented MGET from giving notice of objection within the statutory time limit of one month. 7. The application under s 70A, which has a relevant time limit of 6 years, is confined to the tax year 2003/2004. Under s 70A, a taxpayer may obtain repayment “if, upon application made within 6 years after the end of a year of assessment … it is established to the satisfaction of an assessor that the tax charged for that year of assessment is excessive by reason of an error or omission in any return or statement submitted in respect thereof…”. In respect of the tax year 2003/2004, MGET’s case is that the tax[3] paid was excessive by reason of errors in the return or statement submitted in respect thereof, because MGET’s profits had been deliberately and fraudulently inflated by the management of MGET. 8. The Commissioner rejected these applications. On the liquidators’ application for judicial review, on 15 February 2011, Reyes J decided in favour of the liquidators and ordered the Commissioner to reconsider her decisions. On 21 March 2012, the Court of Appeal allowed the Commissioner’s appeal and set aside Reyes J’s orders. Essentially, the Court of Appeal decided that the fraudulent knowledge of MGET’s management that the profits had been inflated should be attributed to MGET such that MGET had not been prevented from giving notice of objection within time nor was there any error within the meaning of s 70A. This court granted leave to appeal so that the law relating to attribution could be considered. 9. I have had the advantage of reading the judgment of Lord Walker of Gestingthorpe in draft. I am grateful to Lord Walker for his analysis of the concept of attribution and the fraud exception. Lord Walker has summarized (at para 106) his conclusion on the general effect of the leading authorities on attribution and the fraud exception. I shall use Lord Walker’s summary as my guide. Section 64(1)(a) 10. I am of the view that MGET’s reliance on s 64(1)(a) is misplaced. Like Lord Walker, I am inclined to accept Mr Brennan’s submission for the Commissioner, that s 64(1)(a) contemplates some temporary impediment of an external and physical nature, rather than something internal and psychological. Thus, if a taxpayer failed to object within the statutory time limit because of wrong advice from his accountants, the taxpayer could not be said to have been prevented from giving timely notice of objection. Nor when the “cause” is said to be the undiscovered fraud of its management. This is so, even when the fraud is not attributed to the corporate taxpayer. For the purpose of s 64(1)(a) I do not think it necessary to decide whether the fraud of MGET’s previous management should be attributed to it. Should it be necessary to do so, I would for the reasons I give below in respect of s 70A, hold that the guilty knowledge of MGET’s management should not be attributed to it. Section 70A 11. So far as the appellant’s application under s 70A is concerned I am also in respectful agreement with Lord Walker that “a deliberate lie cannot be an error for the purpose of s 70A” (para 128). However, with respect, I do not agree that the fraud exception is not applicable. 12. Lord Walker says, and I respectfully agree, questions of attribution are always sensitive to the factual situation in which they arise, and the language and legislative purposes of any relevant statutory provisions. The specific context here is s 70A. An application under s 70A does not fall within the well known categories of liability cases or redress cases. Nor is it truly comparable to cases where redress is sought from an auditor or an insurance company. In Stone and Rolls Ltd v Moore Stephens (HL(E)) [2009] 1AC 1391, in the context of a claim against an allegedly negligent auditor, Lord Walker said where there are innocent shareholders: “… the court would have to inquire quite closely into the facts in order to see (as Saville LJ put it in Group Josi) whether it would be contrary to justice and common sense to treat the company as complicit ...” 13. I believe justice and common sense is the reason for the fraud exception. The robust language of Lords Halsbury and Macnaghten in Gluckstein v Barnes [1900] AC 240 quoted by Lord Walker at para 81 of his judgment reflected their views on the justice and common sense in that case. 14. It may be helpful to consider what are or may be “errors” under s 70A. The scope of s 70A may assist in deciding what justice and common sense require in the present case. 15. In Extramoney Ltd v Commissioner of Inland Revenue [1997] HKLRD 387, Patrick Chan J (as he then was) would regard “something incorrectly done through ignorance or inadvertence; a mistake” as errors for the purpose of s 70A. I respectfully agree. 16. Radio Picture Ltd v Commissioners of Inland Revenue [1937] 22 TC 106, a case mentioned by Patrick Chan J, was concerned with s 24 of the Finance Act 1923, on which s 70A was based. There, the question was whether the taxpayer’s liability under an agreement between it, an English company, and its parent company, an American company, constituted a dollar debt or a sterling debt. It paid tax on the basis that the liability was a sterling debt although it had paid the parent company on the basis that it was a dollar debt. The difference was substantial and against the taxpayer after the United Kingdom went off the gold standard in 1931. How the taxpayer came to pay tax on the less advantageous basis was described by Sir Wilfrid Greene MR in the Court of Appeal at 128: “… But when its balance sheet was being drawn up, the auditor, for some reason which it is difficult to understand, unless it be some misapprehension on his part or some misconstruction of the document, did not in point of fact bring in these outstanding liabilities on a dollar basis, but brought them in on a sterling basis. The result of that was that the English Company was made to appear as having overpaid, and substantially overpaid, the American company, because, as I have said, the payments which it made were on a dollar basis. That mistake was put right when a new auditor took over the business of auditing the Company’s accounts, and this claim was the result; …” 17. Lawrence J at First Instance held that on the proper construction of the agreement that it was a dollar debt, that there was a mistake and allowed the appeal from the Commissioner. The Court of Appeal unanimously upheld his decision. 18. For my purpose, it is sufficient to quote from the judgment of Scott LJ, who said: “… the case made is the simplest possible case for relief under the Act of 1923, namely, that the Company had made a mistake in the method of making its returns and had paid tax during the years following 1931 (when the UK went off the gold standard) upon higher figures than represented its real income in those years.” 19. In Hong Kong, a similar point arose by way of a preliminary issue before a Board of Review. In Tax Case No D6/91, the taxpayer with full knowledge of all the facts had filed a tax return which had been accepted by the assessor and an assessment was issued. Subsequently a new tax adviser took over the affairs of the taxpayer and took a different view and then sought a correction under s 70A. The new tax adviser took the view that the taxpayer was not subject to tax in Hong Kong because it was not carrying on business in Hong Kong or if it was, its profits were not derived from Hong Kong. Although, the board doubted whether a change or difference of opinion could be an error, it said: “If … there is only one true and correct interpretation then it is not a matter of opinion. In the case now before us it would appear to us that whether or not the profits arose in or were derived from Hong Kong and whether or not the Taxpayer was carrying on business in Hong Kong is a matter of fact and legal interpretation which can have one answer only. The answer is clearly yes or no. The Taxpayer has been assessed to tax and paid tax on the basis that the answer is affirmative. If however, having heard the case and all of the evidence, we were to decide that the answer should be negative then clearly an error has been made in the tax return and the matter is capable of being rectified under section 70A of the Inland Revenue Ordinance.” 20. The Board also mentioned an earlier decision, namely, Exxon Chemical International Supply SA v CIR 3 HKTC 57. In Exxon the taxpayer had for some years been assessed to and paid profits tax on profits from the resale of group products. On 10 March 1987, “with the advice of a different tax adviser whose view was accepted by the Assessor in respect of its 1985/86 Profits Tax Return”[4], it applied under s 70A to correct Profits Tax assessments for the years of assessment 1980/81 to 1984/85, claiming the profits from sales to affiliate companies had been included as assessable income by error and were not properly subject to profits tax, because the profits were sourced offshore. The taxpayer failed before the Board of Review. The Board held that the profits were sourced within Hong Kong. That was upheld on appeal by Godfrey J who concluded that “the profits … arose in, or is derived from, Hong Kong” at 100. What is relevant for the present purpose is that, there, the Commissioner conceded that, if the Board decided that the profits should not be taxed because the profits were not sourced in Hong Kong, a mistake had been made for the purpose of s 70A. I should mention that counsel for the Commissioner took care to say that the concession should not be taken as a precedent. It is against such background that the Board said in Tax Case No D6/91 : “in (Exxon) the Commissioner submitted that if the Board were to find in favour of the taxpayer, then he considered that an error had been made and should be rectified under section 70A. In that case the Board placed a caveat upon the matter because arguments had not been made before the Board and the point was conceded by the Commissioner. We can see little difference between the case before us today and the Exxon case. Had we taken a different view of the appeal now before us we would find it difficult to rationalize our decision in the light of the fact that the Commissioner had considered in the Exxon case that such cases are appropriate for section 70A. Section 70A does not give the Commissioner any discretion to decide whether or not section 70A can or cannot be used.” 21. These decisions show that a genuine mistake made by an auditor could be an error for the purpose of s 70A. I see no difference in principle between a mistake by an auditor and a mistake by the management of a taxpayer. Nor between an individual and a corporation. The history of s 24 of the Finance Act 1923 was discussed in the article “Error Mistake Relief” by David Stopforth in 1989 [BTR] 151. The discussion showed that s 24 was enacted to enable a taxpayer who had made a bona fide mistake in a return so that he had paid income tax on more income than he had received to be repaid. 22. I turn to the legislative history of s 70A. Section 70A was added by amendment in 1956. Before s 70A was enacted, s 79 enabled tax paid in excess to be refunded. s 79 provided, where relevant: “(1) if it is proved to the satisfaction of the Commissioner by claim duly made in writing within 6 years of the end of a year of assessment or within 6 months after the date on which the relevant notice of assessment was served, whichever is the later, that any person has paid tax in excess of the amount with which he was properly chargeable for the year, such person shall be entitled to have refunded the amount so paid in excess:” It is clear from the language used that the purpose of s 79 was to ensure that only tax properly chargeable should be paid and any excess should be refunded.[5] 23. However, because when an assessment had become final and conclusive under s 70 (for example, if no notice of objection had been given against an assessment within one month under s 64(1) and no extension of time obtained under s64(1)(a)), s 79 would not be available, further legislation was considered necessary. That led to the enactment of s 70A. The 1954 Report explained : “95. Section 70 is so universal in its prohibitions that even an obvious error cannot be adjusted after the statutory period for appeal has passed if the adjustment will reduce the income or profits assessed. We do not consider that this somewhat harsh interpretation was intended by the Legislature as, in our view, section 79(1) admits to the possibility of errors and envisages the necessity for provisions to correct them. WE RECOMMEND that provision be made whereby, despite section 70, an assessor may amend an assessment if it is proved to his satisfaction, within the time limit set forth in section 79, that the assessment is incorrect by reason of an error or mistake in any return or statement submitted or in the calculation of the assessed income or profits or of the tax charged thereon.” 24. Section 70 cast a wide net. It provided: “Where no valid objection or appeal has been lodged within the time limited by this Part against an assessment as regards the amount of the assessable income or profits or net assessable value assessed thereby, or where an appeal against an assessment has been withdrawn under section 68(1A)(a) or dismissed under subsection (2B) of that section, or where the amount of the assessable income or profits or net assessable value has been agreed to under section 64(3), or where the amount of such assessable income or profits or net assessable value has been determined on objection or appeal, the assessment as made or agreed to or determined on objection or appeal, as the case may be, shall be final and conclusive for all purposes of this Ordinance as regards the amount of such assessable income or profits or net assessable value: Provided that nothing in this Part shall prevent an assessor from making an assessment or additional assessment for any year of assessment which does not involve re-opening any matter which has been determined on objection or appeal for the year.” 25. It is clear that s 70A was enacted so that, despite the many and various ways in which an assessment could become final and conclusive under s70, a taxpayer may obtain a refund if he could show that “the tax charged … is excessive by reason of an error”. 26. Given the legislative history of s 70A and its language, I believe, like s 79, it was enacted to relieve the hardship and injustice to a taxpayer who had paid more tax than was properly chargeable. Both s 70A and s 79 have the same time limit. Section 79 does not in terms require any error or omission, but mistakes or omissions would be the obvious reasons for the invocation of s 79. 27. Lord Walker says: “The important policy consideration is the balance between promptness and finality, on the one hand, and avoiding hardship and injustice, on the other hand.” (para 116) And: “Annual taxes should be levied so as to ensure prompt payment and so as to achieve finality within a reasonably short time.” (para 119) 28. I respectfully agree. Promptness and finality are provided for in the time limit. The requirement for a genuine error may contribute to finality. I would not read more into these policy considerations. They must not be allowed to undermine the clear object of s 70A, indeed, the object of the Inland Revenue Ordinance, namely, that a taxpayer should pay what is properly chargeable and no more. No doubt payment should be prompt but any genuine mistake could be sorted out within a generous time limit. Since the amount of tax which may have to be refunded under s 70A in any year must be insignificant compared to the annual tax revenue,[6] and the legislature has allowed 6 years, I do not believe the fact that taxes are annual an important consideration. Fraud Exception Applied 29. I turn to consider the facts of the present case. Although it has yet to be established, I proceed on the basis of that tax has been overpaid as a result of the fraud of MGET’s management.[7] In other words, more tax than properly chargeable had been paid. This results in hardship and is unjust. Precisely, the object and purpose of s 70A to relieve. Of course, whether there was an error depends on whether the fraudulent knowledge of MGET’s management should be attributed to MGET. In this context I note that if profits had been overstated due to the negligence or ineptitude of its management or its auditors, s 70A would apply. Looked at with common sense, it is difficult to see why a company whose management was negligent or inept should be better off than one where innocent shareholders had suffered at the hands of fraudulent management. Negligence and ineptitude are more common than fraud and would probably have a greater impact on revenue. If one compares the situation of the taxpayers in Radio Pictures and Tax Case No D6/91 with MGET, it is difficult to conclude that those taxpayers are more deserving. 30. Suppose, a fraudulent management had inflated profit to obtain a higher bonus and the only victims were its innocent shareholders, it is difficult to see why justice and common sense require attribution of the fraud so that s 70A should not apply. Here, MGET was a direct victim of the fraud. It was not the only victim. The main victims were presumably the creditors of MGET and Holdings. Success under s 70A should benefit these creditors. It has not been suggested the fact that MGET was not the only victim makes any difference. 31. The liability of the Commissioner is statutory and depends on the language of s 70A. I have said this is not a liability or redress claim. The issue is not whether MGET is liable to the Commissioner for the dishonesty of MGET’s management. The Commissioner is innocent and we are not concerned with any claim for redress. We are concerned with a statutory scheme under which tax paid in excess of what is properly chargeable may be refunded. The Commissioner is concerned with the proper administration of the scheme. He bears the primary responsibility to decide whether an application satisfies the requirement of the scheme. It is in this very unusual context that the court has to consider the application of the fraud exception. I believe the plain object and purpose of s 70A supply the answer. Has MGET paid more tax than was properly chargeable? If the liquidators could prove that the profits had indeed been inflated, the answer must be yes. Do justice and common sense tell one that they should not have a refund? No, given the purpose of s 70A, I believe justice and common sense require the application of the fraud exception. Subject to the 6 years time limit, the Commissioner has no good policy reason to wish to keep tax paid in excess of what was properly chargeable. 32. Accordingly, I would restore the order of Reyes J in respect the application under s 70 A, and remit the matter to the Commissioner for further consideration. Mr Justice Bokhary NPJ: 33. I should, in fairness to the argument of Mr Robin Dicker QC for the taxpayer and to Mr Justice Tang PJ’s dissent, indicate that I had at one stage leaned in favour of the view that the taxpayer’s application under s 70A ought to succeed. Ultimately, however, I find myself in agreement with Lord Walker of Gestingthorpe NPJ’s judgment by which the taxpayer appeal is dismissed in its entirety. Lord Walker of Gestingthorpe NPJ: Introduction 34. This appeal is the final stage in one phase of complex litigation, comprising criminal, public law and civil proceedings, arising out of the collapse of a group of companies known as the Moulin group. The group’s principal activity was the manufacture, distribution and retail sale of spectacles. Its ultimate holding company was Moulin Global Eyecare Holdings Limited (“Holdings”), a company which was incorporated in Bermuda but registered in Hong Kong. From 1993 the shares of Holdings were listed on the Hong Kong Stock Exchange. A large minority holding (decreasing from about 40 per cent to about 30 per cent during the relevant period) was held by Ma Bo Kee (the founder of the business and chairman of Holdings) and members of his family. The principal trading subsidiary was Moulin Global Eyecare Trading Limited (“MGET”), which was held through intermediate subsidiaries. It was incorporated in Hong Kong in 1980 under the name of Moulin Optical Manufactory Limited. Provisional liquidators were appointed for both companies on 23 June 2005, and winding-up orders were made in respect of both companies on 5 June 2006. The provisional liquidators, Mr Roderick Sutton and Mr Desmond Chung, were appointed as the liquidators. 35. The criminal phase of the matter was the prosecution of Ma Bo Kee, his son Cary Ma (the group chief executive officer) and Ma Bo Kee’s sister-in-law Michelle Lam (the group treasurer) on various charges involving fraud. They were charged in February 2009. Cary Ma and Michelle Lam pleaded guilty on arraignment in September 2010 and Ma Bo Kee was convicted at the conclusion of the trial. Ma Bo Kee was sentenced to 12 years’ imprisonment, Cary Ma to 10 years and Michelle Lam to 9½ years (reduced on appeal to 8½ years). In sentencing them the trial judge, Line J, described their conduct as “commercial crime of the worst kind”. The group's creditor banks were owed about $2.7 bn and investors stood to lose about $1.75 bn. 36. The civil proceedings included a claim for damages for breach of duty brought by MGET's liquidators against KPMG, who were its auditors for the first part of the relevant period, that is from the 1998-99 year of assessment until 2000-01. The liquidators of other group companies made similar claims (see Moulin Global Eyecare Holdings Limited v KPMG HCA 118/2007, in which Barma J gave an interlocutory judgment on 10 February 2010). The Court was told that these claims have now been compromised, and that proposed claims by the liquidators against Ernst & Young (“EY”), the firm that took over the audit in 2001-02, have also been compromised. 37. These criminal and civil proceedings are part of the background to the issues in this appeal. Closer to the issues, but nevertheless distinct as a set of proceedings, has been the dispute between MGET’s liquidators and the Commissioner of Inland Revenue (“the Commissioner”) as to a proof of debt lodged by the Commissioner in respect of unpaid profits tax together with surcharges for late payment. In proceedings in the insolvent winding-up of MGET Kwan JA held on 17 December 2009 that the debt should be admitted to proof: Re Moulin Global Eyecare Trading Limited (Proof of Debt: Tax Assessments) [2010] 4 HKLRD 283. This decision was largely procedural, being based on the general principle that the Inland Revenue Ordinance, Cap 112 (“the IRO”) provides an exclusive code for the determination of tax liabilities. The decision leaves the substantive question of the liabilities covered by the proof to be determined. That question is not formally an issue in this appeal, but the issues of law raised in the appeal may in practical terms be determinative of it. 38. What is central to this appeal is the claim by MGET's liquidators, made by way of judicial review, that the Commissioner has erred in law in refusing to reopen assessments to profits tax, and surcharges for unpaid profits tax, for a succession of years of assessment starting with 1998-99. The liquidators’ case, in very brief summary, is that MGET did not make any taxable profits in any of those years, and that its reported profits were false and non-existent as the result of various frauds perpetrated by Ma Bo Kee, Cary Ma and Michelle Lam. These three individuals were referred to in the first-instance judgment of Reyes J as “the Ma family directors” but it should be mentioned at once that they were not all, at any time, directors of MGET. In her judgment in the Court of Appeal (para 58) Kwan JA recorded that for the year ended 31 March 1999 the directors of MGET were Ma Bo Kee, Ma Bo Fung and Ma Bo Lung, and that the balance sheet for that year was approved and signed by all three directors. Cary Ma was appointed as an additional director in October 2000 and the accounts for subsequent years were likewise approved and signed by all the directors. There is little in the documentary evidence about Ma Bo Fung and Ma Bo Lung but it appears (as Mr Dicker QC told the Court without objection from Mr Brennan QC) that they are brothers of Ma Bo Kee; they managed factories in mainland China and rarely visited Hong Kong; and neither of them was charged with any criminal offence. 39. The sums at stake are large. MGET paid a total of almost $89m in profits tax in respect of the 1998-99 to 2003-04 years of assessment, all of which was, on the liquidators’ case, wrongly paid. The Commissioner resists any repayment and, in the proof of debt already mentioned claims a further sum of over $10m. The details are set out in the next section of this judgment. 40. On 15 February 2011, Reyes J decided in favour of the liquidators and directed the Commissioner to reconsider her decisions ([2011] 3 HKLRD 216). On 21 March 2012 the Court of Appeal allowed the Commissioner’s appeal ([2012] 2 HKLRD 911). The Court gave leave for a further appeal limited to “two broad questions”: “(i) Whether the primary rules of attribution should apply to attribute knowledge of the fraudulent directors to MGET? (ii) If not, whether the special rules of attribution should apply to attribute knowledge of the fraudulent directors to MGET?” The terminology of primary and special rules of attribution is of course derived from the seminal judgment of Lord Hoffmann, giving the advice of the Judicial Committee of the Privy Council, in Meridian Global Funds Management Asia Limited v Securities Commission [1995] 2 AC 500 (“Meridian”), discussed in paras 76 to 79 below. These questions also raise an issue of law as to what is sometimes called the fraud (or fraudulent agent) exception, the discussion of which begins at para 80 below. 41. The appellants, the liquidators of MGET, have applied for leave to rely on wider grounds of appeal. The court did not find it necessary to make a formal ruling on that application. On any view it is necessary to set out the known facts in some detail, and to pay close attention, as Kwan JA did, to the general structure and particular language of the IRO in relation to the assessment of profits tax. One of the fundamental points to be taken from Meridian is the importance of context (including, as in this case, statutory context) in any problem of attribution. Although I differ from Kwan JA’s judgment on some important points, it is, if I may respectfully say so, one of the strengths of her judgment that she recognised this in the full citations from Meridian in paras 21 and 22 of the judgment, and in para 30, introducing the detailed discussion of the relevant provisions of the IRO. The facts 42. There is unchallenged evidence as to the manner in which returns, assessments and surcharges in respect of MGET’s liability to profits tax were made for the 1998-99 year of assessment and later years of assessment. Returns of profits were prepared by reference to MGET’s audited accounts for the basis period (which coincided with the year of assessment). These accounts had been approved and signed by all MGET’s directors, as mentioned in para 38 above. The return, including a declaration that its contents were true, correct and complete to the best of the signatory’s knowledge and belief, was signed by the company secretary or a manager employed by MGET, as follows: 43. The amounts of profits returned, the profits tax assessed and any surcharges paid were as follows: 44. On 29 August 2005, about two months after the appointment of the provisional liquidators, the Commissioner served notice of (i) additional assessments and demands for profits tax for the 1999-00, 2001-02 and 2002-03 years of assessment, (ii) revised assessment and refund of tax for the 2000-01 year of assessment, and (iii) an estimated assessment for the 2004-05 year of assessment. The overall effect of these notices and one further notice was to claim a further net sum of $10,363,532 in unpaid profits tax (that sum being the amount of the Commissioner's disputed proof of debt). 45. None of the assessments summarised in paras 43 and 44 above was called into question by a claim under section 64 or section 70A of the IRO until after the commencement of the winding-up on 5 June 2006. By a letter dated 29 March 2007 to the Commissioner the liquidators stated that that letter, and an earlier letter dated 30 November 2005, should “for the avoidance of doubt” be treated as applications under those two sections. 46. The provisional liquidators’ letter of 30 November 2005 to the Commissioner was the beginning of a protracted correspondence which culminated in the liquidators commencing judicial review proceedings on 10 March 2010. As already mentioned, Reyes J granted such relief in a judgment dated 15 February 2011, quashing the Commissioner's decisions and remitting the matter to her for reconsideration. 47. The protracted and, at times, unfocused nature of the correspondence made it uncertain which document or documents contained the relevant decisions, and the judge gave leave for an amendment to cover this point. His judgment contains a full summary of the correspondence (paras 18 to 40), with a detailed discussion of its legal effect at paras 50 to 80. There is a much shorter summary in the judgment of Kwan JA in the Court of Appeal at paras 9 to 16. There was no appeal on this point and so she did not need to go into it in detail. The correspondence can therefore be taken shortly. Two letters from the Commissioner dated 20 December 2005 and 16 April 2007 amounted, as Reyes J held, to no more than the Commissioner’s statement and repetition of her preliminary view, expressed in the absence of any application under section 64 or section 70A of the IRO. Her reviewable decisions were to be found in her letters of 4 December 2009 and 4 February 2010, taken together. The first of these letters stated, among other things, that section 70A did not apply to the assessment for the 2003-04 year of assessment as there was no error or omission in the return (this assessment was the only relevant assessment not caught by the six-year time limit in section 70A). The operative part of the letter of 4 February 2010 was as follows: “We are instructed that so far the fraud alleged has not yet been proved and that, even if it is proved, it was, as you alleged, perpetrated by the former directors of MGET and as such the tax returns allegedly tainted by fraud would be regarded as being filed with knowledge of MGET of the alleged fraud and MGET cannot now argue that it was prevented from lodging an objection within time under s 64(1)(a) of Cap 112. In the above circumstances, our client is not satisfied that there is a reasonable cause which prevents MGET from lodging an objection within time and therefore cannot accept your request for an extension of time for lodging an objection.” These reasons reflect the central issue on which leave to appeal to this Court has been granted. 48. Throughout the correspondence the liquidators’ stance was that as they tried to investigate the financial affairs of MGET and Holdings they found increasingly strong evidence that their accounts had been prepared fraudulently and did not give a true and fair picture of their profits and net assets. The liquidators stated that because of falsification and destruction of records, it was not possible to reconstruct the financial statements so as to show the true position with any accuracy; but that a “static” accountancy model used by the liquidators indicated that MGET probably made no profits at all during the period from the 1998-99 year of assessment. Three principal forms of falsification were identified: (i) the crediting of proceeds of fictitious sales to customers in North America; (ii) the crediting of interest on fictitious loans shown as made by MGET; and (iii) the anticipation of fictitious profits from a scheme referred to as “Frame Board Space Agreements”. The principal purpose of the frauds seems to have been to persuade the Moulin group’s numerous bank creditors to increase their lending. The liquidators had great difficulty in obtaining documents and information from EY, who were MGET’s auditors until their resignation with effect from 30 December 2004. EY were briefly replaced by Deloittes, whose resignation with effect from 18 April 2005 heralded the Moulin group’s collapse. 49. The Commissioner’s attitude throughout the correspondence was very sceptical, even after Ma Bo Kee and his co-conspirators had been charged with serious offences. As appears above, even at the end of 2009 her attitude was that fraud had not been established. She also took the view that it was not enough for the liquidators to rely on a simulated model to establish the absence of profits; only detailed computations based on actual transactions would suffice. The IRO 50. If and so far as there are any special rules of attribution applicable to this case, they must be derived from the legislative scheme of the IRO, and consistent with its detailed procedural provisions, purposefully construed. Kwan JA discussed the IRO at some length, at paras 31 to 55 of her judgment, and the account that follows owes much to her analysis. As to the general approach to statutory construction she referred to the well-known passage in the judgment of Li CJ in HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568, para 12: “The modern approach is to adopt a purposive interpretation. The statutory language is construed, having regard to its context and purpose. Words are given their natural and ordinary meaning unless the context or purpose points to a different meaning. Context and purpose are considered when interpreting the words used and not only when an ambiguity may be thought to arise.” 51. Part 4 of the IRO (sections 14 to 26G) contains the substantive provisions charging profits tax. It is an annual tax imposed on the profits derived from a trade or business carried on by the taxpayer, computed in accordance with some special statutory rules (for instance, as to allowance for depreciation) but otherwise in accordance with generally accepted accountancy principles. Profits for a year of assessment (commencing on 1 April) are generally computed on the basis of the preceding year’s trading. 52. The statutory scheme for returns of taxable income, assessment of tax and objections and appeals are set out in Parts 9, 10 and 11 respectively. Section 51(1) in Part 9 provides that the taxpayer may be required by notice to make a return of taxable income. Section 51(5) provides as follows: “A return, statement or form purporting to be furnished under this Ordinance by or on behalf of any person shall for all purposes be deemed to have been furnished by that person or by his authority, as the case may be, unless the contrary is proved, and any person signing any such return, statement, or form shall be deemed to be cognizant of all matters therein.” 53. Under sections 59 and 62 in Part 10, the Commissioner has the duty of assessing and giving notice of assessment of the tax payable on taxable income. This is carried out primarily on the basis of taxpayers’ returns. If a taxpayer makes a return and it is satisfactory to the officials of the lnland Revenue, an assessment based on the figures in the return is likely to follow and to be accepted by the taxpayer as a matter of course. If however the taxpayer fails to make a return, or a return is regarded as unsatisfactory, or if the accounts of the taxpayer’s trade or business have not been kept properly, the Inland Revenue will make an estimated assessment under section 59 (2), (3) or (4). There is also the possibility that, without any bad faith on the part of the taxpayer, a dispute will arise as to one or more items in the taxpayer’s profit and loss account (for instance, whether some item of expenditure claimed as a deduction on income account ought to be regarded as capital expenditure). In any of these situations the taxpayer is likely to wish to contest the assessment, and may do so by giving notice under section 64 (in Part 11 of the IRO). 54. Section 64 (1) and proviso (a) are as follows: “(1) Any person aggrieved by an assessment made under this Ordinance may, by notice in writing to the Commissioner, object to the assessment; but no such notice shall be valid unless it states precisely the grounds of objection to the assessment and is received by the Commissioner within 1 month after the date of the notice of assessment: Provided that- (a) if the Commissioner is satisfied that owing to absence from Hong Kong, sickness or other reasonable cause, the person objecting to the assessment was prevented from giving such notice within such period, the Commissioner shall extend the period as may be reasonable in the circumstances;” There are two further provisos that are not material. Subsection (2) requires the Commissioner to consider any valid objection and within a reasonable time to confirm, reduce, increase or annul the assessment objected to, making any necessary adjustment under subsection (3). Subsection (4) provides for an appeal, if the matter is not resolved, to the Board of Review under section 66, with a further possible appeal, on a question of law, to the Court of First Instance under section 69 or, exceptionally, direct to the Court of Appeal under section 69A. 55. Proviso (a) to section 64(1) was considered by the Court of Appeal in Chow Kwong Fai v Commissioner of Inland Revenue [2005] 4 HKLRD 687. Woo V-P stated at para 20 that “prevented” should be best understood (by reference to the Chinese language version of the IRO) as “unable to”: “On the other hand, ‘unable to’ imposes a higher threshold than a mere excuse and would appear to give proper effect to the rigour of time limit imposed by a taxation statute. The rationale for the stringent time limit for raising tax objections and appeals was described in Case U175 87 ATC 1007. Tang J had in the judgment under appeal cited quite extensively from that case. I will thus refer only to one short passage: It seems that the need for taxation revenue to flow in predictable amounts according to projections as to cash flow have considered to be such that dispute as to the claims made by the community upon individuals for payment of tax have been treated as quite unlike any other classes of dispute within the community.” 56. Section 70, in Part 11, provides that where no valid objection to an assessment, and no appeal against an assessment, has been made within the time limited by Part 11, the assessment is to be final and conclusive for all the purposes of the IRO as regards the amount of the taxpayer's assessable profits. This general rule, which is aimed at achieving finality as between the Commissioner and the taxpayer, is subject to a limited exception in section 70A. The section, so far as now material, provides as follows: “(1) Notwithstanding the provisions of section 70, if, upon application made within 6 years after the end of a year of assessment or within 6 months after the date on which the relative notice of assessment was served, whichever is the later, it is established to the satisfaction of an assessor that the tax charged for that year of assessment is excessive by reason of an error or omission in any return or statement submitted in respect thereof, ... the assessor shall correct such assessment: ... [There is a proviso excluding correction of a return or statement made “on the basis of or in accordance with the practice generally prevailing at the time”] (2) Where an assessor refuses to correct an assessment in accordance with an application under this section he shall give notice thereof in writing to the person who made such application and such person shall thereupon have the same rights of objection and appeal under this Part as if such notice of refusal were a notice of assessment.” 57. Because section 70A has its own time limit (normally six years from the end of the relevant year of assessment) it can be relevant in this appeal only as regards the 2003-04 year of assessment. In relation to the 2003-04 year of assessment the liquidators rely on it as a fallback argument, if they do not succeed on their primary argument under proviso (a) to section 64(1). 58. Section 70A, which was introduced into the IRO in 1964, is similar to a provision based on “error or mistake” in section 33 of the United Kingdom Taxes Management Act 1970, which was originally enacted in 1923 and has recently been replaced by more elaborate provisions. There is surprisingly little United Kingdom authority on section 33. In Carrimore Six Wheelers Ltd v CIR (1944) 26 Tax Cases 301 an issue was raised as to the meaning of “error or mistake”, but the appeal was dismissed on a preliminary point as to jurisdiction. The judgment of Lord Greene MR is in terms suggesting that he may have agreed with the Special Commissioners that a return containing a deliberately wrong entry, made with full knowledge that it was wrong, is not an error or mistake within the terms of the section but he refrained from expressing any definite view. 59. Section 70A was considered by Patrick Chan J, as he then was, in Extramoney Ltd v Commissioner of Inland Revenue [1997] HKLRD 387 (“Extramoney”). The facts had some resemblance to those of the present appeal. Extramoney was a trading company within the group headed by Carrian Holdings Limited. Carrian went into liquidation in 1983. In January 1982 Extramoney made a profits tax return reporting profits of over $132m for the 1980-81 year of assessment on which profits tax of over $21m was assessed. This tax was paid by Carrian, with a surcharge, in April 1982. In 1987 Extramoney and the liquidators of Carrian applied for relief under section 70A of the IRO. The application was refused, and appeals to be Board of Review and the High Court was dismissed. 60. In the High Court Chan J made some general observations about the meaning of section 70A (at pp 395-396): “I think it would be unwise to attempt to give a comprehensive definition of what is or is not an error or omission which can cater for all situations. It would be easier to identify cases in which it is not. In my view, for the purpose of s.70A, the meaning of ‘error’ given in the Oxford English Dictionary (p. 277) would be appropriate, that is, ‘something incorrectly done through ignorance or inadvertence; a mistake’. I do not think that a deliberate act in the sense of a conscientious choice of one out of two or more courses which subsequently turns out to be less than advantageous or which does not give the desired effect as previously hoped for can be regarded as an error within s.70A. It is even worse if the deliberate act is motivated by fraud or dishonesty. But the question of fraud or dishonesty need not arise.” The principal difference between that case and the present appeal is that in Extramoney there was evidence that the disputed profits had been made, although probably not by Extramoney. But Chan J’s conclusion (at p 396) was expressed in general terms: “Where a taxpayer has deliberately and consciously made a decision to attribute a certain item, be it an item of profit or expenditure, in the tax return to be submitted to the assessor for assessment, if he subsequently changes his mind, that certainly cannot be an error within the meaning of s.70A of the Ordinance. In my view, the Board of Review had adopted the correct principles.” Attribution: the older authorities 61. Attribution means, in this context, the process of legal reasoning by which the conduct or state of mind of one or more natural persons (that is, human beings) is treated as that of a non-natural person (that is, a company) for the purpose of determining the company’s legal liability or rights in civil proceedings (in particular, its liability or rights in contract, in tort or for unjust enrichment) or its criminal liability. In approaching the topic of attribution it is necessary to recognize that there is a problem of terminology in the word “agent”. A natural person may choose whether or not to appoint an agent to act in the management of his property or the conduct of his business affairs. In principle it is his choice, although there are of course some activities calling for special professional skills in which the appointment of an agent is strongly advisable, or even mandatory. A company, on the other hand, never has a choice. As English company law developed during the nineteenth century, eminent judges repeatedly emphasised that a company, as a persona ficta, a legal construct with no real personality, can act only by its agents. 62. Professor Peter Watts QC of Auckland University, a leading scholar in this field, has referred in a recent note (Audit Contracts and Turpitude, (2010) 126 LQR 14, 17) to the “old wisdom” of these judges: “One might be tempted to think that the 19th century judges who routinely voiced the view that companies act solely by agents had an incomplete conception of company law, only made good by the 20th century developments in Lennard's Carrying Company Limited vAsiatic Petroleum Company Limited [1915] AC 705 [“Lennard”] and Tesco Supermarkets Limited v Nattrass [1972] AC 153 [“Tesco”]. It is not, however, the company law of Lord Cranworth LC, Lord Cairns and Lord Lindley that is deficient.” The first citation referred to is of Lord Cranworth LC in Aberdeen Railway Company v Blaikie Brothers (1854) 1 Macq 461, 471: “The directors are a body to whom is delegated the duty of managing the general affairs of the company. A corporate body can act only by agents.” The second citation is of Cairns LJ in Ferguson v Wilson (1866) LR 2 Ch App 77, 89: “What is the position of directors of a public company? They are merely agents of a company. The company itself cannot act in its own person, for it has no person; it can only act through directors, and the case is, as regards those directors, merely the ordinary case of principal and agent.” The third citation is of Lord Lindley in Citizen's Life Assurance Company Limited v Brown [1904] AC 423, 426: “If it is once granted that corporations are for civil purposes to be regarded as persons, ie as principals acting by agents and servants, it is difficult to see why the ordinary doctrines of agency and of master and servant are not to be applied to corporations as well as to ordinary individuals.” 63. It is certainly right to respect the wisdom of these great judges. But as issues of attribution grew more complex, especially with the proliferation of regulatory legislation (sometimes backed by criminal sanctions), it became apparent that it was sometimes necessary to distinguish between cases in which a company should be treated as directly liable for the conduct of its agents, and those in which it was vicariously liable. Whereas Cairns LJ spoke in 1866 of the directors of a company being "merely agents" of the company, Lord Reid in Tesco [1972] AC 153, 170 put it rather differently: “It must be a question of law whether, once the facts have been ascertained, a person in doing particular things is to be regarded as the company or merely as the company’s servant or agent.” Here the emphasis of “merely” is markedly different. 64. An employee who holds a position of responsibility (for instance, the manager of a large supermarket which is part of a nationwide chain) may have power to bind his corporate employer, but his authority to do so is derived (expressly, impliedly, or ostensibly) from the corporate employer’s board of directors. The directors alone are in the special position of deriving their authority, not from other natural persons, but from the company’s written constitution, that is its memorandum and articles of association. This point was made by Hoffmann LJ in El Ajou v Dollar Land Holdings Ltd [1994] 2 All ER 685, 705, (“El Ajou”), a case which can be seen as a precursor to Meridian. In discussing Viscount Haldane’s speech in Lennard [1915] AC 705, he observed: “It is well known that Viscount Haldane LC derived the concept of the ‘directing mind’ from German Law (see Gower, Principles of Modern Company Law, (5th ed, (1992) p194 n36) which distinguished between the agents and organs of the company. A German company with limited liability (GmbH) is required by law to appoint one or more directors (Geschӓftsführer). They are the company’s organs and for legal purposes represent the company. The knowledge of any one director, however obtained, is the knowledge of the company.” Hoffmann LJ commented that English law does not go as far as that, but it does regard a company’s board of directors, as a body, as the company’s vital organ. That is the basis of the primary rule of attribution which he expounded, soon afterwards, in Meridian. The corresponding passage in the current edition of Gower, 9th ed (2012) para 7-44 footnote 138, states, “Lord Haldane’s dictum was probably influenced by the clear distinction drawn between agents and organs in German company law, Haldane having studied in his youth in Germany”. 65. Lennard [1915] AC 705 is an early example of the problems that began to arise in applying regulatory statutes to trading companies. Under section 502 of the Merchant Shipping Act 1894 a shipowner was not liable for the accidental loss of cargo if it occurred “without his actual fault or privity”. A steamship owned by the Lennard company had gone aground off the Dutch coast. The grounding damaged the ship. Some of its cargo of benzine escaped, and there was an explosion resulting in the loss of the ship and cargo. The grounding occurred because the engines had broken down. The breakdown was the result of negligent management in failing to repair heavily-salted boilers. The cargo-owners’ claim succeeded at first instance and in the Court of Appeal. Mr John Lennard, the company’s general manager, who was in overall charge of the shipping operations, did not give evidence, an omission of which the House of Lords was very critical. 66. That was the factual context in which Viscount Haldane LC spoke of the need to identify a company’s “directing mind and will”. But it would be wrong to suppose that the same person or persons (not being the full board of directors) must perform this function for each and every one of a company’s activities. Lord Hoffmann made this clear in Meridian [1995] 2 AC 500, 509: “Because [the Lennard company] does not seem to have done anything except own ships, there was no need to distinguish between the person who fulfilled the function of running the company’s business in general and the person whose functions corresponded, in relation to the cause of the casualty, to those of an individual owner of a ship. They were one and the same person. It was this coincidence which left Viscount Haldane LC’s speech open to the interpretation that he was expounding a general metaphysic of companies.” This point was also made clear in El Ajou (paras 73 to 75 below). 67. The rather belated recognition of this important qualification of the “directing mind and will” concept considerably reduces its apparent force. Except in the case of very small companies with very simple activities, there will not be a single individual in a company who satisfies the test for all purposes. After Meridian some legal scholars conjectured that the concept might disappear from company law, and it might be better if it had disappeared, as it tends to obscure the underlying importance of the basic principles of agency. To refer instead to “the relevant responsible director or employee”, or some such expression, would be less arresting but a good deal more accurate, especially in view of cases such as Tesco No 2 mentioned in para 72 below, where the responsible employee was a till attendant. The importance of the basic principles of agency seems to be the main thrust of Professor Watts’s invocation of the wisdom of the 19th century judges, and the same concerns have been expressed in his article Corrupt Company Controllers, their Companies and their Companies’ Creditors: Dealing with Plens of Ex Turpi Causa [2014] JBL 161, at p 166, comments in his article “Corrupt Company Controllers, their companies and their companies’ creditors: Dealing with pleas of Ex Turpi Causa” [2014] JBL 161, at pp 166, on the recent case of Bilta (UK) Ltd v Nazir [2014] 1 All ER 168 (para 103 below): “The beauty of this analysis [that even in cases of direct liability, the company is only deemed a wrongdoer because of the actions of its agents] is that it does not in fact involve treating companies any differently from humans. Exactly the same analysis applies to the individual owner of a business who has nothing to do with running it, but lives a life of total ignorance about what his or her employees do. If those employees, including managers, engage in illegal conduct, we impute that conduct to the absentee owner even though he or she was personally innocent of the wrongdoing. An absentee owner should not be able to use that absence to disown the conduct of his or her managers. Companies, therefore, are like human owners of businesses who, to use the metaphor of Staughton L.J. in PCW Syndicates v PCW Reinsurers, spend their days on the grouse moors of Yorkshire, or who while away their days in the south of France, to use the analogy of Lord Hoffmann in Standard Chartered Bank v Pakistan Shipping Corp. It is no more necessary to conceive of companies as having personal guilt than it is in relation to the absentee human owner.” 68. In Tesco [1972] AC 153 the House of Lords considered a more complicated issue of attribution that arose under the Trade Descriptions Act 1968. Section 11(2) of that Act made it an offence for a retailer to offer a misleading “money-off” bargain. A branch of Tesco in Northwich, Cheshire (one of several hundred branches then existing in the United Kingdom) had run out of stocks of “money-off” packs of washing powder but continued to display a “money-off” advertisement while selling packs at the full price. Section 20 made clear that both a corporate retailer and a manager employed by it might be liable. But section 24(1) provided a defence to any person who proved that he “took all reasonable precautions and exercised due diligence” and that the offence was due to “the act or default of another person”. Both Tesco and the store manager were prosecuted and convicted. On appeal Tesco contended that it had exercised due diligence and that its branch manager (on whom the top management seemed willing to put the blame) was “another person”. 69. The issue for the House of Lords was not therefore a straightforward issue of attribution. The issue was the correct construction of the words “another person” in this particular context. The House of Lords allowed the appeal, holding that “another person” was capable of including an employee, and even a branch manager, so long as that person’s acts were not to be treated as acts of the company. That test, if it stood alone, would merely beg the question: what acts of an employee are to be treated as acts of the company? Lord Reid answered that question at pp 174-175: “I have said that a board of directors can delegate part of their functions of management so as to make their delegate an embodiment of the company within the sphere of the delegation. But here the board never delegated any part of their functions. They set up a chain of command through regional and district supervisors, but they remained in control. The shop managers had to obey their general directions and also take orders from their superiors. The acts or omissions of shop managers were not acts of the company itself. The explanations given by the other Law Lords were similar. In the passage from Lord Reid’s speech the phrase “within the sphere of delegation” is important, and its importance comes out even more clearly in El Ajou and Meridian. 70. Lord Reid (at p 171), Viscount Dilhorne (at p 187) and Lord Diplock (at p 200) referred in rather guarded terms to a passage in the judgment of Denning LJ in Bolton (Engineering) Co Ltd v Graham & Sons Ltd [1957] 1 QB 159, 172: “A company may in many ways be likened to a human body. It has a brain and nerve centre which control what it does. It also has hands which hold the tools and act in accordance with directions from the centre. Some of the people in the company are mere servants and agents who are nothing more than hands to do the work and cannot be said to represent the mind or will. Others are directors or managers who represent the directing mind and will of the company, and control what it does. The state of mind of these managers is the state of mind of the company and is treated by the law as such.” Lord Diplock said that this “vivid metaphor” is “not authority for extending the class of persons whose acts are to be regarded in law as the personal acts of the company itself, beyond those who by, or by action taken under, its articles of association are entitled to exercise the powers of the company”. 71. Lord Hoffmann gave a further warning in Meridian [1995] 2 AC 500, 509G against being beguiled by Denning LJ’s imagery in Bolton Engineering: “But this anthropomorphism, by the very power of the image, distracts attention from the purpose for which Viscount Haldane LC said [in Lennard], at p713, he was using the notion of directing mind and will, namely to apply the attribution derived from section 502 to the particular defendant in the case.” This is an area of law in which some element of metaphor is almost unavoidable, since the law is concerned with attributing human characteristics (such as knowledge or intention) to an abstraction which is devoid of any such characteristics. Even “organs” (the term which Lord Hoffmann favoured in EI Ajou) has connotations of body parts (though the original meaning of “organon” in ancient Greek was simply “tool” or “instrument”). But although some element of metaphor is unavoidable it must not be forgotten (see Meridian at p 511C) that “the question is one of construction rather than metaphysics”. 72. The decision of the House of Lords in Tesco has received some criticism both from legal scholars and in later authorities (notably Odyssey Re (London) Ltd v OIC Run-off Ltd mentioned in para 95 below). It must be seen as a decision on the construction of a particular penal statute. It may be contrasted with Tesco Stores Ltd v Brent LBC [1993] 1 WLR 1037 (“Tesco No 2”), in which another Tesco company was convicted of selling to a boy aged 14 a video with an “18” classification, an offence under section 11 of the Video Recordings Act 1984. The sale was made by a till attendant with no managerial status. She was the individual whose mental state was relevant for the purposes of the limited defence set out in section 11(2). Staughton LJ said at pp 1042-1043: “It is the employee that sells the film at the checkout point who will have knowledge or reasonable grounds for belief. It is her knowledge or reasonable grounds that are relevant. Were it otherwise, the statute would be wholly ineffective in the case of a large company, unless by the merest chance a youthful purchaser were known to the board of directors. Yet Parliament contemplated that a company might commit the offence: see section 16. By contrast, the single-handed shopkeeper would be less readily able to rely on the defence section, although he would fare better if he had an assistant serving at the counter while he was in the back of the shop. I cannot believe that Parliament intended the large company to be acquitted but the single-handed shopkeeper convicted.” Attribution: El Ajou and Meridian 73. In El Ajou [1994] 2 All ER 685 the claimant, a wealthy Saudi businessman, was attempting to recover funds which had been misappropriated in a massive share fraud carried out by three Canadian fraudsters. After various money-laundering operations they invested part of their ill-gotten gains in a large-scale property development in London. As an investment vehicle they used a Panamanian company, Yulara Realty Ltd (“Yulara”) to enter into a joint venture with Dollar Land Holdings plc (“DLH”), an English public company. There were several hotly-disputed issues in the case but the only relevant issue, for present purposes, was whether DLH should be treated as having knowledge, in order to establish equitable liability for knowing receipt of trust property, that the Yulara investment represented the proceeds of fraud. The claimant sought to establish this by attributing to DLH the guilty knowledge of Mr Ferdman, a Swiss fiduciary agent who acted for associates of the Canadian fraudsters, and was closely involved in the Yulara transaction. 