text
stringlengths
1.16k
670k
The case concerns the proper approach to deciding who has been responsible for harming a child in proceedings taken to protect that child, and others in the family, from harm and the consequences of such a decision. At a fact finding hearing, the judge decided that either the mother or the father had injured their baby boy. He had therefore suffered significant harm attributable to a lack of reasonable parental care, as required by section 31(2) of the Children Act 1989. The judge did not ask herself which parent was responsible, although she expressed the view that it was 60% likely that the father had injured the child and 40% likely that the mother had. The mother and father were separated and the father played no part in the proceedings. At the later welfare hearing, the judge approved the placement of the child for adoption, together with his younger brother, who had been born during the proceedings and placed with foster parents soon after birth. The mother, who had maintained contact and developed a good relationship with the children, appealed. The Supreme Court unanimously allows the appeal and remits the case for a complete rehearing before a different judge. The judgment of the Court was given by Lady Hale. [48] [50] It is now settled law that the standard of proof in care proceedings is the balance of probabilities, as set out in Re H (Minors) (Sexual Abuse: Standard of Proof) [1996] AC 563 and confirmed in Re B (Children) (Care Proceedings: Standard of Proof) [2008] UKHL 35, [2009] AC 11. [8] [13] It is clear from the observations of Lord Hoffman and Lady Hale in Re B that the same approach is to be applied to the identification of perpetrators as to any other factual issue in the case. It was incorrect to apply a heightened standard consistent with the gravity of the allegations. [34] There is no obligation for a judge to decide who has caused the harm to the child, as long as that harm is attributable to someone having care of the child, although he should do so if the evidence warrants this. In a split hearing, there may be particular benefits of making such a finding, mainly because it will promote clarity in identifying the future risks to the child and the strategies necessary to protect him from them. [35] [38] Where a specific perpetrator cannot be identified, a judge should still, where possible, identify a pool of possible perpetrators. The test for doing so is the likelihood or real possibility that a particular person was involved. A person does not have to prove their innocence to be left out of account[40] [43] Where a judge has been unable to identify a perpetrator, it is positively unhelpful to have the sort of indication of percentages that the judge gave in this case. [44] If the judge is able to identify a perpetrator on the balance of probabilities, all the evidence accepted by the judge which is relevant to identifying the risks to the child remains relevant to deciding where his best interests will lie. The court must also be alive to the possibility that the finding who the perpetrator was is wrong and be prepared to revise it in the light of later evidence. [46] [47] In the circumstances of this case the judge had misdirected herself on the standard of proof in the fact finding hearing. In those circumstances the case ought to be remitted in whole to a different judge who can decide the matter on the right basis. [48] The decision to remove the second child, who had never been harmed, must also be remitted for rehearing. The judge had held that there was a risk of future harm to him because there was a real possibility that the mother had injured the older child. It was held in Re H that this is not the correct approach: predictions of future harm must be based on proven findings of fact. [49]
This case concerns two little girls whom the court refers to as Livi and Milly so as to protect their anonymity. Livi is seven and Milly is four. They were born in Norway to a British mother and a Norwegian father, who married shortly after Livis birth. They had lived all of their lives in Norway until September last year when their mother brought them to England with a view to staying here permanently. The father was not asked and did not consent to their removal from Norway. The mother has an older daughter, Tyler, who is nearly 17 and also lived with the family in Norway, but left Norway for England shortly before her mother. After the removal of Livi and Milly, the father applied to the Norwegian central authority under the Hague Convention on the Civil Aspects of Child Abduction 1980 (the Hague Convention) for the children to be returned to Norway. Article 12 of the Hague Convention requires a requested state to return a child forthwith to her country of habitual residence if she has been wrongfully removed in breach of rights of custody. Article 13 provides three exceptions, of which this case is concerned with the second: the requested state is not bound to return the child if the person, institution or other body which opposes its return establishes that (a) ; or (b) there is a grave risk that his or her return would expose the child to physical or psychological harm