74. Mr Ferdman was the chairman and a non-executive director of DLH, but in those capacities he generally acted as a nominee for the investors behind DLH. For the Court of Appeal (differing from the trial judge, Millett J) the decisive fact was that Mr Ferdman was centrally involved in the receipt of the tainted funds by means of transactions between Yulara and DLH. All three members of the Court of Appeal emphasised that the “directing mind and will” issue must be determined, not at a single, general level but by reference to the particular circumstances with which the court is concerned. This appears from the judgments of Nourse LJ at p 696a (“It is necessary to identify the natural person or persons having management and control in relation to the act or omission in point”); Rose LJ at p 699h (… “a company’s directing mind and will may be found in different persons for different activities of the company”); and Hoffmann LJ at p 706d (“The authorities show clearly that different persons may for different purposes satisfy the requirements of being the company’s directing mind and will”). The Court of Appeal held that the evidence was sufficient to justify Mr Ferdman being treated, for the purposes of the Yulara transaction, as DHL’s directing mind and will. 75. It is appropriate to add a brief footnote to El Ajou as a marker for the discussion of the fraud exception later in this judgment. In El Ajou the claimant relied unsuccessfully on an alternative argument for DLH being treated as having had Mr Ferdman’s guilty knowledge. This was based on principles of agency law (the fact that agency was regarded as an alternative argument underlines the point made at para 67 above). In discussing the cases on this point Hoffmann LJ observed (at p 702j): “Or there may be something about a transaction by which the principal is ‘put on inquiry’. If the principal employs an agent to discharge such a duty, the knowledge of the agent will be imputed to him. (There is an exception, the scope of which it is unnecessary to discuss, in cases in which the agent commits a fraud against the principal.)” By his acts Mr Ferdman exposed DLH to a heavy liability for equitable compensation and costs. But Hoffmann LJ seems not to have regarded that as engaging the fraud exception. 76. Meridian was an appeal to the Judicial Committee of the Privy Council from the Court of Appeal of New Zealand ([1994] 2 NZLR 291), which had upheld the first-instance decision of Heron J. The proceedings concerned a dishonest scheme to obtain control of a listed New Zealand company called Euro-National Corporation Limited (“ENC”). The individuals involved in the scheme included two senior employees of Meridian, a fund management company based in Hong Kong: Koo, who was Meridian’s chief investment officer, and Ng, who was a senior portfolio manager. The scheme failed. Its details are not relevant, except for the fact that at one stage Meridian, as a result of action by Koo which was unauthorised but was within his ostensible authority, purchased a large holding of ENC shares, and so Meridian became a “substantial security holder” within the meaning of a New Zealand regulatory statute, the Securities Amendment Act 1988. Section 20 required any person who becomes a substantial security holder to notify that fact to the Securities Commission of New Zealand, and supply prescribed particulars “as soon as the person knows, or ought to know” of it. The Securities Commission commenced proceedings against Meridian. That was the factual and statutory context in which Meridian argued (unsuccessfully at every stage, though with some variation in judicial reasoning) that it was not liable for the irregular conduct of Koo and Ng. The case was argued and decided on the issue of “knows” rather than “ought to know”. 77. Meridian is now rightly regarded as the leading case on the topic of attribution in company law. By 1995 earlier authorities, including those already mentioned, had made clear that the “directing mind and will” test in Lennard had to be applied to the particular circumstances of the case before the court, including the language and purpose of any relevant statutory provision. Meridian added further emphasis to this point, but its most important contribution to company law is perhaps the clarity of Lord Hoffmann’s tripartite classification of (i) (at p 506C) primary rules of attribution, derived from company law statutes and the articles of association of the company concerned; (ii) (at p 506G): “general rules of attribution which are equally available to natural persons, namely, the principles of agency”; and (iii) (at p 507B) exceptional cases “… when a rule of law, either expressly or impliedly, excludes attribution on the basis of the general principles of agency or vicarious liability.” 78. The third category is sometimes referred to as special rules of attribution, which may give the impression that they are to be found in specific statutory provisions. Any such impression would be misleading. The way in which Lord Hoffmann himself put it (at p 507E) was that “In such a case, the court must fashion a special rule of attribution for the particular substantive rule.” Often the legislature will not have spelled out any rule of attribution in clear terms, and in those cases it is the court’s task to determine the intention of the legislature from the statutory language, purpose and context. Similarly the court has had to fashion a special rule where it is dealing, not with some statutory provision, but with the general maxim ex turpi causa non oritur actio as in Stone & Rolls Limited v Moore Stephens [2009] 1 AC 1391 (“Stone & Rolls”) and Safeway Stores Ltd v Twigger [2011] 2 All ER 841 (“Safeway”). Lord Hoffmann explained this in a passage (at p 507 E-F) immediately following his observations about fashioning a special rule: “This is always a matter of interpretation: given that it was intended to apply to a company, how was it intended to apply? Whose act (or knowledge, or state of mind) was for this purpose intended to count as the act etc. of the company? One finds the answer to this question by applying the usual canons of interpretation, taking into account the language of the rule (if it is a statute) and its content and policy.” 79. The actual decision in Meridian it to be found at p 511D. After referring to the New Zealand regulatory statute Lord Hoffmann asked himself: “In the case of a corporate security holder, what rule should be implied as to the person whose knowledge for this purpose is to count as the knowledge of the company? Surely the person who, with the authority of the company, acquired the relevant interest. Otherwise the policy of the Act would be defeated.” That person was Koo. It is worth noting, in connection with the fraud exception, that Lord Hoffmann added (at p 511F): “The fact that Koo did the deal for a corrupt purpose and did not give such notice because he did not want his employers to find out cannot in their Lordships’ view affect the attribution of knowledge and the consequent duty to notify.” Since Meridian there have been further relevant authorities, but it will be better to refer to them as part of the discussion of the fraud exception. The fraud exception: the principle and the early cases 80. The existence of the fraud exception is not in doubt, but there has been a good deal of debate (both in decided cases and among legal scholars) as to its proper scope and limits. The situation to which it most squarely applies (and some would say, the only situation to which it should properly be applied) is where a director or senior employee of a company seeks to rely on his own knowledge of his own fraud against the company as a defence to a claim by the company against him (or accomplices of his) for compensation for the loss inflicted by his fraud. The injustice and absurdity of such a defence is obvious, and for more than a century judges have had no hesitation in rejecting it. 81. An early example is the decision of the House of Lords in Gluckstein v Barnes [1900] AC 240. In 1893 a syndicate of four businessmen acquired the Olympia exhibition premises in London. They formed a company, of which they were the only directors, to purchase the property, and issued shares from which they raised sufficient money to pay themselves £171,000, together with fully-paid shares in the amount of £9,000. They disclosed that the property had cost the syndicate £140,000, but not that they were making a further secret profit of over £20,000. The company went into liquidation and in 1897 the official liquidator issued a summons against one of the syndicate, Mr Gluckstein. He relied on various arguments, including the contention that the syndicate’s knowledge of the secret profit amounted to the company’s knowledge, and that the company must be taken to have approved it. The Earl of Halsbury LC dealt with this argument with characteristic robustness (at p 247): “My Lords, I decline to discuss the question of disclosure to the company. It is too absurd to suggest that a disclosure to the parties to this transaction is a disclosure to the company of which these directors were the proper guardians and trustees.” Similarly Lord Macnaghten, at p 249, referred to the argument as “absurd” and “mere farce”. 82. The principles underlying these robust judicial observations have been analysed in leading textbooks, including Gower and Davies, Principles of Modern Company Law, 9th ed (2012), and Bowstead on Agency, 19th ed (2010). The editors of Gower and Davies, Professor Paul Davies QC and Professor Sarah Worthington QC, comment at para 7-39 (footnotes omitted): “A potential obstacle to such a claim by the company is the common law doctrine that knowledge of an agent is attributed to the principal, which, if applied to the fraudulent agent, would defeat the company's claim against its agent. Sensibly, however, the courts have generally refused to apply the attribution of knowledge rule to the fraudulent agent. Although this is sometimes said to be inconsistent with the rule that a company is liable to third parties for the fraud of the agent, it is submitted that this is not so, either doctrinally or in principle. It is entirely proper to use different rules to govern the liability of the company to the third party, on the one hand, and of the agent or employee to the company, on the other. There is no reason why rules developed to protect third parties against losses caused by fraudulent agents should also operate to protect those same fraudulent agents against the company.” 83. The editors of Bowstead, Professor Peter Watts QC and Professor Francis Reynolds QC, reach the same conclusion but go rather further in regarding the case, not so much as an exception to a rule of imputation, but as a situation in which imputation is simply irrelevant (para 8-213, footnotes omitted): “But the existence of any general fraud exception is open to question. It can be said first that the deployment of the exception as a defence to property-based or restitutionary claims is not supported by the weight of authority, nor is it consistent in that context with the principles of vicarious liability. Secondly, it is not surprising that the use of such an exception has been deployed to protect claimants against a defendant’s arguments (for example, when sued for knowing assistance to the fraud of the claimant's agent) that the claimant is deemed to know of the fraud of its own agents, and on that account can be assumed to have condoned the wrong (by estoppel or ratification). Such arguments by defendants, though hazarded from time to time, are plainly without merit. However, in such situations imputation has no reason to operate. The rules of imputation do not exist in a state of nature, such that some reason has to be found to disapply them. Whether knowledge is imputed in law turns on the question to be addressed. Hence, a fraud exception is superfluous in these situations. Indeed, the irrelevance of imputation to cases where a principal is pursuing an agent, and implicated third parties, as a result of breach of duty by the agent to the principal is not confined to fraudulent breaches of duty. Were the principal deemed to possess the agent’s knowledge of his own breaches of duty, and thereby to have condoned them, the principal could never successfully vindicate his rights.” 84. The exception is sometimes referred to as the principle or rule in Re Hampshire Land Company [1896] 2 Ch 743 (“Hampshire Land”). That was a case in which the same individual was the secretary of two companies, one of which made an irregular loan to the other. Some legal scholars consider the case to have been rightly decided, but that the true ground of decision should be seen as related to the officer’s dual capacity. The case was however referred to as an authority for the fraud exception by Viscount Dunedin in Houghton and Co v Nothard, Lowe and Wills Limited [1928] AC 1, 14, (“Houghton”) which concerned conflict of interests and breach of fiduciary duty in relation to a joint venture company: “But what if the knowledge of the director is the knowledge of a director who is himself particeps criminis, that is, if the knowledge of an infringement of the right of the company is only brought home to the man who himself was the artificer of such infringement? Common sense suggests the answer, but authority is not wanting.” Viscount Dunedin then referred to Hampshire Land, as did Viscount Sumner at p 19. 85. One of the issues in this appeal is whether the fraud exception can apply in circumstances where the primary rules of attribution are in play. On that point counsel for the liquidators rely on Belmont Finance Corporation Limited v Williams Furniture Limited [1979] Ch 250 (“Belmont”). It was cited to the Court of Appeal but not referred to in the judgments. It may be regarded as the first important case in the modern history of the fraud exception, and it calls for consideration in some detail. It was a curious case in that it was a successful appeal against the decision of the trial judge, Foster J, to dismiss the claim of Belmont at the close of its case on the ground that there was no case to answer. As Buckley LJ put it (at p 255): “On account of the way in which the case went, the judge has heard the evidence of the plaintiff company’s side only; he has heard no evidence of, or on behalf of, any defendant. He made no findings of fact; in view of his judgment it was unnecessary for him to do so. He dealt with the case upon the plaintiff company’s pleading, and on that alone he held that since the agreement is alleged to have formed part of the alleged conspiracy, and since the plaintiff company was a party to the agreement, the plaintiff company was a conspirator. So he held that the claim in conspiracy failed in limine on the ground that one party to a conspiracy to do an unlawful act cannot sue a co-conspirator in relation to that act.” It was a surprising outcome, especially as Belmont was in compulsory liquidation and the action was brought by a receiver at the direction of the Companies Court. Even more surprisingly, there was a second successful appeal, Belmont Finance Corporation Limited v Williams Furniture Limited (No 2) [1980] 1 All ER 393 (“Belmont No 2”), from the same judge’s dismissal of the action after a retrial. The fuller statement of the facts in Belmont No 2 assists in the understanding of the earlier decision. 86. The complicated series of transactions by which the sum of £500,000 was extracted from Belmont, in exchange for a dubious investment in a property company called Maximum, is described in detail in Belmont No 2 at pp397-399. In very brief summary, there were two groups of businessmen, one group consisting of Mr James and his associates, who controlled Williams Furniture Ltd (“Williams”), City Industrial Finance Ltd (“City”) and Belmont. City was a wholly-owned subsidiary of Williams and Belmont was a wholly-owned subsidiary of City. Belmont provided hire-purchase facilities for the group, and its creditors included numerous depositors, for whose benefit the proceedings were taken. The other group consisted of Mr Grosscurth and his associates, who controlled several companies, including Maximum. As Waller LJ put it in Belmont No 2 at p 414, at the end of the transactions: “From the point of view of the two groups it was easy to see that the situation was satisfactory. The Grosscurth group had acquired Belmont at no cost to themselves and the James group had sold Belmont for £489,000 … But the question with which we are concerned is Belmont. Had Belmont benefited from a transaction in which the company had paid £500,000 for Maximum even if Maximum was worth £500,000? ” On the evidence, Maximum was worth very much less than that Belmont had approved the transaction at meetings of its board of directors, two of whom were defendants in the action. The other defendants included Williams, City and Mr Grosscurth. 87. At the time of the first appeal these detailed findings lay in the future, but in Belmont Buckley LJ stated and applied the fraud exception on the basis of the assumed facts (at pp 261-262): “…if the allegations in the statement of claim are made good, the directors of [Belmont] must then have known that the transaction was an illegal transaction. But in my view such knowledge should not be imputed to the company, for the essence of the arrangement was to deprive the company improperly of a large part of its assets. As I have said, the company was a victim of the conspiracy. I think it would be irrational to treat the directors, who were allegedly parties to the conspiracy, notionally as having transmitted this knowledge to the company; and indeed it is a well-recognised exception from the general rule that a principal is affected by notice received by his agent that, if the agent is acting in fraud of his principal and the matter of which he has notice is relevant to the fraud, that knowledge is not to be imputed the principal.” In view of the Court of Appeal’s conclusion about the primary rules of attribution it should be noted that the sealing and the completion of the relevant written agreement were the subject of resolutions at formal board meetings of Belmont (see Belmont No 2 at p 398g-h). Nevertheless the fraud exception applied in relation to Belmont’s claim for redress. Belmont No 2 also illustrates the guilty knowledge of directors in Mr James’s group being attributed to Williams and City so as to make those companies conspirators (at p 404e) and again to City so as to make it liable on the ground of knowing receipt of trust funds (at p 405f). 88. This Court has accepted that the principle applied in Belmont is correct: China Everbright – IHD Pacific Limited v Ch’ng Poh (2002) 5 HKCFAR 630 at paras 49 and 52. The fraud exception : more recent cases: primary and secondary victims 89. The general principle underlying the fraud exception has been discussed and refined in numerous more recent authorities in different jurisdictions. Out of the large number of cases which counsel’s industry has brought to the Court’s attention, the most important cases are those mentioned below, in chronological order. It is the most recent case, Bilta (UK) Ltd v Nazir [2014] 1 All ER 168 (“Bilta”) that is the most illuminating. 90. The decision of the Federal Court of Australian in Beach Petroleum NL v Johnson (1993) 43 FCR 1 (“Beach”) was concerned with the acquisition by Beach, a South Australian company, of Mazeley, a Liberian company with oil fields in Oklahoma. The acquisition was at a grossly inflated price, a fact that was known to three directors and one deemed director of Beach, all of whom had conflicting interests. The lengthy judgment of Von Doussa J contains, at paras 22.20 to 22.34, a careful review of the authorities from Hampshire Land to Belmont. He summarised Belmont as follows (at para 22-30): “What insulated [Belmont], but not [Williams and City] from having the knowledge of those officers imputed to them was the fact that only [Belmont] was treated as a victim. [Williams and City] were parties who benefited by reason of the transaction.” 91. He also referred to the decision of the Supreme Court of Canada in Canadian Dredge and Dock Co Ltd v The Queen (1985) 19 DLR (4th) 314. The Supreme Court of Canada adopted a rather different test involving the notion of an act being “totally in fraud of the corporate employer and where the act is intended to and does result in benefit exclusively to the employee-manager.” That test has had some influence in Australia but has not been adopted in England or Hong Kong. 92. PCW Syndicates v PCW Reinsurers [1996] 1 WLR 1136 (“PCW”) and Group Josi Re v Walbrook Insurance Co Ltd [1996] 1 WLR 1152 (“Group Josi”) were heard together in the English Court of Appeal, but separate judgments were given at different dates since the latter appeal raised some special issues of its own. Both appeals were concerned with corporate knowledge in the context of the statutory obligations of an assured under sections 18 and 19 of the Marine Insurance Act 1906. For present purposes it is sufficient to observe that these cases illustrate and reinforce the importance, in this area, of awareness and understanding of the context in which an issue of imputation or attribution arises (see especially Staughton LJ in PCW at pp 1145F-1146B). 93. Duke Group Limited v Pilmer (1999) 73 SASR 64 (“Duke”) was a decision of the Court of Appeal of South Australia on another claim based on a corporate acquisition at an inflated price. It began as a claim against accountants who had prepared a report for the shareholders of Duke (formerly called Kia-Ora) in relation to its acquisition of Western Union. Later the directors of Duke were added to the proceedings, first as third parties and then as defendants. There was an issue as to whether, in the unusual circumstance of the case, the accountants owed fiduciary duties, as well as a duty of care, to Duke. Part of the very lengthy judgment (paras 616 to 652) discussed the fraud exception, including its application to the accountants’ pleading of contributory negligence by Duke. It considered and approved the approach of Von Doussa J in Beach. It concluded that Duke could recover against the accountants, despite breaches of fiduciary duty on the part of its directors, but that their conduct should be treated as fault for the purposes of the issue of contributory negligence (paras 639 and 649). 94. Arab Bank plc v Zurich Insurance Co [1999] 1 Lloyd’s LR 262 (“Arab Bank”) was a decision on unusual facts. A leading firm of estate agents and valuers (“JDW”) held a professional indemnity policy issued by Zurich Insurance. It contained a term covering any dishonest or fraudulent act of a director, with an exception for “any person knowingly committing, making or condoning such act,” and other general conditions relating to fraud and non-disclosure. The form of the policy was supposed to be tailor-made for firms of estate agents and valuers, but had been drafted at a time when such firms were almost always unincorporated partnerships rather than companies. Mr Browne, the managing director of JDW, made several fraudulent valuations in the name of JDW. Arab Bank and its co-plaintiff obtained judgements against JDW and (on its going into liquidation), sought to enforce the judgments directly against the insurers. The judgment is a decision on ten preliminary points, several of which related to the construction of the policy. After referring to the authorities, including PCW and Group Josi, Rix J stated (at p 282): “In the present case, fraud is also assumed, but the primary victim of the fraud has been the lending institution [Arab Bank] which has relied on the valuation. I would accept, however, the plaintiffs’ submission that JDW was also a victim, even if only a secondary victim, of the assumed fraud. One consequence of that assumed fraud has been JDW’s liability to the plaintiffs, albeit in negligence.” This may be the first, or one of the first references in the reported cases to primary and secondary victims. Rix J held (at p 283) that the knowledge of Mr Browne and Mr Pitts (another director complicit the fraud) should not be attributed to JDW, but the decision turned more on the construction of the policy than on Hampshire Land. 95. In Odyssey Re (London) Ltd v OIC Run-off Ltd [2001] Lloyd’s LR (Ins and Reins) 1 (“Odyssey”) an issue of attribution arose in the very unusual context of an application to set aside a judgment, given after a long trial, on the ground that it had been obtained by perjury. The issue was whether the perjury of Mr Sage, who had given evidence for Orion Insurance Co plc, should be attributed to that company. Nourse LJ (at p 12) saw the issue, in that context, as one of status, rather than authority. Brooke LJ stated (at p 64, after a reference to Tesco): “In principle, I would be anxious if the natural development of the civil law in accordance with the thinking of leading judges in this specialist field on the way in which responsibility should be attributed to corporate bodies today had to be cribbed, cabinned and confined by decisions of the House of Lords in the criminal law arena.” A little later (also at p. 64) he added: “When one comes to look at the criminal cases, one very rapidly identifies a shoal of artificial features that have little to do with the civil law issues with which we are concerned. For a variety of reasons, the identification of the principles on which criminal liability should be attributed to corporate bodies took a long time to mature in this country, and when they did come of age, the birthday party was overshadowed by the judges' reluctance to attribute absolute liability to a corporation for a criminal offence (subject only to any statutory defences that might be available to it).” 96. In McNicholas Construction Co Ltd v Customs and Excise Commissioners [2000] STC 553 (“McNicholas”) the relevant issue, for present purposes, was whether McNicholas, an engineering contractor, had within the meaning of section 60 of the Value Added Tax Act 1994 engaged in conduct involving dishonesty for the purpose of evading VAT. This raised an issue of attribution. Much of McNicholas’ work consisted of digging trenches for cable ducts, and most of that work was done by sub-contractors. Some sub-contractors were registered for VAT purposes but others were not registered. Site managers employed by McNicholas conspired with registered sub-contractors to accept and pay VAT invoices from registered sub-contractors for work actually done by unregistered sub-contractors. This would have been financially neutral for McNicholas if it could lawfully deduct the VAT from its own output tax; as it was, the dishonest scheme exposed it to penalties. McNicholas argued that the dishonest intentions of its site managers should not be attributed to it, because of the fraud exception. Dyson J rejected that argument (para 55): “In my judgment, the tribunal correctly concluded that there should be attribution in the present case, since the company could not sensibly be regarded as a victim of the fraud. They were right to hold that the fraud was 'neutral' from the company's point of view. The circumstances in which the exception to the general rule of attribution will apply are where the person whose acts it is sought to impute to the company knows or believes that his acts are detrimental to the interests of the company in a material respect. This explains, for example, the reference by Viscount Sumner in J C Houghton & Co v Nothard Lowe and Wills Ltd [1928] AC 1 at 19 to making 'a clean breast of their delinquency'. It follows that, in judging whether a company is to be regarded as the victim of the acts of a person, one should consider the effect of the acts themselves, and not what the position would be if those acts eventually prove to be ineffective.” 97. Re Bank of Credit and Commerce International SA (No 15), Morris v Bank of India [2005] 2 BCLC 328 (“Morris”) was, as its title suggests, one of numerous sets of proceedings arising out of the fraudulent activities and massive insolvency of BCCI. Mr Samant, the general manager of Bank of India’s London branch, approached a senior employee of BCCI suggesting that BCCI should place deposits with Bank of India, and several large deposits were made between 1981 and 1983. The deposits were made on unusual terms and were part of a dishonest plan contrived by BCCI’s central treasury department to conceal heavy losses which it had incurred. The trial judge, Patten J, found that Mr Samant had “blind-eye” knowledge of the fraud (that is, that he suspected fraud and deliberately decided to avoid confirmation of his suspicion). He also held that this knowledge should be attributed to Bank of India, and rejected an argument based on the fraud exception. In upholding that decision the Court of Appeal placed importance on the language and purpose of section 213 of the Insolvency Act 1986, under which the liquidators of BCCI were proceeding. The Court cited McNicholas with approval (para 118): “As in McNicholas, the acts of Mr Samant were not in fact targeted at BOI. He was acting for, and in what he apparently believed to be the interests of, BOI in seeking to gross up the balance sheet for the purposes of the year end accounts. The potential liability of BOI under s 213 is irrelevant in deciding whether BOI was a victim of Mr Samant and whether his knowledge should be attributed to it for the purposes of s 213.” The Court also distinguished Arab Bank (para 124): “In our judgment, the facts and the contractual context make Arab Bank a different case. It did not lay down a general principle of attribution of knowledge which governs this case of statutory liability to make compensation to victims of fraudulent trading. Arab Bank is not, as Mr Moss contended, authority for the proposition that knowledge of fraud can only be attributed to a company if the individual with the relevant knowledge was a director or directing mind of the company, or where it can be inferred from all the circumstances that the individual transferred his knowledge to the company or to its directing mind and will; nor is it authority for the proposition that there can be no attribution of knowledge where the company is a ‘secondary victim’ of the individual’s wrongdoing or breach of duty.” 98. MAN Nutzfahrzeuge AG v Freightliner Ltd [2005] EWHC 2347 (Comm) (“MAN”) was an unusual case in which Mr Ellis, the financial controller of ERF, a manufacturer of heavy vehicles, persistently falsified its accounts, and defrauded Customs and Excise of VAT, throughout a period when ownership and control of ERF changed twice. The eventual purchaser and ERF sued the intermediate purchaser, which joined ERF’s auditors as a third party. That was the complex situation considered in the monumental judgement of Moore-Bick LJ. It is unnecessary to go into all the complexities, but it is worth noting that the auditors seem to have made no attempt to rely on the fraud exemption as a reason for avoiding liability. They were cleared of liability, but on other grounds. 99. It is also worth noting that in para 154 the judge pointed out that in this area there are three distinct sets of rules. The first set governs vicarious liability, which does not depend on the attribution to an employer of an employee’s state of mind. The second set does involve such attribution. The third set “governs the attribution of the acts and omissions of natural persons to juridical persons” (emphasis supplied) regardless of the state of mind of those natural persons. So for example the decision of the board of Belmont to acquire Maximum (paras 85 to 87 above) was attributed to Belmont, but the guilty knowledge of the members of the board was not attributed to Belmont. This point is implicit in many of the cases, but has not often been so clearly spelled out. 100. The decision of the House of Lords in Stone & Rolls [2009] 1 AC 1391, to which I was a party, has been the subject of a good deal of academic commentary, mostly critical (see for instance Professor Eilis Ferran, Corporate Attribution and the Directing Mind and Will (2011) 127 LQR 239; and the article by Professor Peter Watts mentioned in para 62 above). The issue of attribution arose in the context of a defence of ex turpi causa pleaded by auditors sued for breach of duty. The Law Lords were split three-two, and it is difficult to extract a clear ratio from the speeches of the majority. But it is not necessary or appropriate to discuss the case at any length, since it was on any view an extreme case on its facts. The claimant company had been formed and run exclusively for the purpose of fraud. It never had any legitimate business or assets. In this it resembled the company called Scadlynn Ltd in Brink’s-MAT Ltd v Noye [1991] 1 Bank LR 68, a company formed and used exclusively for the criminal purpose of smelting, recasting and selling stolen gold bullion. In that case the Court of Appeal treated Scadlynn as a victim (see Mustill LJ at p 72 and Nicholls LJ at p 73). I agree with the comment of Lord Phillips of Worth Matravers, in Stone & Rolls at para 5 that “if a person starts with nothing and never legitimately acquires anything he cannot realistically be said to have suffered any loss.” 101. Although lengthy discussion of Stone & Rolls is unnecessary and inappropriate, I should state clearly that I now see that I was wrong, in para 145, to regard the fraud exception as being of general application, regardless of the nature of the proceedings (Lord Mance, dissenting, dealt with this point at para 234). The recent Bilta decision (para 103 below), and the fuller citation and exposition of academic material put before this Court, demonstrate the limited scope of the exception. 102. The defence of ex turpi causa was revisited in Safeway [2011] 2 All ER 841. Safeway, a large supermarket chain, had exchanged commercially sensitive information with other supermarkets and food processors, resulting in increases in the price of milk. This anti-competitive conduct infringed section 2 of the Competition Act 1998 and exposed Safeway and other group companies to penalties. They brought proceedings against Mr Twigger and ten other defendants, all directors or senior employees of Safeway Group companies, seeking an indemnity. The defendants succeeded in striking out the proceedings by reliance on the ex turpi causa principle. The Safeway companies were “personally” at fault in the sense that their liability was not vicarious. They were, under the statute, the only persons liable to a penalty: see the judgment of Longmore LJ at paras 19 to 29. At para 29 he observed: “Once it is appreciated that the claimant companies are (personally and not vicariously) liable to pay the penalties exigible under the 1998 Act, those companies cannot invoke the Re Hampshire Land principle to say that they were not 'truly' liable. The principle gives them no defence to the OFT's claim for the penalties; they are personally liable to pay those penalties and it would be inconsistent with that liability for them to be able to recover those penalties in the civil courts from the defendants. The statutory scheme has attributed responsibility to the claimant companies and the Re Hampshire Land exception to the ordinary rule of attribution can have no import on the application of the ex turpi maxim.” 103. The most recent case is the decision of the Court of Appeal in Bilta [2014] 1 All ER 168. The Court consisted of Lord Dyson MR (who had at first instance decided McNicholas), Rimer LJ (who gave the leading judgment in the Court of Appeal in Stone & Rolls) and Patten LJ (whose first-instance judgment in Morris was upheld by the Court of Appeal.) Bilta had two directors, one of whom owned all its issued shares. It participated in a conspiracy in the form of a VAT “carousel” fraud on a massive scale. The details of the fraud (set out at paras 9 to 13) are technical and complex, but the end result was that Bilta sold carbon credits on terms that the proceeds were transmitted to a Swiss company, Jetivia, controlled by Mr Brunschweiler, leaving Bilta with no assets and a liability of over £38m to the British revenue authorities. Bilta and its liquidators took proceedings against its directors and several other defendants, including Jetivia and Mr Brunschweiler, who applied to have the claims struck out on the ex turpi causa principle. 104. Patten LJ noted that at first instance there had been argument as to whether the conspiracy had Bilta, or the revenue authorities (HM Revenue and Customs), as its intended victim. In his view that question was immaterial. The issue depended (paras 26 to 27) on whether Bilta was a co-conspirator. This was an issue of attribution. El Ajou, Meridian, McNicholas and Morris were all examples of civil liability to a third party being imposed on a company because of the dishonest conduct of its employees. In McNicholas and Morris it was argued that the dishonest conduct also harmed the corporate employer and made it “at least a secondary victim”. That, it was argued, should exclude the company’s liability to the third party. But that argument was rejected. After citing the judgment of Mummery LJ in Morris (para 114) Patten LJ made some observations that merit full quotation (paras 34 and 35): “[34] The point being made in this passage is that attribution of the conduct of an agent so as to create a personal liability on the part of the company depends very much on the context in which the issue arises. In what I propose to refer to as the liability cases like El Ajou v Dollar Land Holdings plc, Royal Brunei Airlines Sdn Bhd v Tan, McNicholas Construction Co Ltd v Customs and Excise Comm and Morris v Bank of India, reliance on the consequences to the company of attributing to it the conduct of its managers or directors is not enough to prevent attribution because, as Mummery LJ pointed out, it would prevent liability ever being imposed. As between the company and the defrauded third party, the former is not to be treated as a victim of the wrongdoing on which the third party sues but one of the perpetrators. The consequences of liability are therefore insufficient to prevent the actions of the agent being treated as those of the company. The interests of the third party who is the intended victim of the unlawful conduct take priority over the loss which the company will suffer through the actions of its own directors. [35] But, in a different context, the position of the company as victim ought to be paramount. Although the loss caused to the company by its director's conduct will be no answer to the claim against the company by the injured third party, it will and ought to have very different consequences when the company seeks to recover from the director the loss which it has suffered through his actions. In such cases the company will itself be seeking compensation by an award of damages or equitable compensation for a breach of the fiduciary duty which the director or agent owes to the company. As between it and the director, it is the victim of a legal wrong. To allow the defendant to defeat that claim by seeking to attribute to the company the unlawful conduct for which he is responsible so as to make it the company's own conduct as well would be to allow the defaulting director to rely upon his own breach of duty to defeat the operation of the provisions of ss 172 and 239 of the 2006 Act whose very purpose is to protect the company against unlawful breaches of duty of this kind. For this purpose and (it should be stressed) in this context, it ought therefore not to matter whether the loss which the company seeks to recover arises out of the fraudulent conduct of its directors towards a third party (as in the McNicholas case and [Morris]) or out of fraudulent conduct directed at the company itself which Sir Andrew Morritt C accepted was what is alleged in the present case. There is a breach of fiduciary duty towards the company in both cases.” 105. Patten LJ then subjected Stone & Rolls to a detailed analysis, especially in relation to “one-man” companies, an issue that does not arise in this appeal. But his observations on the duty owed by an insolvent or failing company’s directors to the company’s creditors carry great weight. He concluded (para 77) that Bilta’s position would have been the same: “…even if the true object of the conspiracy had been HMRC. In the context of a claim against the directors and the appellants [who were conspirators with the directors] for breach of fiduciary duty, the company is the victim regardless of whether its loss was consequential on that to a third party.” Lord Dyson MR and Rimer LJ agreed with his judgment. Attribution and the fraud exception: conclusions 106. The decision of the Court of Appeal in Bilta has achieved a welcome clarification of the law in this area. The general effect of the authorities discussed above can in my view be summarized in some short propositions. (1) Questions of attribution are always sensitive to the factual situation in which they arise, and the language and legislative purpose of any relevant statutory provisions: Tesco at pp 169-170, 194-195, 203; Meridian at pp 507, 511-512; Tesco No 2 at pp 1042-1043; PCW at p 1145; Group Josi at p 1169; Duke at para 615; McNicholas at paras 48-50; Morris at paras 116-124; Safeway at paras 29, 44-46; Bilta at paras 33-35, 45. (2) The “directing mind and will” concept in Lennard, although still often referred to in judgments, has been greatly attenuated by recognition of the importance of the factual and legislative context: El Ajou at pp 151, 154, 159; Meridian at pp 507-509 and 511; and numerous later cases. It might be better if it were to fade away as a general concept. (3) In some cases acts of directors and employees will be attributed to the corporate employer without their state of mind being so attributed: Duke at para 625, 641; MAN at para 154, illustrated by eg Belmont No 2 at p 398 in juxtaposition with Belmont at pp 261-262. (4) The underlying rationale of the fraud exception is to avoid the injustice and absurdity of directors or employees relying on their own awareness of their own wrongdoing as a defence to a claim against them by their own corporate employer: Gluckstein v Barnes at pp 247 and 249; Houghton at pp 14 and 19; Belmont at pp 261-262; Beach at para 22.30; Duke at paras 619 to 622; McNicholas at para 56; Morris at para 114; Bilta at paras 36 to 45. (5) The exception applies even if the wrongdoing consists of a transaction formally approved by the whole board of directors, and completed under the company seal: Belmont No 2 at p 398. In other words the exception can apply even when the primary rules of attribution are in play. (6) But the exception does not apply to protect a company where the issue is whether the company is liable to a third party for the dishonest conduct of a director or employee: El Ajou at p 702 (see para 75 above); Meridian at p 511 (see para 79 above); Duke at para 629; Morris at para 114; Bilta at para 34. (7) The supposed distinction between primary and secondary victims, although sometimes a useful analytical tool, is ultimately much less important than the distinction between third party claims against a company for loss to the third party caused by the misconduct of a director or employee, and claims by a company against its director or employee (or an accomplice) for loss to the company caused by the misconduct of that director or employee: Bilta at paras 45 and 77. (8) In cases concerned with insurance the terms of the policy are likely to be decisive, especially where a company has obtained cover against the risk of breach of duty, including fraud, by directors or employees: Arab Bank at p 283, and the comments on that case in Morris at paras 122-124. Internal fraud was the “very thing” from which the insurance cover was intended to protect the company. (9) The fraud exception does not appear to have been even raised as a defence, still less successfully relied on, in a claim by a company against its auditors for failure to detect internal fraud (as in Duke and MAN) with the sole exception of the extreme “one-man” company case of Stone & Rolls (see that case at paras 175 and 176). Again, internal fraud was the “very thing” from which the auditors had a duty to protect the company. (10) Criminal law cases are of little assistance in determining issues of attribution in civil law cases, because of the reluctance of the court, especially in the earlier cases, to treat offences as carrying strict liability: Odyssey at p 64; Tesco is an example, but Tesco No 2 and Safeway show the more modern approach. The judgments below 107. In paras 44 to 49 of his judgment Reyes J set out five issues that were before him. Those issues have narrowed as there has been no appeal against his conclusions on (i) the identification of the Commissioner’s decisions amenable to judicial review and (ii) her first ground of decision (prematurity of any conclusion about fraud). Moreover no more needs to be said about the proof of debt issue. The relevant part of the judgment of Reyes J is therefore paras 91 to 156, dealing with sections 64 and 70A of the IRO, and with issues of attribution so far as pertinent to the application of those sections. 108. Reyes J went at once to the issue of attribution, which is discussed at paras 94 to 134 of his judgment. He concluded that the Hampshire Land principle applied, and was not ousted by anything in Stone & Rolls, since MGET could not be described as a one-man company (para 133). He concluded that the Commissioner “wrongly attributed the Ma family directors’ mindset to MGET” (para 134). In reaching this conclusion he did not have the benefit (nor did the Court of Appeal have the benefit) of the illuminating judgment of Patten LJ in Bilta. 109. Reyes J treated his conclusion on attribution as decisive, without the need for further discussion, of the issue under section 64. He did not set out his views on the place of section 64 in legislative scheme of the IRO, or on the meaning of “prevented” and “reasonable cause”. That may have been because the amended notice of application referred rather loosely to the issue as being whether “MGET had reasonable cause to object out of time under s 64.” 110. On the view that he took as to section 64, it was not strictly necessary to decide the issue under section 70A, but he expressed his view at paras 144 to 156. Again he treated his conclusion on attribution as decisive of the issue under section 70A. He held that the making of the relevant false returns was an “error” and not (in the words of Chan J in Extramoney [1997] HKLRD 387, 396) “a deliberate act in the sense of a conscientious choice”. He referred to Chan J’s observation (also at p 396): “I accept that in some cases where it can be proved that the profits stated in the accounts of a taxpayer had in fact not been made, this may be sufficient to show that there has been an error justifying a correction in the assessment.” 111. In the Court of Appeal the leading judgment was given by Kwan JA. Fok JA and Lam J gave short concurring judgments. Lam J, in his concurring judgment, expressed doubt as to whether an objection could be made under section 64 to “an assessment based entirely on the calculation set out in a tax return”. Some reference has already been made (paras 38, 41 and 50 above) to the judgment of Kwan JA. After her preliminary exposition of Meridian and the scheme of the IRO she discussed the “key question” of attribution of knowledge at paras 57 to 77, stating her conclusions on sections 64 and 70A at paras 78 to 80 respectively. 112. Kwan JA first addressed the primary rules of attribution. MGET’s articles of association adopted Table A. Its business and affairs were therefore, as usual, to be managed by the directors, and the directors were required to prepare financial statements in accordance with the Companies Ordinance, Cap 32, and lay them before the company in general meeting. She referred to the approval of the accounts by the individuals who were directors of MGET for the year ended 31 March 1999 and subsequent years. She then summarized the competing arguments as to whether the Hampshire Land principle could operate to oust the primary rules of attribution. She held (paras 62 to 64) that it could not operate in that way. She relied for this conclusion on El Ajou, MAN, Safeway and Moore v Bresler Ltd [1944] 2 All ER 515. She did not refer to Belmont, although it had been cited in argument. 113. In my view the authorities which Kwan JA relied on do not establish the conclusion that she reached, and Belmont is decisively against that conclusion. Like Reyes J she did not have the benefit of Bilta. That case has clearly demonstrated that the crucial matter of context includes not only the factual and statutory background, but also the nature of the proceedings in which the question arises. In what Patten LJ called a “liability” case the fraud exception has no part to play, because the company must take responsibility for the fraudulent conduct of its director or employee even if the company is in some sense a victim: see eg McNicholas and Morris. El Ajou was a liability case, since the claimant was making a claim against DLH on the ground of knowing receipt of the proceeds of a fraud on the claimant. At the same time it was also, as it happens, what might be called a “redress” case, but the claimant was an individual and there was no question of Mr Ferdman’s guilty knowledge being attributed to Mr El Ajou. It was rightly attributed to DLH. Nourse LJ made a passing reference (at p 153) to Hampshire Land but as an authority on dual capacity. I have already referred (at para 75 above) to the observations of Hoffmann LJ on this point. 114. The passage in MAN (para 154, referred to in para 99 above) seems to me to lead to a different conclusion from that which Kwan JA arrived at, as Belmont demonstrates. Legal transactions resolved on and formally recorded at a board meeting (in that case, the acquisition of the shares of Maximum) were valid under the primary rules of attribution, but without the dishonest intentions of some or all of the members of the board being attributed to Belmont when its receivers, acting under the direction of the Companies Court, sought redress from two of its directors and their co-conspirators. They were finally successful in Belmont No 2. Safeway and Moore v Bresler Ltd were decisions on criminal liability which do not give much assistance on this point. The Court of Appeal was wrong to conclude that the fraud exception can never apply to decisions taken, with all due formality, by a board of directors. 115. In paras 68 to 70 of her judgment Kwan JA discussed primary and secondary victims. In view of Bilta, it is not necessary to go into this point at any length. Reyes J (paras 128 to 130) had been sceptical about the distinction and Kwan JA was inclined to agree with him that MGET could be regarded, along with numerous banks, as a primary victim of the fraudulent conduct of the Ma family directors. But in the light of Bilta that cannot be determinative of the issue as to the fraud exception. 116. Kwan JA then considered the possible application of some special rule of attribution. She asked herself, correctly, “whether the attribution of knowledge is required to promote the policy of the substantive rule, or (to put it negatively) whether, if attribution is denied, that policy will be frustrated” (para 72, with a reference to McNicholas, para 44). The legislative policy behind sections 64 and 70A is to strike a balance between finality and avoidance of delay, on the one hand, and the need to avoid hardship, on the other hand. Kwan JA considered that if a company was not fixed with knowledge of its conduct by the primary rules of attribution, denial of attribution would not promote, but would frustrate the legislative policy. She added some further points on policy which Mr Dicker has criticised, and these criticisms carry weight. The important policy consideration is the balance between promptness and finality, on the one hand, and avoiding hardship and injustice, on the other hand. A secondary consideration is that individual businessmen and corporate businesses should, so far as possible, be taxed by the same rules. 117. As to section 64 Kwan JA concluded (para 78): “I hold that by the special rules of attribution, the knowledge of the Ma Directors in causing the falsified accounts to be prepared should be attributed to MGET. As the company is primarily liable for its conduct, it cannot be said to have been acting reasonably or to have been “prevented” from giving notice of objection within time by its own fraud.” 118. As to section 70A, she followed Extramoney. The submission of incorrect tax returns must be treated as a deliberate act of MGET and could not be an error within the meaning of section 70A. The scheme and policy of the IRO 119. For any government, faced with ever-increasing financial responsibilities and obligations, it is of the highest importance to have a fair and efficient tax system which can be expected, year on year, to produce public revenue to a more or less predictable level. Annual taxes should be levied so as to ensure prompt payment and so as to achieve finality within a reasonably short time. As Arden LJ observed in Monro v HMRC [2009] Ch 69, para 32, “The state has a legitimate interest in ensuring finality in fiscal transactions”; see also the citation from Chow Kwong Fai v CIR in para 55 above. Those are the policy aims of Parts 9 (Returns, etc), 10 (Assessments), 11 (Objections and Appeals), 12 (Payment and Recovery of Tax) 13 (Repayment) and 14 (Penalties and Offences) of the IRO. They apply for the purposes of property tax, salaries tax and profits tax, all annual taxes. 120. Section 51(5) in Part 9 (set out at para 52 above) might appear to contain one rebuttable and one irrebuttable presumption. Mr Brennan submitted that that is indeed its correct construction. But that is a submission which would lead to extraordinary results, which the legislature cannot have intended. In The Queen v Ng Wing Keung [1997] HKLRD 142 the Court of Appeal, in a judgment delivered by Bokhary JA, rejected the argument that the second presumption is irrebuttable (at p 146): “The word ‘and’ immediately preceding the second deeming provision ties the same to what goes before. It does that so as to incorporate, albeit not in the most felicitous language, the allowance of proof to the contrary.” The case was a criminal case but the decision on statutory construction is of general application. The effect of section 51(5) is therefore limited, in this case, to a presumption (not in dispute) that the company secretary or manager who signed the tax returns did so with the authority of MGET’s board of directors, and was aware of the contents of the returns. 121. Section 64 is the first section in Part 11, relating to objections and appeals. A notice of objection under section 64(1) is in effect the first step in the appeal process, in that it leads to a reconsideration by the Commissioner under subsection (2), at which the Commissioner may ask for books and papers and examine witnesses. If the reconsideration does not lead to agreement, the way is open for the taxpayer to appeal to the Board of Review under section 66, with the possibility of further appeals. 122. Section 70A has a six-year time limit for any “error or omission” application which would otherwise be barred by the stringent terms of section 70. Mr Dicker drew attention to the absence of any such overall time limit in the proviso to section 64(1). The only time limit, he said, was in the indefinite requirement that any extension of the one-month period should be “reasonable in the circumstances.” Mr Dicker relied on this as a point in his favour but I would draw the opposite conclusion. The more likely explanation is that the legislature regarded it as inconceivable that a reasonable extension of the short period of one month could ever amount to years, rather than a few weeks or (at most) a few months. The Court of Appeal seems to have taken a similar view in Lam Ying Bor Investment Co [1979] 1 HKLR 571. In that case the taxpayer company had three directors, only one of whom was active in its affairs. When he was taken ill with cardiac trouble the other two directors failed to take charge of company’s business for a period of just under three months. Huggins JA described this as “negligence…of a very high order.” The result of the delay was that a notice of objection to an additional assessment was lodged about seven weeks out of time, and the Court of Appeal followed the Full Bench in upholding the Commissioner’s refusal to grant an extension. 123. Mr Dicker submitted the ejusdem generis rule of construction does not apply to the proviso to section 64(1) because the words “absence from Hong Kong, sickness” do not establish a genus (that is, a category of related items) capable of limiting the meaning of “other reasonable cause”. In this submission he has the support of the notes to the Annotated Ordinances of Hong Kong (2013 Reissue, p 830), although no authority is cited. Mr Brennan submitted that the rule does apply, although he realistically recognised that it is not a particularly clear or compelling instance of it. 124. It is not a clear or compelling instance, but the language used, coupled with the context of a short time limit for a step that is potentially the start of an appeal process, inclines me to accept Mr Brennan’s submission that the proviso contemplates some temporary impediment of an external and physical nature, rather than something internal and psychological. If for instance a taxpayer received an assessment and thought that it was excessive, but refrained from lodging an objection because that would involve too much trouble and expense, and made an application for an extension of time only when it was explained to him that a large amount of tax was at stake, it is difficult to say that he had been prevented from acting sooner (and still less that he had been prevented by a reasonable cause). There must be a clear causal connection between the reasonable cause and the prevention. 125. Even in the absence of attribution to MGET of the guilty knowledge of Ma Bo Kee and Cary Ma, there is a difficulty about Mr Dicker’s invitation to the Court to regard the real MGET as having been “hijacked” by these fraudulent directors so as to be prevented from putting in proper returns of profits. Apart from the point made in the previous paragraph, there is the difficulty that this approach involves not merely metaphor (“hijacking”) but also metaphysics (“the real MGET”). As Lord Hoffmann put it in Meridian (at p 507), displaying his knowledge of the German philosopher Immanuel Kant: “There is in fact no such thing as the company as such, no ding an sich, only the applicable rules.” The difficulty becomes insuperable if the rules point to attribution to MGET of the directors’ guilty knowledge. 126. Section 70A raises different, but not wholly dissimilar considerations. The section was introduced in 1956, following a recommendation of Inland Revenue Ordinance Committee, to alleviate the draconian effect of section 70. Its scope is restricted by the need for an error in a return or an accompanying statement, by the proviso for an error which was nevertheless “the practice generally prevailing at the time”, and by the six-year time limit, which is a reasonably generous one. These restrictions represent the legislature’s striking of the balance between finality and fairness. 127. Even if the guilty knowledge of the fraudulent directors is not attributed to MGET, there is some difficulty in viewing the proposed substitution of an entirely new return and set of statements (which could not, it seems, include audited accounts). It would not be the correction of one specific error (such as excluding a bad debt, or a duplicated credit of a single receipt). Everything would have to be rewritten. 128. Again, this difficulty becomes insuperable if the guilty knowledge of the fraudulent directors is attributed to MGET. A deliberate lie cannot be an error for the purpose of section 70A. It is therefore necessary to revisit the issue of attribution. Conclusions on attribution 129. The most important authorities on the topic of attribution, and the views of some legal scholars with a special interest in this topic, having already been considered at some length (paras 61 to 105 above) and I have summarised my conclusions at para 106. These conclusions owe much to the decision of the English Court of Appeal in Bilta. It has resolved several difficulties which the decision of the House of Lords in Stone & Rolls regrettably failed to resolve. I can therefore state my conclusions quite briefly. 130. Although the Court of Appeal was wrong to think that application of the primary rules of attribution must automatically trump and exclude the fraud exception, much of the Court’s analysis was correct. The profits tax returns were signed and submitted by an official of MGET with the authority of MGET’s board of directors. The returns were based on MGET’s audited accounts, which were included in the statements submitted with the returns. Ma Bo Kee was a director for all six years during the relevant period, and Cary Ma was a director for five of those years. All the available evidence indicates that it was they who were the active perpetrators of the frauds, and in particular the preparation of false accounts (it is immaterial whether or not Michelle Lam was also involved). No one suggests that Ma Bo Fung or Ma Bo Lung, who seem to have attended board meetings and signed the accounts on the occasions of their infrequent visits from the mainland, played any active part in the preparation of the false accounts. Ma Bo Kee and Cary Ma were, both under the primary rules and by the Lennard test as explained in later cases, the obvious persons whose knowledge should be attributed to MGET, unless the fraud exception applies to exclude such attribution. 131. In the light of Bilta, which goes a long way (but not the whole way) to confirming the views of Professor Watts, the true purpose and limits of the fraud exception have become much clearer. The gradual accretion of learning about primary and secondary victims, with or without additional refinements such as “targeting” or “vehicle of fraud”, can be seen as having missed the point. The crucial distinction depends on the nature of the proceedings in which the issue of attribution arises. On one side there are what Patten LJ (in Bilta, para 34) called the liability cases, such as El Ajou, Meridian, McNicholas and Morris. In them a company is being sued by a third party (which may be an official body) because the company is responsible for dishonest conduct on the part of one or more of its directors or employees. Here the fraud exception does not apply, even if the company is in some sense a victim. On the other side are what may be called the redress cases, such as Gluckstein v Barnes, Belmont, Beach and Bilta itself. In cases of this sort a company is seeking to make its own delinquent director or employee (probably by then an ex-director or ex-employee), or an accomplice of such a person, accountable for the loss that the company has suffered. That is the situation in which the fraud exception applies, because it would be absurd and unjust to permit a fraudulent director or employee to be able to use his own serious breach of duty to his corporate employer as a defence. 132. Mr Dicker has argued that the present case should be recognized as a redress case, in which the liquidators of MGET are trying to recover profits tax which ought not to have been paid, and would not have been paid but for the fraudulent directors’ serious breaches of their duty to MGET. Mr Brennan has disputed that, likening the position of the Commissioner to someone who has acquired title in good faith and without notice (he did not, as I recollect, add “for value”). 133. The authorities cited to the Court include some cases which do not fit comfortably into either of the “liability” and “redress” categories. These are cases of a claim under a fidelity policy covering an employee’s fraud (Arab Bank) and claims against auditors who have failed to uncover fraud (Duke and MAN, with Stone & Rolls as the controversial exception): see para 106(8) and (9) above. In those situations the claim is not for redress from a fraudster or his accomplice. It is against insurers or auditors who have, for value, undertaken to provide protection against the risk of internal fraud, or to use reasonable professional skill to uncover internal fraud. Such insurers and auditors must be supposed to have had ample opportunity to acquaint themselves with the relevant business before undertaking these obligations. There is no reason for the law to apply the fraud exception so as to absolve the contractual obligations of the insurers or the auditors (except, as the House of Lords held, in the extreme and exceptional circumstances of Stone & Rolls). 134. The liquidators’ claim against the Commissioner fits even less comfortably into either of the “liability” and “redress” categories. The Commissioner is not the accomplice of a fraudster. The Commissioner’s position has a faint resemblance to that of insurers or auditors in that they are all third parties from whom a corporate employer hopes to recoup, indirectly, part of a loss that it has suffered as the result of a director’s or employee’s misconduct. But there are obvious differences. The Commissioner, unlike insurers or auditors, cannot be expected to make inquiries about a taxpayer’s business in such a way as to become well acquainted with it. Furthermore, the Commissioner’s functions, powers and obligations are to be found wholly in the sphere of public law, and in particular in the IRO. An essential part of the scheme of the IRO is that the Commissioner should be able to make assessments on the basis of the taxpayer’s returns. It would frustrate this statutory purpose if the fraud exception were to intrude into this scheme. The fraud exception must be limited to its proper, limited role, that is of barring an unmeritorious defence in claims by corporate employers against dishonest directors or employees, or accomplices who have conspired with them. 135. The liquidators cannot therefore rely on the proviso to section 64(1) of the IRO, because MGET was not prevented from lodging an objection within time; it chose not to do so. Nor can the liquidators rely on section 70A, because MGET must be taken as having known that its returns were false, and (to summarize Extramoney), a deliberate lie is not an “error” for the purposes of that section. 136. The Commissioner’s decision was not flawed. I would dismiss the appeal with an order nisi for costs against the appellant, such costs to be taxed if not agreed. Any contrary submissions as to costs should be in writing, and should be served on the other party and lodged with the Court within 14 days of the handing down of this judgment, with liberty for the other party to serve and lodge submissions within 14 days thereafter. In the absence of such written submissions, the order nisi will stand absolute at the expiry of the time limited for those submissions. Chief Justice Ma: 137. For the above reasons, by a majority (given the dissenting judgment of Mr Justice Tang PJ as to a part of the appeal), the appeal is dismissed. The Court also makes the further orders as set out in para 136 above. Mr Robin Dicker QC, Mr Ashley Burns SC, instructed by Lipman Karas, and Mr Jason Karas of Lipman Karas, for the Appellant Mr Timothy Brennan QC and Mr Roger Beresford, instructed by the Department of Justice, for the Respondent [1] All references to the management of MGET are to the management during these tax years. [2] $88,972,757 [3] $21,503,428 [4] At p 79. [5] The Inland Revenue Ordinance Committee 1954 considered the time limit of 3 years in s 79(1) as originally enacted too short and recommended that it be extended to 6 years. Section 79(1) was amended by Ordinance 49 of 1956 and the time limit extended to 6 years. Para 104 of the Report of the first Inland Revenue Ordinance Committee 1954. (1954 Report) [6] The total tax revenue HKSAR Government for the financial year 2012-13 was $242.2 billion (Press release of the HKSAR government 2 May 2013). [7] Whether, and how that might be established is not the subject of this appeal. 
Chief Justice Ma: 1. I agree with the Reasons for Judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the Reasons for Judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: A. Introduction 3. Over the past decade, a form of transport has developed in many cities around the world from the use of new computer and internet technologies, including mapping and payment software programmes, that enable drivers of cars and passengers to locate one another, whether or not through third parties, and to request, provide and pay for carriage through the medium of their smartphones. The question that arises in this appeal is whether, in this jurisdiction, drivers who carry passengers in this manner can be guilty of the offence of contravening s.52(3) of the Road Traffic Ordinance (Cap.374) (“the RTO”). On the facts of this case, for the reasons which follow, an affirmative answer must be given to that question. Consequently, at the hearing, after hearing counsel for the appellants, the Court dismissed the appeal for reasons to be given in due course. These are my reasons for dismissing the appeal. 4. By s.52(3) of the RTO, it is provided that: “(3) No person shall— (a) drive or use a motor vehicle; or (b) suffer or permit a motor vehicle to be driven or used, for the carriage of passengers for hire or reward unless – (i) the vehicle is licensed as a public bus, public light bus or taxi; (ii) the vehicle is licensed as a private bus and the passengers are students, teachers or employees of an educational institution, disabled persons, or persons employed to assist disabled persons; (iia) the vehicle is licensed as a private light bus and is used— (A) as a school private light bus; or (B) exclusively for the carriage of persons who are disabled persons and persons assisting them; or (iii) a hire car permit is in force in respect of the vehicle.” 5. Contravention of the prohibition in s.52(3) is an offence under s.52(10) of the RTO which relevantly provides: “(10) Any person who contravenes— (a) this section, other than subsection (2) [concerning user of an unlicensed rickshaw], commits an offence and is liable in the case of a first conviction for that offence to a fine of $5,000 and to imprisonment for 3 months, and in the case of a second or subsequent conviction for that offence to a fine of $10,000 and to imprisonment for 6 months”. 6. The appellants each faced a single charge of “driving a motor vehicle for the carriage of passengers for hire or reward without a hire car permit” contrary to s.52(3) and s.52(10) of the RTO. 7. On the relevant dates charged, the appellants each drove and carried passengers in a private car in the following circumstances. The passengers, 21 of whom were undercover police officers and three of whom were civilians, had each used the Uber App on their smartphones to request a ride from a specified location to another specified location. The appellants respectively drove to the appointed places to pick up the passengers and then drove the passengers to their specified destinations. At the end of each trip, the fare was paid for by the passenger by a credit card transfer to an Uber entity and the passenger was notified by the Uber App of the amount paid. Although there was no direct evidence of the receipt by the appellants of any payments relating directly to the particular trips, there was some evidence of Uber paying remuneration to some of them. There was no hire car permit in force in respect of any of the motor vehicles driven by the appellants. 8. On 17 July 2018, the appellants were each convicted by a magistrate of the offence charged and sentenced to fines ranging from HK$3,000 to HK$4,500.[1] On 13 September 2019, their appeals against conviction to the Court of First Instance were dismissed.[2] On 29 November 2019, the judge refused to certify that points of law of great and general importance were involved in his judgment.[3] On 20 March 2020, the Appeal Committee granted leave to appeal to the Court of Final Appeal on the following question of law: “What, on the true construction of the Road Traffic Ordinance, Cap.374, are the elements of the s.52(3) offence with which the applicants were charged, and in particular, what is the proper construction of the phrase ‘for the carriage of passengers for hire or reward’ as used in that sub-section?” [4] B. The legislative scheme 9. The issue to be addressed in this judgment, as will be seen when the appellants’ case is discussed below, is one of statutory construction of s.52(3) of the RTO. It is therefore convenient to begin by setting out the legislative scheme in which that provision falls. B.1 Prohibition on certain types of user of vehicles 10. As might be expected from its subject matter, the RTO is an ordinance of considerable breadth. Its long title states that it is “[t]o provide for the regulation of road traffic and the use of vehicles and roads (including private roads) and for other purposes connected therewith.” In addition to the principal ordinance, there are regulations in subsidiary legislation covering activities including the construction and maintenance of vehicles, driving licences, parking, public service vehicles, the registration and licensing of vehicles, safety equipment, traffic control and so forth. 11. Part 6 of the RTO is headed “Use, Sale and Hire of Vehicles” and s.52 is the first section in that part. The heading of s.52 itself is “Restriction on the use of vehicles” and it contains a series of prohibitions in respect of various types of vehicles and activities. The prohibitions apply unless the vehicles concerned have been properly licensed. Section 52 essentially imposes a licensing regime as its regulatory purpose. (1) In HKSAR v Cheung Wai Kwong, the Court addressed the prohibition in s.52(1) and made certain observations about the legislative purpose of licensing under that provision.[5] (2) In this appeal, we are concerned with the prohibitions in s.52(3), namely on (a) the driving or use of a motor vehicle for the carriage of passengers for hire or reward, and (b) the suffering or permitting of a motor vehicle to be driven or used for the carriage of passengers for hire or reward, unless the vehicle is licensed as a public bus, public light bus or taxi (s.52(3)(i)), or is licensed as a private bus or private light bus and used for particular types of carriage (s.52(3)(ii) and (iia)), or is a private car and there is a hire car permit[6] in force in respect of it (s.52(3)(iii)). (3) It is relevant to note one of the other prohibitions in s.52, namely s.52(6) which provides that: “No person shall permit or suffer a motor vehicle which is licensed as a private car, private light bus or private bus to stand or ply for hire or reward.” (4) As will be seen, these or similar prohibitions have been part of the road traffic legislation in Hong Kong for many years. B.2 The hire car permit regime 12. Section 52(3) applies to motor vehicles, which, as defined,[7] covers any mechanically propelled vehicle. The present case is concerned, within that definition, with private cars. The prohibition in s.52(3) does not apply to a private car in respect of which a hire car permit is in force. 13. The regulations concerning hire car permits are contained in Part III of the Road Traffic (Public Service Vehicles) Regulations (Cap.374D). A “hire car permit” is defined in reg. 13 as “a hire car permit referred to in regulation 14(1)” and the Commissioner for Transport is authorised under reg. 14(1) to issue, in respect of private cars, hire car permits for various types of services, including a private hire car service. The other hire car services, namely hotel, tour, airport or school hire car services are not relevant to the present case. 14. Restrictions on the issue of hire car permits and the considerations for their issue are contained in reg. 15. The matters to which the Commissioner for Transport may have regard to in determining whether to issue a hire car permit for a private hire car service include: “(a) the extent to which the area from which the applicant proposes to operate the private hire car service is served by public transport; (b) whether the applicant is able reasonably to demonstrate that a private hire car service is required in the area from which he proposes to operate; and (c) whether the applicant has, in the area from which he proposes to operate the private hire car service, a place which, in the opinion of the Commissioner, is a suitable place to park the private car when it is available for hire.”[8] 15. The Commissioner for Transport has power to limit the number of permits issued at any time for any type of hire car permit (reg. 19), fees are payable in respect of hire car permits (reg. 20) and provision is made for offences and penalties for breach of the conditions to which the hire car permit is subject (reg. 21), those conditions being set out in Schedule 3 (reg. 14(5)). B.3 Legislative history 16. Since the parties each made submissions in their respective printed cases on the legislative history of the prohibition in s.52(3) and the hire car permit system, it is convenient now to set out some of the history. 17. There have been prohibitions on the carriage of passengers for hire or reward and on plying for hire for many years. These prohibitions date back as long ago as the Vehicles and Public Traffic Ordinance 1883,[9] s.2 of which provided: “[n]o vehicle shall ply or be let for hire for the carriage of passengers, unless the owner thereof has obtained a licence for the same …”. That ordinance was replaced by the Vehicles and Traffic Regulation Ordinance 1912,[10] which introduced a distinction between “private” and “public” vehicles. A “public vehicle”, as provided in s.2(h), included “every vehicle which plies for hire or is from time to time let out for hire or is intended to be let out for hire”. 18. The prohibitions on the carriage of passengers for hire or reward and on plying for hire continued to apply after 1912 in respect of private vehicles but were contained in various iterations of the subsidiary legislation rather than the principal ordinance. For example, the Road Traffic (Registration and Licensing of Vehicles) (Amendment) Regulations 1961[11] provided, by reg. 26, that: “(1) No person shall drive or use, or permit or suffer to be driven or used, any motor vehicle for a purpose other than that for which it is registered. (2) No person shall drive, or use, or permit or suffer to be driven or used, for the carriage of passengers for hire or reward, any motor vehicle which is registered as a private car, as a private omnibus, as a goods vehicle or as a dual-purpose vehicle for use as a private car and as a goods vehicle. (3) No person shall solicit, or attempt to solicit, any person for hire or reward to travel in any vehicle which is registered as a private car, as a private omnibus, as a goods vehicle or as a dual-purpose vehicle for use as a private car and as a goods vehicle. (4) No person shall permit or suffer a motor vehicle which is registered as a private car or as a private omnibus to stand or ply for hire or reward.” 19. In 1977, significant amendments to the legislation were made by the Road Traffic (Amendment) (No.2) Ordinance 1977.[12] These were introduced to address two mischiefs, namely (i) illegal public car hiring (colloquially known as “hung pais”) and (ii) illegal private car hiring (colloquially known as “pak pais”). This was achieved by: (1) providing for a route to “reregister as a taxi a motor vehicle which immediately prior thereto was registered as a public car” upon payment of a premium;[13] (2) abolishing the former class of motor vehicles known as “public cars” by providing that the registration of a vehicle as a public car would cease to have effect from the expiry of the vehicle licence in force for that car at the commencement of the amending section;[14] and (3) introducing regulations under which the Commissioner for Transport might grant or cancel “permits authorizing the use of private cars for the carriage of passengers for hire or reward”.[15] 20. Therefore, from the date of those amendments in 1977, public cars had to be re-registered as taxis, or they would become private cars when their then current vehicle licence expired, and private cars could carry passengers for hire or reward if authorised by a permit issued by the Commissioner for Transport. 21. Moving the second reading of the Road Traffic (Amendment) (No.2) Bill 1977, the Secretary for the Environment explained that: “… this bill has four main purposes. First, it seeks to abolish the present category of public hire car and to provide for these cars to be converted to taxis on payment of a premium if their owners so wish. Secondly, it will provide for the issue of contract hire permits to owners of private cars who wish to operate hire services on a strictly controlled basis, or to public hire car owners who do not wish to convert their vehicles to taxis. …”[16] 22. Hung pais, registered as “public cars”, were undesirable because, with red licence plates, they were well placed to impersonate taxis and to enjoy the privileges of being a taxi without having to observe the same controls and costs. As the Secretary explained: “… these cars are distinguishable by the fact that they carry a red number plate. But many of them are painted to resemble taxis and carry some form of meter. Although they are intended to be hired by pre-arranged contract only and it is illegal for them to ply for hire on the streets as if they were taxis, the great majority of them do act as if they were taxis for at least a part of the day. Not being taxis, however, they charge what fares they like and pick and choose the journeys they are willing to undertake. Nevertheless, the public are encouraged to support the illegal operations of these vehicles, partly because they look like taxis and partly because of the shortage of legal taxis in relation to the demand. At present there are more than 1,300 of these public hire cars and it is obviously wrong that the majority of them should continue to be permitted to ply for hire as taxis, without having paid the proper premium for the privilege and without being subject to the controls which are placed on taxi operators, including the carrying of the appropriate insurance for the protection of passengers. Nevertheless, the success of these illegal operations demonstrates a real public demand for the services which they provide. Having reviewed the situation, therefore, the Government has decided that public hire car operators should be given the opportunity to convert their vehicles to taxis on payment of a premium …”.[17] 23. Similar reasoning also applied to pak pais offering themselves for hire. In the same speech, the Secretary stated: “I turn now, Sir, to a further type of illegal operation which it is intended, by this bill, to bring under control. I refer here to the socalled [(sic)] ‘pak pai’ or private car offering itself for hire. While many pak pais perform a useful social function by providing personalized transport on a regular basis between homes, offices and schools, often operating from a particular building or block of flats, others function almost wholly as pirate taxis by plying for hire on the streets. None of them are effectively under any form of control as regards their mode of operation, the mechanical state of their vehicle or the insurance they carry for the protection of their passengers. Clause 3 of the bill seeks to regularize this situation by providing for the issue by the Commissioner for Transport of contract hire car permits. If this bill is passed into law, draft regulations laying down conditions for the issue of these permits will subsequently be put to the Governor in Council to be made. What is intended, however, is that, for a fee of $500 per annum, or $175 for four months, a private car owner, whose vehicle has been inspected and found to be mechanically sound and who is fully insured against third party risks, can be issued with a contract hire car permit. The conditions of the permit will be that the vehicle should only be operated from a designated address, that it should have no distinctive markings and that its permit should be displayed but not be visible from outside the vehicle. The object of these last two conditions will be to discourage the driver from plying for hire and the public from hailing the car on the street. Public hire car operators who do not opt to convert to taxis will also be eligible to apply for contract hire car permits. I should add that, in certain cases, for instance cars run by hotels to provide a service for their guests, the Commissioner for Transport will be empowered to permit distinctive markings to be used, as these would not mislead the general public into trying to hail the car. … Sir, if this bill is enacted it will, I hope, bring a much needed measure of control and regulation to the operation of hire cars and the taxi trade, to the benefit of the legitimate operators themselves as well as to the general public who make use of their services.”[18] 24. The hire car permit regime therefore introduced a distinction between legitimate and illegitimate pak pais, that is those with, and those without, a hire car permit. This distinction is important because it underlies the subsequent enactment of regulations to refine the regime. 25. The statutory regime for hire car permits was introduced by amendments to subsidiary legislation. By the Road Traffic (Registration and Licensing of Vehicles) (Amendment) Regulations 1977,[19] a new Part IIA (consisting of regulations 29B, 29C and 29D) was added to those regulations which provided for the issue of hire car permits by application, and upon payment of the prescribed fee, to the Commissioner for Transport. There was also added to those regulations a prohibition in these terms: “(2A) No person shall drive or use, or suffer or permit to be driven or used, any motor vehicle which is registered as a private car for the carriage of passengers for hire or reward unless there is in force in respect of that vehicle a hire car permit issued under regulation 29C.” 26. The Road Traffic (Registration and Licensing of Vehicles) Regulations were then amended in 1981[20] to include more detailed provisions for hire car permits. These included the introduction of separate types of hire car services for hotel, tour, airport, school and private hire car services. Restrictions on the issue of hire car permits and the considerations for their issue were contained in regulation 29D. In respect of hire car permits for a private hire car service, regulation 29D(5) provided: “A hire car permit for a private hire car service may only be issued to the registered owner of the private car and, in determining whether to issue such a hire car permit, the Commissioner may have regard to, amongst other matters – (a) the extent to which the area from which the applicant proposes to operate the private hire car service is served by public transport; (b) whether the applicant is able reasonably to demonstrate that a private hire car service is required in the area from which he proposes to operate; and (c) whether the applicant has, in the area from which he proposes to operate the private hire car service, a place which, in the opinion of the Commissioner, is a suitable place to park the private car when it is available for hire.” 27. In moving the second reading of the Road Traffic (Amendment) Bill 1981, the Secretary for the Environment re-stated the objective of the hire car permit regime and explained: “The purpose of this Bill is to augment and clarify the provisions for the issue of contract hire car permits which were included in the amendments to the Road Traffic Ordinance enacted in June 1977. The intention behind those amendments was to provide existing pak pai operators with the general opportunity to legalize their operations. This Bill, by spelling out the proposed arrangements for the issue of permits, seeks to provide safeguards for the contract hire care scheme when it is brought fully into effect to ensure that this original purpose is maintained. … Here I must emphasize again that the primary objective of the scheme, in accordance with the intentions of previous legislation, is to provide an opportunity for existing and bona fide pak pai operators to legalize their business under certain controlled conditions. The operators concerned would be well advised to avail themselves of this opportunity because, thereafter, all illegal carrying without a contract hire car permit and all plying and soliciting for hire or reward by vehicles other than taxis will be subject to stiffer legal sanctions, such as the impounding of vehicles and the suspension of vehicle licences, in addition to the existing penalties of a $1,000 fine and imprisonment for three months for a first offence and six months for subsequent convictions. …”.[21] 28. In 1982, a general revision of the Road Traffic Ordinance and its subsidiary legislation was carried out in response to the fragmentary structure of the old legislation which was considered to have become “unnecessarily complicated, confusing and cumbersome.”[22] This led to the enactment of the current RTO. C. The appellants’ contentions on this appeal 29. The appellants challenge their convictions on the basis that, on the facts found, the offence in s.52(3) of the RTO was not established. Essentially, the contention advanced on behalf of the appellants by Mr Jonathan Caplan QC[23] is that, on its true construction, the phrase “for the carriage of passengers for hire or reward” in s.52(3) requires that, for the offence to be proved against them, each of the appellant drivers had to be shown to be driving for the purpose of fulfilling a direct agreement for carriage between themselves and the passenger carried from which each driver was to be rewarded. 30. The magistrate[24] and the judge[25] below both rejected this contention and, in this Court, Mr William Tam SC,[26] for the respondent, submitted that they were correct to do so. 31. In this Court, the appellants advanced four bases for such a construction of s.52(3): (i) the wording of s.52(3) of the RTO; (ii) the legislative purpose of the provision; (iii) previous case law; and (iv) the technological advances since the enactment of the provision. D. The proper construction of “for the carriage of passengers for hire or reward” in s.52(3) 32. It is now long-established in Hong Kong that the exercise of statutory construction involves “an integrated consideration of text, context and purpose”[27] of the provision to be construed. The context of a statutory provision includes other provisions of the statute itself and, in seeking to ascertain its purpose, a statement made by the responsible official of the Government in relation to the bill in the Legislative Council may, sometimes, be used.[28] 33. Applying those principles of statutory construction, and for the reasons that follow, I would reject the appellants’ construction of s.52(3), namely that the phrase “for the carriage of passengers for hire or reward” requires a direct agreement for carriage between the driver and passenger concerned. Whether considered individually or cumulatively, the bases advanced by the appellants do not provide a sound support for their construction of s.52(3). D.1 The wording of s.52(3) 34. In the printed Case for the Appellants, it was contended[29] that the use of the connecting word “for” twice in the phrase “for the carriage of passengers for hire or reward” in s.52(3) links the “hire or reward” and the “carriage of passengers”, therefore requiring a direct causative link between (i) the defendant’s act of driving, (ii) his carriage of passengers, and (iii) his reward “that is to be derived directly from that singular act of driving passengers”. He must, it was submitted, be driving for the sole purpose of the carriage of passengers for reward and “the offence is proven only if the Appellants were driving for the sole purpose of fulfilling a direct agreement for carriage between himself and the passenger (i.e. for the carriage of passengers) from which he was to be rewarded (i.e. for hire or reward)”.[30] 35. Whilst there can be little doubt that s.52(3) requires there to be a link between the carriage of passengers and the hire or reward, it does not follow that the link must take the form of a single bipartite contract between the driver and his passenger. On the contrary, such a construction of s.52(3) would make no sense in the context of the driver of a public bus or public light bus (s.52(3)(i)), since the contract for the carriage would be between the passenger and the bus company; similarly, for a private bus (s.52(3)(ii)) or private light bus (s.52(3)(iia)). In addition, it would not sit with the distinction between the persons who drive the motor vehicle and the persons who use, suffer or permit the vehicle to be driven. Those words (“use”, “suffer or permit”) indicate that someone other than the driver may be involved in the carriage. 36. The context of s.52(3) therefore suggests that the words “for hire or reward” are to be read as referring to the nature or circumstance of the carriage rather than requiring the direct agreement contended for by the appellants. These words focus on the nature of the carriage of the passenger in the motor vehicle. The essential inquiry in most, if not, all cases is this: is the carriage of the passenger for hire or reward, whether by or from the passenger (the usual case) or someone else? 37. To illustrate this, one may take the example, put to counsel for the appellants by Lord Sumption NPJ at the hearing, of a taxi company that operates a fleet of 20 taxis and which takes telephone bookings for carriage from members of the public. If a caller telephoned and ordered a taxi to collect them (or their child) from their home and to drive them to a given destination on the basis that payment would be made to the taxi company by cash tendered at the end of the journey to the driver as the company’s agent, there is no reason why, as a matter of the language, the driver would not be properly described as driving the vehicle for the carriage of passengers for hire or reward. D.2 The legislative purpose of s.52(3) 38. The appellants also contended in their printed case[31] that the legislative purpose of s.52(3) was to prevent all cars, other than taxis, from plying for hire on the streets and that the abolition of “public hire cars” as a class of vehicle was to stop them plying for hire on the streets as if they were taxis. This mischief was also the reason for the controls imposed over pak pais by the introduction of the hire car permit regime. 39. It was further contended that, adopting a purposive approach, s.52(3) must be construed restrictively otherwise a driver under an extraneous contract, such as a chauffeur driving under an employment contract, would also be caught when carrying his employer’s friends of family or his employer’s employees or work associates. It was clearly not the intention of the legislature, the appellants contended, to target this sort of normal business or economic activity.[32] 40. That the legislative purpose of s.52(3) is limited solely to a prohibition on plying for hire is, however, too narrow a description of the legislative purpose of s.52(3). I have already touched on this in Sections B.2 and B.3 above. 41. There are clearly multiple legislative purposes served by the RTO and its various provisions. In Cheung Wai Kwong, where the Court was concerned with the prohibition in s.52(1) of the RTO on the use of a vehicle unless registered and licensed, the legislative purposes were identified as including revenue generation, administrative regulation and road safety.[33] The prohibitions in s.52(3) on the carriage of passengers for hire or reward unless a relevant licence or hire car permit is in force for the vehicle in question also serve those same legislative purposes. There is also an additional legislative purpose of s.52(3), read in the context of the RTO as a whole, which is the regulation of transport services businesses provided by a range of vehicles, including buses, taxis and private cars. 42. In this regard, it is relevant to note that there are various statutory provisions regulating such transport services businesses. Thus: (1) Public service vehicles are defined in s.2 of the RTO as “any motor vehicle registered as a public bus, public light bus or taxi, or as a private car in respect of which a hire car permit is in force” and such vehicles are subject to regulations for their licensing and operations in the Public Bus Services Ordinance (Cap.230), Public Bus Services Regulations (Cap.230A), and Road Traffic (Public Service Vehicles) Regulations (Cap.374D). (2) Taxis are subject to the requirement of a taxi licence (RTO, ss.22(1) and 25), and licensing and public safety requirements, namely: (a) the requirement in Part III of the Road Traffic (Construction and Maintenance of Vehicles) Regulations (Cap.374A) that the taxi pass a vehicle examination every year for the licence to be renewed; (b) the provision that a person cannot obtain a taxi driving licence unless he has at least three years’ driving experience and does not have any serious driving conviction within the previous five years (Road Traffic (Driving Licences) Regulations, Cap.374B, reg. 8), and the requirement that he be tested on his familiarity with the taxi regulations and local places and routes (Cap.374B, reg. 33 and Schedule 8); and (c) the requirements that taxis take the most direct practicable route and charge a fare according to a prescribed schedule of fares (Cap.374D, regs. 37(d) and 47); the fare must be calculated by taximeter, to be tested every six months (Cap.374A, reg. 44(1)). (3) Public buses are operated under franchises granted under s.5 of the Public Bus Services Ordinance (Cap.230), which requires that: the franchisee must provide a proper and efficient service (s.12) and comply with the condition of the franchise (s.5(3)(c)); the bus service may only operate on specified routes (s.11) and its operation is subject to controls over matters such as fares, routes and frequency (ss.13-16A). (4) Non-franchised public buses and public light buses are operated under passenger service licences granted under s.27 of the RTO and the Road Traffic (Public Service Vehicles) Regulations (Cap.374D), Part II, pursuant to which: (a) each passenger service licence is granted at a fee payable to the Government (Cap.374D, reg. 10) and is valid for no more than five years (Cap.374D, reg. 6); (b) the issue and renewal of passenger service licences are subject to policy considerations such as road traffic conditions and operator’s standard of service (Cap.374, s.28); and (c) conditions may be imposed on passenger service licences regarding the use of vehicles, such as the number or type of vehicles, the areas and routes, stopping and pick-up points, fares to be charged and frequency and period of service (Cap.374, s.29(1)(b)). (5) Public buses and public light buses are subject to detailed requirements regarding: (a) their construction and maintenance (Road Traffic (Construction and Maintenance of Vehicles) Regulations (Cap.374A), Part III); (b) the picking up and setting down of passengers (Cap.374D, reg. 33); (c) the cleanliness and conduct of drivers and passengers (Cap.374D, regs. 44-46); and (d) the need for an applicant for a driving licence for a public light bus to complete a pre-service course (Cap.374B, reg. 8A). (6) Hire car permits are issued only if the Commissioner for Transport is satisfied that the requisite policy of insurance and vehicle licence are in force and that the type of hire car service is reasonably required (Cap.374D, reg. 14(3)). In determining whether to issue a hire car permit, the Commissioner will have regard to the various matters stipulated in relation to each type of hire car service (Cap.374D, reg. 15). The matters relevant to the issue of a hire car permit for a private hire car service are those specified in reg. 15(5) (set out at [14] above). The respondent informed the Court that there are currently 1,040 private hire car permits in force, as against 18,163 taxis currently in operation in Hong Kong.[34] 43. The various regulations referred to in the preceding paragraph demonstrate clearly that vehicular passenger carriage in Hong Kong is intended to be operated on a strictly controlled basis. A comprehensive statutory regime has been established for the regulation and control of the operational standards, safety and quality of the various services in question. These regulations form part of the context and purpose in which s.52(3) is to be construed. As the Court noted in HKSAR v Ho Loy,[35] where the issue concerned regulations requiring compliance with a traffic sign, there is an important public interest in the statutory purpose of RTO provisions concerning the maintenance of road safety.[36] This public interest is also engaged in relation to the provision of passenger transport services. 44. That the legislative purpose of the hire car permit regime is not limited to private vehicles plying for hire on the streets is also reinforced by reference to the Secretary for the Environment’s speech moving the second reading of the Road Traffic (Amendment) Bill 1981 (set out at [27] above), in particular his reference to the fact that, after the implementation of the scheme under which pak pai operators could legalise their businesses under “controlled conditions”: “… all illegal carrying without a contract hire car permit and all plying and soliciting for hire or reward by vehicles other than taxis will be subject to stiffer legal sanctions”. (Emphasis added) 45. In any event, that the provisions of s.52(3) are not limited to the pak pai plying for hire on the streets is also demonstrated by the use of the prohibition against certain vehicles standing or plying for hire or reward in s.52(6). If s.52(3) were limited to the situation of a private hire car that was plying for hire or reward on the streets, there would be no need specifically to include a redundant prohibition of that activity in s.52(6). It must be assumed that the legislature intended a distinction between the two types of carriage activities in the two sub-sections. Since s.52(3) must include more than just the activity of plying for hire in the streets, the natural contextual meaning is that it extends to cover commercial carriage services to passengers for payment which are operated by a business entity even where the individual drivers have no separate agreement with the passengers being carried. 46. The appellants made the point[37] that there is no complaint here about the appellants not having valid third party insurance or their vehicles not being mechanically sound. Whilst this may be so, it does not detract from the lack of supervision of the mechanical soundness of their vehicles as compared with taxis. Nor does it necessarily mean that the more expensive third party insurance for commercial carriage, as opposed to domestic carriage, will have been arranged. 47. The appellants’ argument (see [39] above) that the construction of s.52(3) must be interpreted restrictively or else the offence could be committed by employed chauffeurs carrying passengers for their employers is unconvincing. There will, of course, be cases which raise a legitimate question as to whether a driver is using a vehicle for the carriage of passengers for hire or reward. An employed chauffeur who drives his employer’s spouse or child in the course of his employment will not usually be carrying a passenger for hire or reward in the sense that phrase should be understood in s.52(3). This is because, although he is paid to carry the passengers, his carriage of the employer’s spouse or child is a natural way of his employer using the car as a private car and not part of a separate business arrangement for carriage. 48. The hire or reward arrangement envisaged in s.52(3) is a business one. In this context, it is useful to refer to the House of Lords’ decision in Albert v Motor Insurers’ Bureau.[38] Although the issue in that case concerned whether a particular arrangement was carriage that required the driver to have compulsory third party insurance, the ratio of the decision was that “‘a vehicle in which passengers are carried for hire or reward’ meant a vehicle used for the systematic carrying of passengers for reward, not necessarily on a contractual basis, going beyond the bounds of mere social kindness and amounting to a business activity”.[39] The point is perhaps most helpfully expressed by Lord Pearson in his speech where he said: “One cannot fail to observe that a private motor car has passenger seats. The owner-driver of a private motor car can very easily be helpful and obliging to friends and acquaintances by giving them lifts in his car. He may himself like to have company on his journeys, but still he is conferring a favour. The passengers may, especially when this happens frequently, think it fitting that they should in return for the favour confer some benefit on the owner-driver. Many co-operative or reciprocal arrangements, which are natural uses of a private motor car, were suggested in the course of the argument. For instance, A and B may for their weekly game of golf travel to the golf course in A’s car driven by A, and B make his contribution by paying for A’s lunch or green fee or for the petrol that is bought on the journey. Mothers of children going to the same school may take turns at driving the children to and from the school. A party of men living in the same village and going to work in a city may take turns at driving the party in their respective cars. So long as such arrangements do not acquire the character of business arrangements they should be regarded as natural ways of using a private car as such and should not be regarded as involving the carriage of passengers for hire or reward.”[40] 49. Carriage will be for hire or reward, therefore, only where it constitutes “something more than a friendly arrangement such as was found to exist in Coward’s Case”.[41] The example of the employed chauffeur carrying passengers at the direction of his employer for his private purposes (which is not the case before us) can therefore be contrasted with the present case, where the appellants were providing a service to the public of on demand pick up of passengers and carriage to a destination of their choice, in return for payment from Uber for that carriage. This transportation service had the clear character of a separate business arrangement for carriage and was not merely ancillary to some other form of employment as a driver. D.3 The case law argument 50. The appellants cited two cases concerning prosecutions of pak pai drivers as being consistent with their construction of s.52(3). These were Kwong Pak-yam v The Queen[42] and Ngai Kam-chung v The Queen.[43] 51. Those were cases of drivers who solicited passengers and carried them from one point to another in return for payment of a fare. They were held to have carried the passengers for reward. The cases were therefore both cases of bipartite direct contracts between the passengers and the drivers. The comments of Huggins J (as he then was) relied upon by the appellants as to the need for a contract between the driver and his passenger were case specific and, since he was not deciding the issue based on the facts of the present appeals, do not lend support to the appellants’ construction of s.52(3). He was certainly not purporting to construe s.52(3) as only being applicable in a bipartite direct contractual situation. D.4 The new technology contention 52. The appellants contended that the business model of the Uber App was not in the contemplation of the legislature when enacting s.52(3), so it cannot have been the legislative intent to criminalise the conduct leading to the charges against the appellants. Reliance was placed on the case of Transport for London v Uber London Ltd[44] in support of the contention that an updating interpretation of s.52(3) to take into account this new technology was not warranted. 53. Whilst it may be true to say that the Uber App was not in the specific contemplation of the legislature in 1982, it does not follow that the activity in which the appellants were engaged was not part of the mischief which the provisions of the RTO were intended to address. Since the offence also catches the user of a vehicle, as distinct from its driver, as well as anyone who suffers or permits it to be driven for the carriage of passengers for hire or reward, it does not follow that, if a pak pai driver had used a third party dispatcher to receive requests for rides and to act as an accounting middleman in the commercial arrangement, that such activity would not have been within the mischief of the offence. On the contrary, there is no obvious reason why that activity would not have been within the prohibition. 54. The Transport for London case is distinguishable. The issue there concerned the meaning of the word “taximeter” in s.11 of the Private Hire Vehicles (London) Act 1998 and the question was whether the Uber smartphone application provided to drivers that enabled information to be relayed to a server to calculate a fare was a taximeter within that provision. It was held that the essence of a taximeter was a device for calculating fares whereas the smartphone was a device for recording time and distance and not for calculating the fare. Accordingly, a smartphone was not a taximeter. 55. Instead, the present case raises the question of whether, merely because the Uber platform was used by passengers to request the appellants to pick them up and to give them the destination to which they wished to be carried, in return for a payment by the passenger to Uber and by Uber to the appellants, this falls outside the legislative intent of s.52(3) because that platform did not exist in 1982. For the reasons set out above, there is no good reason to hold that it does. This is simply an application of the principle that a statute is “always speaking”: see HKSAR v Wong Yuk Man (2012) 15 HKCFAR 712 at [27] and R (Quintavalle) v Health Secretary [2003] 2 AC 687 per Lord Bingham at [9]-[10]. It is also an example of the situation described by Lord Jowitt LC in Joyce v Director of Public Prosecutions [1946] AC 347 at p.366: “[i]t is not an extension of a penal law to apply its principle to circumstances unforeseen at the time of its enactment, so long as the case is fairly brought within its language.” Notwithstanding that smartphone applications enabling the Uber business model were not specifically in contemplation when the relevant statutory prohibition was enacted, the activities of the appellants constituted “carriage of passengers for hire or reward” within the meaning of s.52(3) and well within the mischief that the statutory scheme seeks to address. E. The elements of the offence under s.52(3) clearly established here E.1 The actus reus of the offence 56. Taking the above legislative purpose and context into account, the offence under s.52(3) is made out where it is established, to the requisite standard of proof, that: (1) a person has driven or used a motor vehicle, or suffered or permitted another to do so; (2) the driving or use of the vehicle has been for the purpose of carrying passengers for hire or reward in that the carriage is undertaken as a business or commercial arrangement whereby payment is made by the passenger or on his behalf (whether to the driver or some third party) and that payment is received (whether by the driver or some third party) in respect of the provision of the carriage in question. 57. In relation to sub-paragraph (1) of the preceding paragraph, the terms “drive”, “use”, “suffer” or “permit” another to “drive” or “use” will have the same meanings as those terms were held to have in Cheung Wai Kwong: at [26]-[29] for “drive”; at [30]-[34] for “use”, but subject to [35]-[37] in relation to vicarious user; and at [45] in relation to “suffer” or “permit”. 58. In the present case, the appellant drivers each drove a vehicle, in respect of which a hire car permit was not in force, for the carriage of passengers for hire or reward. It was sufficient that the carriage was undertaken as part of the Uber ride service business and unnecessary that there be a direct contract between each driver and their respective passengers. E.2 The mens rea of the offence 59. The respondent accepted[45] that it is necessary for the prosecution to prove that the person charged has driven or used the vehicle “for the carriage of passengers” with knowledge, intention or recklessness, i.e. with full mens rea. 60. However, in relation to the element of the offence that the carriage of passengers be “for hire or reward”, the respondent contended[46] that the presumption of mens rea has been displaced and that, instead, the offence is in the third alternative of cases described in Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at [96]. The respondent therefore contended that the prosecution is not obliged to prove the accused’s knowledge, intention or recklessness in relation to the carriage of passengers being “for hire or reward” but he will have a good defence if he can prove on the balance of probabilities that the prohibited act was done in the honest and reasonable belief that the circumstances were such that, if true, he would not be guilty of the offence. 61. This point is not open to the respondent on this appeal. The magistrate ruled against the prosecution on its submission that the offence was one of strict liability.[47] As the judge pointed out, if the respondent had wanted to appeal against that ruling, it should have appealed by way of case stated. In any event, the judge agreed with the magistrate and rejected the prosecution contention that the offence under s.52(3) was one of strict liability.[48] 62. In the present case, the magistrate drew as inferences based on his findings of primary fact that the appellants had allowed the passengers, who were strangers, to get in their cars solely for the purpose of the car rides that were to be paid, and the appellants must have known and intended that the journeys were to be paid.[49] It was also an agreed fact that there was no hire car permit in force in respect of any of the vehicles driven by the appellants at the time of the offences in question. In those circumstances, there can be no doubt that offence was clearly established even if the presumption of mens rea in relation to the element of “for hire or reward” is not displaced. F. Disposition 63. The proper construction of the phrase “for the carriage of passengers for hire or reward” is as set out above and, applying that construction to the facts of the present case, the appeal was accordingly dismissed. 64. This appeal concerned the ambit of the prohibition in s.52(3) of the RTO which is a pure question of law. Whether ride hailing services should be permitted to operate in Hong Kong, on the other hand, is a question of transport policy, not a question of law, and is not a matter for the Court to determine. Mr Justice Cheung PJ: 65. I agree with the Reasons for Judgment of Mr Justice Fok PJ. Lord Sumption NPJ: 66. I agree with the Reasons for Judgment of Mr Justice Fok PJ. Mr Jonathan Caplan QC, Mr Derek Chan SC and Ms Katrina KW Lee, instructed by Haldanes, for the Appellants Mr William Tam SC, DDPP, Mr Marcus Lee SPP and Ms Cherry Chong PP, of the Department of Justice, for the Respondent [1] KCCC 3412-3430, 3432, 3496, 3969, 3971/2017 & 502/2018, [2018] HKMagC 3, Statement of Findings dated 8 August 2018 (Mr Joseph To, Magistrate). [2] HCMA 381-399, 401, 402, 404, 405 & 415/2018 (consolidated), [2019] HKCFI 2280, Judgment dated 13 September 2019 (Alex Lee J). [3] [2019] HKCFI 2899, Ruling dated 29 November 2019. [4] FAMC 58/2019, [2020] HKCFA 10, Determination dated 20 March 2020 (Ma CJ, Ribeiro & Cheung PJJ) at [4]. [5] (2017) 20 HKCFAR 524 at [20]. [6] RTO s.2 defines “hire car permit” as “a permit issued in accordance with this Ordinance authorizing the use of a private car for the carriage of passengers for hire or reward”. [7] RTO s.2. [8] (Cap.374D), reg. 15(5). [9] Ord. No. 5 of 1883. [10] Ord. No. 40 of 1912. [11] Made under the Road Traffic Ordinance 1957 (Ord. No. 39 of 1957). [12] Ord. No. 45 of 1977. [13] Road Traffic Ordinance (Cap.220) (“Cap.220”), s.4A. [14] Cap.220, s.4A(5). [15] Cap.220, s.3(1)(ga). [16] Legislative Council Official Report of Proceedings, 1 June 1977, p.966. [17] Ibid. [18] Ibid. pp.967-968. [19] L.N. 161 of 1977. [20] Road Traffic (Registration and Licensing of Vehicles) (Amendment) Regulations 1981 (L.N. 111 of 1981). [21] Legislative Council Official Report of Proceedings, 11 February 1981, pp.407-408. [22] Legislative Council Official Report of Proceedings, 28 July 1982, p.1107. [23] Appearing with Mr Derek Chan SC and Ms Katrina K.W. Lee. [24] Statement of Findings at [192]. [25] CFI Judgment at [65]-[67]. [26] Deputy Director of Public Prosecutions, appearing with Mr Marcus Lee, Senior Public Prosecutor, and Ms Cherry Chong, Public Prosecutor. [27] HKSAR v Chui Shu Shing (2017) 20 HKCFAR 333 at [42] (French NPJ). [28] HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at [13]-[14] (Li CJ). [29] At [29]-[31]. [30] Case for the Appellants at [31] (italics in original). [31] At [27] and [34]. [32] Case for the Appellants at [32]-[33]. [33] (2017) 20 HKCFAR 524 at [20]. [34] Case for the Respondent at [33]. [35] (2016) 19 HKCFAR 110. [36] Ibid. at [20]. [37] Case for the Appellants at [28]. [38] [1972] AC 301. [39] Ibid. at p.302B. [40] Ibid. at pp.332F-333A. [41] Kwong Pak-yam v The Queen [1965] HKLR 931 at p.936 (referring to Coward v Motor Insurers’ Bureau [1963] 1 QB 259). [42] [1965] HKLR 931. [43] [1965] HKLR 941. [44] [2015] EWHC 2918 (Admin), [2016] RTR 12 per Ouseley J at [39]. [45] Case for the Respondent at [58]. [46] Ibid. at [61]. [47] Statement of Findings at [172]-[185]. [48] CFI Judgment at [88]-[95]. [49] Statement of Findings at [191]. 