or otherwise place the child in an intolerable situation . (emphasis supplied) The mother, with Tylers support, argues that this exception applies. She makes allegations against the father which, if true, amount to a classic case of serious psychological abuse. She recounts incidents of physical violence towards other people, property and the ill treatment of the family pets. She alleges that the father was domineering and controlling and that she and the children were frightened of him. There is also psychiatric evidence that the mother is suffering from a mental disorder which will deteriorate if she has to return with the children to Norway unless certain protective measures are in place. She is and always has been the childrens primary carer. Thus she argues that that there is a grave risk that the children would be placed in an intolerable situation if returned to Norway. She also argues that article 13(b) should be interpreted in the light of article 3.1 of the United Nations Convention on the Rights of the Child, which requires that in all actions concerning children the best interests of the child shall be a primary consideration. The father denies the mothers allegations, although he accepts that he can get angry and that he did kill a family cat and Tylers rabbit. He has made arrangements and given undertakings to reassure the mother pending any order in the Norwegian courts: in particular he would vacate the family home and not go within 500 metres of it; he would pay household costs and money for child support; and he would not remove the children from the mothers care. The trial judge in England decided that it was overwhelmingly in the childrens best interests for them to return to Norway for their futures to be decided there. The Court of Appeal rejected the mothers appeal. Both the Court of Appeal and the Supreme Court agreed to hear the case principally because of concerns about the impact of the decision of the Grand Chamber of the European Court of Human Rights in Neulinger and Shuruk v Switzerland [2011] 1 FLR 122. The Supreme Court unanimously dismisses the mothers appeal. The best interests of the child are at the forefront of the whole process under the Hague Convention. The aim is to serve the interests of children generally, by deterring wrongful abduction and restoring abducted children to their home country, but also to serve the interests of the individual child by making certain assumptions about what will be best for her [14]. These assumptions can be rebutted in a limited range of circumstances, all inspired by the best interests of that child [16]. The case law of the European Court of Human Rights indicates that the right to respect for family life in article 8 of the European Convention must be interpreted in the light of the Hague Convention and the UN Convention on the Rights of the Child. The best interests of children have two aspects: to be reunited with their parents as soon as possible so that one parent does not gain an unfair advantage over the other through the passage of time; and to be brought up in a sound environment in which they are not at risk of harm [52]. In Neulinger and Shuruk v Switzerland, the Grand Chamber held that the return of a child from Switzerland to Israel would breach article 8 of the European Convention on Human Rights. In doing so, the Grand Chamber gave the appearance of turning the swift, summary decision making procedure which is envisaged by the Hague Convention into a full blown examination of the childs future in the requested state, the avoidance of which was the very object of the Hague Convention [22]. However, the President of the Strasbourg court has recently acknowledged extra judicially that the logic of the Hague Convention is that a child who has been abducted should be returned to the jurisdiction best placed to protect his interests and welfare, and it is only there that his situation should be reviewed in full[25]. Hence Neulinger does not require a departure from the normal process, provided that the decision is not arbitrary or mechanical. The exceptions to the obligation to return are by their nature restricted in scope and should be applied without extra interpretation or gloss. Violence and abuse between parents may constitute a grave risk to the children. But where there are disputed allegations which can neither be tried nor objectively verified, the focus of the inquiry is bound to be on the sufficiency of any protective measures which can be put in place to reduce the risk. The clearer the need for protection, the more effective the measures will have to be [52]. In this case, the trial judge was satisfied that medical treatment would be available for the mother and that there were legal remedies to protect the children should they be needed. It is not the task of an appellate court to disagree with the trial judges assessment [49]. The Supreme Court urges the Hague Conference to consider whether machinery can be put in place whereby, when the courts of the requested state identify specific protective measures as necessary if the Article 13(b) exception is to be rejected, those measures can become enforceable in the requesting state [37].