Chief Justice Ma: 1. For the reasons contained in the judgment of Mr Justice Fok PJ, with which I agree, the appeal was allowed with disbursements awarded to the appellant in the sum of $12,538.50. Mr Justice Ribeiro PJ: 2. I agree with the Reasons for Judgment provided by Mr Justice Fok PJ. Mr Justice Fok PJ: 3. The appellant was convicted, after trial before a magistrate,[1] of one charge of common assault contrary to section 40 of the Offences Against the Person Ordinance (Cap.212), for which he was sentenced to four weeks’ imprisonment, suspended for 18 months. His conviction was upheld on appeal to the Court of First Instance.[2] 4. The Appeal Committee granted leave to appeal to this Court on the ground that it was reasonably arguable that there had been a substantial and grave injustice in that the conduct of the trial and the treatment of the evidence might have miscarried so that the appellant was denied a fair trial.[3] 5. After hearing the parties, the Court allowed the appeal, quashed the appellant’s conviction and ordered the respondent to pay the appellant his disbursements incurred before the Deputy Judge and before this Court,[4] indicating that the reasons for doing so would be handed down at a later date. These are my reasons for making those orders. A. The trial and intermediate appeal 6. The case in the magistracy was a simple one. The appellant was a chef who operated a restaurant in Tai Po. He was accused of assaulting an inspector of the Environmental Protection Department (“EPD”) by punching him in the face. The EPD inspector, who gave evidence as PW1 at trial, had visited the appellant’s restaurant premises in the early evening of 3 March 2017 together with his supervisor, who gave evidence as PW2 at trial. Their visit was in the course of their duties to follow up on a notice issued by the EPD requiring the installation of an electrostatic precipitator (i.e. an air filter) to reduce the greasy fumes emitted by the restaurant’s kitchen. It was the prosecution case that the appellant, coming out of the kitchen and recognising PW1, was angered by his presence and punched him on the nose. The following evening, PW1 went to the Accident and Emergency Department of the Tseung Kwan O Hospital, where he was examined and his nose was found to be swollen. 7. The appellant was unrepresented at trial. He elected to give evidence but called no other witnesses to testify. He denied that he came out of the kitchen to confront the EPD officers at all or that the assault occurred as alleged and suggested that the EPD officers had concocted their evidence of the assault. 8. The magistrate accepted the evidence of the two EPD officers, PW1 and PW2, and disbelieved the appellant. She found that the appellant struck PW1 with his right fist once on the bridge of the nose causing pain, redness and swelling. She accordingly convicted the appellant. 9. On intermediate appeal to the Court of First Instance, the appellant, now represented, advanced four grounds of appeal, which included those for which leave to appeal to this Court was granted. The Deputy Judge dismissed the appeal and upheld the appellant’s conviction.[5] B. The accepted norms relevant to this appeal 10. The arguments in the parties’ written cases have focused on the question of the conduct of the trial and the treatment of the evidence, those being the bases on which the Appeal Committee considered it was reasonably arguable that the appellant was denied a fair trial. 11. How a trial is conducted fairly and the evidence properly treated, being relevant to the way issues of fact are resolved at trial, are both fundamentally important aspects of the criminal justice system. As this Court has previously explained, in Ko Man Chun v HKSAR:[6] “Most criminal cases in Hong Kong are tried in the Magistrates’ Court. The issues at such trials are generally of fact rather than of law. Where (as quite often happens) there is an appeal to a single judge of the High Court, the issues are still generally of fact. In practical terms, therefore, one of the most important aspects of our criminal justice system is how issues of fact are resolved in the Magistrates’ Court and then on appeal therefrom to a single judge of the High Court. As to this, the accepted norms include these. The issues of fact must of course be approached with due care. And the conclusions thereon must be expressed in terms indicative of such care. This is not to say that lengthy reasons are always or even generally required. It depends on the circumstances.” 12. A number of accepted norms concerning the conduct of a trial, in particular where a defendant is unrepresented, apply and are relevant here. Since the appellant in this case was unrepresented at trial, it was incumbent on the magistrate to assist him in the conduct of his defence. This is by no means an easy task, as Barnes J correctly observed in HKSAR v Chan Shu Hung:[7] “A magistrate is faced with an arduous task when he is dealing with an unrepresented defendant. The Magistrate must remain neutral but must also provide appropriate assistance to the defendant. Most likely a defendant has no legal knowledge. When he cross-examines a witness, he may ‘make statements’ instead of ‘asking questions’. The Magistrate should be patient and try to understand as much as possible what the defendant wants to say, and then translate the statements made by the defendant into questions and put them to the witnesses concerned.” 13. In particular, the conduct of the cross-examination of a prosecution witness may give rise to difficulties since the unrepresented defendant may be imprecise or inarticulate in expressing his case and how he wishes to dispute the prosecution evidence. In this regard, in The Queen v Chan Chun Kit,[8] Fuad VP provided helpful advice as to how a magistrate should properly assist an unrepresented defendant in this task, in the following terms: “I think I can do no better than to set out what Sir Alison Russell had to say (at pp.64-65) in the Second Edition of his work ‘The Magistrate’ which has been a source of wise counsel to lay and professional magistrates ever since it first appeared: ‘Question put to witness on behalf of accused. The following matter arises when the accused is [unrepresented]: it makes demands on the utmost skill and patience of the magistrate - (1) … (2) … (3) … (4) … (5) … (6) … (7) it can only be suggested that the magistrate should listen and do his best to find out what the question is that is implied by what the accused is saying, and then formulate that question, ask the accused whether that is the question he wishes to ask the witness, and, if the accused says that it is the question, then the magistrate should put it to the witness. (8) … (9) what the accused says for this purpose should not be recorded except in the form of relevant questions to the witness put by the magistrate on behalf of the accused, since it is really ‘cross-examination in the rough’ which is frequently elicited from the accused by questions put to him by the magistrate and so cannot properly be regarded as a ‘statement’ by the accused.’” 14. Regardless of the difficulty, it is incumbent on the magistrate to conduct the trial fairly and impartially and he must not lose the appearance of neutrality in the dispute. The principle was stated by Bokhary PJ, writing for the Appeal Committee, in HKSAR v Leung Kwok Hung:[9] “… In a criminal case tried by a judge without a jury or by a magistrate, if the judge or magistrate’s conduct of the trial has lost him the appearance of impartiality, the question, at least in general, is whether there is a real danger, as opposed to a merely fanciful one, that his conduct had actually influenced his decision adverse to the defendant so as to have deprived the defendant of a fair trial. A real danger of that is a basis for quashing the conviction or convictions concerned and, at least in general, the same should then be quashed.” 15. In addition to these norms relating to the conduct of the trial, it is also incumbent on a magistrate to address and explain material inconsistencies in the evidence of prosecution witnesses: see the helpful summary of citations provided by Nguyen J in support of this trite proposition in HKSAR v Lo Cheuk Him.[10] C. Was the appellant denied a fair trial? 16. The issue for the Court on this appeal was whether it was “shown that there has been to the appellant’s disadvantage a departure from accepted norms which departure is so serious as to constitute a substantial and grave injustice”: So Yiu Fung v HKSAR.[11] 17. For the following reasons, I conclude that there was such a departure from the accepted norms set out in Section B above that the appellant was denied a fair trial, so that his conviction was unsafe and unsatisfactory and should therefore be quashed. 18. This was a straight forward case in which either the prosecution case that there had been an assault was accepted or the defence case that the assault never took place and was concocted was accepted. The credibility and reliability of the evidence of the protagonists were crucial matters for the magistrate to weigh and consider in resolving this simple issue of fact. 19. In their testimony at trial, there were a number of discrepancies in the evidence of PW1 and PW2.[12] In particular: (1) PW1 testified that the appellant came out of the kitchen twice before the alleged assault, whereas PW2 testified that he only came out once. (2) PW1 testified that PW2 told the appellant the purpose of their visit when the appellant came out of the kitchen the first time at which point the appellant scolded them and swung a frying spatula in his right hand in front of PW1 before going back into the kitchen. PW2, however, testified that the appellant punched PW1 before he had a chance to reveal his identity. (3) PW1 testified that he was taking photographs when the appellant walked out of the kitchen to punch him, whereas PW2 testified that the appellant punched PW1 when he was holding a camera but had not yet started shooting. (4) PW1 testified that, before the appellant punched him, the appellant used his body to push PW1 away and that, after punching him, the appellant wrenched off PW1’s warrant card, which was hanging around PW1’s neck, with his hand. PW2 did not mention these details. 20. The magistrate did not address these differences in the evidence in her Statement of Findings, stating (at [34]) in relation to PW1’s testimony, “I notice that his evidence and PW2’s is consistent without discrepancies.” For his part, in respect of the discrepancies identified in the previous paragraph, the Deputy Judge described these (in the CFI Judgment at [33]) as “precisely the comforting crumbs in the wording of Stock J, which are insufficient to lead the magistrate to doubt their credibility or reliability.” 21. The reference by the Deputy Judge to “comforting crumbs” is likely a reference to the judgment of Stock J (as he then was) in The Queen v Kwong Wing On and Another, where he cautioned:[13] “… that microscopic dissection of a transcript will always uncover a discrepancy, a failure to answer a question, some inherent improbability or other, a piece of evidence not included in statements to the police, and a myriad of bits and pieces upon which to build pages of grounds of appeal. In the real world, and even with truthful witnesses, these discrepancies, improbabilities, and omissions will occur. Indeed if they do not, then the evidence is attacked as being artificial or collusive. A magistrate is not expected to deal expressly with every comforting crumb to which the defence may be able to point. A realistic attitude must be encouraged, and the approach to such attacks is to ask whether there have been material and significant discrepancies, improbabilities or omissions, such as would lead or should lead a tribunal to doubt credibility on central facts.” 22. Whether the discrepancies in the prosecution evidence to which the appellant points are material discrepancies that the magistrate was duty bound to deal with or peripheral factual issues which she did not need to resolve is clearly arguable. The evidence of two witnesses may differ in some respects but still be inherently accurate and true as to the central issue of fact that needs to be proved. As regards such inconsistencies, Stock VP rightly observed in HKSAR v Okafor [2012] 1 HKLRD 1041 at [33(3)] that: “… Whilst inconsistencies may be of the kind that truly affect reliability, inconsistencies should be approached realistically recognising that some inconsistencies are, with the passage of time, inevitable; some are more apparent than real especially when one remembers that witnesses when testifying are often apt to use loose language; sometimes the questions to which answers are given are obscure or double-barrelled; sometimes the witness is merely confused; although, of course, there are occasions when an inconsistency is such that the evidence should not be relied upon. The reliability of a witness's testimony is more often truly tested by inherent probabilities and improbabilities, a direction of cross-examination which is too seldom utilised precisely because it is too often considered that establishing an inconsistency is the be-all and end-all of cross-examination.” 23. Notwithstanding those observations, the fact remains that the appellant’s only defence was an outright denial of the prosecution evidence and an accusation that the prosecution witnesses were lying about the assault. It was therefore clearly relevant to that defence that inconsistencies in the evidence of PW1 and PW2 should be put to them to test the reliability of their recollection of events and their credibility. Although, if the appellant had been represented, a failure to put these matters to the prosecution witnesses would not necessarily have cast doubt on the safety of a conviction, in this case, the magistrate, being bound to assist the appellant as an unrepresented defendant, should fairly have put these matters to the prosecution witnesses and, to that extent, to determine if these were in fact material inconsistencies that might undermine the prosecution evidence. 24. Instead of putting these inconsistencies to the prosecution witnesses, though, (recalling PW1 after the testimony of PW2, as necessary) the magistrate’s assistance to the appellant in their cross-examination, in particular that of PW1, was lacklustre to say the least. Granted the appellant was not clear in his version of events, and the magistrate had to clarify this but, as was pointed out in the Printed Case for the Appellant,[14] in the course of the cross-examination of PW1, the magistrate asked only 10 questions of the witness but asked 60 questions of the appellant. 25. Whether the extent of the magistrate’s interventions crossed the line of impartiality is a matter of impression and the number of interruptions by a judge in the course of testimony cannot be decisive: see The Queen v Yeung Mau-lam;[15] HKSAR v Tam Chi Hang;[16] and HKSAR v Lai Oi Yan.[17] However, a consideration of the transcript as a whole, in particular the tenor as well as the number of questions put to him, leaves the distinct impression the magistrate was using the exercise of clarifying what the appellant wanted to put to PW1 and PW2 as an opportunity to challenge the reliability and credibility of his own version of events rather that of the prosecution witnesses. Some of the questions demonstrated that the magistrate was openly sceptical of what the appellant was saying and she appears to have proceeded in effect to cross-examine the appellant, occasionally in a hostile manner. 26. On appeal, the Deputy Judge was unimpressed with the appellant’s grounds of appeal that the magistrate had failed to assist him as an unrepresented defendant when cross-examining the prosecution witnesses and that the magistrate failed to adopt an attitude of impartiality when examining the evidence. In respect of the former ground of appeal, the Deputy Judge simply held: “I have read the transcript referred to by Mr. Cheung. Apparently, the appellant was incapable of cross-examining the prosecution witnesses by himself. Therefore, the magistrate must first understand the appellant’s stance before assisting him in putting the defence case to the witnesses. I disagree with Mr. Cheung’s view. The magistrate was not cross-examining the appellant, nor was she intervening his cross-examination. The magistrate’s approach did not prejudice the appellant.”[18] And in respect of the latter ground, the Deputy Judge held: “As regards Ground of appeal 4, I have read the relevant parts of the trial transcript. In my judgment, the allegations are not made out and they are unfair to the magistrate. I agree with the analysis and arguments made by Mr. Lau that the magistrate was only clarifying and confirming the appellant’s stance. The magistrate’s approach is proper.”[19] 27. Contrary to those conclusions, the transcript shows that the magistrate’s assistance to the appellant in the conduct of the cross-examination of the prosecution witnesses did prejudice him since she did not put the various inconsistencies in their evidence to those witnesses. In the context of this case, that omission was substantially to the appellant’s disadvantage. Similarly, even if it was arguable that the magistrate erred in failing to address or resolve the inconsistencies between the evidence of PW1 and PW2, the magistrate proceeded on the erroneous basis that PW1’s and PW2’s evidence was mutually consistent when in fact there were significant inconsistencies in that evidence. Furthermore, the appearance of a fair trial was put in doubt by the magistrate’s conduct of the questioning of the appellant when she was assisting him to cross-examine the prosecution witnesses, in particular PW1. D. Conclusion and disposition 28. For these reasons, I conclude that the appellant was denied a fair trial and that his appeal must be allowed, his conviction for assault quashed and his disbursements (in the agreed sum) be paid to him by the respondent. Mr Justice Cheung PJ: 29. I agree with the Reasons for Judgment provided by Mr Justice Fok PJ. Lord Neuberger of Abbotsbury NPJ: 30. I agree with the Reasons for Judgment provided by Mr Justice Fok PJ. Mr Eric TM Cheung, Solicitor Advocate and Mr Carter Chim, instructed by ONC Lawyers, on a pro bono basis, for the Appellant Mr Derek Lau SPP, of the Department of Justice, for the Respondent [1] FLCC 2580/2017 (Ms Debbie Ng, Magistrate). [2] HCMA 583/2017, [2019] HKCFI 1108 (Deputy High Court Judge Gary Lam). [3] FAMC 24/2019, [2020] HKCFA 20 (Ribeiro, Fok & Cheung PJJ). [4] In the agreed sum of HK$12,538.50. [5] HCMA 583/2017, [2019] HKCFI 1108, Judgment dated 30 April 2019 (“CFI Judgment”). [6] (2010) 13 HKCFAR 123 per Bokhary PJ at [1]. [7] [2012] 2 HKLRD 437 at [40]. [8] HCMA 40/1991, unrep., 27 February 1991, at pp.6-7. [9] (2007) 10 HKCFAR 148 at [17]. [10] HCMA 384/2007, unrep., 20 September 2007, at [11]-[14]. [11] (1999) 2 HKCFAR 539 per Bokhary PJ at p.543G. [12] Printed Case for the Appellant at [49]; Case for the Respondent at [58]. [13] HCMA 574/1996, unrep., 9 August 1996, at p.4. [14] At [29]. [15] [1991] 2 HKLR 468 per Power JA (as he then was) at p.473C-F. [16] CACC 475/2003, unrep., 21 July 2004, per Stock JA (as he then was) at [19]. [17] [2016] 3 HKLRD 273 per McWalters JA at [76]-[77]. [18] CFI Judgment at [36]. [19] Ibid. at [38]. 
Mr Justice Ribeiro PJ: 1. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Tang PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: 3. Section 103(1) of the Securities and Futures Ordinance[1] makes it an offence to issue advertisements, invitations or documents relating to investments in certain cases but there are various exemptions specified in section 103(3). This appeal concerns the proper construction of section 103(3)(k), specifically the ambit of the exemption it provides from the application of the offence-creating section. Does the exemption apply where an advertisement relates to a product which, as a matter of fact, is only disposed of or intended to be disposed of to professional investors? Or, does the advertisement have to actually state that the product is intended for professional investors before the exemption can apply? A. Section 103 of the SFO 4. Section 103(1) of the SFO creates an offence in the following terms: “(1) Subject to subsections (2), (3) and (5) to (9), a person commits an offence if he issues, or has in his possession for the purposes of issue, whether in Hong Kong or elsewhere, an advertisement, invitation or document which to his knowledge is or contains an invitation to the public – (a) to enter into or offer to enter into – (i) an agreement to acquire, dispose of, subscribe for or underwrite securities; or (ii) a regulated investment agreement or an agreement to acquire, dispose of, subscribe for or underwrite any other structured product; or (b) to acquire an interest in or participate in, or offer to acquire an interest in or participate in, a collective investment scheme, unless the issue is authorized by the Commission under section 105(1).” 5. A person who commits an offence under section 103(1) is liable on summary conviction to a fine at level 6[2] and to imprisonment for 6 months.[3] 6. Section 103(3) provides a list of exemptions from the application of section 103(1), which includes that in sub-section 103(3)(k), as follows: “(3) Subsection (1) does not apply to the issue, or the possession for the purposes of issue – … (k) of any advertisement, invitation or document made in respect of securities or structured products, or interests in any collective investment scheme, that are or are intended to be disposed of only to professional investors.” 7. The term “professional investors” is defined at some length in section 1 of Part 1 of Schedule 1 of the SFO and includes persons prescribed as being within the meaning of that definition in section 3 of the Securities and Futures (Professional Investor) Rules[4] made by the Securities and Futures Commission (“the Commission”)[5]. 8. It is common ground, and it is self-evident from the structure of the provisions, that the burden of proof of the material elements of the offence in section 103(1) rests on the prosecuting authority, whilst the burden of proof that the exemption in section 103(3)(k) applies rests on the person charged with the offence. The issue in the present case relates to what that person must show to pray in aid the exemption. B. The material facts 9. The 1st appellant is a company of which the 2nd appellant is the Chief Executive Officer and provides investment advice. On 2 November 2011, the 1st appellant sent an email prepared by the 2nd appellant to various recipients. The email announced the launch of a fund called the Pacific Sun Greater China Equities Fund (“the Fund”) and attached a press release to the same effect (exhibit P8 at trial). On 25 November 2011, the 1st appellant’s website published three documents relating to the Fund (exhibit P9 at trial). 10. It is not disputed that the Fund is a collective investment scheme[6], which was launched by the 1st appellant on the 2nd appellant’s instruction, and that exhibits P8 and P9 were advertisements for the Fund issued to the public or a sector of the public. It is also common ground that authorisation had not been obtained from the Commission in respect of the issue of the advertisements. 11. The key facts for the purposes of the present appeal are that: (1) as found by the Magistrate[7], the Fund was or was intended to be available solely to professional investors and was not one for investment by the general public; and (2) this fact is not expressly stated in the advertisements themselves. C. The course of the proceedings below 12. The email attaching exhibit P8 formed the basis of two charges[8] brought by the Commission against the appellants for breach of section 103(1) of the SFO. The publication on the 1st appellant’s website of the documents which were exhibit P9 formed the basis of two further charges[9] brought by the Commission under section 103(1). 13. After trial, the Magistrate acquitted the appellants on two bases. (1) First, based on disclaimers in the exhibits and subsequent screening procedures adopted by the 1st appellant to ensure that the Fund was available to professional investors only, he held that he was unable to conclude beyond reasonable doubt that the advertisement contained an invitation to the public to invest in the Fund. Instead, he found that it constituted an invitation to the public to seek further information from the appellants if required.[10] (2) Secondly, he found that the Fund was or was intended to be available solely to professional investors[11] and was not one for investment by the general public, so that the exemption in section 103(3)(k) applied.[12] 14. The Commission appealed by way of Case Stated against the Magistrate’s verdict and the Judge allowed the appeal, ruling that the Magistrate had erred in respect of both bases on which he had acquitted the appellants.[13] In this appeal, we are only concerned with the Judge’s holding that the Magistrate erred in respect of the applicability of the section 103(3)(k) exemption. In that regard, the Judge held that for the exemption in section 103(3)(k) to apply it was necessary for it be seen from the advertisement itself whether it was, by its terms, confined to professional investors to the exclusion of other members of the investing public. She also held that the carrying out of a screening process to ensure that all investors in the Fund were professional investors was irrelevant. 15. Her Ladyship remitted the matter to the Magistrate with her opinion on the questions raised in the Case Stated to determine whether to convict or acquit on the facts he found. On the remitter, the Magistrate has since convicted the appellants of the offences charged and sentenced the 1st appellant to a total fine of HK$20,000 and the 2nd appellant to two concurrent terms of 4 weeks’ imprisonment suspended for 12 months.[14] Those convictions are the subject of an appeal to the Court of First Instance by the appellants but that appeal has been stayed pending the outcome of this appeal.[15] D. The issue in this appeal and the parties’ rival contentions 16. As already stated, in this appeal, the Court is concerned only with the Judge’s holding that the exemption in section 103(3)(k) does not apply. The questions for which leave to appeal was granted[16] are: (1) Whether for section 103(3)(k) of the SFO to be applicable, it must be seen from the advertisement, invitation or document itself, that it is, by its terms, confined to professional investors to the exclusion of other members of the investing public; and (2) Whether the carrying out of a screening process to ensure that all investors investing in the collective investment scheme are professional investors is irrelevant. 17. The Commission’s case is that, for the exemption in section 103(3)(k) to apply, the advertisement must make it clear that the relevant advertised investment product is or is intended only for professional investors. It was the submission of Mr Jat Sew Tong SC[17], leading counsel for the Commission, that the advertisement has to state expressly that the investment product is or is intended to be disposed of only to professional investors. He submitted that any other construction would defeat the statutory purpose and would be inimical to the protection of the investing public intended by the statutory scheme. 18. For their part, the appellants contend that the exemption applies if, as a matter of fact, the relevant investment product is or is intended to be sold only to professional investors. Mr Laurence Li, counsel for the appellants, submitted the burden will be on the person issuing or possessing the advertisement to prove this fact, for example by proving that there is a screening process to exclude persons who are not professional investors. E. The proper construction of section 103(3)(k) 19. The issue raised and the parties’ respective submissions give rise to a short point of construction of section 103(3)(k). For the reasons that follow, I would reject the Commission’s construction of that section and, instead, accept the appellants’ construction. E.1 The natural and ordinary meaning of section 103(3)(k) 20. By reason of section 103(3)(k), section 103(1) does not apply to the issue or possession for the purposes of issue of an advertisement, invitation or document: “made in respect of securities or structured products, or interests in any collective investment scheme, that are or are intended to be disposed of only to professional investors.” 21. Since the other elements of section 103(3)(k) (namely (i) “the issue or possession for the purposes of issue of …”, (ii) “… any advertisement, invitation or document …” and (iii) “… made in respect of securities or structured products, or interests in any collective investment scheme …”) are already constituent parts of the offence created by section 103(1), it is the additional concept of disposal only to professional investors that provides the substance of the exemption created by section 103(3)(k). 22. It is clear, as a matter of grammar, that the concept of being disposed of or intended to be disposed of only to professional investors qualifies the securities, structured products or interests in a collective investment scheme that are being advertised rather than “the advertisement, invitation or document”. First, the use of the plural “are or are intended” must relate the concept of disposal to the (plural) “securities or structured products, or interests in any collective investment scheme” rather than the (singular) “advertisement, invitation or document”. Secondly, the entire clause “that are or are intended to be disposed of only to professional investors”[18] must qualify the “securities or structured products, or interests in any collective investment scheme” rather than “the advertisement, invitation or document” since it is only the former and not the latter which can be relevantly disposed of to anyone. 23. For the exemption to come into play, “the advertisement, invitation or document” must be “made in respect of” one of the relevant products. The words “in respect of” mean having a relation, connection, reference or regard to something[19]. In Cunard’s Trustees v Commissioners of Inland Revenue, they were described by Lord Greene MR as “colourless” [20]. Mann CJ in The Trustees Executors & Agency Co. Ltd. v Reilly[21] said of them: “The words ‘in respect of’ are difficult of definition, but they have the widest possible meaning of any expression intended to convey some connection or relation between the two subject-matters to which the words refer.” 24. The use of this wide expression to frame the exemption provided by section 103(3)(k) does not support the construction advanced by the Commission. The exemption simply applies to the issue of an advertisement “made in respect of [the relevant investment products] that are or are intended to be disposed of only to professional investors”; that is to say, having some connection or relation to the investment products disposed of only to professional investors. It is not drafted in terms, for example, of disapplying section 103(1) to the issue of an advertisement “expressed to be for”, or “containing a statement that it is for”, the exempted investment. E.2 Construing section 103(3)(k) in context 25. Since, however, the proper construction of the words “in respect of” will necessarily depend on their context[22], one must look to the context of their use in section 103(3)(k) to determine whether any conclusions can be drawn as to their legislative intent. 26. In this regard, when one construes section 103(3)(k) in its context, it is abundantly clear that, where the Legislature chose to express an exemption to section 103(1) in terms of express statements contained in the relevant advertisement, invitation or document, it said so in clear, express and unambiguous terms. 27. A clear example of this can be seen in section 103(2)(ga), which provides: “(2) Subsection (1) does not apply to the issue, or the possession for the purposes of issue, of any advertisement, invitation or document – … (ga) to the extent that the advertisement, invitation or document relates to an offer falling within paragraph (b)(ii) of the definition of prospectus in section 2(1) of the Companies (Winding Up and Miscellaneous Provisions) Ordinance (Cap.32)”. (Italics in original) 28. Section 2(1)(b)(ii) of Cap.32 provides that “prospectus”: “(b) does not include any prospectus, notice, circular, brochure, advertisement, or other document – … (ii) to the extent that it contains or relates to an offer specified in Part 1 of the Seventeenth Schedule as read with the other Parts of that Schedule”. 29. The Seventeenth Schedule is headed “Offers Specified for the Purposes of Paragraph (b)(ii) of the Definition of Prospectus in Section 2(1) of this Ordinance” and Part 1 of that schedule contains a list of 12 offers not falling within the definition. Paragraph 2 in Part 1 of the Seventeenth Schedule lists: “An offer – (a) to not more than 50 persons; and (b) containing a statement specified in Part 3 of the Eighteenth Schedule to this Ordinance.”[23] 30. Part 3 of the Eighteenth Schedule is headed “Statement to Be Contained in Certain Offers Specified in Part 1 of the Seventeenth Schedule” and provides: “A statement, in a prominent position, if in the English language, in the following form or a form to the like effect – ‘WARNING The contents of this document have not been reviewed by any regulatory authority in Hong Kong. You are advised to exercise caution in relation to the offer. If you are in any doubt about any of the contents of this document, you should obtain independent professional advice.’; and, if in the Chinese language, a statement in the following form or a form to the like effect – ‘警告 本文件的內容未經在香港的規管當局審核。你應就有關要約謹慎行事。如你對本文件的任何內容有任何疑問，你應尋求獨立專業意見。’.” 31. Thus, when an advertisement, invitation or document must contain a particular express statement in order for an exemption to the application of section 103(1) to apply, the Legislature has clearly and expressly stated this and identified the substance of the particular form of words to be used. 32. In the face of this compelling contextual argument, Mr Jat submitted, first, that Cap.32 is directed to offers made to certain persons and so expressly regulates the wording of such offers to those persons whereas section 103(3) is dealing with advertisements, invitations or documents seeking to elicit investment interest from their recipients. With respect, this is a distinction without any difference. Section 103(1) is directed to advertisements, invitations or documents relating to investments in certain cases. There would not appear to be any good reason for recipients of offers to be afforded greater protection by the legislation than recipients of advertisements, invitations or documents relating to investment products. 33. Mr Jat’s second submission in answer to this contextual argument was that there is no defined category of situations in sections 103(2) and (3); rather, the two sub-sections contain a collection of discrete exemptions and there is no rationale for thinking that the same requirements apply to different exemptions. However, while it is correct that the various situations in sections 103(2) and (3) disapplying section 103(1) constitute a disparate list of exemptions, Mr Jat’s argument does not answer the more significant fact that, within the same division of the same part of the same ordinance, where an exemption to the application of an offence-creating section is provided for, insofar as the exemption depends on the inclusion in a document of a particular form of words, the legislation expressly states the fact that particular wording must be used and specifies the substance of that wording. The absence of this in section 103(3)(k) points strongly to the conclusion that, as a matter of context, the Commission’s construction of that exemption is erroneous. E.3 The purposive construction of section 103(3)(k) 34. There was no dispute that the modern approach to statutory construction is to adopt a purposive interpretation and that the language of a statutory provision is to be construed having regard to its context and purpose[24]. It is this principle of construction that the Commission primarily relies upon in support of its construction of section 103(3)(k). 35. Section 103 is to be found in Division 2 of Part IV of the SFO. Part IV of the SFO is headed “Offers of Investments” and Division 2 of Part IV carries the sub-heading “Regulation of offers of investments, etc.” Section 103 itself is headed “Offence to issue advertisements, invitations or documents relating to investments in certain cases”. 36. Broadly, section 103(1) regulates the issue of advertisements containing an invitation to the public to acquire certain investments. It is that invitation which constitutes the prescribed content which section 103(1) seeks to regulate and the system of regulation thereunder is subject to the overarching authority of the Commission to authorise the issue of an advertisement containing such an invitation under section 105(1). However, the system of regulation in section 103(1) does not stand alone and must be read in the context of the opening words which expressly make section 103(1) subject to sub-sections (2), (3) and (5) to (9). Thus, for example, section 103(2) provides that: “Subsection (1) does not apply to the issue, or the possession for the purposes of issue, of any advertisement, invitation or document –” falling within the terms of any of paragraphs (a) to (i) under section 103(2) and section 103(3) provides a further list of circumstances (in paragraphs (a) to (k)) in which section 103(1) does not apply. 37. It was the Commission’s contention that, in the light of its regulatory objects and functions[25], the key object of the SFO is to ensure that the interests of the investing public are protected. In this context, it was submitted that the purpose of section 103(1) is to regulate the issue of advertisements that contain the prescribed content (set out in that section) in relation to certain investment products unless authorised by the Commission. Thus, it was submitted, since it is not an offence to launch or sell a collective investment scheme that has not been authorised by the Commission, the Legislature has chosen to protect the investing public by regulating the advertising of the investment products as distinct from their subsequent sale. 38. Mr Jat submitted that, since section 103(1) was designed to protect the investing public from being exposed to unauthorised advertisements with a particular type of content, it was at the point of issue of the advertisement that the protection was required, as distinct from and prior to any actual sale activity. This was also consistent, it was submitted, with the Commission’s duty to exercise regulatory oversight, by way of the grant of authorisation for the issue of an advertisement, to ensure that the contents of advertisements are accurate and not misleading so far as practicable. As a development of this argument, it was submitted that criminal liability under section 103(1) attaches at the point in time when an advertisement is issued, or when a person has such an advertisement in his possession for issue, and if the content of the advertisement contains the invitation prescribed by section 103(1)(a) or (b). It does not, it was submitted, depend on whether any investment product was actually sold subsequently to the issue of the advertisement. 39. It was therefore submitted that it is only by construing section 103(3)(k) in the manner contended by the Commission that effect could be given to the legislative purpose of protecting the public from exposure to the prescribed type of unauthorised advertisements, which by definition would not previously have been vetted by the Commission, and in ensuring that the public is protected before the actual sale of the investment products. 40. An obvious flaw in the Commission’s argument is that if the investment products are not in fact sold or intended to be sold to the general public and instead are sold or intended to be sold only to professional investors, there is no necessity for protection to be afforded to the general public since they are not exposed to any material risk. It was accepted by the Commission that the purpose of the exemption in section 103(3)(k) is clear, namely that professional investors, as opposed to the general investing public, do not require statutory protection under section 103(1) so that advertisements of products intended only for them are exempted from the prohibition. 41. Mr Jat sought to side-step the force of this point by referring to the advertisements in this case and drawing attention to the fact that one of the documents in exhibit P9 indicated a minimum investment of US$100,000 and that the presentation slides in that exhibit did not indicate the Fund was only to be disposed of to professional investors, merely containing the statement “preference for professional investors reduces trading volatility”. Since the prescribed portfolio threshold for professional investors is HK$8 million[26], the minimum investment amount was well within that threshold and therefore non-professional, or retail, investors might think the Fund was intended for disposal to them. 42. The thrust of the Commission’s argument in this respect would appear to be that, because the advertisements did not expressly state that the Fund was or was intended to be disposed of only to professional investors, retail investors might have their interest piqued and, relying on the minimum investment amount, might wrongly think the Fund was intended for them. But, if that were so, a retail investor would soon be disabused of this misapprehension upon his expressing an interest in the Fund to the appellants. Once telephone or other inquiries informed a retail investor that the Fund was not intended for disposal to him, there would be no interest of the retail investor for the statutory regime to protect, save possibly from a waste of the time necessary to discover that the Fund was only for professional investors. There is nothing to suggest the legislative purpose of section 103(3)(k) is to protect retail investors from wasting their time or, alternatively, against having their investment appetites whetted. 43. What appears to lie behind the Commission’s contention that retail investors might think the Fund was intended for them is a complaint about the Magistrate’s finding of fact that the Fund was or was intended only for professional investors and not for the general public. However, although it is a fair point that the presence of express wording in the advertisement might go towards satisfying the burden of establishing that the exemption applies, the opposite by no means follows, and, in any event, this complaint is not open to the Commission on this appeal. 44. A further weakness in the Commission’s purposive construction argument is that, as Mr Jat rightly accepted, the exemption cannot properly be claimed merely because an advertisement states that a particular investment product is intended only for professional investors. The exemption goes to the substance of the investment and it is therefore necessary for a person claiming its benefit to demonstrate that the relevant investment is in fact intended solely for professional investors. It is the demonstration of this fact that shows that a retail investor is protected against exposure to an unsuitable investment product. Once it is accepted that the substance of the question of whether the investment product is or is not intended for professional investors has to be investigated in order to invoke the exemption, the statutory purpose of investor protection is achieved, without reference to the express wording of the advertisement in question. 45. As to Mr Jat’s point in time argument, namely that criminal liability must attach at the time of the issue of the advertisement, an acceptance of the appellants’ construction argument does not necessarily result in criminal liability attaching at a subsequent time. The burden will be on the issuer of the advertisement to establish that the investment product is intended only for professional investors. If, as a matter of fact, the product has already been disposed of to a non-professional investor, the offence will have been committed at the point in time of the issue of the advertisement. Similarly, if the issuer of the advertisement is unable to establish that the product is intended to be disposed of only to professional investors, the offence will likewise have been committed at the time of the issue of the advertisement. 46. It was further contended in the Commission’s printed case that the appellants’ construction sought to inject an unwarranted additional subject matter into section 103(3)(k), requiring separate proof that the investment product is or is intended to be disposed of only to professional investors and thus requiring an additional element of mens rea. However, the answer to this contention is that the burden of proving the fact of disposal or intention to dispose only to professional investors rests with the issuer of the advertisement so that the Commission would not have any additional prosecutorial burden. Insofar as an issuer of an advertisement acted through the medium of a public relations or marketing company, the latter parties might potentially be liable under section 103(1) but would, if they satisfied the relevant conditions, be entitled to claim the protection of the exemptions provided in sections 103(7) or 103(8) and so the Commission’s contention that a further party’s mens rea might need to be examined is already inherent in the statutory scheme. 47. For the above reasons, I do not accept the Commission’s contention that the purpose of section 103(3)(k) would be frustrated by the appellant’s construction of the exemption. On the contrary, to accept the Commission’s construction of the exemption would require the Court “to attribute to a statutory provision a meaning which the language of the statute, understood in the light of its context and the statutory purpose, is incapable of bearing”[27]. The statutory purpose of investor protection is achieved on the appellants’ construction since the burden remains on the issuer of the advertisement to demonstrate that the investment product is or is intended for disposal only to professional investors. E.4 Penalisation under a doubtful provision 48. Where a statutory provision is ambiguous, unclear or open to two reasonable interpretations, its penal effect may indicate a narrower construction[28]. Since, however, I have concluded that, on its proper construction, the exemption in section 103(3)(k) clearly applies to an advertisement of a relevant investment product that, as a matter of fact, is or is intended to be disposed of only to professional investors whether or not this fact is actually stated in the advertisement itself, it is not necessary for the appellants to invoke this principle in support of their appeal. 49. Had the proper construction of section 103(3)(k) not been clear, there would have been some scope for the application of the principle against penalisation under a doubtful law. In this regard it is relevant that, whether tried summarily or on indictment, the offence may result in a term of imprisonment and, in the present case, the 2nd appellant has been sentenced on conviction to a term of imprisonment, albeit suspended for 12 months. 50. In the respondent’s printed Case[29], it was contended that a reason not to apply the principle in the present case is that the appellants’ construction of section 103(3)(k) is less favourable to an advertiser since, on the respondent’s construction, the requirement to prove to whom the investment products are actually sold or intended to be sold is an additional requirement. This contention, however, is inconsistent with and negated by the respondent’s acceptance of the need to have regard, on its construction, to the substance of a claim in an advertisement that the investment product is or is intended to be disposed of only to professional investors in order for the exemption to apply[30]. E.5 The Judge’s construction of section 103(3)(k) 51. The two questions in the Case Stated which related to section 103(3)(k) were: “(iii) For s.103(3)(k) to be applicable, as a matter of statutory interpretation, does the advertisement, invitation or document itself need to contain a reference to the fact that the securities or structured products, or interests in any collective investment scheme being advertised or promoted are, or are intended to be disposed of only to professional investors? (iv) Given the finding that the Fund launched by [PSA Ltd] was or was intended to be available solely to professional investors and was not one for investment by the general public, did I err in law in finding that s.103(3)(k) applied to exhibits P8 and P9?” 52. The Judge’s analysis of these questions was brief. She said: “30. The SFC has a duty, to be exercised so as effectively to protect the investing public, in relation to such advertisement, invitation or document. Except possibly in wholly exceptional circumstances, and it is plain that no such circumstances exist in the present case, the SFC, in order to be able to exercise this duty effectively, has to be able to see from the advertisement, invitation or document itself whether it is, by its terms, confined to professional investors to the exclusion of other members of the investing public. Section103(3)(k) is concerned with the advertisement, invitation or document itself and not with what may or may not be the arrangement or intended arrangement behind it. 31. At best for PSA Ltd and Mr Mantel, his evidence that he would carry out a screening process to ensure that all investors in the fund were professional investors is irrelevant. If anything, the idea of such a screening process suggests that the advertisement will result in some offers to acquire an interest in or participate in the fund coming from non-professional investors who would then have to be screened out.” 53. The answers the Judge therefore gave to the two questions were: “32. … (E) Answer to Question (iii): Subject to the possibility of a different answer in wholly exceptional circumstances (which certainly do not exist in the present case), the answer to this question is that it has to be apparent from the advertisement, invitation or document itself that it is confined to professional investors. (F) Answer to Question (iv): Yes.” 54. The Judge would appear simply to have accepted, although somewhat equivocally, the respondent’s construction of section 103(3)(k) on the basis that retail investors require protection against having their investment appetites whetted or in wasting their time in pursuing an interest in investing only to be told they are not eligible to do so. With respect, for the reasons already stated, this is not the statutory purpose of section 103(3)(k). Insofar as, additionally, the Judge seems to have accepted the respondent’s construction for prosecutorial ease and convenience, this is not a good reason for supporting the respondent’s construction. On the contrary, it is a clear example of the vice referred to by Lord Millett of a court distorting or ignoring the plain meaning of the text of a statute in order to achieve a perceived desirable result[31]. F. Section 103(10)(a) and (b) 55. As indicated above, this appeal is only concerned with the application of the exemption in section 103(3)(k). We have not examined the other basis on which the Magistrate acquitted the appellants, which the Judge held to have been erroneous. The Magistrate held that because of the disclaimers in them and the screening procedures adopted by the 1st appellant, he could not conclude that the advertisements contained an invitation to the public to invest in the Fund. The Judge disagreed and held that the deeming provision in section 103(10)(a)[32] applied. Since the proper scope of section 103(10)(a) has not been the subject of any argument before us, nothing in this judgment should be taken to indicate support for, or disagreement with, the Judge’s interpretation of that provision. G. Disposition 56. For these reasons, I would allow the appellants’ appeal, set aside the Judge’s answers to Questions (iii) and (iv) in the Case Stated and substitute the answers “No” to each question. It follows also that I would set aside the Judge’s order remitting the matter to the Magistrate to continue the case since his original decision acquitting the appellants was correct and should not have been reversed. The Magistrate’s subsequent convictions following the remitter are under appeal and although the disposition of that appeal is not a matter before us, its outcome will now follow this judgment. Finally, I would direct that any submissions as to costs be lodged with the Registrar in writing within 14 days from the date of this judgment. Mr Justice Stock NPJ: 57. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Gleeson NPJ: 58. I agree with the judgment of Mr Justice Fok PJ. Mr Jat Sew-Tong SC and Mr Derek C.L. Chan, instructed by the Securities and Futures Commission, for the Respondent Mr Laurence Li, instructed by Timothy Loh Solicitors, for the 1st and 2nd Appellants [1] (Cap.571) (“the SFO”) [2] i.e. a maximum fine of HK$100,000 [3] Section 103(4)(b). Conviction on indictment gives rise to a liability to a fine of HK$500,000 and to imprisonment for 3 years: section 103(4)(a). [4] (Cap.571D) (“the PI Rules”) [5] Respondent in this appeal [6] As defined in Schedule 1, Part 1, s.1 of the SFO [7] Mr Joseph To Ho-shing [8] ESS 30881/2012 & 30882/2012 [9] ESS 30883/2012 & 30884/2012 [10] ESS 30881-84/2012, Statement of Findings, 21 March 2013 (“SOF”) §51 [11] As defined in Schedule 1, Part 1, s.1 of the SFO [12] SOF §52 [13] HCMA 490/2013, V. Bokhary J, Judgment dated 24 January 2014 [14] ESS 30881-84/2012, Statement of Findings, 11 July 2014 [15] By direction of the Appeal Committee in its Determination granting leave to appeal (§5) [16] FAMC 39/2014, Determination, 26 September 2014 [17] Appearing with Mr Derek C.L. Chan [18] Prior to the enactment of the Securities and Futures Companies Legislation (Structured Products Amendment) Ordinance 2011 section 103(3)(k) was identical to its present wording save that the word “that” before “are or are intended to be disposed of …” read “which”. That particular change of wording made no material difference to the section. [19] Shorter Oxford English Dictionary (6th Ed.) Vol.2 at p.2549 [20] (1946) 174 L.T. 133 at p.136; see also Saab v Saudi American Bank [1999] 1 WLR 1861 at §24 where they were similarly described by Clarke LJ (now Lord Clarke of Stone-cum-Ebony NPJ). [21] [1941] VLR 110 at 111; this view is reflected also in the decision of the Supreme Court of Canada in R v Nowegijick [1983] SCJ No.5 at p.7, followed by the High Court of Australia in Smith v Federal Commissioner of Taxation (1987) 164 CLR 513 at §21. [22] See: R(on the application of Geologistics Ltd) v Financial Services Compensation Scheme [2004] 3 All ER 39 at §17 per Waller LJ; Saab v Saudi American Bank [1999] 1 WLR 1861 at §24; similarly, Technical Products Pty Ltd v State Government Insurance Office (1989) 85 ALR 173 at 175 where the High Court of Australia described the words as having “a chameleon-like quality in that they commonly reflect the context in which they appear”. [23] The reference to an offer “containing a statement specified in Part 3 of the Eighteenth Schedule to this Ordinance” is repeated in paragraphs 3, 4, 7, 8, 9, 10 and 11 of Part 1 of the Seventeenth Schedule. [24] HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at §§11-12; Leung Chun Ying v Ho Chun Yan Albert (2013) 16 HKCFAR 735 at §12 [25] Set out, for example, in sections 4 and 5 of the SFO [26] PI Rules, ss.3(b) & 3(c) [27] China Field Ltd v Appeal Tribunal (Buildings) (No.2) (2009) 12 HKCFAR 342 per Lord Millett NPJ at §36 [28] Baker v Quantum Clothing Group Ltd [2011] 1 WLR 1003 per Lord Mance JSC at §46, citing Franklin v Gramophone Co Ltd [1948] 1 KB 542 at p.557; see also, Bennion on Statutory Interpretation (6th Ed.), Section 271 (p.749) stating the principle against penalisation under a doubtful law. [29] Respondent’s Case, §4.23 [30] Respondent’s Case, §4.6 [31] China Field Ltd v Appeal Tribunal (Buildings)(No.2), supra., at §36 [32] Section 103(10)(a) provides: “For the purposes of any proceedings under this section – (a) an advertisement, invitation or document which consists of or contains information likely to lead, directly or indirectly, to the doing of any act referred to in subsection (1)(a) or (b) shall be regarded as an advertisement, invitation or document (as the case may be) which is or contains an invitation to do such act;”. 
Mr Justice Chan PJ: 1. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Ribeiro PJ: 2. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Tang PJ: 3. The appellant pleaded guilty to two offences of possession of dangerous drugs. The first (charge 16) involved possession of a total of 1.46 grammes of a powder containing 0.85 grammes of cocaine inside 3 plastic bags at the Diesel’s Bar on Lamma Island. The second (charge 17) involved 1.05 grammes of cannabis resin, 9.94 grammes of a solid containing 5.51 grammes of cocaine and 0.25 grammes of a powder containing 0.12 grammes of cocaine found at his home. 4. Deputy District Judge Dufton regarded 6 months imprisonment as appropriate for the first offence and 12 months for the second. However, he said: “9. … The cocaine being found in a total of five plastic bags, four of which contained small quantities of cocaine together with the fact the defendant had taken three of those bags with him to the bar, there was in my view a risk some of the cocaine may fall into the hands of others.[1] I therefore increase the starting points by 3 months to 9 months and 15 months imprisonment.” 5. After giving the appellant full credit for his pleas of guilty the Judge sentenced him to six months imprisonment on charge 16 and 10 months imprisonment on charge 17, to be served concurrently. 6. The appellant argued on appeal that the latent risk sentencing principle is unconstitutional since it imputes to a person convicted of a possession offence an unproven intent, predilection or propensity to commit the more serious offence of trafficking contrary to s4(1)(a) and (3) of the Dangerous Drugs Ordinance, Cap 134. It is said that that contravenes the presumption of innocence protected in Article 87 of the Basic Law and Article 11(1) of the Hong Kong Bill of Rights Ordinance, Cap 383. 7. His appeal was dismissed by the Court of Appeal.[2] The Appeal Committee has granted leave to appeal on a certified point of law namely “to what (if any) extent a sentencing court should take into account the risk that the dangerous drugs the subject-matter of a charge of simple possession will find their way into hands of others, and how the court should approach the question of any such risk? ” 8. Mr Dykes SC for the appellant accepts that the court may have regard to aggravating factors. One such aggravating factor, he accepts, is, leaving drugs in a place where others have access to them. Mr Dykes also accepts that taking drugs to a pub is an aggravating factor, he says because consumption of drugs in public may encourage drug taking. He contends, however, that it is impermissible to have to regard to the risk of dissemination[3]. That is not because the risk is non-existent or unreal, but, because, to do so, one would be punishing the possessor for his “present conditional intention” to traffic in dangerous drugs. That, Mr Dykes submits, is tantamount to punishing him for an offence (trafficking in dangerous drugs) for which he has not been charged. 9. One of the first cases in which the latent risk factor in drug offences was considered is R v Chiu Hung-wong And Another [1994] 1 HKCLR 184, an appeal from District Judge Lugar-Mawson, as he then was. Silke VP who gave the judgment of the Court approved the following statement by Judge Lugar-Mawason: “That is not to say however that a Judge should close his eyes to the obvious dangers and temptations that the possession of a large quantity of drugs in the hands of an addict may present both to himself and society. It is known and I would be ignoring over 12 years experience in criminal practice in Hong Kong if I ignored it, that addicts are persons who are easily tempted and who do sell their stock of dangerous drugs to others in order to fund not only their own addiction, but also their daily living expenses. If their stock is small the risk of sale is slight, if it is large the risk is greater. Formerly the law recognised this regrettable feature of human behaviour by providing an evidential presumption based on weight or packaging that went to determine the offender’s guilt. Such a presumption, going as it did to establish guilt, was unsustainable under the Bill of Rights Ordinance. That is not to say however that the Judge when it comes to sentencing, where the question of guilt or innocence have been disposed of, should not have regard to this feature and recognise the potential harm to society that possession of large quantities of dangerous drugs poses and take the weight of the dangerous drugs possessed by the offender into account when determining the length of his sentence.” 10. Silke VP went on to say there is “no offence known to law as ‘unproven trafficking’” and added: “The latent risk of danger to the public was substantial. But that having been said the applicant should not be sentenced as if the charge had been brought under section 4.” 11. In R v Lee Siu-lung [1995] 2 HKCLR 247, an appeal against sentence was allowed because the Judge, after saying that the applicant had said that the drugs in question were for his personal consumption, added: “I have no doubt that this is true for some of the drugs seized.” (my emphasis) Power VP said on behalf of the Court: “… It is one thing for a judge, when sentencing, to say that he accepts that the accused had drugs in his possession for his own consumption but takes into account the damage to the community posed by the possession of that quantity of drugs in the unstable hands of an addict, even one who intends them for his own consumption. It is quite another for a judge to say that he does not accept that the accused had possession of all of the drugs for his own consumption and to sentence him upon that basis. The latter approach would only be permissible upon a plea to a charge of trafficking.” 12. In HKSAR v Mok Cho Tik [2001] 1 HKC 261, on the subject of latent risk, Stock JA (as he then was) said: “Sentencing is an art, and we emphasize that the starting points and the degrees of enhancement for the risk factor must bend to the circumstances of each case. We remain of the view that the starting point for mere possession of a quantity of dangerous drugs which a bona fide user would normally have in his possession should be in the range of 12 to 18 months, (and that is the range that has been adopted particularly for cases in the District and High Court). But that is not the be all and end all of the appropriate starting point. It assumes cases in which a sentence of imprisonment rather than a rehabilitative measure is deemed necessary; and it does not seek to exclude magistrates from imposing lesser terms for very small quantities where the circumstances so warrant. Obviously the quantity will be the main governing factor which will determine where the starting point will lie. If an offender is a persistent offender he can expect the starting point to be higher than it otherwise would be. The existence of the risk factor and the degree of risk is not a matter of mathematics upon which this court should provide a tariff. Amongst the obvious factors to be taken into account are whether the offender is in employment; whether the drugs are kept in a place to which others have access; whether the offender has convictions for trafficking; and of course the quantity of drugs. The full circumstances of the risk must be assessed.” 13. In HKSAR v Wan Sheung Sum [2000] 1 HKLRD 405, the Court of Appeal suggested a three-step approach to sentencing for an offender who is a bona fide user and is convicted of simple possession of a dangerous drug. The three steps are: “Step 1: The judge should normally determine a starting point of between one year to eighteen months’ imprisonment. Step 2: The starting point is increased or enhanced to take account of the latent risk factor in order to arrive at a total sentence. This will reflect the risk to society of the drugs being redistributed and finding their way into other hands apart from the offender’s. The latent risk will be determined from all the circumstances in any particular case, including of course the quantity of drugs possessed and the personal circumstances of the offender. Step 3: The total sentence is then adjusted to take account of the mitigating factors, such as a guilty plea.” 14. These authorities show clearly an awareness that whilst a sentencing court may take into account the risk that the drug will fall into the hands of others, the possessor must not be sentenced as if he had been convicted of trafficking. 15. Here, Fok JA said: “28. … there can be no objection to a sentencing court taking into account the relevant circumstances of the case in determining whether the possession of the drugs leading to the conviction are such as to give rise to a real risk that some of those drugs might end up being redistributed and finding their way into the hands of others apart from the offender’s. That is not to say that the court then attributes to the defendant an intention to traffic in the drugs but simply reflects the fact that the quantity and circumstances of their possession are such as to pose a risk to society.” 16. With respect, I agree. The assessment of the risk of dissemination depends on the drawing of the correct inference from the circumstances of the possession as established on the evidence. Before any such inference is drawn, the sentencing court must be sure that in all the circumstances of the particular case, the risk of dissemination is real. The sentencing judge is best placed to decide whether there is such a real risk. Often times, it is just a matter of common sense. 17. The sentencing judge must also bear in mind Fok JA’s reminder that: “31. …… (1) if a judge is minded to enhance sentence on the basis of risk of dissemination, he must forewarn counsel for the accused, to enable the accused to challenge the issue, if necessary by the giving of evidence; (2) a judge is not to sentence for unproved trafficking, namely an unproved actual intention to traffic.” 18. The authorities discussed above show that the court’s approach to latent risk does not in anyway contravene the presumption of innocence. As Fok JA explained: “29. … It is not a question of punishing a person for a crime he has not committed. It is not a question of punishing him for a crime he may commit. It is a question of punishing him for the crime which he has committed taking into account the circumstances of its commission and the dangers to society which those circumstances create.” 19. I would dismiss the appeal. Mr Justice Hartmann NPJ: 20. I agree with the Judgment of Mr Justice Tang PJ. Lord Clarke of Stone-cum-EbonyNPJ: 21. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Chan PJ: 22. The Court unanimously dismisses the appeal. Mr P J Dykes SC & Mr G Surman, instructed by Vidler & Co, for the appellant Mr Kevin P Zervos SC, DPP & Mr Ira Lui, SPP of Department of Justice,for the respondent [1] This is often referred to in the authorities as the latent risk. [2] Stock VP, Fok JA and Line J. Fok JA delivered the Reasons of the Court. [3] For example, the possessor being prevailed upon by his friends to share what he has. 
Mr Justice Chan PJ: 1. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Ribeiro PJ: 2. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Tang PJ: 3. In 1998, the plaintiff, a married man, met the defendant, an unmarried masseuse, at a massage establishment in Jordan. Their relationship between late 2004 and October 2007 was described in the statement of claim as “romantic”[1]. 4. The plaintiff claims that between January 2005 and October 2007 he lent a total of HK$3,350,701 to the defendant, of which he said the defendant repaid HK$153,000. This action was commenced in January 2008 for the balance of HK$3,197,701. 5. The defendant’s case is that the plaintiff gave her money as gifts on various occasions between early 2005 and September 2007, and during the same period, the plaintiff on various occasions asked the defendant to allow him to deposit his “private monies” into her bank account. Part of these monies were returned to the plaintiff and part were used with the plaintiff’s consent as the defendant’s pocket money. The defendant also said that plaintiff asked her to give up her job and that he would look after her[2]. 6. At trial, the plaintiff relied on his own evidence[3], an Acknowledgment of Debt dated 31 July 2007 and certain taped conversations[4] with the defendant. After trial, on 30 September 2010 Chung J dismissed the claim. 7. The learned trial judge was “not fully satisfied with the evidence of either the plaintiff and the defendant” and took the view that the monies paid by the plaintiff to the defendant might have been:- “ (a) gifts; or (b) joint investment and/or asset; or (c) monies for temporary custody and/or loans, and not necessarily be loans.” 8. He dismissed the plaintiff’s claim because the plaintiff had failed to prove how much of these payments[5] fell into which of the above categories. 9. In the Court of Appeal[6], the plaintiff relied on Seldon v Davidson [1968] 1WLR 1083, a decision of the English Court of Appeal, and argued that in the circumstances of the present case, once payment was proved, it was prima facie repayable on demand, and the burden was on the defendant to show otherwise. He also contended that the trial judge ought to have given effect to the Acknowledgement of Debt and that on the evidence (including the taped conversations) he should have found that the payments were loans made by him to the defendant. The Court of Appeal dismissed his appeal. 10. The plaintiff appealed to us as of right. Although he was represented at trial as well as on appeal, he appeared before us in person.[7] He relies on 3 grounds of appeal: (1) the Court of Appeal erred in finding that the principles in Seldon v Davidson did not apply to this case and that presumption of advancement applied in favour of the defendant. (2) the Court of Appeal erred in giving no weight to the Acknowledgment of Debt dated 31 July 2007. (3) the trial judge had not correctly or fairly treated the evidence, and he had not received a fair trial. The Court of Appeal had failed to correct such mistakes. Seldon v Davidson 11. On Seldon v Davidson, the Court of Appeal said: “53. The facts in the case of Seldon v Davidson show that while a defendant admits having received money from a plaintiff in the absence of any circumstances giving rise to any presumption of advancement, the defendant has the onus of proving that the money is not repayable. However, the above principle is applicable only in the absence of presumption of advancement or anything of that sort.” 12. The plaintiff submits, rightly in my view, that his relationship with the defendant would not give rise to any presumption of advancement. However, I do not agree with his submission that the Court of Appeal had decided against him because they thought that the defendant could rely on a presumption of advancement. 13. In Seldon v Davidson, the plaintiff claimed the return of £1550 as money lent to the defendant who was employed by her as a chauffeur and handyman. The defendant had admitted the receipt of the money but claimed that the money was a gift. The County Court Judge ruled that in such circumstances it was for the defendant to begin. The defendant appealed and the ruling was upheld by the Court of Appeal. Willmer LJ said at 1088C that it was admitted on the defendant’s behalf that: “… the house which had been bought with the aid of the money paid by the plaintiff was no doubt prima facie subject to a resulting trust in favour of the plaintiff. That being so, it would be strange indeed if the same considerations did not apply to the money paid by the plaintiff to the defendant to assist him in the purchase of the house.” 14. Here we are not concerned with a resulting trust. It is clear that on the plaintiff’s case the defendant was free to use the money lent although she was obliged to repay on demand. 15. Edmund Davies LJ (as he then was) agreed with Willmer LJ. He pointed out at 1089G: “ The important fact in the present case is that there is no blood relationship of any kind between the plaintiff and the defendant.” (my emphasis)[8] His lordship asked at 1090F: “…what is to be inferred as to the nature of the transaction when the simple payment of money is proved or admitted between strangers.” He went on to say that he agreed with Willmer LJ that: “…on that bald state of affairs, proof of payment imports a prima facie obligation to repay the advancement in the absence of circumstances from which presumption of advancement can or may arise.” 16. I believe the proper inference to draw depends on the circumstances of the particular case and not on who has the burden of proof.[9] 17. In Mak Ka Hing v Pang Ming Chung, [2011] 1 HKLRD 347, Le Pichon JA[10] said at para 23: “A defendant seeking to evade repayment of the money which was paid to him had the onus of proving the facts which he alleges show that the money was not repayable.”[11] 18. Here, we are not concerned with transactions between strangers. The plaintiff and the defendant had an intimate relationship between late 2004 and late 2007. The payments that are the subject of the plaintiff’s claim were made during this period. The plaintiff relied on a total of 178 alleged loans in his statement of claim. The payments commenced on 17 January 2005 with a cash payment of $5,000 and ended on 15 September 2007 with a cash payment of $4,000. The first 116 of these loans were made between 17 January 2005 and 21 November 2006. They were all made in cash, 84 of which were for $5,000 or less. Then between 14 December 2006 and 8 August 2007, there were a total of 13 remittances or bank transfers ranging from $2,000 to $500,000. Between 31 December 2006 and 15 September 2007, there were a total of 49 cash payments, of which 21 were for $5,000 or less. 19. The Court of Appeal said: “36. …The ways in which those payments were made were not consistent at all with Mr Lui’s allegation that they were loans, and also contrary to the content of the Letter.”[12] 20. The Court of Appeal added: “55. Mr Lui’s allegation that Ms Chui has, over a period of about two and a half years, borrowed money from him for 140-150 times (several times a month) in amounts ranging from several thousand to several ten thousand dollars and it had been made clear that the monies were to be repaid in full upon Mr Lui’s demand, is not credible at all in light of the relationship between the two parties at the material time. In view of Mr Lui’s financial ability and his sentimental attachment to Ms Chui, it was extremely possible that he, as she alleged, was willing to ‘sacrifice’ so that her ‘days can be happy’, and thus gave monetary gifts to her. Mr Lui was not honest when he said that between the two parties there was a common understanding by which the monies were to be repaid in full by Ms Chui. The trial judge was definitely justified to reject it and found that Mr Lui was not an honest or reliable witness.” 21. In the circumstances of this case, the proper inference to draw is that the payments were gifts. Certainly so, in the case of the cash payments. But, the plaintiff did not distinguish between the cash payments and the remittances. His case was that they were all loans. Given his relationship with the defendant, it is incredible that he was not supporting the defendant. Moreover, although he had denied ever making any cash gift to the defendant, it is clear from his witness statement that he had made expensive gifts to the defendant.[13] 22. Moreover, as the Court of Appeal said: “37. Mr Lui was financially much more capable than Ms Chui, and in fact, Mr Lui agreed that Ms Chui had from time to time indicated to him that she did not have sufficient income to cover the expenses. Under such circumstances, the suggestion that Mr Lui and Ms Chui had a common understanding which required Ms Chui to make full repayment of the financial assistance Mr Lui provided to her is still not convincing.” 23. I believe the proper inference is that all the payments were gifts. 24. I turn to the letter the plaintiff wrote to the defendant on 3 March 2008, in relation to which the Court of Appeal said: “4. Although Mr Lui had already filed a claim in the Court against Ms Chui for the monies owed in January 2008, on 3 March 2008 he wrote Ms Chui a letter (“the Letter”) and the content showed that he still loved her. In our view, it is necessary to excerpt some parts from the Letter because the content is relevant to the merits of the matter. The excerpt is as follows: ‘… I must make it known to you that I did not betray, and what’s more, I have not ceased to be faithful. The development of the matter is different from what you understand/imagine and I am also a victim. This matter has been used to attack the two of us so as to separate us. I have all along been resisting pressure from various directions. I have been striving to protect you … The technical errors made in dealing with this matter have caused some problems, others have taken the advantages of the situation. However, the current situation is not hopeless, I think it is still possible to turn the tide. … I would rather sacrifice myself instead of making you suffer. Regrettably, you did not go deep into the matter to acquire a better understanding, but made a wrong decision and broke off the contact completely. This is our tragedy which should not have happened… I am looking forward to your contacting me again for the promise to be fulfilled, hoping that your days can be happy. …’ ” 25. I believe this letter is more consistent with the payments being gifts than loans. Acknowledgment of Debt 26. The plaintiff also relied on an Acknowledgement of Debt in Chinese dated 31 July 2007. It bears what was purported to be the defendant’s signature. As translated it reads : “I, Chui Kin Man, borrowed from Lui Fai Yeung HK$ one million dollars. I undertake to repay firstly the debts of not less than HK$ five hundred thousand dollars within 3 months from the date of signing this Acknowledgement of Debts, and then to repay the balance within the subsequent three months, totaling to HK$ one million dollars. If any outstanding amounts overdue remain unpaid, Mr. Lui Fai Yeung shall institute legal proceedings against me to recover.” 27. In her evidence, the defendant said the plaintiff had tried to persuade her to sign some such documents but she refused. According to the defendant, the plaintiff was under pressure to seek repayment from her. 28. The plaintiff was a director and shareholder of International Food Ingredients (Asia) Limited, (“International Food”) which carried on the business of food trading. His wife has substantial wealth.[14] A cousin of the plaintiff’s wife was the major owner of International Food and the money paid by the plaintiff for the defendant came at least in part from International Food. There is a Deed dated 14 November 2007 made between International Food, Mr Wee Lee Hiong T (the cousin), the plaintiff and his mother-in-law Wong Suet Ying in which the plaintiff admitted indebtedness to International Food and Mr Wee of HK$1,500,000. The plaintiff agreed to repay HK$1,500,000 on or before 14 May 2008 and as security Ms Wong agreed to provide a property as security. 29. There is a Recital in the Deed that: “Mr. Lui shall recover a debt in the approximate sum of HK2,300,000 from 崔健敏 (the Defendant) on or before 1 March 2008, otherwise the Company and Mr. Wee shall institute legal proceedings against Mr. Lui to recover the Debt (HK$1,500,000).” 30. It is against this background that one should consider the Acknowledgment of Debt. 31. The judge was not satisfied with the plaintiff’s evidence regarding the Acknowledgement of Debt and placed no weight on it. The Court of Appeal agreed. With respect I agree with them. 32. By the Acknowledgment of Debt, the defendant had promised to repay $500,000 within 3 months of 31 July 2007 and the balance within the following 3 months. However, not only had the defendant made no repayment, the plaintiff made a cash payment to the defendant on 31 July 2007 of $7,000. There were then five further cash payments from 3 August 2007 and 15 September 2007 ranging from $3,000 to $9,000, as well as a remittance of $80,258 on 8 August 2007. The total paid by the plaintiff to the defendant after the signing of the Acknowledgment of Debt was $112,258. These payments form part of the Plaintiff’s claim. As noted above, the plaintiff had written a love letter to the defendant in March 2008, after the present action was commenced on 23 January 2008. Such conduct casts grave doubt on its genuineness or the reliability of the document as an acknowledgment of debt. There is also a letter signed by the plaintiff for International Food saying that the defendant had been working there since 1 July 2004 at a salary of $23,500. It seems clear that it is not what it purported to be. So the plaintiff was not above producing documents which cannot be relied on. 33. Moreover, although the plaintiff had a habit of taping conversations, no taped conversation was available regarding the Acknowledgment of Debts, or its signing or the discussion(s) which led to it. That I must say is highly surprising. 34. In such circumstances, and also for the reasons given by the courts below, I agree that no weight should be given to the Acknowledgement of Debt. Fair Treatment of Evidence Taped Conversation 35. Seven transcripts were produced of conversations which took place on: 1. 25 May 2005 (whilst being massaged) 2. 25 May 2005 (whilst being massaged) 3. 31 May 2005 (whilst being massaged) 4. 2 August 2007 (face to face) 5. 18 September 2007 (telephone conversation) 6. 1 October 2007 (telephone conversation) 7. 1 October 2007 (telephone conversation) 36. The recorded conversation of 2 August 2007 was between the plaintiff, the defendant and the defendant’s sister. The others were between the plaintiff and the defendant only. 37. According to the plaintiff the conversations were recorded with a MP3 and that there were other taped conversations which he had not produced. 38. The background to these conversations can be briefly stated. The defendant and her sister had jointly purchased a “relocation property” in Shanghai in 2005, each contributing $50,000. The defendant’s contribution came from money provided by the Plaintiff, probably the cash sum of $50,000 provided on 21 March 2005[15]. The 2005 conversations were mostly about the purchase. 39. The 2007 conversations show that the plaintiff was under pressure from the cousin to repay money which the plaintiff had obtained (possibly improperly)[16] from International Food. He was trying to persuade the defendant (possibly her sister as well) to help him by selling a property (probably the relocation property in Shanghai), to charge the defendant’s shop in Hong Kong to the cousin as security for repayment, as well as to liquidate a small investment of $70,000[17] with the Bank of China and pay him the proceeds. It is clear that the plaintiff was desperate. In the taped conversation of 2 August 2007, he said: “I am dying. Even don’t want to answer the call from cousin.”[18] 40. The plaintiff relied on the following passage in the transcript of the 2 August 2007 conversation, where the plaintiff said at 94 “If those problems still cannot be resolved, I will be in big trouble. I borrowed the money for lending to her.” (借來借給她的) The plaintiff’s statement was not contradicted by the defendant. 41. However, I do not regard the absence of contradiction as an admission by the defendant. After all, her elder sister was present and she might not wish her to know her true relationship with the plaintiff. Throughout this conversation there was no express admission by the defendant that she had ever borrowed any money from the plaintiff. The fact that she was willing to raise money and to pay such money to the plaintiff is also equivocal. After all, they were still on intimate terms, and according to the plaintiff, as late as June 2007 she had wanted the plaintiff to divorce his wife and marry her[19]. 42. Since it is clear that by this time, the plaintiff would want evidence of an admission from the defendant, his failure to obtain a clear admission from her on tape is telling. None of these conversations contain an unequivocal admission by the defendant that she had borrowed money from the plaintiff. 43. I turn to the taped conversations of May 2005. They mainly concern the purchase of a relocation property in Shanghai. In the second transcript of a conversation on 25 May 2005, the plaintiff relied on the following passage in the translation provided by him: “174 Chui: For me is not that matter. For the worst, I will get back the money to repay to you. 175 Lui: well, okay, not that critical. 176 Chui: For the worst, I will get back the money and repay to you. I will still be okay. I just get back the money and repay to you. After all, the money was borrowed from you for his house” 44. The Chinese original reads: “ 174 崔：我沒關係，我最多把錢拿回來還給你。 175呂：喔好，唔緊要。 176 崔：我最多把錢還給你，我沒所謂，我最多把錢拿回來還給你，反正本來就是拿你的錢去搞這層樓的。” 45. In the Chinese original the word “borrowed” was not used. Instead of “After all, the money was borrowed from you for this house”, I believe the Chinese should be translated as “After all, your money was used for this house”. 46. In any event, I would not regard this as a sufficiently unequivocal admission by the defendant that she had borrowed $50,000 from the plaintiff. 47. It is remarkable that, despite the numerous loans that the plaintiff relied on[20], and his practice of taping conversations with the defendant, the plaintiff has been unable to produce any taped conversation of any request for loan by the defendant or the making of a loan by him. 48. In all the circumstances, I believe the correct and natural inference is that the payments were gifts and not loans. I therefore agree with the courts below that the plaintiff’s claim should be dismissed. I would dismiss the plaintiff’s appeal. 49. I would also make an order nisi that the plaintiff pays the costs of this appeal to the defendant, such costs to be taxed unless agreed. Any contrary submission on costs should be made within 14 days from the date of this judgment. In default of such submissions, the order nisi should stand as an order absolute without further order. Mr Justice Bokhary NPJ: 50. In civil cases, issues of fact are normally resolved by way of a finding one way or the other. Disposing of such an issue on the burden of proof is far from ideal. But it is not a course which the law absolutely forbids. The trial judge dismissed a claim for the repayment of what the plaintiff said were loans but the defendant said were gifts. In dismissing the claim, the trial judge proceeded on the basis that the plaintiff had not discharged his burden of proving the loans which he alleged and the defendant denied. The Court of Appeal affirmed the trial judge’s judgment. I see no ground on which we can interfere. 51. Having called the course adopted by the trial judge far from ideal, I should in fairness to him add this. If one were to go by the inherent probabilities, there were some payments by the plaintiff to the defendant that one might be disposed to think were loans. But the plaintiff’s case — and his testimony — was that all the payments were loans. Neither at trial, nor in the Court of Appeal, nor indeed before us, was the defendant ever called upon to deal with a case that there were some loans even though there were also some gifts. 52. In the result, I, too, would dismiss the appeal with the order as to costs proposed by Mr Justice Tang PJ. Lord Walker of Gestingthorpe NPJ: 53. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Chan PJ: 54. The appeal is dismissed with a cost order nisi as in para.49. The appellant appeared in person The respondent appeared in person [1] The defendant married in March 2005 and was divorced in February 2007. Her intimate relationship with the plaintiff continued during her marriage. According to the plaintiff, the defendant’s “masquerade of being single” was only unmasked in mid-2007. [2] It appears that she had given up her job as a masseuse and started a foot massage business but it is not clear when she did so. [3] There was also evidence from Lu Cheung Pi, the plaintiff’s nephew, whose evidence was relevant to when the plaintiff’s wife (Mrs Lui) discovered the plaintiff’s affair with the defendant. [4] There were a total of 7 taped conversations. The taped communications were produced “to support what I have stated in the present action.” Para 2, plaintiff’s supplemental witness statement. They will be dealt with in some detail below. [5] Chung J, 		however, made no finding on the amount of money actually paid by the 		plaintiff to the defendant. According to the defendant, it was HK$2,000,000 odd. [6] Both the trial and the appeal were conducted in Chinese. Their judgments were also in Chinese, we have been provided with translations. [7] The Defendant was represented until 18 December 2009. Thereafter, she acted in person. [8] In other words, his lordship did not have in mind only blood relationship such as between father and son which would give rise to a presumption of advancement. [9] It is not necessary to consider whether Seldon v Davidson was concerned with persuasive or evidential burden. [10] With the agreement of Ma CJHC (as he then was) and Rogers VP [11] Mak Ka Hing was concerned with commercial transactions between strangers. [12] See para 24 below. [13] Which include running machines, massage beds, crystal accessories, presumably for her foot massage business. [14]Para 13 plaintiff witness statement. It was an issue at trial when Mrs Lui discovered the plaintiff’s affair with the defendant. The plaintiff’s nephew gave evidence that the plaintiff’s wife telephoned him on 14 September 2007 and told him that the plaintiff had lent a lot of money to the defendant that she suspected that the plaintiff had an affair with the defendant. This is what the Court of Appeal said about such evidence : “41. The judge has accepted the nephew’s evidence which was that Mr Lui’s wife called him in mid-September 2007, expressing her suspicion of an extra-marital relationship between Mr Lui and Ms Chui. However the judge made it clear that there was no evidence showing how long before September 2007 that Mr Lui’s wife became aware of such situation. He also found that it was around August 2007 that Mr Lui told Ms Chui that his wife had already noticed the relationship between the two of them, and asked Ms Chui to leave Hong Kong, and it was also around August in 2007 that he told Ms Chui that his wife’s cousin would come to Hong Kong to check the accounts and told her to admit having borrowed money from him when being asked.” [15] Item 15, para 6 the statement of claim. [16] It appears from the taped conversation on 1 October 2007 that the plaintiff had requested the defendant to sign a (mortgage or charge) “按揭”on 4 Oct 2007 and that she must do so “不然表哥會告我，到時候連你都接賘什麼的，都有問題了” [“otherwise cousin will (sue/prosecute) me, at that time, you will have problem, for receiving stolen properties”] [17] At 361 2 August 2007 transcript. [18] At 370 2 August 2007 transcript. [19] Though according to the plaintiff that was because the defendant did not want to repay him. Para 13 plaintiff’s witness statement. [20] Which include a remittance of $500,000 on 14 December 2006, and a remittance of $270,000 on 9 January 2007. 