Cleveland Meat Company Ltd (CMC) bought a bull at auction. It was passed fit for slaughter by the Official Veterinarian (OV) stationed at its slaughterhouse. After a post mortem inspection of the carcass, and discussion with a Meat Hygiene Inspector, the OV declared the meat unfit for human consumption. It did not therefore acquire a health mark. CMC took the advice of another veterinarian surgeon and challenged the OVs opinion. It contended that in the event of a dispute, and of its refusal to surrender the carcass voluntarily, the OV would have to seize it under s 9 of the Food Safety Act 1990 (the 1990 Act) and take it before a Justice of the Peace to determine whether or not it should be condemned. The respondent (the FSA) did not accept that it needed to use this procedure. It maintained that the carcass should be disposed of as an animal by product and served a notice for such disposal. CMC, together with Association of Independent Meat Suppliers, issued a claim for judicial review to challenge the FSAs assertion that it did not have to use the s 9 procedure. They claimed in the alternative that it was incumbent on the UK to provide some means for challenging the decisions of an OV in such cases. The claim failed in the High Court and Court of Appeal. Before the Supreme Court there were three main issues. The first was a matter of domestic law: whether the s 9 procedure was available or mandatory in these circumstances. The second issue was whether the use of the s 9 procedure was compatible with the food safety regime laid down by European Union law, specifically Regulations (EC) 178/2002, 852/2004, 853/2004, 854/2004, 882/2004, and 1069/2009. The third issue was whether Regulation 882/2004 mandates an appeal procedure and, if so, whether such an appeal should allow a challenge to the full factual merits of the OVs decision or whether the limited scope of challenge in a judicial review claim is sufficient to comply with the regulations requirements. The Supreme Court decides to refer two questions to the Court of Justice of the European Union. The terms of the reference are set out by Lady Hale and Lord Sales in their joint judgment, with whom Lord Hodge, Lady Black and Lord Lloyd Jones agree. For the purposes of this reference, the Court of Justice of the European Union is asked to assume that the claimant appellants interpretation of section 9 of the 1990 Act is correct, and that a Justice of the Peace has power to give a ruling which may result in an award of compensation if he considers that a health mark ought to have been applied to a carcass. In order to determine this appeal, this Court refers the following questions to the Court of Justice of the European Union [22]: (1) Do Regulations (EC) Nos 854 and 882 preclude a procedure whereby pursuant to section 9 of the 1990 Act a Justice of the Peace decides on the merits of the case and on the basis of the evidence of experts called by each side whether a carcass fails to comply with food safety requirements? (2) Does Regulation (EC) No 882 mandate a right of appeal in relation to a decision of an OV under article 5.2 of Regulation (EC) No 854 that the meat of a carcass was unfit for human consumption and, if it does, what approach should be applied in reviewing the merits of the decision taken by the OV on an appeal in such a case?
The issue in these appeals is whether the mistreatment of migrant domestic workers who are vulnerable because of their precarious immigration status amounts to direct or indirect race discrimination. The appellant in the first appeal, Ms Taiwo, is a Nigerian national who entered the United Kingdom lawfully in February 2010 to work for the respondents. She had a migrant domestic workers visa obtained for her on the false basis that she had previously been employed by Mr Olaigbes parents in Nigeria. Ms Taiwos passport was taken from her and she was expected to work during most of her waking hours for minimal wages. She was starved and subject to physical and mental abuse. She escaped and brought successful claims in the employment tribunal for the failure to pay her the minimum wage, for unlawful deductions from wages, for failure to provide rest periods and to give her written terms of employment. She was awarded compensation in respect of these claims but her claim for race discrimination, which would have entitled her to damages for the fear and distress she suffered, was dismissed. The tribunal found that her mistreatment was because she was a vulnerable migrant worker who was reliant on the respondents for her continued employment and residence in the UK, not because she was Nigerian. Ms Onu, the appellant in the second appeal, suffered a similar experience. She had worked for her employers in Nigeria and came to the UK on a domestic workers visa. She worked on average for 84 hours a week, without the required rest periods, nor was she paid the minimum wage and she was threatened and abused by her employers. She brought similar claims in the employment tribunal, which all succeeded including her claim for direct race discrimination. The latter finding was reversed by the Employment Appeal Tribunal. The Court of Appeal heard Ms Taiwo and Ms Onus appeals together and upheld the dismissal of their discrimination claims on the grounds that immigration status was not to be equated with nationality for the purpose of the Equality Act 2010. Ms Taiwo appealed (and Ms Onu applied for permission to appeal) to the Supreme Court. The Supreme Court unanimously grants permission to appeal to Ms Onu but dismisses both Ms Taiwo and Ms Onus appeals. It holds that neither appellant has suffered race discrimination because the reason for their abuse by the respondents was not nationality but their vulnerability as a particular kind of migrant worker. Lady Hale gives the only substantive judgment. Under s 13(1) Equality Act 2010 (EA) a person (A) discriminates against another (B) if, because of a protected characteristic, A treats B less favourably than A treats or would treat others. Race is a protected characteristic and includes colour, nationality and ethnic origins [13]. There is no doubt in these cases that the appellants were treated disgracefully by their employers in a way which employees who did not share the appellants vulnerable immigration status would not have been treated. The question is whether discrimination on grounds of immigration status amounts to discrimination on grounds of nationality [14]. Generally speaking employers are free to choose whom to employ, subject to the limits under the EA (and earlier legislation) to protect specified groups, who have historically suffered discrimination, from being shut out of access to employment for irrelevant reasons which they can do nothing about [21]. Parliament could have chosen to include immigration status in the list of protected characteristics but it did not do so [22]. Immigration status is a function of nationality in that non British nationals (other than Irish citizens) are subject to immigration control, but there is a wide variety of immigration statuses [23]. The appellants were particularly vulnerable to the abuse they suffered because of the terms of their domestic workers visas which meant they were dependent on their current employers for their continued right to live and work in the UK [24]. But there are many non British nationals living and working in the UK who do not share this vulnerability and would not have been abused in the same way. The treatment of the appellants had nothing to do with the fact they were Nigerian and they were not the subject of direct discrimination [26]. This was not a case of indirect discrimination. There was no provision, criterion or practice as defined in s 19 EA applied by the respondents to all their employees regardless of their immigration status [32]. The present law does not therefore offer redress for all the harm suffered by the appellants. Parliament might wish to consider extending the remedy available under the Modern Slavery Act 2015 to give employment tribunals jurisdiction to grant compensation for ill treatment meted out to workers [34].