Chief Justice Cheung : 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ : 2. This appeal is concerned with what constitutes falsity for the purposes of forgery-related offences involving use of a false instrument under the Crimes Ordinance (“CO”).[1] It also concerns the principles governing substitution of a conviction of another offence in place of an offence for which the accused was originally charged and convicted. In examining the latter question, this judgment will have occasion to consider whether dishonesty is an element of the fraud offence created by section 16A of the Theft Ordinance.[2] I. THE CO OFFENCES 3. The appellant, Chan Kam Ching, a solicitor practising under the name of John Chan & Co, was involved with his clients in dealings relating to a property in Tuen Mun (“the property”). Those dealings occurred in connection with the Government’s Small House Policy which enables eligible indigenous inhabitants of the New Territories to build village houses for their own occupation on concessionary terms. The appellant was convicted after trial before HH Judge K Lo[3] of using a false instrument and using a copy of a false instrument contrary to sections 73 and 74 of the CO. He was sentenced to a total of 8 months’ imprisonment suspended for 2 years. His appeal to the Court of Appeal was dismissed.[4] A. The applicable CO provisions 4. Section 73 provides: “A person who uses an instrument which is, and which he knows or believes to be, false, with the intention of inducing somebody to accept it as genuine, and by reason of so accepting it to do or not to do some act to his own or any other person’s prejudice, commits an offence and is liable on conviction on indictment to imprisonment for 14 years.” 5. Section 74 is to the same effect, the difference being that it refers to using “a copy of an instrument which is ... false (etc)” intending to induce somebody “to accept it as a copy of a genuine instrument”. 6. By its Determination dated 13 December 2021, the Appeal Committee[5] granted leave to appeal in respect of the following question, namely: “What on the true construction of the CO, is the meaning of ‘false instrument’ where section 69(a)(vii) is relied on?” 7. CO section 69 defines falsity for the purposes of Part IX of the CO which includes sections 73 and 74, as follows: “For the purpose of this Part— (a) an instrument is false if it purports to have been— (i) made in the form in which it is made by a person who did not in fact make it in that form; (ii) made in the form in which it is made on the authority of a person who did not in fact authorize its making in that form; (iii) made in the terms in which it is made by a person who did not in fact make it in those terms; (iv) made in the terms in which it is made on the authority of a person who did not in fact authorize its making on those terms; (v) altered in any respect by a person who did not in fact alter it in that respect; (vi) altered in any respect on the authority of any person who did not in fact authorize its alteration in that respect; (vii) made or altered on a date on which or place at which or otherwise in circumstances in which it was not in fact made or altered; or (viii) made or altered by an existing person but he did not in fact exist; ...” B. The charges 8. The appellant faced three charges. Charge 1, brought under section 74, alleges that he used a copy of a sale and purchase agreement (“SPA”) dated 4 June 2007 relating to the property which was, and which he knew or believed to be, a false instrument, with the intention of inducing GE Capital (Hong Kong) Limited (a finance company) (“GE”) to accept it as a copy of a genuine instrument, and by reason of so accepting it, to do or not to do some act to its own or any other person’s prejudice.[6] 9. Charges 2 and 3 relate to his using the said SPA and an assignment of the property dated 21 June 2007 (“Assignment”), both allegedly being instruments which were false and known or believed by him to be false, intending to cause staff at the Land Registry to accept them as genuine and by reason of so accepting them, to do or not to do some act to their own or any other person’s prejudice.[7] 10. In response to a request by the defence, the prosecution provided by letter dated 14 August 2018 particulars of the alleged falsity of the two instruments in question. Those particulars are considered in the discussion which follows.[8] 11. Particulars of the prejudice alleged were not stated in the charge sheet, but the prosecution’s case at the trial was that the intended prejudice under Charge 1 involved GE’s grant of a mortgage loan; and that such prejudice under Charges 2 and 3 involved causing the Land Registry officers to register the false SPA and Assignment instruments proffered.[9] C. The facts C.1 The Small House Policy 12. It is necessary to appreciate certain features of the Small House Policy which inform the events which occurred. The Policy was recently described in this Court’s decision in Kwok Cheuk Kin v Director of Lands[10]as follows: “The beneficiaries of the Policy are ‘indigenous villagers’ of the New Territories. This means adult males descended through the male line from a resident in 1898 of a recognised village. Recognised villages are villages included in a list approved by the Director of Lands. There are currently 642 recognised villages. The Policy relates to land in the village or its immediate environs which are not affected by any impending development or future planning or development proposals. Its object is to enable an eligible villager to build, once in his lifetime, a small house in his own village for his own occupation. It allows him to apply for three kinds of grant: (i) a free building licence, which is a licence to build on private land at a nil premium in the case of pre-1898 villages or a full premium in the case of villages recognised since that date; (ii) a private treaty grant of government land at a reduced premium; and (iii) an exchange, involving the surrender of an existing title in exchange for the grant of a new title at a nil premium so far as it was private land and a reduced premium so far as it was government land. ... Over the years, benefits under the Policy have come to be known as ‘Ding rights’.”[11] 13. Thus, only a person who qualifies as a New Territories indigenous villager (a “Ding”) is eligible for the grant of a right to build, once in his lifetime, a small house on land which he owns or acquires in his village for his own occupation on such concessionary terms. The building licence granted imposes a restriction on alienation of the land generally (as in the present case) for a period of five years from the date when the District Lands Office (“DLO”) issues the certificate of compliance upon completion of the building.[12] C.2 The initial acquisition of the land 14. The relevant events unfolded in three phases. The first spans the period from November 1996 to 2000. On 15 November 1996, the property (which consisted of agricultural land) was acquired by So Chi (“So”) and Lau Chi Wai (“Lau”) for $380,000. They caused it to be conveyed on the same day to Chan Wai Man (“Chan”) ostensibly for $200,000. Chan was a Ding and was thereupon registered as the property’s owner. 15. The land was within Chan’s village and solicitors (Leung Kin & Co) were instructed to act on his behalf in applying on 18 November 1996 to the Tuen Mun DLO for a building licence, Chan declaring himself eligible under the Small House Policy. A year later, the DLO accepted his application in principle and, on 2 January 1998, Chan executed a power of attorney in favour of So, authorising So to act in dealings with the DLO. Chan was obviously a mere nominee, with So and Lau being the beneficial owners of the property. 16. On 17 April 1998, the DLO issued a building licence in favour of Chan which So executed as Chan’s attorney. As indicated earlier, it was a condition of the licence that there was a restriction on alienation of the property for five years from the eventual issue of the certificate of compliance.[13] C.3 Shum, Lai and the appellant enter the scene 17. The next phase spans the period from October 2000 to May 2007. On 26 October 2000, So, ostensibly acting as Chan’s attorney, signed an agreement to transfer what was stated to be the beneficial interest of So and Lau in the property to Lai Yee Kum, Kate (“Lai”) for $1,050,000. Lai was described as the “de facto wife” of Shum Kin Wing (“Shum”) who funded the acquisition and regarded himself as beneficially entitled to the interest agreed to be assigned to Lai. The appellant acted as solicitor on behalf of Shum and Lai and witnessed the execution of the agreement which was not registered in the Land Registry. So and Lau then dropped out of the picture with Chan remaining the registered owner of the property. Chan had to retain that status so that the concessionary terms of the Small House Policy could continue to be pursued. 18. Shum and Lai proceeded to build a house on the property. In December 2001, the appellant’s firm, stating that it was acting on Chan’s behalf, informed the DLO that the building had been completed and applied for a certificate of compliance which was granted on 15 March 2002. Shum and Lai then took up residence in the house in 2002. I note in passing that no charges were laid against anyone in respect of the representations made to the DLO in connection with the Small House Policy and will say no more on that matter. C.4 Lai acquires registered title and the impugned transactions are effected 19. The third phase occurred after the five-year restriction on alienation expired in March 2007. On 3 May 2007, Shum acquired a power of attorney from Chan (drawn up and attested by the appellant) enabling Shum to deal with the property. Shum and Lai had decided to raise a mortgage loan on the property and instructed the appellant to take the necessary steps, including effecting transfer of the registered title into Lai’s name. 20. The appellant’s firm prepared a draft of the SPA, naming Chan, putatively acting through Shum as his attorney, as vendor and Lai as purchaser and, on 21 May 2007, faxed a copy to GE, from whom it was proposed to obtain the mortgage loan. According to the draft SPA, Lai was to acquire the property for $3 million and had already paid $1.5 million towards the purchase. The SPA was executed on 4 June 2007. 21. On 28 May 2007, Lai applied for a mortgage loan of $1.5 million which GE’s management approved two days later. On 18 June 2007, the appellant’s firm submitted documentation in support of the loan application, including a copy of the SPA (the subject of Charge 1). GE then instructed the appellant to prepare a first legal charge on the property and on 20 June 2007, the SPA was registered with the Land Registry, with the appellant signing the attached memorial verifying its content (the subject of Charge 2). An assignment of the property dated 21 June 2007, prepared by the appellant’s firm, was executed by Shum as Chan’s attorney, transferring title to Lai stating the consideration to be $3 million. On the same day, GE transferred $1,498,500 (having deducted an amount for fire insurance) to the appellant’s firm, of which $1,442,670 was passed on to Shum. The assignment was then registered in the Land Registry on 5 July 2007 (the subject of Charge 3), the appellant signing the attached verifying memorial. D. The decisions of the courts below[14] D.1 HH Judge K Lo[15] 22. On the aforesaid findings, the trial Judge held that the prosecution had proved all the elements of the offences charged, namely, that the appellant had used a copy of a false instrument and two false instruments which he knew were false, with the intention of inducing GE and the Land Registry to accept them as genuine, and by reason of so accepting them to do the respective acts of granting a mortgage loan and accepting registration of the documents to their prejudice. This conclusion was upheld by the Court of Appeal. The element of the two offences with which we are concerned is the requirement that the instruments used had to be false. 23. Counsel for the prosecution had provided particulars of such falsity by letter dated 14 August 2018, stating: “[The] Prosecution says that the documents referred in the 3 charges were false because they contained false representation of ‘payment of 3 million as consideration’ and ‘the manner of payment stated’, which your client (Defendant) admitted to be untrue when he testified [in earlier proceedings[16]]. The Prosecution says that the documents are false within the meaning of Section 69(a)(vii) of [the CO], ie, an instrument is false if it purports to have been made in circumstances in which it was not in fact made.” 24. As we have seen, section 69(a)(vii) specifies that “... an instrument is false if it purports to have been ... made or altered on a date on which or place at which or otherwise in circumstances in which it was not in fact made or altered”. 25. The prosecution’s case was accepted by the trial Judge who stated: “In my view, the [SPA] dated 4 June 2007 and the Assignment dated 21 June 2007 are clearly untrue in their content, and they are also in line with (the definition of) a false instrument under section 69(a)(vii) of the Ordinance.”[17] 26. By way of elaboration, Her Honour stated: “According to the Accused’s testimony in the other case, the Accused did not dispute that LAI did not pay $3,000,000 to CHAN for the purchase of the Property. Therefore, in the content of the instrument referred to in Charge 1, namely the copy of the [SPA], and the instruments referred to in Charges 2 and 3, namely the [SPA] and the Assignment submitted to the Land Registry, the statements and descriptions relating to the purchase price of the Property are all false. Therefore, the Agreement dated 4 June 2007 submitted to [GE] is a copy of a false instrument, and (those) submitted to the Land Registry, namely the [SPA] dated 4 June 2007 and the Assignment dated 21 June 2007, are false instruments. The circumstances of those instruments or the copy of those instruments are in line with section 69(a)(vii) of the Crimes Ordinance, and (they) are therefore false instruments.”[18] 27. She found that the appellant well knew of such falsity, again focusing on non-payment of the stated consideration: “I am of the view that at the material time the Accused was absolutely aware of the existence and the content of the instrument of 4 June 2007, in particular the consideration of $3,000,000, LAI and CHAN as the contracting parties and the payment mode as well as the Assignment of $3,000,000 were all untrue. These are false instruments, and fall into the ambit of section 69(a)(vii) of the Ordinance.”[19] 28. The Judge accordingly held that the SPA and Assignment were “false” within the meaning of section 69(a)(vii) because they contained false representations as to the fact and manner of payment of the $3 million consideration, consistent with the particulars supplied by the prosecution and with the prosecution’s written opening which dealt in detail with non-payment of the $3 million consideration.[20] D.2 The Court of Appeal[21] 29. While endorsing the Judge’s decision, the Court of Appeal went further and concluded that the SPA and Assignment were false on a significantly different basis. Anthea Pang J,[22] writing for the Court of Appeal, held that “leaving aside the veracity of the sale price and the payment condition as stated in the [SPA]”,[23] those documents were false because the underlying sale and purchase transaction did not, and was never intended to, transfer property in the Land to the Ding and so was a sham. 30. Her Ladyship stated: “... the sale and purchase negotiations and transactions by both parties in relation to the Land and the Property had to exist before the preparation of the [SPA]; and the nature of the document should reflect the sale and purchase in question. Hence, leaving aside the veracity of the sale price and the payment condition as stated in the [SPA] and by solely considering the facts admitted by the Defence, i.e., LAI and CHAN had never performed any transaction at any time, that would be sufficient to show that the [SPA] is a false instrument, because that instrument was ostensibly prepared under the circumstances of LAI and CHAN performing a sale and purchase transaction, but in fact it was not so. Because both parties did not enter into negotiations or reach any agreement in respect of that transaction, nor was there the intention to do so, the transaction did not exist in nature. Hence at all material times when CHAN (or SHUM as his attorney) and LAI executed the [SPA], the identity of a real vendor and that of a real purchaser never existed, nor was there the intention to exercise the related legal rights or shoulder the related obligations arising from the provisions stated in the [SPA].”[24] 31. In summary, she likened the documents to a blank sheet of paper: “In short, the circumstances of that sale and purchase stated in the instrument in question did not exist, and the terms on the price and payment set out by both parties were not facts either; to put it more directly, the instrument is just like a sheet of blank paper and does not reflect the truth, no matter in nature or in terms of the clauses.”[25] 32. The following observations regarding the Court of Appeal’s decision may be made at this point. First, the “sham” approach as the basis for establishing falsity departs substantially from the prosecution’s case particularised in the letter of 14 August 2018 and outlined in the prosecution’s written opening. They both alleged that the documents were false because of misrepresentations they contained about the fact and manner of payment of the consideration. As we have seen, the trial Judge convicted the appellant on that footing. 33. Secondly, the Court of Appeal largely founded its “sham” conclusion on the authority of HKSAR v Lau Kam Ying[26] which, as Anthea Pang J acknowledged, was not mentioned or relied on by the trial judge.[27] However, the Lau case did not involve a forgery offence. It had its origins in an attempt fraudulently to manipulate the Small House Policy, but the case evolved into a prosecution for conspiracy to defraud after that attempt had failed. A company (“Howin”) had acquired land and assigned sections of the land to 17 Dings with a view to their acquiring concessionary building rights under the Policy. The scheme failed because of objections to such building and Howin was wound up. The Dings, who had earlier executed declarations of trust in favour of Howin, registered the assignments and became ostensibly the registered owners of the land. Subsequently, the Government moved to resume the land and the original promoters of Howin (who with the solicitor concerned, later faced prosecution) formed a new company (“CGG”) and persuaded 12 of the Dings to assign their rights as registered owners to CGG which then claimed the right to compensation in the resumption exercise. This meant that insofar as Howin remained the owner of the land – whether beneficially or both legally and beneficially – its rights and those of its creditors to the compensation would be usurped by CGG and its promoters. 34. The Court of Appeal in Lau Kam Ying concluded that Howin had retained the right to claim the compensation and was the victim of a conspiracy to defraud. It was in this context that it held that either the assignments to the Dings were a sham and of no effect so that Howin remained both legal and beneficial owner; or, if the assignments had vested legal title in the Dings, that Howin nonetheless retained beneficial ownership so that the scheme to use CGG to obtain the compensation while concealing Howin’s interest in the land constituted a conspiracy to defraud Howin, its creditors and the Government. Thus, Lau Kam Ying shows that schemes of this type may be characterised as sham transactions for certain purposes. However, it does not address the question whether characterising an underlying transaction as a “sham” supports the conclusion that the documents generated as part of that scheme are false instruments for the purposes of section 69(a)(vii). That was not an issue in the case. 35. Thirdly, as the Court of Appeal’s theory of falsity was not part of the prosecution’s case and was not explored at the trial, it is not at all clear how far Anthea Pang J’s statements that “LAI and CHAN had never performed any transaction at any time”; that “... both parties did not enter into negotiations or reach any agreement in respect of that transaction, nor was there the intention to do so”; and that “when CHAN (or SHUM as his attorney) and LAI executed the [SPA], the identity of a real vendor and that of a real purchaser never existed, nor was there the intention to exercise the related legal rights”; are justified. I do not accept the repeated assertions by Ms Maggie Yang,[28] counsel appearing for the respondent on this appeal, that those were matters of “undisputed fact”. Given that they were not part of the prosecution’s case, it is not surprising that such allegations were not the subject of scrutiny and dispute. There must be serious reservations as to whether it was open to the Court of Appeal as a matter of procedural fairness to hold that falsity was established on the basis identified by Anthea Pang J. 36. The acts legally attributable to Chan were done pursuant to powers of attorney that he had executed on 2 January 1998 and on 3 May 2007. The latter power authorised Shum to deal with the property after expiration of the five-year restriction on alienation in order to transfer Chan’s registered legal title to Lai who thereby acquired those rights, enabling her to obtain a mortgage loan. It is therefore at least questionable whether it was accurate to say that Chan had played no role at all in the scheme and that a real vendor and purchaser and an intention to create legal rights did not exist. There is no doubt that Chan was a nominee but it is difficult to see why he did not have legal title to the property which he transferred to Lai. The prosecution’s and the trial Judge’s description of the transaction as false because of non-payment of the consideration was factually more secure than the Court of Appeal’s characterisation of it as a “blank sheet”, non-existent transaction with no legal consequences. Counsel appearing for the respondent sought to support the “sham” approach but retreated somewhat from the proposition that Chan had no title to give when that was queried by the Court. 37. Fourthly, what the Judge’s and the Court of Appeal’s decisions have in common is that they both base the falsity of the SPA and Assignment for the purposes of the CO section 73 offence on the lack of genuineness of the underlying transaction: non-payment of the stated consideration or its allegedly “sham” character. Those features of the transaction are relied on as bringing the instruments within section 69(a)(vii). I return to consider the validity of that approach after examining the applicable law. E. The competing interpretations of CO section 69(a)(vii) 38. In holding as aforesaid, the courts below favoured a wide interpretation of section 69(a)(vii). They held, by virtue of the words “otherwise in circumstances in which it was not in fact made”, that a document which contains a lie about some extraneous fact or circumstance, such as an underlying transaction, qualifies it as a “false instrument”. I shall call this “the wide construction”. 39. The competing view is that sections 73 and 74, read together with section 69(a)(vii), do not encompass lies concerning events or matters extraneous to the making or alteration of the instruments themselves. It asserts that the traditional requirement in the law of forgery that the document must “tell a lie about itself” – sometimes referred to as “the automendacity principle”[29] – is preserved in the CO’s scheme. A person who creates or uses a document which tells lies about an underlying transaction or other extraneous matters may well be guilty of other offences, such as deception or fraud under the Theft Ordinance or may be part of a conspiracy to defraud. But forgery-related offences under CO Part IX are concerned with a particular type of fraud that involves making or using an instrument which, on its face, tells a lie about the circumstances of its own making or alteration, for example, lies as to who made it, the signatory’s authority to sign it, when and where it was made, and so forth. I shall refer to this as “the automendacity approach”. 40. Choosing between the two approaches depends on the true construction of the relevant provisions. They represent a codification of the law of forgery and related offences since CO section 79 abolishes the offence of forgery at common law for all purposes. Two lines of authority and commentary have developed, supporting respectively the wide construction and automendacity approaches. Since the CO provisions are derived from and mirror the provisions of the United Kingdom’s Forgery and Counterfeiting Act 1981 (“the 1981 Act”), English case-law interpreting sections 3 and 9(1) of that Act is pertinent, CO sections 73 and 69(a)(vii) being their direct equivalents. F. The wide construction 41. The wide construction is usually traced back to the 1984 judgment of Lawton LJ in R v Donnelly,[30] where the manager of a jeweller’s shop made out and signed what purported to be a written valuation he had made of certain items of jewellery. In fact, the jewellery did not exist and he had performed no such valuation, the certificate having been made with a view to defrauding an insurance company. On a charge of making a false instrument contrary to section 1 of the 1981 Act, the issue was whether, as the defence submitted, the purported valuation certificate was not a false instrument but “just a lying one, and that an instrument was not a forgery if it did no more than on its face tell a lie, not being a lie as to what it was”; or whether, on the contrary, the certificate qualified as a false instrument within section 9(1)(g) of the 1981 Act. The prosecution conceded that the certificate would not be a forgery at common law or under the Forgery Act 1913 but submitted that section 9(1)(g) had made new law and greatly expanded the scope of the offence. 42. Lawton LJ accepted the prosecution’s wide construction, stating: “In our judgment the words coming at the end of paragraph (g) ‘otherwise in circumstances …’ expand its ambit beyond dates and places to any case in which an instrument purports to be made when it was not in fact made. This valuation purported to be made after the appellant had examined the items of jewellery set out in the schedule. He did not make it after examining these items because they did not exist. That which purported to be a valuation after examination of items was nothing of the kind: it was a worthless piece of paper. In our judgment the trial judge’s direction was correct. This purported valuation was a forgery.”[31] (Emphasis in the original) 43. Donnelly was held to be binding and was followed by the English Court of Appeal in R v Jeraj.[32] In that case, an employee of a bank signed a document on bank notepaper which purported to be verification that he had received a letter of credit issued by a named Bolivian bank and that he, on behalf of his bank, fully endorsed it. Neither the letter of credit nor the Bolivian bank existed and it was held that this circumstance brought the case within section 9(1)(g) of the 1981 Act, qualifying the document he had signed as a false instrument. 44. The competing views were canvassed in Attorney-General’s Reference (No 1 of 2000),[33] where a coach driver was alleged to have tampered with a tachograph to show that he was taking a break as required by law, and that another driver was driving when there was in fact no other driver. He was charged with forgery of the tachograph record sheet. Noting academic criticisms that Donnelly had excessively broadened the scope of the offence, Lord Woolf CJ held that that decision remained binding but added: “Both Donnelly and Jeraj should however be restricted in their application so that they apply only where circumstances need to exist before the document can be properly made or altered. If those circumstances do not exist there will then be a false instrument for the purposes of section 9(1)(g). If the circumstances do not exist the document is telling a lie about itself because it is saying it was made in circumstances which do not exist. ... The need for the existence of these circumstances prior to the making of the instrument explains why if the circumstances do not exist the document is telling a lie about itself.”[34] (Emphasis in the original) 45. It was held that the tachograph record was a false instrument: “... [t]he circumstance which was false was that the record was being made during a period when there wrongly purported to be a second driver who was driving.”[35] Lord Woolf CJ’s view was therefore that the Court’s approach was consistent with the requirement that a document should “tell a lie about itself” even though the “circumstances” in question involved facts extraneous to the document (the defendant’s continuous driving and the absence of any other driver) rather than circumstances attending the making or alteration of the document itself. 46. This line of authority, and in particular, A-G’s Reference (No 1 of 2000), was adopted by the Hong Kong Court of Appeal in Secretary for Justice v Yeung Hon Keung Larry.[36] The accused in that case applied for admission as a part-time student in a university post-graduate law course, one of the requirements being that he had to produce a letter from his employer supporting his application and confirming that he would be released for such tuition. Apprehending that such a letter might be refused by his true employer, the accused proffered two letters which falsely stated that he was an employee of a named hotel and that his employer supported his application. He had never been so employed. However, the letters were signed and provided by persons friendly to the accused who were in fact the manager and a shareholder of the company owning the hotel. He was charged inter alia with using a false instrument contrary to CO section 73. 47. The trial Judge[37] held that the letters were not false instruments since they were written on the headed notepaper of the hotel and were signed by the persons who had purported to sign them, correctly described in the letters and with authority so to sign. The accused was acquitted on that charge. 48. That decision was overturned on an appeal by way of case stated. Stuart-Moore VP, writing for the Court, reviewed the authorities and concluded that the legislation substantially broadens the scope of the offence: “Whilst not binding on this court, we consider that the reasoning given in Attorney-General’s Reference (No 1 of 2000) to be powerfully persuasive. There can be no doubt that the law relating to forgery has been widened in its scope since the time of earlier legislation which confined its application to much narrower issues. The deliberate addition of the words ‘in circumstances in which it (the instrument) was not in fact made’ to s 69(a)(vii) of the Ordinance, mirroring the same addition made to the 1981 Act in England and Wales, can only have been for this purpose. The legislative intent seems to us to have been plain enough.”[38] 49. The aforesaid authorities espouse the wide construction focusing on the words in the latter part of CO section 69(a)(vii)[39] which state: “made or altered ... in circumstances in which it was not in fact made or altered”. They regard those words as applicable to any instrument whose content involves a lie about the existence of certain extraneous facts, especially facts whose prior existence is implied or referred to in its content, treating such extraneous facts (or their absence) as the relevant “circumstances”. 50. That approach was adopted by the respondent in this appeal. Ms Yang argued that the SPA and Assignment were “false” for the purposes of section 69(a)(vii) because there was no underlying legally binding agreement for the sale and purchase of the property. Such absence, she submitted, was the circumstance which brought those instruments within section 69(a)(vii). G. The automendacity approach in construing section 69(a)(vii) 51. Plainly, if one construes section 69(a)(vii) by focusing on what the words “made or altered ... in circumstances in which it was not in fact made or altered” are capable of encompassing simply as a matter of language, one is likely to arrive at a construction attributing a very wide scope indeed to that provision. Taken in isolation, those words give no indication that any limits exist regarding the types of extraneous circumstances which may trigger operation of the provision. It was by focusing on those words that Lawton LJ in R v Donnelly,[40] and Stuart-More VP in Secretary for Justice v Yeung Hon Keung Larry[41]concluded that the ambit of forgery offences had been greatly widened. Lord Woolf CJ sought to limit the compass of those words by holding that they covered situations “where circumstances need to exist before the document can be properly made or altered”. But applying that construction, the resultant rule is still very wide since a lie about an extraneous fact or underlying transaction contained in an instrument almost always suggests the prior existence (or absence) of such fact or transaction, and so would suffice to constitute falsity on Lord Woolf CJ’s approach. 52. With respect, a more valid approach requires the relevant provisions of section 69(a)(vii) to be construed contextually and purposively. G.1 Law Commission Report 53. It is appropriate to begin with the 1973 Report of the English Law Commission on Forgery and Counterfeit Currency (“LC Report”).[42] It formed the basis of the 1981 Act which enacted the material provisions of the Draft Bill that the Law Commission had put forward, which was in turn enacted in this jurisdiction as Part IX of the CO.[43] 54. The LC Report explains that the purpose of criminalising forgery and kindred offences is to address a particular class of fraud involving the use of false documents designed to exude a spurious air of authenticity: “In the many and varied activities of modern society it is necessary to rely to a large extent on the authenticity of documents as authority for the truth of the statements which they contain. Indeed, in the vast majority of forgery cases the purpose of the forger is to lull the person to whom the document is presented into a false position in which he will be unlikely, because of the apparent authenticity of the document, to make further enquiry into the correctness of the facts related. The same is not true of false statements contained in a document which carries no spurious authenticity. A letter by an applicant for an appointment setting out falsely his qualifications is in quite a different category from a letter of recommendation purporting to come from a previous employer.”[44] 55. The spurious quality targeted is reflected in the way that falsity in the instrument is approached, namely, by asking whether the document “tells a lie about itself”,[45] ie, the automendacity approach. As the LC Report observes: “By the middle of the nineteenth century it was established that for the purposes of the law of forgery the fact that determined whether a document was false was not that it contained lies, but that it told a lie about itself. It was in R v Windsor (1865) 10 Cox CC 118 at 123, that Blackburn, J. said- ‘Forgery is the false making of an instrument purporting to be that which it is not, it is not the making of an instrument which purports to be what it really is, but which contains false statements. Telling a lie does not become a forgery because it is reduced into writing’.”[46] 56. As the LC Report makes clear,[47] this approach is implemented by clause 6(2) of its Draft Bill which defines “falsity” in the terms subsequently adopted in Hong Kong by CO section 69(a): “An instrument is false if, to adopt the words used in the authorities ‘it tells a lie about itself’. This phrase is put into precise language by clause 6(2). There is thus preserved the distinction between a document which merely tells a lie, and one which gives a false idea as to its source or the circumstances of its creation.”[48] 57. The Law Commission makes it clear that it was not proposing to extend the scope of the offence to cover lies reduced to writing: “We do not think that there is any need for the extension of forgery to cover falsehoods that are reduced to writing, and we do not propose any change in the law in this regard.”[49] 58. It elaborates as follows: “The essential feature of a false instrument in relation to forgery is that it is an instrument which ‘tells a lie about itself’ in the sense that it purports to be made by a person who did not make it (or altered by a person who did not alter it) or otherwise purports to be made or altered in circumstances in which it was not made or altered. Falsity needs to be defined in these terms to cover not only, for example, the obvious case of forging a testator’s signature to a will, but also the case where the date of a genuine will is altered to make it appear that the will was executed later than it in fact was, and therefore after what in truth was the testator’s last will.”[50] 59. As Professor J C Smith pointed out, preservation of the automendacity principle was the intention stated in the Parliamentary debates: “In introducing the present section 9 in the House of Lords, Viscount Colville of Culross stated that ‘we have here a comprehensive list of the range of circumstances in which an instrument is said in law to be false. This substantially retains two things; first, the definitions in the 1913 Act and then also, the common law’ (Hansard, Vol. 416, col. 607). This statement was repeated in substance by Mr. Garel-Jones in the House of Commons (Hansard, Vol. 6, col. 1304). This explanation was accepted by both Houses without comment.”[51] G.2 Construction of CO Part IX 60. It is clear that CO Part IX faithfully reflects the English Law Commission’s approach. The offences created target use of instruments of spurious authenticity. Thus, the offence of forgery created by section 71 is constituted as follows: “A person who makes a false instrument, with the intention that he or another shall use it to induce somebody to accept it as genuine, and by reason of so accepting it to do or not to do some act to his own or any other person’s prejudice, commits the offence of forgery ...” In other words, a forger is someone who seeks to confer spurious authenticity on the forged instrument. 61. The charges in the present case under CO sections 73 and 74[52] involve the use of such a false instrument or a copy thereof, intending others to accept the authenticity of the instruments or copies used. The offences are thus not about intending others to accept the truth of whatever lies may be told in the contents of the instruments concerning extraneous events or transactions, but to accept the genuineness of the instruments themselves. 62. Section 69(a)[53] defines exhaustively how such falsity may arise. Three significant features may be noted. First, it defines falsity by reference to what the instrument “purports”, in other words, what the instrument on its face conveys. Or as Mr David Perry QC,[54] appearing for the appellant put it, “what the instrument appears to be”. Secondly, in each case, what the instrument purports relates to the circumstances of its making or alteration. Thirdly, such characteristics are designed to confer spurious authenticity on the instrument itself, such as the identity or authority of its maker. Throughout section 69(a), the focus is on the making or alteration of the document itself. 63. Taken in its proper context, section 69(a)(vii) creates no exception. The reference to the “circumstances” in which the instrument was not in fact made or altered should be construed ejusdem generis with the other two instances of falsity enumerated in section 69(a)(vii) itself, namely, “the date on which or place at which” the instrument was made or altered. As with those two instances, the circumstances in question relate to falsity in the purported circumstances of the making or alteration of the instrument – not to extraneous facts. 64. Sub‑paragraph (vii) must also be read ejusdem generis with the other sub-paragraphs in section 69(a) each of which is consistent with the automendacity approach, defining an instrument as false if it purports to be made or altered by a person who did not in fact make or authorise its making in the form or terms the document assumes. There is no basis for thinking that sub‑paragraph (vii) (by the words coming at its end), in contrast with the other sub-paragraphs, radically alters the entire fabric of the offence, so as to characterise a lie (such as a lie about some past fact or event) reduced to writing as a forgery. The tail must not be allowed to wag the dog. G.3 The authorities and commentaries supporting the automendacity approach 65. The automendacity approach is consistent with authorities both in England and Wales and in Hong Kong, and with the view taken almost unanimously by distinguished academic commentators, disapproving of the wide construction. 66. Professor J C Smith, commenting on Donnelly,[55]observed: “If the present case is followed, the law of forgery is extended far beyond its proper bounds. Whenever an instrument tells a lie about some alleged past fact it purports to be made after the fact occurred. There was nothing peculiar about the alleged fact of having examined and valued the items of jewellery. Thus, virtually all documents telling lies would be capable of amounting to forgery.” 67. He advocated an ejusdem generis construction, arguing: “... that the words ‘or otherwise in circumstances in which, it was not in fact made or altered’ must be read ejusdem generis with the preceding words, ‘made or altered on a date on which, or at a place at which.’ and indeed with all the other instances of falsity set out in section 9. There was little point in setting out all these particular instances if the words in issue in the present case were in effect, to embrace virtually all of them - and much more.”[56] 68. Stressing that the falsity envisaged by the equivalent of section 69(a)(vii) concerns the making or alteration of the instrument, he gave the examples of such circumstances: “The sort of circumstances which the provision is intended to cover is the case where an instrument purports to be made in the presence of a witness who was not in fact present. It then purports to be something - an instrument made in the presence of a witness - which it is not; and that is the characteristic of falsity in the law of forgery.”[57] 69. In contrast, the jeweller’s valuation certificate in Donnelly “was exactly what it purported to be, except that it purported to be a true statement of facts when it was not”,[58] such falsity being insufficient for the purposes of forgery. 70. R v More,[59] was a case in which the accused, who had obtained a cheque drawn in favour of “M R Jessell”, opened an account in the name “Mark Richard Jessell” and then paid the cheque into it. He later withdrew £5000 from the account in the form of a cheque payable to “M R Jessel,” signing the withdrawal form “M R Jessell”. 71. The prosecution contended that the withdrawal form was a forgery, relying on section 9(1)(h) of the 1981 Act,[60] arguing that it purported to have been made by a non-existent person. The trial Judge rejected those allegations, holding that the document: “... was undoubtedly made by the appellant and it was undoubtedly made in the form of a withdrawal form. It was undoubtedly signed by the person making it, and that signature was undoubtedly the signature of the holder of the account in the name ‘Mark Richard Jessell’.”[61] 72. However, the English Court of Appeal held that section 9(1)(h) applied “since it purported to have been made by an existing person but he did not, in fact, exist.”[62] But its judgment was reversed by the House of Lords, Lord Ackner (with whom the other Law Lords agreed) stating: “It is common ground that the consistent use of the word ‘purports’ in each of the paragraphs (a) to (h) inclusive of section 9(1) of the Act imports a requirement that for an instrument to be false it must tell a lie about itself, in the sense that it purports to be made by a person who did not make it (or altered by a person who did not alter it) or otherwise purports to be made or altered in circumstances in which it was not made or altered.”[63] (Italics supplied) His Lordship made no exception regarding section 9(1)(h), the equivalent of our section 69(a)(viii): the automendacity approach was applicable to all the heads of falsity enumerated. 73. Referring to the withdrawal form, his Lordship held, quashing the forgery conviction: “... the appellant was a real person. It was he who was the holder of the account and in that capacity had signed the withdrawal form. The withdrawal form clearly purported to be signed by the person who originally opened the account and in this respect it was wholly accurate. Thus, in my judgment, it cannot be validly contended that the document told a lie about itself ...”[64] 74. Laws J, writing for the English Court of Appeal, provided a compelling analysis of that head of falsity in Warneford and Gibbs.[65] It was a case where the defendants obtained mortgage loans proffering what purported to be references from their employers when they were not in gainful employment (one was unemployed, the other a student). The Crown’s case was that the instruments came within section 9(1)(g) of the 1981 Act,[66] the argument being that the documents, not being in truth references from employers, were not what on their face they purported to be so that they told lies about themselves. 75. Laws J construed the provision differently: “It is entirely plain ... that section 9(1)(g) is not to be construed so as to bring within its compass every document which contains a falsehood. It has long been accepted, in cases decided both before and after the coming into force of the Act, that to be a forgery a document must ‘tell a lie about itself’ - as indeed the judge told the jury. That formulation chimes with the correct construction of section 9(1)(g). The expression ‘otherwise in circumstances in which it was not in fact made’ must, in our judgment, refer to the circumstances of the making of the document just as surely as the references in the sub-paragraph to date and place concern the date and place on or at which the document is made.”[67] 76. His Lordship emphasised that the provision concerned the circumstances of the document’s making and not circumstances involving extraneous facts: “If, for example, the document on its face purports to have been made in the presence of certain named individuals who in fact were not present at all, it would fall within the sub-paragraph. Likewise, a document whose words purported to have been dictated by a particular person, when it was in fact entirely the work of the writer, would come within the definition. So would a document purporting to have been made at a different time of day from its actual making. Other examples may suggest themselves. But in every case the lie in the document must relate to the actual circumstances of the document’s making. A lie about other facts, extraneous to the document, does not suffice; such a lie may go in proof of other offences (notably under the Theft Act), but cannot establish forgery. The offences which, under the Act of 1981, require proof of forgery within section 9 are exclusively concerned with the document itself.”[68] 77. He pointed out that this approach was in line with the decision in R v More and added: “... the decision in Donnelly cannot, in our judgment, stand with More. It offers a mistaken construction of section 9(1)(g). To say that the document purported to be a true valuation, but was not, is to urge no more than that it purported to be a true representation of its contents, but was not; it is, in effect, to embrace within the concept of a forgery every document which tells a lie about its subject-matter. That cannot be the right approach. It fails to ask and answer the question whether the document tells a lie about itself. It amounts to a misconstruction of the words ‘…purports to have been made… in circumstances in which it was not in fact made…’ in the sub-paragraph.”[69] 78. Professor J C Smith argued that Warneford and Gibbs is correct, noting that it is in conflict with Donnelly and Jeraj.[70] Laws J’s comment that Donnelly cannot stand with More is echoed by the learned editors of Smith, Hogan and Ormerod’s Criminal Law.[71] They state: “Obviously, the valuation certificate told a lie, but did it tell a lie about the circumstances in which it was made? If it did, then a begging letter in which the beggar, or someone on his behalf, falsely states that he is homeless or unemployed, is equally a forgery because the circumstances to which the writer alludes do not exist. This would be a remarkable extension of the law of forgery as previously understood. It is submitted, however, that it cannot stand with the decision of the House of Lords in More (in which Donnelly was not mentioned).”[72] 79. Similarly, Professor A T H Smith’s view[73] was that Donnelly did not pay sufficient regard to the meaning of “purports” as explained by Lord Ackner and that insofar as the decisions were inconsistent with each other, the latter decision was to be preferred. 80. Lord Woolf CJ in A-G’s Reference (No 1 of 2000) noted that Laws J thought that Donnelly could not stand with More but evidently did not agree, holding[74] that the Court was bound by Donnelly and Jeraj and that Warneford and Gibbs must be regarded as coming to the wrong conclusion on the facts. His Lordship did however add that the Court was “in agreement with certain of Laws J’s remarks, to which we referred earlier” which appears to have been a reference to Laws J’s conclusion that falsity under section 9(1) was concerned with the making of the document. How that was reconcilable with the approach in Donnelly and Jeraj was not discussed. 81. In Hong Kong, the Court of Appeal considered the effect of CO section 69(a)(vii) in HKSAR v Huynh Bat Muoi,[75] and reached a conclusion contrary to that reached six years later in Secretary for Justice v Yeung Hon Keung Larry,[76] which, as we have seen, adopted the wide construction. Huynh was concerned with a defendant charged with aiding, abetting, counselling or procuring the use of a forged travel document contrary to section 42(2)(b) of the Immigration Ordinance.[77] Section 42(5) of that Ordinance specified that “forged” in that provision had the meaning assigned to that term by CO Part IX, making CO section 69(a)(vii) directly relevant. 82. The instrument in question was a Belgian passport which was issued to X, a Belgian citizen who had reported his passport stolen. A replacement passport was issued by the proper authority stating all of X’s particulars correctly but bearing a photograph of someone else which X had supplied to be used in the passport. This was part of a fraudulent scheme orchestrated by the defendant to obtain the passport to be used by the person in the photograph. The question was whether that passport was a false instrument within the meaning of CO section 69(a)(vii). 83. Woo JA, writing for the Court of Appeal, declined to adopt the English line of authorities culminating in A-G’s Reference (No 1 of 2000), holding firmly to the automendacity approach and focusing on whether the falsity related to the circumstances of the making of the document: “The part of s 69(a)(vii) relevant for our consideration is that ‘an instrument is false if it purports to have been made in circumstances in which it was not in fact made’. The significant words are ‘the circumstances in which’ the instrument was ‘in fact made’. What is it that the passport purports to say about itself and about the circumstances in which it was made? Only that it has been issued by an official on such and such a date, and at such and such a place, pursuant to the authority which that officer had to issue it. The passport was in fact made in the form and upon the terms in which the maker made it. It has not since been altered, and was made on the date and place and in the circumstances in which it purports to have been made. If it also purports to represent that it was made after an application for a passport, then that too is an accurate representation about itself. The subject passport contains no lie about itself, although it contains a lie. It also contains no lie about the circumstances in which it was made, whether as to time, or place, or authority. Although it is an instrument containing a falsehood in the photograph in it not being that of the true holder …, we are not persuaded that it qualifies as a forgery under s 69(a)(vii) of the Crimes Ordinance or any of the categories of false instrument within the ambit of any other category under s 69.”[78] 84. Although the aforesaid passage in Huynh was cited by Stuart-Moore VP in Secretary for Justice v Yeung Hon Keung Larry,[79] he did not discuss the obvious inconsistency between the wide construction adopted by the Court in Yeung Hon Keung Larry, following A-G’s Reference (No 1 of 2000), and adherence to the automendacity approach by the Court in Huynh (of which Stuart-Moore VP had also been a member). H. The proper construction of section 69(a)(vii) 85. In my view, the proper construction of section 69(a)(vii) compels acceptance of the automendacity approach. The foregoing analysis strongly militates against the wide construction which the Court of Appeal here embraced, applying Yeung Hon Keung Larry.[80] It failed in particular to construe the relevant words in section 69(a)(vii) ejusdem generis with the rest of that sub-paragraph and also with section 69(a) as a whole. The approach in Huynh Bat Muoi[81] is to be preferred. 86. By way of summary, the wide construction concentrates too narrowly on the words at the end of section 69(a)(vii). A purposive and contextual construction of that sub-paragraph, of section 69(a) and CO Part IX as a whole is needed. The statutory purpose is to target the perpetration of frauds using instruments of spurious authenticity, documents which tell lies about themselves. That purpose is implemented by the statutory scheme’s focus on the making, alteration, copying and use of such instruments intending others to accept them as genuine. The core emphasis is on what the instruments purport to convey as to the circumstances of their making or alteration. Section 69(a)(vii) does not create an exception. The words at the end of that provision should be construed ejusdem generis with the preceding words; and section 69(a)(vii) should be construed ejusdem generis with the other sub-paragraphs of section 69(a), leading to the conclusion that sub-paragraph (vii), like the other provisions, is directed against lies which relate to the circumstances of the document’s making or alteration. Forgery offences are centred on the bogus qualities of the instrument itself, and not on the truth or falsity of representations made as part of its contents assessed against any extraneous facts which may expressly or impliedly be referred to. The authorities supporting this approach, especially R v More, Warneford and Gibbs and Huynh Bat Muoi, are preferable as a matter of principle to the cases adopting the wide construction. It is an approach that has the general support of distinguished academic commentators. The submission of counsel for the respondent that there is no authority to contradict the line of decisions culminating in A-G’s Reference (No 1 of 2000),[82] is unfounded. I. Application of the provisions to the facts 87. The appellant was charged with using false instruments consisting of the draft SPA faxed to GE, the SPA dated 4 June 2007 and the Assignment dated 21 June 2007 presented for registration at the Land Registry as set out in Section B above. As we have seen,[83] the prosecution’s case and the basis of the Judge’s conviction at first instance was that the said instruments were false because they contained false representations concerning the fact and manner of payment of the $3 million stated to be the consideration for the sale and purchase of the property. The Court of Appeal upheld the conviction on the footing that the documents were false because the underlying transaction did not and was never intended to transfer property to the Ding and so was a sham. As noted above, I have grave reservations as to whether it was open to the Court of Appeal to adopt that approach which significantly departed from the prosecution case and was not explored at the trial. 88. In any event, it is plain that the decisions below rest upon the wide construction which I have rejected. Falsity under section 69(a)(vii) was held by the Judge and the Court of Appeal to have been established by reference to lack of genuineness in the underlying transaction when, on its true construction, falsity depends on the instrument itself being shown to tell a lie about the circumstances of its making rather than about such extraneous matters. 89. The prosecution did not advance any case based on the automendacity principle. Nor did either Court below base the appellant’s conviction thereon. The SPA and Assignment told lies about the underlying sale and purchase transaction, particularly in respect of the fact and manner of payment of the consideration, but it was not established that those documents told a lie about the circumstances of their making. The SPA was executed by Lai and by Shum on Chan’s behalf, pursuant to a power of attorney created and attested by the appellant, granted by Chan to Shum, authorising him to execute the agreement, witnessed by a clerk in the appellant’s firm and accepted for registration by the Land Registry with the appellant signing the attached memorial.[84] Similar findings were made in relation to the Assignment which was also verified by the appellant and registered.[85] It follows that the falsity within the meaning of section 69(a)(vii) properly construed was not made out and the appellant’s convictions under sections 73 and 74 must be quashed. 90. The complications and difficulties in the present case flow from the prosecutorial decision to charge the forgery-related offences under CO sections 73 and 74 rather than bringing more straightforward charges which might well have led to a different outcome. It must not be thought that in holding that the offences charged were not proved, this judgment in any way condones schemes of the type that was hatched and carried out in the present case. II. SUBSTITUTION OF A CONVICTION FOR ANOTHER OFFENCE 91. On the footing that the appellant’s conviction under section 73 must be quashed because the instrument was not “false”, the question arises as to whether a conviction for some other offence should be substituted. CO section 74 does not require separate treatment as it raises identical issues regarding use of a copy of a false instrument. 92. The prosecution has submitted that substitution of a conviction for conspiracy to defraud at common law or for fraud contrary to section 16A of the Theft Ordinance (“TO section 16A”) would be appropriate. It is therefore necessary to examine the principles which govern such substitution and how they should be applied in the present case. J. The statutory provisions on substitution 93. The Court of Appeal’s power to substitute a conviction for an alternative offence is contained in section 83A of the Criminal Procedure Ordinance (“CPO section 83A”)[86] which states: “(1) This section applies on an appeal against conviction, where the appellant has been convicted of an offence and the jury could on the indictment have found him guilty of some other offence, and on the finding of the jury it appears to the Court of Appeal that the jury must have been satisfied of facts which proved him guilty of the other offence. (2) The Court of Appeal may, instead of allowing or dismissing the appeal, substitute for the verdict found by the jury a verdict of guilty of the other offence, and pass such sentence in substitution for the sentence passed at the trial as may be authorized by law for the other offence, not being a sentence of greater severity.” 94. This Court’s power to exercise a like power of substitution flows from section 17(2) of the Court’s statute.[87] 95. CPO section 83A must be read together with section 51(2) of the same Ordinance (“CPO section 51(2)”)[88] which lays down the conditions for returning an alternative verdict of guilty against an accused at the trial: “If on the trial of any information, charge or indictment for any offence other than treason it is proved that the accused is not guilty of that offence but the allegations in the information, charge or indictment amount to or include, whether expressly or by implication, an allegation of another offence falling within the jurisdiction of the court of trial, he may be found guilty of that other offence or of an offence of which he could be found guilty on an information, charge or indictment specifically charging that other offence.” 96. Section 83 of the District Court Ordinance[89] (“DCO section 83”) provides for the application (inter alia) of CPO section 83A (and so necessarily also CPO section 51(2)) to trials before a judge sitting without a jury in the District Court: “Sections 80 to 83Y inclusive of the Criminal Procedure Ordinance (Cap 221) shall apply to criminal proceedings in the Court with such verbal alterations and modifications not affecting the substance thereof as may be necessary to render the same conveniently applicable, and, in particular, any reference to ‘indictment’ shall be understood to refer to a charge sheet, and any reference to ‘jury’ shall be understood to refer to a judge as a judge of fact.” K. The principles governing substitution K.1 Two conditions and a discretion 97. It is clear that the power to order substitution is discretionary. CPO section 83A provides that the Court of Appeal may order substitution.[90] But before it considers exercise of that discretion, two conditions laid down by CPO section 83A(1) must be satisfied. They impose constraints on the exercise of the power, requiring it to be apparent to the Court of Appeal that: (a) “the jury could on the indictment have found him guilty of some other offence” (“the first condition”); and (b) “on the finding of the jury ... the jury must have been satisfied of facts which proved him guilty of the other offence” (“the second condition”). K.2 The first condition: the link between CPO section 83A and CPO section 51(2) 98. As noted above, CPO section 83A must be read together with CPO section 51(2). The words of the first condition which require the case to be one where “the jury could on the indictment have found [the accused] guilty of some other offence” take their context and meaning from that provision. 99. CPO section 51(2) only allows such an alternative conviction if “the allegations in the information, charge or indictment amount to or include, whether expressly or by implication, an allegation of another offence”. It is only in such cases that the Court of Appeal might consider it apparent that “the jury could on the indictment” have found the accused guilty of some other offence, satisfying the first condition. I shall refer to the offence for which the accused has been found not guilty as “the original offence” and the offence potentially to be substituted as either “the other offence” or “the alternative offence”. 100. The limits on the power of substitution cater for the elementary proposition that “accused persons come to court to contest the charge which is in the indictment. They may be put at a disadvantage by finding during the trial that they have to meet an allegation that they have committed another offence.”[91] The restrictions safeguard the accused’s right to a fair trial. K.3 The requirements of the first condition 101. The following principles are established in connection with the first condition. It is concerned with “the allegations in the information, charge or indictment”. For brevity, I will simply refer to “the charge”. The court (whether at first instance or on appeal) therefore looks at the content of the charge as laid and not at the evidence in deciding whether the first condition is met.[92] 102. As CPO section 51(2) provides, substitution may be considered only if the allegations charging the original offence “amount to or include, whether expressly or by implication, an allegation” of the proposed alternative offence. As Lord Roskill pointed out[93] four possibilities are envisaged, ie, that the allegation in the original charge expressly or impliedly includes or amounts to an allegation of the proposed alternative offence. 103. The allegation of another offence is express where, for instance, that other offence is specifically charged as an alternative. Thus, for example, in Hau Tung Ying v HKSAR,[94]the accused were charged with conspiracy to defraud and, in the alternative, with theft. The Court held that conviction on the conspiracy charge could not be sustained because of a material irregularity, but held that the first condition of CPO section 83A was met since allegations of theft were expressly made in the charge.[95] Referring to the first condition, Chan PJ explained: “... the substituted verdict must be one for an offence of which ‘the jury could on the indictment have found him (the accused) guilty’; that is to say, it must be for an offence which is open on the indictment. This is often the case where the substituted offence is laid as an alternative charge in the indictment or where there are alternatives available under the law to the convicted offence. Normally, these would be lesser offences, such as manslaughter in a case of murder or theft in a case of robbery, but that is not necessarily the case and would require consideration of the nature of both the convicted and substituted charges in question and the alleged facts sought to be proved in support of either charge.”[96] 104. Cases where an alternative verdict is available under the law include those falling within the list of Theft Ordinance offences specified by TO section 32 and the TO Schedule.[97] 105. One approach to assessing whether the first condition is satisfied by an express allegation of the other offence involves what was called “the red pencil test” in R v Lillis.[98] That was a case where the accused was “arraigned on an indictment charging him with burglary in entering a building as a trespasser and stealing therein”.[99] It stated as particulars of the charge that the accused “having entered as a trespasser part of a building, ... stole therein a rotary grass mower”.[100] That charge failed because the evidence showed that he had been given permission to enter and so was not a trespasser. However, it was held that he could properly be convicted of theft of the lawn mower since, having struck out from the charge references to entering as a trespasser, there expressly remained in the charge as laid the allegation that he had stolen the object in question. As Lawton LJ explained: “If what is left after striking out all the averments which have not been proved leaves particulars of another offence within the jurisdiction of the court of trial which the accused can then and there defend, the judge can and should ask the jury to consider whether that other offence has been proved.”[101] 106. Less clear-cut are occasions when the allegations in the charge are said impliedly to amount to or include an allegation of the alternative offence. In Graham, Lord Bingham CJ held that: “A count charging offence A impliedly contains an allegation of offence B if the allegation in the particular count would ordinarily involve an allegation of offence B and on the facts of the particular case did so.”[102] 107. As Chan PJ pointed out, the alternative offence is often a lesser offence. Indeed, as Lord Roskill noted, “... if A includes B, it must be because A is sufficiently comprehensive to include B”.[103] And in Graham,[104] Lord Bingham CJ pointed out that the power to substitute has “usually been exercised in relation to offences of violence or public order offences by substituting a lesser offence for the offence charged, there being in such instances a clear hierarchy of offences at common law or by statute.” 108. Thus, examples of possibly (but obviously not necessarily) complementary pairs of greater and lesser offences might include murder/manslaughter; rape/indecent assault or some other sexual offence; assault occasioning actual bodily harm/common assault; robbery/theft; burglary/theft; theft/handling stolen goods;[105] and riot/unlawful assembly. 109. It is evident that in considering whether the allegations in the charge meet the first condition, it is necessary to examine the nature and elements of the offence charged as well as the particulars alleged, to decide whether the allegations in the original charge expressly or impliedly amount to or include an allegation of the proposed alternative offence, taking account of its elements. K.4 The requirements of the second condition 110. The second condition for substitution is that “on the finding of the jury ... the jury must have been satisfied of facts which proved him guilty of the other offence”. The words “on the finding of the jury” are important. An appellate court can only substitute an alternative conviction, if it appears that the jury must actually have made findings sufficient to establish such guilt. Thus, in Hau Tung Ying,[106] Chan PJ cited with approval Widgery CJ’s statement in R v Deacon,[107] that: “The basis of the power to substitute a verdict for a different offence must … be based on the finding of the jury. It is only when it appears to the court from the finding of the jury that the facts essential to establish the alternative offences were proved, that the court may substitute the alternative verdict.” 111. As Chan PJ pointed out:[108] “... the court’s power under s.83A is quite different from the power to apply the proviso under s.83 of the same Ordinance which is governed by different considerations.” 112. In the same case, Lord Hoffmann NPJ[109] (with whom all the other members of the Court also agreed) highlighted the difference between the two exercises, citing Lord Mustill in Moses v The State,[110] as follows: “The question whether an appellate court should use the proviso to uphold the verdict which the jury actually returned is quite different from the issue whether the court should substitute a verdict which the jury did not return. The former is a question of assessing the evidence; the latter involves an assessment of the verdict ...” 113. Thus, the proviso in CPO section 83 lays it down: “... that the Court of Appeal may, notwithstanding that it is of opinion that the point raised in the appeal might be decided in favour of the appellant, dismiss the appeal if it considers that no miscarriage of justice has actually occurred.”[111] 114. Its effect has been reiterated in many cases. Thus, in Tam King Hon v HKSAR,[112] Li CJ stated: “The test for the application of the proviso is whether a reasonable hypothetical jury, properly instructed, would, on the evidence, without doubt convict or would inevitably come to the same conclusion.” 115. Application of the proviso therefore involves the Court’s assessment of the evidence, asking whether a hypothetical jury would have undoubtedly convicted if properly directed. But in deciding whether conviction of an alternative offence should be substituted, the Court examines whether the jury sitting in the case did actually make the findings necessary to prove the accused guilty of that alternative offence. 116. Moreover, for the appellate court to find it apparent “that the jury must have been satisfied of facts which proved him guilty of the other offence” the jury’s finding has to be soundly based. As Chan PJ emphasised: “To make a finding upon which an accused can be convicted there must be a factual and legal basis for such finding: it must be based on properly admitted evidence and upon a correct direction on the law. If there was no evidence to support the finding or the jury’s finding was based on inadmissible evidence, or they were misdirected on the law in respect of the substituted offence, the finding was simply not a proper finding of fact upon which the appellate court can convict. In such case, there is no room for the application of s.83A.”[113] 117. Lord Hoffmann NPJ went on to highlight certain differences arising where the appellate court is concerned (as in the present case) with the Reasons for Verdict of a Judge after trial without a jury. In the case of a jury trial his Lordship pointed out: “As the question is what facts the jury must have found, and the jury gives no reasons for its decision, the application of s.83A requires an analysis of the facts of which the jury had necessarily to be satisfied if they were to reach the verdict which they did, and then to consider whether those facts, or more usually a subset of those facts, constitute another offence of which he could have been convicted.”[114] 118. And in relation to the verdict of a Judge sitting alone: “In a trial by judge alone, however, the verdict can be penetrated a little further. The section itself gives no guidance on how it should be adapted to such a case. It is for this Court to do so. The Court is in my opinion entitled to consider the reasoning by which the judge arrived at his conclusion and treat the facts which were essential to his conclusion that the appellant was guilty as facts which can form the basis of a substituted conviction. But I wish to guard against any suggestion that the court could substitute a conviction on the basis of facts which the judge said he found but which were not essential to his conclusion that the appellant was guilty. A verdict should not be substituted on the basis of an obiter dictum.”[115] 119. In Hau Tung Ying itself, the issue was whether alternative convictions for theft should be substituted for the failed prosecution for conspiracy to defraud. It was held that both CPO section 83A conditions were satisfied and the Court proceeded to exercise its discretion to order substitution. As noted above, offences of theft were expressly included as alternatives in the charge sheet, satisfying the first condition. The Judge also clearly made findings that theft had been established. He did not enter verdicts on those charges because he thought they were encompassed by the conviction for conspiracy to defraud. As Lord Hoffmann NPJ held: “What justifies a substitution in this case is that the judge (contrary to the submissions of prosecuting counsel, but in my view rightly) treated the theft allegations as overt acts of the conspiracy and treated his findings that they had committed the thefts as an essential basis of his finding that they were guilty of conspiracy. It is because the shape of the case took this very unusual form that I consider that the convictions of theft should be substituted.”[116] 120. Thus, it was clear from the Judge’s Reasons for Verdict that he had made findings establishing guilt for the alternative offences charged and that his findings were not merely obiter dicta but were an essential part of his decision to convict for conspiracy to defraud. K.5 The Discretion 121. As Chan PJ pointed out in Hau Tung Ying:[117] “If the conditions in s.83A are met, the appellate court has a discretion whether to substitute a conviction on the other offence for the conviction on the offence which it has quashed. This depends on the justice of the case, including on the one hand, the public interest in having a guilty person properly convicted and on the other hand, whether it is fair to the accused to convict him on the other offence.” 122. It is unsurprising that the discretion is narrowly circumscribed. As has been pointed out, defendants should obviously not be put at a disadvantage either by finding during the trial that they have to meet an allegation that they have committed an offence other than the ones charged or finding themselves convicted of an alternative offence upon which the jury has not been specifically directed or where that other offence has not properly been the focus of the judge’s decision to convict. 123. In many cases, the appellate court might consider ordering a retrial instead of substituting a verdict. It is noteworthy that the power to order a retrial is similarly circumscribed. Thus, CPO section 83E provides: “(1) Where the Court of Appeal allows an appeal against conviction and it appears to the Court of Appeal that the interests of justice so require, it may order the appellant to be retried. (2) A person shall not under this section be ordered to be retried for any offence other than— (a) the offence of which he was convicted at the original trial and in respect of which his appeal is allowed as mentioned in subsection (1); (b) an offence of which he could have been convicted at the original trial on an indictment for the first-mentioned offence; or (c) an offence charged in an alternative count of the indictment in respect of which the jury were discharged from giving a verdict in consequence of convicting him of the first-mentioned offence.” 124. The limits on the power to order a retrial set by CPO section 83E(2) as to the classes of offence on which an accused may be retried echo the rules constraining the power of substitution, balancing in this context fairness and the avoidance of double jeopardy against the need to bring guilty parties to justice. 125. The discretion to substitute is broad and allows the appellate court to cater for the defendant’s right to a fair trial without being confined to strictly legal considerations. As was pointed out in Spies v The Queen,[118] a decision of the Australian High Court cited in Hau Tung Ying, even where the equivalent of the two CPO section 83A conditions have been satisfied: “... the legal error may have put the appellant at some forensic, as opposed to legal, disadvantage. In such a case, it would be proper not to substitute a verdict.” L. The substitution principles applied 126. In my judgment, the submission that there should be substitution of a conviction for conspiracy to defraud or for fraud under TO section 16A must be rejected. The case for substitution falls at the first hurdle, failing to satisfy the first condition of CPO section 83A. It is unnecessary to consider the second condition and the question of discretion. L.1 Substituting conspiracy to defraud 127. As explained in the foregoing discussion, the Court must examine the allegations contained in the charges (rather than the evidence) to see whether they expressly or impliedly amount to or include either of the other offences advocated by the prosecution. I will begin by considering conspiracy to defraud. 128. As will be recalled, the essential features of the charges faced by the appellant are as follows. Charge 1, preferred under CO section 74, alleges that he used a copy of the SPA dated 4 June 2007, a false instrument known by him to be false, with the intention of inducing GE to accept it as genuine and, by reason of so accepting it, acting to its own or another’s prejudice. Charges 2 and 3, both laid under CO section 73, make similar allegations in respect of his using the SPA and the Assignment dated 4 June 2007 and 21 June 2007 respectively, with knowledge of their falsity and with the intention of inducing staff of the Land Registry to accept them as genuine and, by reason of so accepting them, acting to their own or another’s prejudice. 129. Plainly, the charges laid against the appellant do not “amount to or include, whether expressly or by implication, an allegation of” a conspiracy to defraud. As this Court held in Mo Yuk Ping v HKSAR:[119] “...the offence is constituted by becoming a party to an agreement with another or others to use dishonest means (a) with the purpose of causing economic loss to, or putting at risk the economic interests of, another; or (b) with the realization that the use of those means may cause such loss or put such interests at risk. The offence extends also to cases in which the dishonest means cause a person to act contrary to his public duty, this being a class of case which requires no further consideration by the Court in the circumstances of this case.” 130. The present charges do not allege any agreement between the appellant and any other person, let alone an agreement with the abovementioned characteristics. So the first condition is not met. The Judge could not, on the charges laid, have found him guilty of conspiracy to defraud. L.2 Substituting fraud under TO section 16A 131. The elements of the CO section 73 offence charged,[120] as we have seen, consist of: (1) a person’s use of an instrument; (2) which is false; (3) which he knows or believes to be false; (4) with the intention of inducing somebody to accept it as genuine, and because of such acceptance, to act or omit to act to that person’s or someone else’s prejudice. The actus reus is use of the false instrument in (1) and (2); and the mens rea involves knowledge of falsity and the further intention in (3) and (4). 132. It is important to note that neither the charges laid, nor the elements of the CO section 73 offence, involve any allegation or requirement that use of the false instruments produced a prohibited result. The allegations stop at describing the accused’s act and intention. The offence under CO section 73 is committed without proof that anyone was in fact induced to act as intended by the accused. 133. The elements of the fraud offence, in contrast, do require as an essential element, the occurrence of a prohibited result. TO section 16A(1) materially states: “If any person by any deceit (whether or not the deceit is the sole or main inducement) and with intent to defraud induces another person to commit an act or make an omission, which results either— (a) in benefit to any person other than the second-mentioned person; or (b) in prejudice or a substantial risk of prejudice to any person other than the first-mentioned person, the first-mentioned person commits the offence of fraud ...” 134. Section 16A(1) therefore requires proof that the accused’s deceit induced the victim to act or omit to act with either of the results specified in paragraphs (a) and (b). TO section 16A(2)[121] makes it clear that the accused must have a corresponding mens rea, defining “with intent to defraud” as intending to induce the victim to act, etc, with either or both of the stated consequences. Furthermore, section 16A(3) defines “benefit” as “any financial or proprietary gain, whether temporary or permanent” and “prejudice” as “any financial or proprietary loss, whether temporary or permanent”. 135. It follows that, looking at the allegations in the CO section 73 charges (and not at the evidence), one must conclude that they do not “expressly or impliedly amount to or include” an allegation of a TO section 16A offence. They lack any allegation of the prohibited results which are an essential ingredient of the latter offence so that the first condition is not met. L.3 TO section 16A and dishonesty 136. That disposes of the proposal to substitute a conviction under TO section 16A. However, it was submitted on behalf of the appellant that another reason exists for holding that the first condition is not met in relation to the fraud offence. As the appellant was able to rely on the Court of Appeal’s decision in HKSAR v Ho Ka Keung[122] in support, that submission deserves to be substantively examined. 137. The appellant argues that whereas dishonesty is not required for proof of an offence under CO section 73, it is an essential ingredient of the TO section 16A offence. If that contention is correct, it would follow that a conviction for the latter offence cannot be substituted since it requires proof of dishonesty which forms no part of the forgery-related offences charged. For the reasons which follow, I am unable to accept that argument. 138. TO section 16A has its origins in the Report of the Law Reform Commission of Hong Kong on Creation of a Substantive Offence of Fraud[123] and enacts without alteration the provisions in the Draft Fraud Bill proposed by the Commission.[124] The question of whether dishonesty should be prescribed as a separate ingredient was raised as part of the consultation process and examined in detail in the Report. Having noted the inclusion of the requirement of an intent to defraud, the Commission concluded: “We have carefully considered the arguments for and against such an approach and have, by a majority, concluded that there should be no separate requirement of dishonesty.”[125] 139. “Dishonesty” is thus not mentioned as an ingredient in the offence-creating provisions of either CO section 73 or TO section 16A. Leaving aside the point previously made about the difference regarding a “result” ingredient as part of the actus reus in the fraud offence, the two offences can be seen to be similarly structured in that they both involve the accused practising a form of deceit (“any deceit” under TO section 16A, and knowing use of a false instrument to induce its acceptance as genuine in the CO sections), in order to get someone else to act to their prejudice (or in the case of fraud, alternatively to the benefit of the accused). Given such similarity, one might naturally be inclined to think that either both offences involve no separate ingredient of dishonesty or, if dishonesty is to be an implied requirement, that this should apply equally to both offences. Of course, as a matter of popular speech, both offences can quite properly be described as offences of dishonesty. But that is not the same thing as saying that as a matter of law they require proof of dishonesty as a technically independent element of each offence. 140. Existing authority supports the proposition that forgery offences including that under CO section 73 do not require proof of dishonesty. This was accepted by the Court of Appeal below, albeit without analysis.[126] The learned authors of Smith, Hogan & Ormerod’s Criminal Law[127] endorse that proposition relying on the judgment of Ackner LJ in R v Campbell[128] in which his Lordship recorded that one of the grounds of appeal was that the trial judge had erred in ruling that dishonesty was no longer a necessary element in the offence of forgery. His rejection of that ground of appeal was implicit since there is no discussion excluding dishonesty in the short reported judgment. His Lordship nevertheless held that all the ingredients of the offence had been established so that the aforesaid ground of appeal was necessarily rejected by implication. 141. With respect, there are good reasons for the Campbell approach. The Ordinance spells out the elements of CO section 73 in detail. They require proof of the accused’s use of a false instrument, knowing or believing it to be false with the intention of inducing somebody to accept it as genuine and for that reason, to act or omit to act to that person’s prejudice, without mentioning dishonesty. It would therefore be wrong to suggest (as the appellant unsuccessfully sought to do in Campbell) that dishonesty should be introduced as an independent element of the offence. Nothing in the section supports such a requirement. 142. The same arguments are germane to the offence created by TO section 16A. In line with the recommendations and Draft Bill prepared by the Law Reform Commission, that section also spells out the culpable mental state, defining with precision the meaning of “with intent to defraud” in section 16A(2): “For the purposes of subsection (1), a person shall be treated as having an intent to defraud if, at the time when he practises the deceit, he intends that he will by the deceit (whether or not the deceit is the sole or main inducement) induce another person to commit an act or make an omission, which will result in either or both of the consequences referred to in paragraphs (a) and (b) of that subsection.” 143. There is no mention of dishonesty. As indicated above, one can well see that in general parlance, the offence may properly be described as involving dishonesty by the accused, but this does not justify the court in introducing a requirement of proof of “dishonesty” as a separate element of the offence. 144. That is what the Court of Appeal did in Ho Ka Keung.[129] It was an appeal against a conviction under TO section 16A on facts which are not presently relevant. One of the grounds of appeal was that the trial judge had “applied the wrong test in relation to dishonesty”, which implied that dishonesty was a necessary ingredient of the offence. In dealing with that ground, Yuen JA set out the elements of the offence as enacted in section 16A(1) and (2)[130] and continued as follows: “It would be noted that the word ‘dishonestly’ does not appear as a specific element in s 16A, unlike in s 2 (theft), s 17 (obtaining property by deception), s 18 (obtaining pecuniary advantage by deception), or s 18A (obtaining services by deception) and other offences. However, ‘deceit’ incorporates the element of dishonesty and the judge was well aware of the requirement that the prosecution must prove that the Defendant had been dishonest ...”[131] 145. Thus, although her Ladyship noted that, unlike the other TO offences listed, section 16A did not specify dishonesty as an element, she held that it was a necessary element because “deceit incorporates the element of dishonesty” (without citing any authority). She rejected the aforesaid ground of appeal holding that the judge had been well aware of and had properly dealt with that requirement. Wright J, concurring, held that the Judge was required to apply “the appropriate test for dishonesty in accordance with the well-known decision in R v Ghosh (1982) 75 Cr App R 154.”[132] He was satisfied that the Judge “was alive to the Ghosh test”[133] and had not applied some erroneous test. 146. With respect, the Court of Appeal’s decision on this point cannot be supported. When the TO creates an offence requiring proof that the accused acted dishonestly, it expressly so prescribes. It has done this in respect of at least twelve different offences.[134] Its approach to section 16A is different. It does not mention “dishonesty” but, as we have seen, prescribes and defines in detail the culpable mental state required, consisting of the accused’s “intent to defraud”. Thus, construing TO section 16A in the context of the Ordinance as a whole, there is no justification for concluding that the legislative intent was impliedly to incorporate an element of dishonesty as part of the offence. 147. The suggestion that “deceit incorporates the element of dishonesty” is, with respect, untenable. “Deceit” and “dishonesty” are entirely different and independent concepts. Thus, the classic definition of “deceit” is provided by Buckley J in In Re London And Globe Finance Corporation, Limited,[135] stating: “To deceive is, I apprehend, to induce a man to believe that a thing is true which is false, and which the person practising the deceit knows or believes to be false.” The application of this definition in the context of conspiracy to defraud was accepted in Scott v Metropolitan Police Commissioner.[136] 148. “Dishonesty” addresses a wholly different matter. As Sir Anthony Mason NPJ noted in Mo Yuk Ping v HKSAR,[137] it “describes a state of mind”. He pointed out[138] that the House of Lords in Scott,[139] when considering the offence of conspiracy to defraud, emphasised the element of dishonesty while rejecting deceit as an indispensable element of that offence. This highlights the difference between the two concepts. 149. Sir Anthony Mason NPJ noted that the Ghosh test for dishonesty represents the law in Hong Kong at present,[140] expressing the two-stage test as follows:[141] “In determining whether the prosecution has proved that the defendant was acting dishonestly, a jury must first of all decide whether according to the ordinary standards of reasonable and honest people what was done was dishonest. If it was not dishonest by those standards, that is the end of the matter and the prosecution fails. If it was dishonest by those standards, then the jury must consider whether the defendant himself must have realised that what he was doing was by those standards dishonest. In most cases, where the actions are obviously dishonest by ordinary standards, there will be no doubt about it. It will be obvious that the defendant himself knew that he was acting dishonestly. It is dishonest for a defendant to act in a way which he knows ordinary people consider to be dishonest, even if he asserts or genuinely believes that he is morally justified in acting as he did ...” 150. It is therefore obvious that “deceit” and “dishonesty” are quite distinct concepts. No basis exists for the proposition that “‘deceit’ incorporates the element of dishonesty”, thereby importing the Ghosh test into TO section 16A. Ho Ka Keung should not be followed on this point. M. Disposition of this appeal 151. For the aforesaid reasons, I would allow the appeal and quash the appellant’s convictions for offences under CO sections 73 and 74. I would also reject the submission that substitution of an alternative offence pursuant to CPO section 83A would be appropriate. Mr Justice Fok PJ : 152. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Lam PJ : 153. I agree with the judgment of Mr Justice Ribeiro PJ. Lord Phillips of Worth Matravers NPJ : 154. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Cheung : 155. The Court unanimously allows the appeal and quashes the appellant’s convictions. It also declines to substitute alternative convictions in the present case. Mr David Perry QC, Ms Maggie Wong SC and Ms Karry Lau, instructed by Chui & Lau, for the Appellant Ms Maggie Yang, DPP, Mr Derek Lau, SADPP (Ag) and Ms Sabra Lo, SPP, of the Department of Justice, for the Respondent [1] Cap 200. [2] Cap 210. [3] Reasons for Verdict (“RFV”) [2019] HKDC 611 (29 March 2019). [4] Yeung VP, Pang JA and Anthea Pang J [2021] HKCA 7 (7 January 2021). [5] Ribeiro and Lam PJJ, Stock NPJ [2021] HKCFA 43. [6] RFV§1. [7] RFV§§2 and 3. [8] In Section D of this judgment. [9] RFV§§207 and 208. [10] (2021) 24 HKCFAR 349. [11] Ibid at §3. [12] Details of the Small House Policy are also set out in Best Sheen Development Ltd v Official Receiver [2001] 1 HKLRD 866; and HKSAR v Lau Kam Ying, CACC 422/2010 (26 March 2013). [13] Clause 4 stipulated that the licensee should not, either directly or indirectly, reach any agreement or arrangement with anyone to convey the Ding House within a period of 5 years from the issuance of the Certificate of Compliance, without consent of the officers of the Lands Department. [14] Quotations in this judgment from the Judge’s Reasons for Verdict and the Court of Appeal’s Judgment are from translations of the original Chinese in which those decisions were delivered. [15] [2019] HKDC 611 (29 March 2019). [16] A District Court prosecution (DCCC 175/2013) of Shum and Lai for money laundering, in which the appellant testified as a defence witness: CA§6. [17] RFV§148. [18] RFV§165. [19] RFV§178. [20] Amended Prosecution Opening dated 10 September 2018, §§16, 20, 24, 26, 27, 29, 30(d) and 30(f). [21] Yeung VP, Pang JA and Anthea Pang J [2021] HKCA 7 (7 January 2021). [22] As Anthea Pang JA then was. [23] CA§33. [24] Ibid. [25] CA§34. [26] Stock VP, Yeung VP and Kwan JA, unreported, CACC 422/2010 (26 March 2013). [27] CA§43. [28] Appearing with Mr Derek Lau and Ms Sabra Lo. [29] Apparently coined by Professor Griew in his discussion of the English Law Commission’s Working Paper on Forgery: E Griew, [1970] Crim L R 548 at 549. [30] (1984) 79 Cr App R 76. [31] Ibid at 78. It was argued in an article by R Leng “Falsity in forgery” [1989] Crim L R 687 that Donnelly was not inconsistent with the automendacity approach. However, subsequent decisions basing themselves on Donnelly, have not viewed that decision in that light. It has, however, been pointed out in A-G’s Reference (No 1 of 2000) [2001] 1 Cr App R 218 and in Warneford and Gibbs 1994 WL 1061976 (English Court of Appeal Criminal Division, 28 April 1994), that subsequently in Lack (1987) 84 Cr App R 342, Lawton LJ puzzlingly appears to have left the question open whether section 9(1)(g) excluded the automendacity approach. [32] [1994] Crim L R 595. [33] [2001] 1 Cr App R 218. [34] Ibid at §24. Emphasis in the original. [35] Ibid at §25. [36] Stuart-Moore VP, Burrell and Barnes JJ [2007] 4 HKC 397. [37] Deputy District Judge DI Thomas. [38] [2007] 4 HKC 397 at §44. Emphasis in the original. In the subsequent case of HKSAR v Yeung Ko Moon, HCMA 479/2013 (15 November 2013), DHCJ Yau followed the approach of Stuart-Moore VP. [39] Section 9(1)(g) of the 1981 Act. [40] (1984) 79 Cr App R 76. [41] [2007] 4 HKC 397. [42] The Law Commission, (LAW COM. No. 55) Report on Forgery and Counterfeit Currency (17th July 1973). [43] CO sections 71, 73 and 74 creating the offences of forgery, using a false instrument and using a copy of a false instrument reproduce sections 1, 3 and 4 of the 1981 Act which in turn reproduce clauses 1, 3 and 4 of the Draft Bill. CO section 69(a)(vii) reproduces section 9(1)(g) of the 1981 Act which does the same in respect of clause 6(2)(a)(iii) of the Draft Bill. [44] LC Report §14. [45] Generally traced back to Kenny, Outlines of Criminal Law (14th ed, 1932) at 264. [46] LC Report§41. R v Dodge and Harris [1972] 1 QB 416 was the leading authority at the time of the LC Report. It involved use of bonds signed by a purported debtor putatively showing debts owed to the accused where such debts did not exist. The bonds were held not to be forgeries. [47] Enacted in the same terms by CO section 69(a)(vii) and section 9(1)(g) of the 1981 Act. [48] Paragraph 3 of the Explanatory Notes to clause 1, the offence-creating provision (p 53). [49] LC Report §42. [50] Ibid §43. [51] In his commentary on Donnelly [1984] Crim L R 490. Those being the days before the decision in Pepper v Hart [1993] AC 593, the learned Professor added: “The courts decline (in theory at least) to look at Parliamentary debates and there are strong arguments in support of this practice. There is something wrong. however, when a court can use as a step in its argument that a provision was intended to make new law when, in fact, that was not the intention of anyone concerned with its enactment.” [52] Set out in Section A above. [53] Also set out in Section A above. [54] Appearing with Ms Maggie Wong SC and Ms Karry Lau. [55] [1984] Crim L R 490 at 492. [56] Ibid. [57] Ibid. [58] Ibid. [59] [1987] 1 WLR 1578. [60] Our equivalent being section 69(a)(viii). [61] [1987] 1 WLR 1578at 1585. [62] Ibid. [63] Ibid. [64] Ibid. [65] 1994 WL 1061976 (English Court of Appeal Criminal Division (28 April 1994)). [66] Our section 69(a)(vii). [67] 1994 WL 1061976 (English Court of Appeal Criminal Division (28 April 1994)) at pp　5-6. [68] Ibid at p 6. Other examples were provided by the Law Commission including documents where the number or any distinguishing mark identifying the document is falsely stated in it (LC Report §41). As the Law Commission pointed out, such falsity has been found in “false documents such as passports, credit cards, railway season tickets or Cup Final tickets, as well as many items which it has already been found necessary to cover by specific legislation such as road fund and driving licences”. (LC Report §16). [69] Ibid at p 8. [70] [1994] Crim L R 753 at 755. [71] Ormerod and Laird, Smith, Hogan and Ormerod’s Criminal Law (OUP 16th Ed) Ch 29 at p 8. [72] Ibid, footnotes omitted. [73] Property Offences (Sweet & Maxwell, 1994) at §23-16, §23-23. [74] [2001] 1 Cr App R 218 at §24. [75] Stuart-Moore VP, Woo and Stock JJA [2001] 4 HKC 643. [76] Stuart-Moore VP, Burrell and Barnes JJ [2007] 4 HKC 397. [77] Cap 115. [78] [2001] 4 HKC 643 at 657. [79] [2007] 4 HKC 397 at §§42 and 43. [80] [2007] 4 HKC 397. [81] [2001] 4 HKC 643. [82] [2001] 1 Cr App R 218. [83] Section D.1 above. [84] RFV§§24, 29 and 33. [85] RFV§§34-36. [86] Cap 221. This section is derived from section 3 of the Criminal Appeal Act 1968 which is materially in the same terms, so that English authorities on that provision are of assistance. [87] Hong Kong Court of Final Appeal Ordinance (Cap 484), section 17(2): “For the purpose of disposing of an appeal, the Court may exercise any powers of the court from which the appeal lies (including the power to order a retrial), or may remit the case to that court.” [88] This is materially the equivalent of section 6(3) of the Criminal Law Act 1967 in England. [89] Cap 336. [90] As recognised by this Court in Hau Tung Ying v HKSAR (2011) 14 HKCFAR 453 at §68. Similarly, in relation to the English equivalents, see R v Graham [1997] 1 Cr App R 302 at 312-313. [91] R v Lillis [1972] 2 QB 236 at 241-242. [92] R v Graham [1997] 1 Cr App R 302 per Lord Bingham CJ at 313. [93] Commissioner of Police of the Metropolis v Wilson [1984] AC 242 at 258, in connection with section 6(3) of the Criminal Law Act 1967 which is directly comparable. [94] (2011) 14 HKCFAR 453. [95] This authority is further discussed in Section K.4 below in relation to the second condition, where the situation in that case was described as unusual. [96] (2011) 14 HKCFAR 453at §63. [97] Section 32: “If on the trial of any information, charge or indictment for an offence specified in the first column of the Schedule it is proved that the accused is not guilty of that offence but guilty of one of the offences specified opposite thereto in the second column of that Schedule or of attempting or being a party to an offence so specified, the accused shall be acquitted of the offence originally charged and shall be convicted of such other offence or of attempting or being a party to such other offence and be liable to be punished accordingly.” [98] [1972] 2 QB 236, a decision of a five-judge English Court of Appeal. [99] Ibid at 237. [100] Ibid at 239. [101] Ibid at 241-242. [102] R v Graham [1997] 1 Cr App R 302 at 313. [103] Commissioner of Police of the Metropolis v Wilson [1984] AC 242 at 259. [104] [1997] 1 Cr App R 302 at 313. [105] However, as pointed out above, alternative verdicts for TO offences are specifically provided for under TO section 32 and the TO Schedule. [106] (2011) 14 HKCFAR 453 at §64. [107] [1973] 1 WLR 696 at 699. [108] (2011) 14 HKCFAR 453 at §65. [109] Ibid at §116. [110] [1997] AC 53 at 68. [111] CPO section 83. [112] (2006) 9 HKCFAR 206 at §40. [113] Hau Tung Ying v HKSAR (2011) 14 HKCFAR 453 at §67. [114] Ibid at §117. [115] Ibid at §119. [116] Ibid at §120. [117] Ibid at §68. [118] (2000) 201 CLR 603 at §48, in the joint judgment of Gaudron, McHugh, Gummow and Hayne JJ, applying section 7(2) of the Criminal Appeal Act 1912 (NSW) which contains provisions directly comparable to CPO section 83A (although no equivalent to CPO section 51(2) was mentioned). [119] (2007) 10 HKCFAR 386 at §40, per Sir Anthony Mason NPJ. [120] Those of CO section 74 being materially the same. [121] Set out in Section L.3 below. [122] [2009] 1 HKC 61 (CA). [123] July 1996. [124] Ibid Annexure 4, Clause 2. [125] Ibid §§5.31 and 5.32. [126] CA§51. [127] (OUP 16th Ed), Additional Chapter 29, §29.1.4.3. [128] (1985) 80 Cr App R 47. [129] [2009] 1 HKC 61 (CA) by the majority judgments of Yuen JA and Wright J. Tang VP (as he then was) dissented on different grounds. A similar approach had been taken in the earlier decision in HKSAR v Yeung Hon Keung Larry [2007] 4 HKC 397, Burrell J writing for the Court of Appeal. [130] Ibid at §36. [131] Ibid at§37. [132] Ibid at §98. [133] Ibid at§99. [134] Theft (ss 2, 3 and 9); abstracting of electricity (s 15); dishonest use of public telephone system (s 16); obtaining property, pecuniary advantage or services by deception (ss 17, 18 and 18A); evasion of liability by deception (s 18B); making off without payment (s 18C); procuring entry in certain records by deception (s 18D); false accounting (s 19); suppression, etc of documents (s 22) and handling stolen goods (s 24). [135] [1903] 1 Ch 728 at 732. [136] [1975] AC 819 at 835-836. The House of Lords found it necessary to explain that the relationship between fraud and deceit referred to elsewhere in Buckley J’s judgment was not an exhaustive statement of the law but Buckley J’s definition of deceit was unaffected. [137] (2007) 10 HKCFAR 386 at §45. [138] Ibid at §42. [139] [1975] AC 819. [140] (2007) 10 HKCFAR 386 at §51. [141] Ibid at §46, citing Ghosh [1982] QB 1053 at 1064. 
Chief Justice Ma: 1. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Mr Justice Ribeiro and Mr Justice Fok PJJ: 2. These appeals are brought by members of two families. In each case, the 1st appellant is a mother who is a foreign national who has no right of abode and no right to enter or remain in Hong Kong. The other appellants in each case are their respective minor children who have either Hong Kong resident (“HKR”) or Hong Kong permanent resident (“HKPR”) status. Each of the mothers has applied for a further extension of permission to remain in Hong Kong to take care of the children concerned. Their applications were refused by the Director of Immigration (“the Director”) on the basis that they do not fall within any of the categories recognised under his immigration policy and that no exceptional circumstances exist to justify an extension on humanitarian or compassionate grounds. 3. The appellants challenge the Director’s refusal arguing that, in reaching his decision, he wrongly failed to take into account and give effect to a series of rights under the Basic Law,[1] the International Covenant on Civil and Political Rights (“ICCPR”),[2] the International Covenant on Economic, Social and Cultural Rights (“ICESCR”),[3] the Convention on the Rights of the Child (“CRC”)[4] and the best interests of the child principle at common law (collectively “the asserted rights”). 4. The appellants’ challenge failed before Au J[5] and the Court of Appeal,[6] both Courts holding that, by virtue of the immigration reservation contained in section 11 of the Hong Kong Bill of Rights Ordinance (“HKBORO”),[7] which enjoys constitutional status by virtue of Article 39 of the Basic Law, the asserted rights are not engaged. 5. Leave to bring the present appeal was granted by the Appeal Committee[8] which held that the following questions of law involved in the appeal are of the requisite general or public importance, namely: When a parent applicant, who is a foreign national with no right of abode in Hong Kong, makes an application for permission to remain in Hong Kong to take care of his/her minor child, who is a HKSAR permanent resident having the right of abode in Hong Kong, whether the Director of Immigration in considering the application is obliged, as a matter of law, to take into account the parent-and-child family’s enjoyment of applicable fundamental rights while living in Hong Kong. (Question 1) Whether section 11 of the Hong Kong Bill of Rights Ordinance (Cap 383) exempts the immigration authorities from having to take into account rights protected under the Basic Law of a child member of the family, when decisions are made by immigration authorities under immigration legislation in respect of the non-Hong Kong resident family member impacting on the integrity of the family and the enjoyment of family life in Hong Kong. (Question 2) A. The facts A.1 The Comilang appeal 6. The 1st appellant, Ms Milagros Tecson Comilang (“Ms Comilang”), is a national of the Philippines and came to Hong Kong in 1997 as a foreign domestic helper. After her contract of employment came to an end in July 2005, she married Mr Ahmed, a Pakistani national, on 7 October 2005. Mr Ahmed had acquired HKPR status in 2004. On 2 February 2006, Ms Comilang gave birth to a daughter, Zahrah, the 2nd appellant, whose status as a HKPR was established at birth. 7. On 10 October 2005, three days after her marriage to Mr Ahmed and prior to Zahrah’s birth, Ms Comilang’s permission to stay expired. She thereupon applied to the Director for a change of status to enable her to remain in Hong Kong as Mr Ahmed’s dependant. However, it was discovered that Mr Ahmed was still married to another woman in Pakistan when he purported to marry Ms Comilang. On 8 May 2007, Mr Ahmed withdrew his support for her change of status application and Ms Comilang left him for good on 4 June 2007, keeping Zahrah with her. 8. Between September 2007 and April 2011, a number of applications by Ms Comilang for extensions of stay to take care of Zahrah were rejected by the Director. She succeeded in judicial review proceedings against those rejections when the Court of Appeal held that the Director had adopted a wrong approach to her applications for extension.[9] However, on reconsideration, the Director again refused her application for an extension of stay, finding that it did not fall within any of the recognised categories under the Department’s immigration policy and that no exceptional circumstances justified an extension on humanitarian or compassionate grounds. As noted above, Ms Comilang’s judicial review proceedings against this second refusal failed in the Courts below. A.2 The Luis appeal 9. The 1st appellant, Ms Desiree Rante Luis (“Ms Luis”), is also a national of the Philippines and came to Hong Kong as a foreign domestic helper in 1991. She married Mr Luis, a fellow Filipino and foreign domestic helper, in Hong Kong in 1997. The 2nd, 3rd and 4th appellants (David, Carl and Mark) are their sons, born in Hong Kong in June 2002, August 2004 and December 2009 respectively, all thus being minors. 10. In April 2006, after David and Carl were born, Ms Luis’s employment ended and she left Hong Kong on 3 June 2006. Between that date and January 2012, she made numerous trips to Hong Kong, entering as a visitor, and obtained many extensions of stay. Mark was born in Hong Kong in December 2009 on one of those visits. David acquired HKPR status in 2010 while Carl and Mark remain in Hong Kong as dependants of Mr Luis and have HKR status. There was evidence that Carl and Mark have medical conditions that require regular medical monitoring. 11. On 26 January 2012, the Director refused Ms Luis permission to stay to take care of her three sons living here. After a further application for an extension of stay was refused on 26 July 2013 on grounds similar to those applied to Ms Comilang, Ms Luis applied for judicial review of that decision in May 2014. As noted above, her application failed in the Courts below. B. The asserted rights B.1 The claims of the mothers and of the children 12. The constitutional, statutory and common law rights asserted by the appellants are set out below. They are relied on in two ways. 13. First, the 1st appellants, the mothers, contend that they are entitled to assert relevant rights directly on their own behalf as a basis for challenging the Director’s refusal of an extension of stay, or at least for contending that the Director is legally obliged to take those rights into account in deciding whether to grant them extensions of stay. 14. Secondly, the other appellants, the minor children (acting by their respective mothers as their next friends), assert relevant rights on their own behalf, contending that they confer an entitlement to have their mothers granted permission to stay in Hong Kong to take care of them, or at least that the Director is legally obliged to take the asserted rights into account in deciding whether to grant the extensions of stay sought by their mothers. B.2 The relevant provisions 15. Provisions material to the asserted rights comprise the following: B.2a Under the Basic Law[10] BL 4 The Hong Kong Special Administrative Region shall safeguard the rights and freedoms of the residents of the Hong Kong Special Administrative Region and of other persons in the Region in accordance with law. BL 24 1. Residents of the Hong Kong Special Administrative Region (“Hong Kong residents”) shall include permanent residents and non-permanent residents. 2. The permanent residents of the Hong Kong Special Administrative Region shall be: ... (4) Persons not of Chinese nationality who have entered Hong Kong with valid travel documents, have ordinarily resided in Hong Kong for a continuous period of not less than seven years and have taken Hong Kong as their place of permanent residence before or after the establishment of the Hong Kong Special Administrative Region; (5) Persons under 21 years of age born in Hong Kong of those residents listed in category (4) before or after the establishment of the Hong Kong Special Administrative Region; … 3. The above-mentioned residents shall have the right of abode in the Hong Kong Special Administrative Region and shall be qualified to obtain, in accordance with the laws of the Region, permanent identity cards which state their right of abode. 4. The non-permanent residents of the Hong Kong Special Administrative Region shall be persons who are qualified to obtain Hong Kong identity cards in accordance with the laws of the Region but have no right of abode. BL 37 The freedom of marriage of Hong Kong residents and their right to raise a family freely shall be protected by law. BL 39 (1) The provisions of the International Covenant on Civil and Political Rights, the International Covenant on Economic, Social and Cultural Rights, and international labour conventions as applied to Hong Kong shall remain in force and shall be implemented through the laws of the Hong Kong Special Administrative Region. (2) The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law. Such restrictions shall not contravene the provisions of the preceding paragraph of this Article. BL 41 Persons in the Hong Kong Special Administrative Region other than Hong Kong residents shall, in accordance with law, enjoy the rights and freedoms of Hong Kong residents prescribed in this Chapter. BL 154(2) The Government of the [HKSAR] may apply immigration controls on entry into, stay in and departure from the [HKSAR] by persons from foreign states and regions. B.2b Under the HKBORO[11] Section 11 As regards persons not having the right to enter and remain in Hong Kong, this Ordinance does not affect any immigration legislation governing entry into, stay in and departure from Hong Kong, or the application of any such legislation. B.2c Under the Bill of Rights (“BOR”) implementing the ICCPR BOR 14 [ICCPR 17] (1) No one shall be subjected to arbitrary or unlawful interference with his privacy, family, home or correspondence, nor to unlawful attacks on his honour and reputation. (2) Everyone has the right to the protection of the law against such interference or attacks. BOR 19(1)-(2) [ICCPR 23(1)-(2)] (1) The family is the natural and fundamental group unit of society and is entitled to protection by society and the State. (2) The right of men and women of marriageable age to marry and to found a family shall be recognized. ... BOR 20(1) [ICCPR 24(1)] (1) Every child shall have, without any discrimination as to race, colour, sex, language, religion, national or social origin, property or birth, the right to such measures of protection as are required by his status as a minor, on the part of his family, society and the State. B.2d Under the ICESCR ICESCR 10 The States Parties to the present Covenant recognize that: 1. The widest possible protection and assistance should be accorded to the family, which is the natural and fundamental group unit of society, particularly for its establishment and while it is responsible for the care and education of dependent children. ... B.2e Under the CRC CRC 3 (1) In all actions concerning children, whether undertaken by public or private social welfare institutions, courts of law, administrative authorities or legislative bodies, the best interests of the child shall be a primary consideration. (2) States Parties undertake to ensure the child such protection and care as is necessary for his or her well-being, taking into account the rights and duties of his or her parents, legal guardians, or other individuals legally responsible for him or her, and, to this end, shall take all appropriate legislative and administrative measures. B.2f The common law principle of the best interests of the child This is dealt with in Section E.3 below. C. The immigration reservation 16. The immigration reservation contained in section 11 of HKBORO set out above has a centrally important bearing on all the issues in this appeal. The constitutional status, scope and effect of that provision were closely examined in three decisions of the Court, namely, in Ubamaka v Secretary for Security;[12] GA v Director of Immigration;[13] and Ghulam Rbani v Secretary for Justice.[14] The legal effect of section 11 is thus well-established. However, because of its crucial importance in the present case, the analysis bears reiteration. C.1 Application of the immigration reservation to Hong Kong 17. When, on 20 May 1976, the United Kingdom Government ratified the ICCPR, extending it to its dependent territories including Hong Kong, it did so with a number of reservations, among them an immigration reservation which stipulated as follows: “The Government of the United Kingdom reserve the right to continue to apply such immigration legislation governing entry into, stay in and departure from the United Kingdom as they may deem necessary from time to time and, accordingly, their acceptance of article 12(4) and of the other provisions of the Covenant is subject to the provisions of any such legislation as regards persons not at the time having the right under the law of the United Kingdom to enter and remain in the United Kingdom. The United Kingdom also reserves a similar right in regard to each of its dependent territories.” 18. It is clear that, as a matter of international law, such a reservation is consistent with the rights of sovereign states to control the entry, residence and expulsion of aliens.[15] 19. As pointed out in Ubamaka:[16] “The extension to Hong Kong of the ICCPR subject to the immigration reservation therefore meant that the Hong Kong Government reserved mutatis mutandis the right to continue to apply such immigration legislation governing entry into, stay in and departure from Hong Kong as it might deem necessary from time to time and, accordingly, that its acceptance of art.12(4) and of the other provisions of the Covenant was subject to the provisions of any such legislation as regards persons not at the time having the right under the law of Hong Kong to enter and remain in Hong Kong.” 20. In the Joint Declaration executed on 19 December 1984 by the Central People’s Government and the United Kingdom Government regarding China’s resumption of the exercise of sovereignty over Hong Kong, it was agreed that the ICCPR “as applied to Hong Kong shall remain in force”.[17] 21. Accordingly, the Basic Law promulgated on 4 April 1990 (to come into effect on 1 July 1997) contained BL 39 set out above, providing that the ICCPR “as applied to Hong Kong shall remain in force and shall be implemented through the laws of the Hong Kong Special Administrative Region.” 22. Subsequently, on 8 June 1991, HKBORO was enacted. Its long title states that it is “An Ordinance to provide for the incorporation into the law of Hong Kong of provisions of the [ICCPR] as applied to Hong Kong; and for ancillary and connected matters”. Its section 11 (set out above) mirrors and gives effect to the immigration reservation as applied to Hong Kong. Referring to immigration legislation governing “entry into, stay in and departure from Hong Kong”, it also reflects BL 154(2) which authorises the HKSAR to exercise immigration “controls on entry into, stay in and departure from the [Region]”.[18] 23. HKBORO was subjected to the vetting process prescribed by BL 160. The Ordinance,[19] including section 11, was adopted as consistent with the Basic Law and thus as part of the law of the HKSAR with effect from 1 July 1997.[20] C.2 Incorporation of the BOR subject to section 11 24. Against this background, our courts have consistently held that by enacting HKBORO, the fundamental rights guaranteed by the BOR set out in Part II of the Ordinance, have not merely been incorporated domestically but, by virtue of BL 39, incorporated as part of the Basic Law and given constitutional effect. This is the consequence of BL 39(2) which protects the rights and freedoms of Hong Kong residents against restrictions which “contravene the provisions of the preceding paragraph of this Article”, such provisions being for present purposes the ICCPR as applied to Hong Kong and incorporated via HKBORO. 25. Thus, in HKSAR v Ng Kung Siu,[21] Li CJ noted that BOR 16, dealing with freedom of expression, is incorporated into the Basic Law by virtue of BL 39. In Shum Kwok Sher v HKSAR,[22] Sir Anthony Mason NPJ held that the BOR provisions should be interpreted in the same way as BL 39 and given a purposive and generous interpretation since their object is to guarantee the fundamental rights and freedoms of HKSAR residents. And in Swire Properties Ltd v Secretary for Justice,[23] Bokhary PJ stated that the whole of the BOR “is entrenched by art.39 of our constitution the Basic Law”. Similarly, in Ubamaka[24] it was recognised that HKBORO is given constitutional force by BL 39. Thus, by virtue of BL 39, the rights in the BOR take effect in the law of the HKSAR as constitutional guarantees. 26. By tracking the wording of the immigration reservation, the intention of section 11 is plainly to give effect to BL 39’s stipulation that the content of the ICCPR “as applied to Hong Kong” – in other words, as applied subject to the immigration reservation – is to be incorporated. In Gurung Kesh Bahadur v Director of Immigration,[25] Li CJ put this as follows: “The provisions of the [ICCPR] as applied to Hong Kong were implemented through the [HKBORO], which contains [the BOR]. That Ordinance effects the incorporation of the ICCPR as applied to Hong Kong into our laws [citing Shum Kwok Sher, and Ng Kung Siu].The ICCPR as applied to Hong Kong was subject to the reservation, originally made by the United Kingdom, that immigration legislation as regards persons not having the right to enter and remain could continue to apply. It is unnecessary to set out the terms of the reservation in full since it is reflected in [section 11 of HKBORO].” 27. In giving effect to the immigration reservation as part of the implementation process mandated by BL 39, section 11 lays down a specific exception limiting the scope of the BOR rights incorporated in the Basic Law. As Ma CJ points out in GA:[26] “The intention of s.11 is to except the applicability of the BOR to the aforesaid aspects of immigration control [i.e. entry into, stay in and departure from Hong Kong]. … … the provision is intended to except immigration legislation that deals with each stage of a person’s stay in Hong Kong, as stated earlier, from entry through his or her stay in Hong Kong, to departure.” C.3 Constitutional status of section 11 28. Section 11 therefore operates at the constitutional level. It excludes from the scope of the provisions of the BOR given constitutional effect by BL 39, immigration legislation governing entry into, stay in and departure from Hong Kong and the application of such legislation. 29. This is subject to the qualification identified in Ubamaka which is not presently relevant. The Court there held that, as a matter of construction, section 11 read together with section 5 of HKBORO, does not operate to exclude the protection of BOR 3 which confers an absolute and non-derogable right against torture and cruel, inhuman or degrading treatment or punishment. C.4 A consistent scheme of constitutional rights 30. While the limitation created by section 11 expressly addresses the scope of HKBORO, this Court has held that the scheme of constitutional rights laid down by the Basic Law, both in its Chapter III and in the BOR incorporated via BL 39, must be interpreted as a coherent whole, consistently with section 11. 31. Thus, in Rbani, a case involving detention of a non-resident person by the exercise of the Director’s powers to regulate termination of his stay in Hong Kong, it was held[27] that section 11 excludes such person’s reliance on BOR 5(1) which provides: “Everyone has the right to liberty and security of person. No one shall be subjected to arbitrary arrest or detention. No one shall be deprived of his liberty except on such grounds and in accordance with such procedure as are established by law.” 32. The appellant relied also on similar rights under BL 28 which protects freedom of the person and prohibits arbitrary or unlawful detention[28] via his reliance on BL 41 which provides: “Persons in the Hong Kong Special Administrative Region other than Hong Kong residents shall, in accordance with law, enjoy the rights and freedoms of Hong Kong residents prescribed in this Chapter.”[29] 33. Stressing that section 11 operates at the constitutional level, the Court held that the rights conferred on non-residents by BL 41 “in accordance with law” are to be understood as constitutional rights operating as a coherent scheme consistently with the immigration reservation. It would not be coherent for section 11, given constitutional force by BL 39, to exclude non-residents from relying on BOR 5(1) (as it clearly does) while construing BL 41 to permit such non-residents to rely on similar rights under BL 28 in the same excepted immigration context. 34. Thus, it was held in Rbani[30]that: “... in giving constitutional status to the Hong Kong Bill of Rights Ordinance including s.11, BL art.39 gives constitutional status to a specific exception to relevant provisions of the Hong Kong Bill of Rights in relation to persons not having the right to enter and remain in Hong Kong and in respect of immigration legislation governing entry into, stay in and departure from Hong Kong, or the application of such legislation. ... [That] exception precludes resort to the right to liberty and security of person under BOR art.5(1) by the aforesaid class of persons in relation to the application of the specified categories of immigration legislation. Giving effect to the words ‘in accordance with law’ in BL art.41, it is necessary to read BL art.28, which is concerned with freedom of the person in general, as subject to that specific exception provided for by s.11, given constitutional status by BL art.39.” 35. It was therefore held that the exception created by the immigration reservation, given constitutional force by BL 39, is not confined in its operation to rights in the BOR but extends by necessary implication to cognate rights in the Basic Law, requiring them to be interpreted consistently with section 11 as laying down a coherent scheme in the specified immigration context. D. The appellants’ case 36. The two-fold nature of the claims separately asserted by, on the one hand, the appellant mothers (1st appellants in each appeal) and, on the other hand, the appellant children (the 2nd and other appellants in each appeal), and the various rights asserted, are described in Sections B.1 and B.2 above. 37. It is the contention of the appellants, both the mothers and the children, that the Director was under a duty to take into account the various rights asserted by them in considering whether to grant an extension of stay to the mothers to remain in Hong Kong. The complaint of the appellants is that the Director did not start from the premise that they, both the mothers and the children, had these rights when making the decisions under challenge, namely the decisions to refuse to allow the mothers to remain in Hong Kong. It was the appellants’ case that the Director, being obliged to take those rights into account, should have given effect to them by recognising that his decision to refuse permission to the mothers to stay would infringe those rights. Therefore, it was contended, the Court should declare that such infringement of the appellants’ rights should be subject to a proportionality analysis conducted by the Director upon a remission of the challenged decisions to him. D.1 The rights asserted by the mothers D.1a The mothers’ rights under the BOR 38. So far as the rights contained in the BOR are concerned, section 11 of HKBORO excludes reliance on those rights by the mothers. That this was the effect of section 11 was not disputed by the appellants. This was a correct concession on their part. The rights relied upon were not of the character excluded from the operation of section 11 (see [29] above). As a matter of clear language, the specific exception of relevant provisions of the BOR provided by section 11 plainly applies to the appellant mothers. They are persons not having the right to enter and remain in Hong Kong and the Director’s decisions under challenge were made pursuant to the application of immigration legislation governing entry into, stay in and departure from Hong Kong. D.1b The mothers’ rights under the Basic Law 39. Faced with the obstacle of section 11 to reliance on rights under the BOR, the appellant mothers contended that their reliance on BL 37 was not excluded by section 11. In addition, they relied on section 2(5) of HKBORO to contend that the immigration reservation in section 11 could not restrict or derogate from rights arising under the Basic Law. Further, they relied on the decision of this Court in Gurung Kesh Bahadur v Director of Immigration[31] to support the proposition that rights arising under the Basic Law were not affected by any restriction or limit in HKBORO. D.1b(i) Scope of the Basic Law rights 40. The appellants’ reliance on BL 37 raises the question of the scope of the rights conferred by that article. The appellants’ contention was that the right “to raise a family freely” included the right to family unity. There is a question as to whether BL 37 is to be so interpreted and the Director disputed that interpretation contending instead that the right in BL 37 was a right to procreate freely and not to be subject to any policy in the PRC limiting the number of children which a couple may lawfully have. 41. Regardless of that, since the appellant mothers are not Hong Kong residents but only have visitor status here, their rights under Chapter III of the Basic Law are rights enjoyed pursuant to BL 41 and the scope and effect of those rights is qualified by section 11 which, as explained in Section C above (in particular the citation from Rbani at [97]), operates at the constitutional level. 42. Thus, the rights claimed by the appellant mothers under BL 37 through BL 41 are constitutionally subject, via BL 39, to the section 11 exception. The appellants were therefore forced to argue that Rbani was wrongly decided in this respect. We reject that submission. As pointed out in Section C.4 above and as decided in Rbani, it would be incoherent to hold that section 11, given constitutional effect by BL 39, excludes reliance on the relevant rights in the BOR (BOR 14, 19 and 20) while permitting reliance on cognate rights under BL 37 via BL 41. There is no indication that the drafters of the Basic Law were intending to give greater rights to challenge immigration decisions than were available under the provisions of the BOR prior to the Basic Law coming into effect on 1 July 1997. D.1b(ii) HKBORO section 2(5) 43. Section 2(5) of HKBORO, reflecting ICCPR 5(2),[32] provides: “There shall be no restriction upon or derogation from any of the fundamental human rights recognized or existing in Hong Kong pursuant to law, conventions, regulations or custom on the pretext that the Bill of Rights does not recognize such rights or that it recognizes them to a lesser extent.” 44. The appellants rely on section 2(5) to argue that, even if it excludes or limits any of the rights under the BOR, section 11 cannot affect rights under the Basic Law. The premise of this argument is that there are other relevant rights recognised or existing in Hong Kong pursuant to law (etc.) upon which the appellants can rely which are therefore not to be restricted or derogated from. Those rights, it is said, include BL 37. In this way, the appellants seek to insulate Basic Law rights from section 11. 45. We do not accept that argument. Section 2(5) and section 11 are both provisions in HKBORO and there is no reason to regard them as mutually inconsistent. What section 2(5) does is to prevent the BOR from derogating from any relevant “fundamental human rights recognized or existing in Hong Kong”. Properly construed, section 2(5) operates on the footing that such “recognized or existing rights” are rights which are qualified by section 11 given constitutional status by BL 39 in the specified immigration context. As discussed in Section C.4 above, the rights conferred by Chapter III of the Basic Law and by the BOR incorporated by BL 39 are to be interpreted as a coherent and consistent scheme. Section 2(5) is not to be given a construction bringing incoherence to that scheme. 46. There is nothing in section 2(5) comparable to section 5 of HKBORO capable of negating the effect of section 11 by reference to any absolute and non-derogable rights as discussed in Ubamaka (see [29] above). In other contexts, unaffected by section 11, section 2(5) has an important role to play in protecting rights against any restriction or derogation. D.1b(iii) Gurung Kesh Bahadur v Director of Immigration 47. The appellants relied on Gurung Kesh Bahadur at [28] to contend that, since they were relying in this case on rights under the Basic Law that are separate from and additional to those contained in the BOR, section 11 could not operate to qualify any of those rights. We do not accept that the passage cited has any bearing on the issues in the present case. 48. It is important to note, however, that the respondent in Gurung Kesh Bahadur was a Hong Kong non-permanent resident who had been granted permission to stay and was still within his limit of stay when he sought to exercise his right to re-enter Hong Kong pursuant to that permission. As such, the Court held (at [40]) that the application of section 11(10) of the Immigration Ordinance[33] to such a non-permanent resident, whose permitted limit of stay had not expired, was inconsistent with his freedom of movement rights under BL 31.[34] The Court noted that section 11(10): “… continues validly to apply to persons who are not non-permanent residents with an unexpired limit of stay, for example, to visitors.” 49. Thus, the Court emphasised (at [42]) that the judgment in that case only concerned a non-permanent resident whose permitted limit of stay had not expired and that: “A person whose limit of stay has expired would cease to be a non-permanent resident and would be in a very different position. Further, as has been pointed out, as regards a non-permanent resident in the position of the respondent, the [Immigration] Ordinance contains powers for curtailing the period of his permitted limit of stay as well as for his removal and deportation.” 50. The appellants’ reliance on Gurung Kesh Bahadur cannot assist them because the appellant mothers are not Hong Kong residents and do not have any right to enter or remain in Hong Kong. Their positions are materially different to that of the respondent in Gurung Kesh Bahadur. D.2 The rights asserted by the children 51. The appellants’ contention, in respect of the appellant children, is that when the Director makes a decision concerning the non-HKR mothers, he has to take into account the family rights of the children, under both the BOR and also the Basic Law. 52. In considering this contention, it is important to keep in mind the nature of the decisions under challenge in this case. Here, the appellants are seeking to challenge immigration decisions governing entry into, stay in or departure from Hong Kong in respect of persons with no right to enter or remain in Hong Kong. Whether viewed from the perspective of the appellant children or the appellant mothers, one is still concerned with a decision not to allow a person to stay in Hong Kong who has no right to enter or remain in Hong Kong. D.2a The children’s rights under the Bill of Rights 53. As with the mothers, section 11 also bars the appellant children’s reliance on rights under the BOR. This is so both as a matter of construction of the language of section 11 and also as a matter of purposive construction of that section. 54. First, as a matter of plain language, section 11 precludes reliance on rights under the BOR (subject to the qualification noted in Ubamaka) so long as the impugned decision involves the application of immigration legislation governing entry into, stay in or departure from Hong Kong as regards a person not having the right to enter or remain. The exception in section 11 does not focus on who has the fundamental rights but rather on the content of the decision itself and to whom the decision specifically relates. 55. This construction was adopted in the judgment of Nazareth JA in In re Hai Ho Tak and Cheng Chun-heung,[35] where (at p.209) he held: “It seems to me also, that the meaning of s. 11 is quite clear: the Bill of Rights (as part of the Ordinance) does not affect immigration legislation governing the entry into, stay in and departure from Hong Kong of a person who does not have the right to enter and remain in Hong Kong; nor, more to the point, does the Bill affect the application of such legislation. That means not that members of the family of such a person (whether or not they have the right to enter and remain) do not have ‘family’ rights under the Bill, but that their family rights do not affect such legislation or its application in relation to that person; likewise that even if such family members have locus standi in proceedings such as this, their family rights would not avail in respect of the operation of such immigration legislation as regards that person.” (Emphasis added) 56. The appellants submitted that Hai Ho Tak was wrongly decided and should be overruled. This argument is addressed in Section D.4 below. 57. As a matter of purpose, we accept the Director’s submission that it would frustrate the evident purpose of section 11 and BL 39 if a person who has no right to enter and remain is able to circumvent that position by saying: “I’m relying on someone else’s rights”. An interpretation of the children’s rights to such effect would lead to incoherence in the constitutional scheme. The need for a coherent approach is especially cogent given the recognised necessity for strict and effective immigration control that has long been the policy adopted in Hong Kong. As noted by Ma CJ in his judgment in GA v Director of Immigration,[36] in the context of claims by asylum seekers to be entitled to work whilst in Hong Kong pending resettlement elsewhere (at [30]): “Particularly in a place like Hong Kong, which has considerable economic attractions to many people, the need to control immigration and to control the number or type of people who may wish to work here, can easily be seen.” D.2b The children’s rights under the Basic Law 58. Even if they are excluded from relying on rights under the BOR, the children appellants’ case proceeds on the basis that they can nevertheless rely on independent family rights arising under the Basic Law. These Basic Law rights are not, they contend, subject to the section 11 exception and so are rights which the Director is duty bound to take into account when exercising any discretion under the Immigration Ordinance in respect of the mothers which may have an effect on the children’s enjoyment of those family rights. 59. The correctness of this contention depends on the interpretation of the Basic Law and the particular rights relied upon. We shall address the appellants’ particular reliance on BL 24 below (in Section D.3). For the following reasons, we would reject the appellants’ contention so far as it was based on BL 37. 60. First, the approach to interpretation of the Basic Law is now well-established: see Ng Ka Ling v Director of Immigration (1999) 2 HKCFAR 4 at pp.28-29 and Vallejos v Commissioner of Registration (2013) 16 HKCFAR 45 at [76]-[77]. Provisions of the Basic Law are to be construed in the light of their context and purpose. The context of a provision of the Basic Law includes other provisions of the Basic Law and the provisions of the ICCPR as applied to Hong Kong. Context and purpose are to be considered in the first instance and not merely in the case of ambiguity. 61. Applying those settled principles, the Basic Law rights relied upon must be construed as a coherent whole together with BL 39 and section 11 which, for the reasons explained above, is given constitutional status. Although section 11 in terms only applies textually to the rights set out in HKBORO, by necessary implication it limits the application of cognate rights in the BL whether they are invoked directly or in connection with the enjoyment of another right (as further discussed in relation to BL 24 below). To seek to isolate section 11 would frustrate BL 39, especially when the family rights relied upon under the Basic Law (in particular BL 37) are also contained in the ICCPR (Articles 17, 23 and 24) and BOR (Articles 14, 19 and 20). 62. Secondly, there is a clear link between section 11 and BL 154(2) as reflected in the constitutional jurisprudence of this Court. The provisions were read together in Ubamaka at [102]. In GA v Director of Immigration (supra), Ma CJ held (at [29(1)]) that it was clear from section 11 that it was dealing with immigration control on entry into, stay in and departure from Hong Kong “as reflected in art.154(2) of the Basic Law”. And at [29(3)], Ma CJ further held that: “The intention of art.154(2) of the Basic Law and the Reservation, both of which are couched in general terms, must have been, one would have thought, for the purpose for enabling effective immigration control to be exercised.” See also, Rbani at [86], where these passages from GA were referred to with approval. 63. Contrary to the appellants’ submission that the geographical circumstances of Hong Kong are only relevant and to be considered at the stage of conducting a proportionality exercise to determine if an infringement of a constitutional right is justified, the linkage between BL 154(2) and section 11 demonstrates that, save for non-derogable rights, the drafters of the Basic Law thought it appropriate to exclude all other rights, be they in the BOR or the Basic Law, in the context of a decision relating to entry into, stay in or departure from Hong Kong by someone without the right to enter and remain. 64. Thirdly, the appellants’ argument entails accepting that the coming into effect of the Basic Law effected a radical change to the BOR. From 1991, when HKBORO was enacted, until 1 July 1997, the rights under the BOR did not apply in the immigration context falling within section 11 and the law before 1 July 1997 was reflected in Hai Ho Tak. There is no discernible basis for concluding that the intention of the drafters of the Basic Law was, through the coming into effect of the Basic Law, to impose new and stricter limits on the discretion of the Director of Immigration in making immigration decisions. Such a conclusion would fly in the face of the clear theme of continuity reflected in the Basic Law: see Secretary for Justice v Lau Kwok Fai & Another (2005) 8 HKCFAR 304 per Sir Anthony Mason NPJ at [35]. 65. Even on the appellants’ case, the family rights under BL 37 are no greater than the rights conferred under the BOR 14, 19 and 20. That being the case, given the constitutional status of section 11 through BL 39, it is untenable to contend that BL 37, viewed as part of a coherent scheme of rights, is not subject to the immigration reservation. In the circumstances, it is unnecessary to consider the Director’s alternative argument as to the scope of BL 37. D.3 BL 24 66. Faced with these difficulties in respect of reliance on Basic Law rights which reflect similar rights to those protected under the BOR, the appellants submitted that their reliance on BL 24 was different in that there is no parallel between the rights conferred by BL 24 and those conferred in the BOR. 67. BL 24 is concerned with the categories of persons who are qualified to be permanent residents of the HKSAR. BL 24(3) confers the substantive right of abode on such residents and their right to be issued with a permanent identity card stating their right of abode. BL 24 has no equivalent in the BOR since none of the provisions of the ICCPR are directly concerned with the right of abode or the categories of nationals and non-nationals who should be entitled to that right. 68. On this basis, the appellants contended that the refusal of entry to the mother appellants had the effect of interfering with the right of abode of the children appellants since it had or could have the practical consequence that the children, being dependent on their mothers, would not be able to exercise their right of abode because they would have to leave Hong Kong in order to be cared for by their mothers. 69. This novel argument in respect of BL 24 cannot be accepted. The rights conferred by BL 24 do not enable a permanent resident to require the Director to permit any other person to enter Hong Kong. As pointed out above, the decision being challenged is the Director’s decision refusing an extension of stay to the non-resident mothers, whether viewed from the perspective of the children or their mothers. While cast as an argument based on BL 24, the appellants’ case depends in reality on an asserted family unity right necessarily incidental to the enjoyment of the right of abode. The challenge both by the mothers and their children is to the Director’s exercise of his powers of immigration control covered by section 11 on the footing that they have a right (under BOR 14, 19 or 20 and BL 37) not to have their family relationship disrupted, the practical consequence of which is said to be the endangering of their children’s BL 24 right to permanent residence. As indicated above, by necessary implication, section 11 limits the application of rights in the BOR and cognate rights in the Basic Law (here the family unity rights) whether they are invoked directly or in connection with the enjoyment of another right (here BL 24). The appellants’ attempt to hermetically seal BL 24 from the BOR and the rest of the Basic Law, including BL 37, BL 39 and BL 154(2) must fail. D.4 In re Hai Ho Tak and Cheng Chun-heung 70. As noted above, the appellants argued that the Court of Appeal’s decision in Hai Ho Tak was wrong and that it should be overruled. Three reasons were advanced in support of this argument, namely:[37] (1) It was contrary to the principle that rights are to be construed generously and reservations narrowly: it amounted to reading the words “or their family members” into section 11 after “persons not having the right to enter and remain in Hong Kong”; (2) It failed to regard the HKPR and HKR family members as rights-holders in their own right and, contrary to the approach in Beoku-Betts v Secretary of State for the Home Department,[38] failed to consider family rights holistically and not separately; and (3) The appellants’ contrary construction of section 11 did not empty the section of meaning or permit a free-for-all for a family one of whose members is a non-HKR because the Director’s immigration concerns would be addressed on a proportionality analysis. 71. The answer to these criticisms of Hai Ho Tak lies in the constitutional status given to section 11 as explained in Ubamaka, GA and Rbani. Non-resident family members cannot rely on the family rights of HKPR or HKR children to challenge immigration decisions refusing them permission to enter or remain in Hong Kong to be with their children, as a matter of Hong Kong constitutional law. Reliance on Beoku-Betts to support the contention that immigration authorities should consider the rights of close family members of those affected by their decisions does not avail the appellants since there is no equivalent of section 11, or BL 39 or BL 154(2), in the relevant laws of the United Kingdom. 72. Hai Ho Tak is consistent in particular with Rbani. The passage from Nazareth JA’s judgment in Hai Ho Tak (at p.209) quoted above remains a correct statement of the law since the coming into effect of the Basic Law on 1 July 1997. E. Other rights relied upon by the appellants 73. The appellants also sought to rely on various other rights arising under the ICESCR and the CRC and also the common law principle of the best interests of the child. These may be disposed of briefly. E.1 ICESCR 10 74. The appellants’ reliance on ICESCR 10 is unsustainable for the simple reason that the ICESCR is an international treaty and under the common law dualist principle is not self-executing. Unless and until made part of Hong Kong domestic law by legislation, the provisions of such a treaty do not confer or impose any rights or obligations on individual citizens. This principle has been clearly stated and applied in Ubamaka at [42]-[44] and in GA at [58]. On the application of that principle, the appellants simply cannot rely on the ICESCR unless they can show that its provisions have been incorporated into domestic legislation. 75. Where it is said that a provision of an international treaty has been incorporated into domestic legislation, then it is largely a matter of statutory construction to determine whether it has actually done so and to what extent: see GA at [60(2)]. 76. In the present case, the appellants contended that ICESCR 10 was domesticated through BL 37 and the BOR (by BOR 19 and 20) and relied on the statement of the Government of the HKSAR to this effect.[39] Relying on those statements of the Government of the HKSAR, the appellants advanced a further argument that they had a legitimate expectation that restrictions on their rights under ICESCR 10 would be fully justified in accordance with the provisions of ICESCR 4.[40] 77. Even accepting that the provisions of ICESCR 10 have been incorporated into domestic legislation through BL 37, BOR 19 and BOR 20, as explained above, those rights, properly construed, are subject to the immigration reservation in section 11 and therefore reliance on ICESCR 10 cannot give the appellants any greater rights than under those provisions so construed. It is an established principle that “any legitimate expectation has to give way to contrary statutory provisions”: see Ng Siu Tung & Others v Director of Immigration (2002) 5 HKCFAR 1 at [132]. In the present case, there can be no greater expectation than that arising under the provisions of BL 37, BOR 19 or BOR 20, all of which are qualified by section 11. This excludes the appellants’ reliance on ICESCR 10 whether directly or indirectly by way of legitimate expectation. E.2 CRC 3 78. Similarly, the appellants’ reliance on CRC 3 does not assist them. The CRC, like the ICESCR, is an unincorporated international convention and, on the dualist principle, does not give rise to any enforceable rights or obligations. 79. Insofar as the provisions of CRC 3 are said to have been implemented through BL 37 or provisions of the BOR, those rights are all subject to section 11 and, in the context of the immigration decisions under challenge, are excluded. As a matter of construction, CRC 3 could not be relied on to extend the ambit of those rights which, interpreting the Basic Law as a coherent whole, are subject to section 11. 80. In any event, the Government of the PRC, when notifying the Secretary-General of the United Nations that the CRC would apply to the HKSAR, declared that: “The Government of the People’s Republic of China reserves, for the Hong Kong Special Administrative Region, the right to apply such legislation, in so far as it relates to the entry into, stay in and departure from the Hong Kong Special Administrative Region of those who do not have the right under the laws of the Hong Kong Special Administrative Region to enter and remain in the Hong Kong Special Administrative Region, and to the acquisition and possession of residentship as it may deem necessary from time to time.”[41] 81. That reservation, which is in very similar terms to that entered in respect of the ICCPR as reflected in section 11, would clearly preclude reliance on CRC 3 to defeat the application of immigration legislation to the decision to refuse to permit the appellant mothers to remain in the HKSAR. The appellants’ submission that the above reservation is limited in its scope to the appellant mothers, who do not have the right to enter and remain in Hong Kong, but not the appellant children, who do have such rights, is inconsistent with the reservation construed as a whole. 82. The appellants’ reliance on the approach of the UK Supreme Court in ZH (Tanzania) v Secretary of State for the Home Department[42] does not assist the appellants since a reservation entered by the UK Government in respect of the CRC similar to that entered in respect of the HKSAR was lifted in 2008 and led to the introduction of section 55 of the Borders, Citizenship and Immigration Act 2009 which provides that the Secretary of State must ensure that functions in relation to immigration “are discharged having regard to the need to safeguard and promote the welfare of children who are in the United Kingdom”. The position in Hong Kong is materially different since the reservation to the CRC remains in place. 83. The appellants’ further argument seeking to rely on a legitimate expectation that the “interests of the child are primary in the formulation and dispensation of all legislation and policies relating to or affecting children”[43] faces the same difficulty as that in respect of ICESCR 10. E.3 Common law rights 84. The appellants also relied on the common law principle of the best interests of the child in support of their case that the Director had a duty to take the position of the children into account when making immigration decisions in respect of their mothers. This common law principle, the appellants argued, in reliance on Rbani, was not affected by section 11. 85. There are a number of difficulties with that argument. 86. The common law principle of the best interests of the child is one which operates in the context of custody and wardship issues: see, for example, the case of Re G (A Minor)[44] cited by the appellants in their printed case.[45] It has no application in the immigration context in relation to decisions concerning the entry, stay or departure of family members. 87. Rbani does not assist the appellants in this regard. In that case, the Court considered challenges to the applicant’s immigration detention pending the execution of a removal order. Those challenges were made on the basis of constitutional arguments and traditional judicial review grounds relying on the common law Hardial Singh principles concerning the power to detain. The Court rejected the constitutional argument on the basis of the effect of section 11 precluding reliance on BOR 5 or BL 28. In respect of the common law complaint, however, the Court held, at [98]: “As we have seen, section 11 does not preclude persons who come within its terms from relying on the common law right to personal freedom and to protection against arbitrary arrest or detention. Nor does section 11 preclude such persons from relying on Chapter III and Bill of Rights guarantees in respect of legislation other than ‘immigration legislation governing entry into, stay in and departure from Hong Kong’.” 88. There are no comparable common law rights available to the appellants in the present case. 89. Customary international law was relied upon in passing in the appellants’ printed case but was not pursued and requires no further discussion. F. Conclusion 90. For the above reasons, we would dismiss the appeals. The Director is not duty bound to take into account the various rights relied upon when exercising his discretion to refuse permission to stay to the appellant mothers since such rights are disengaged by section 11. His exercise of discretion is amenable to judicial review grounds on the traditional bases available to affected parties. However, the appellants’ challenges to the Director’s decisions on such traditional judicial review grounds have failed and are not pursued on these appeals. 91. We would add, however, that the availability of the traditional judicial review jurisdiction over the decisions of the Directors demonstrates the unsustainability of the appellants’ concluding submission in reply that, if the Court were to dismiss the appeals, there would be a “rule of law deficit”. That submission is, with respect, extravagant and potentially misleading. There is no rule of law deficit in the context of this case or otherwise. The Director does not enjoy an unfettered discretion in making immigration decisions. On the contrary, as the jurisprudence of the Hong Kong courts amply demonstrates, the Director’s exercise of discretion is subject to review. That the appellants are unable to rely on particular rights because of a decision by the drafters of the Basic Law to make those rights subject to the immigration reservation reflected in section 11 does not amount to a “rule of law deficit”. 92. We would direct that any submissions as to costs be lodged in writing within 14 days after the handing down of this judgment to be dealt with on the papers. Mr Justice Stock NPJ: 93. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Mr Justice French NPJ: 94. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Chief Justice Ma: 95. For the above reasons, the appeals are unanimously dismissed and a direction is made as to the costs as set out in paragraph [92] above. Ms Gladys Li SC, Mr Raza Husain QC and Mr P.Y. Lo, instructed by Daly & Associates, assigned by the Director of Legal Aid, for the 1st and 2nd Applicants (1st and 2nd Appellants) in FACV 9/2018 and the 1st to 4th Applicants (1st to 4th Appellants) in FACV 10/2018 Lord Pannick QC, Mr Abraham Chan SC and Ms Grace Chow, instructed by the Department of Justice, for the Respondent (Respondent) in FACV 9 & 10/2018 [1] Articles 24 and 37. [2] Articles 17, 23(1) and 24(1), enacted as Articles 14, 19(1) and 20(1) of the Hong Kong Bill of Rights. [3] Article 10. [4] Articles 1-7, 9 and 18. [5] HCAL 13, 45 and 56/2014 (12 January 2016). The present appellants brought two of the three applications for judicial review heard together by his Lordship. The third applicant did not appeal Au J’s decision. [6] Cheung CJHC, Lam VP and Poon JA [2018] HKCA 175 (26 March 2018). The present appellants brought two of the three appeals heard together by the Court of Appeal. Leave to appeal to this Court was refused: [2018] HKCA 423 (24 July 2018). [7] (Cap 383). Section 11 and BL 39 are set out in Section B.2 below. [8] Ribeiro and Fok PJJ, Stock NPJ [2018] HKCFA 53 (7 November 2018). [9] Cheung CJHC, Stock VP and Fok JA, CACV 183/2012 (2 May 2013). [10] In this judgment, the abbreviation “BL” followed by the number of the relevant article in question will hereinafter be used for references to articles in the Basic Law. The same convention will also apply in respect of the “ICCPR”, “BOR”, “ICESCR” and “CRC”. [11] (Cap 383). [12] (2012) 15 HKCFAR 743. [13] (2014) 17 HKCFAR 60. [14] (2014) 17 HKCFAR 138. [15] For instance, in the European Court of Human Rights (“ECtHR”): Chahal v United Kingdom (1996) 23 EHRR 413 at [73]; F v United Kingdom [2004] ECHR 723 (Application No 17341/03), 22 June 2004 ; and Al Husin v Bosnia and Herzegovina [2012] ECHR 232. Recognised domestically in the UK: R (Saadi) v Secretary of State for the Home Department [2002] 1 WLR 3131, R (Ullah) v Special Adjudicator [2004] 2 AC 323. And in this Court: Ubamaka at [103]-[104]; GA at [29(3)]. [16] At [51]. [17] Annex I, Section XIII of the Joint Declaration, coming into force on 30 June 1985. [18] As noted in GA at [29(2) and (3)]. Section 11 and BL 154(2) are set out in Section B.2. [19] Except for provisions not presently relevant. [20] Decision adopted at the Twenty Fourth Session of the Standing Committee of the Eighth National People's Congress on 23 February 1997. See Ubamaka at [74]-[76]. [21] (1999) 2 HKCFAR 442 at p.455. [22] (2002) 5 HKCFAR 381 at [58]. [23] (2003) 6 HKCFAR 236 at [53]. [24] At [113]. [25] (2002) 5 HKCFAR 480 at [21]-[22]. [26] At [29(2)] and [30]. [27] At [82]-[84]. [28] BL 28: “The freedom of the person of Hong Kong residents shall be inviolable. No Hong Kong resident shall be subjected to arbitrary or unlawful arrest, detention or imprisonment …” [29] Including BL 28. [30] At [97]. [31] (2002) 5 HKCFAR 480. [32] ICCPR 5(2) provides: “There shall be no restriction upon or derogation from any of the fundamental human rights recognized or existing in any State Party to the present Covenant pursuant to law, conventions, regulations or custom on the pretext that the present Covenant does not recognize such rights or that it recognizes them to a lesser extent.” [33] (Cap 115). [34] BL 31 provides: “Hong Kong residents shall have freedom of movement within the Hong Kong Special Administrative Region and freedom of emigration to other countries and regions. They shall have freedom to travel and to enter or leave the Region. Unless restrained by law, holders of valid travel documents shall be free to leave the Region without special authorization.” [35] [1994] 2 HKLR 202. [36] (2014) 17 HKCFAR 60. [37] The Case of the Appellants at [4.9]. [38] [2009] 1 AC 115. [39] Second Report of the HKSAR of the PRC in the light of the ICESCR (2003), at [2.3] and Annex 2A; see also the statement of the Secretary for Home Affairs in Record of the Proceedings of the Legislative Council (2004/2005) for 6 April 2005 at p.5680 and pp.5687-5688. [40] ICESCR 4 provides: “The States Parties to the present Covenant recognize that, in the enjoyment of those rights provided by the State in conformity with the present Covenant, the State may subject such rights only to such limitations as are determined by law only in so far as this may be compatible with the nature of these rights and solely for the purpose of promoting the general welfare in a democratic society.” [41] Notifications by China and the United Kingdom of Great Britain and Northern Ireland Relating to Hong Kong, C.N.277.1997.TREATIES (Depository Notification) for 22 August 1997 at p 4. [42] [2011] 2 AC 166. [43] Report of the Hong Kong Special Administrative Region under the Convention on the Rights of the Child (June 2003) at [52]; and Combined Third and Fourth Reports of the People’s Republic of China under the Convention on the Rights of the Child: Part Two: Hong Kong Special Administrative Region (May 2012) at [105]. [44] [2018] 4 HKC 518. [45] The Case of the Appellants at [3.6].