Traders who wish to appeal against assessments to Value Added Tax (VAT) in the United Kingdom are required, by section 84 of the Value Added Tax Act 1994, first to pay or deposit the tax notified by the assessment with HMRC, unless they can demonstrate that to do so would cause them to suffer hardship. This pay first requirement is a feature of the procedural regime for appealing assessments to a number of other types of tax including Insurance Premium Tax and Landfill Tax. It is not a condition for appealing assessments to Income Tax, Capital Gains Tax, Corporation Tax or Stamp Duty Land Tax. VAT is regulated by the EU VAT Directive 2016/112. An appeal against a VAT assessment is therefore a claim based on EU law. The appellant, Totel Ltd (Totel), seeks to appeal a number of assessments to VAT but has been unable to demonstrate that a requirement to pay or deposit the tax in dispute would cause the company hardship. Totel claims the requirement to pay or deposit the disputed tax, as a pre condition for an appeal, offends against the EU law principle of equivalence. Totel first raised its challenge based upon the principle of equivalence when it successfully sought permission to appeal to the Court of Appeal. The Court of Appeal dismissed Totels appeal. The Supreme Court unanimously dismisses the appeal. Lord Briggs gives the lead judgment with which the other Justices agree. The principle of equivalence requires that the procedural rules of member states applicable to claims based on EU law are no less favourable than those governing similar domestic claims [3]. The principle of equivalence and its qualifying no most favourable treatment proviso (the Proviso) are creations of the Court of Justice of the European Union (CJEU) jurisprudence and take effect within the general context that it is for each member state to establish its own national procedures for the vindication of rights conferred by EU law [6]. The principle of equivalence requires a true comparator for it to be able to operate at all. Identification of one or more true comparators is therefore the essential first step [7]. Whether any proposed domestic claim is a true comparator with an EU law claim is context specific [9]. The domestic court must focus on the purpose and essential characteristics of allegedly similar claims [10]. Of particular importance is the specific procedural provision that is alleged to constitute less favourable treatment of the EU law claim. Differences in procedural rules are frequently attributable to differences in the underlying claim [11]. It is not necessary or appropriate to treat VAT claims as unique with no possibility of having a true comparator. Such a general rule would run counter to the context specific basis which underpins the principle of equivalence. In Reemsta Cigarettenfabriken GmbH v Ministero delle Finanze (Case C 35/05) [2007] ECR1 2452, the CJEU considered equivalence as a mere fringe issue and, had Reemsta actually established such a rule, this would have provided a simple solution for the question before the CJEU in the Littlewoods Retail Ltd v Revenue & Customs Comrs (Case C 591/10) [2012] STC 1714 [18 21]. Applying the context specific analysis, the Court of Appeal was correct to conclude that none of the domestic taxes constituted true comparators with VAT. A trader seeking to appeal a VAT assessment is typically in a significantly different position from a taxpayer seeking to appeal an assessment to any of those other taxes [22]. VATs economic burden falls upon the consumer, but it is collected by the trader from the consumer and accounted for by the trader to HMRC. Taxpayers appealing Income Tax, for example, are being required to pay something of which the economic burden falls on them and which they have not collected from anyone else. Therefore, it is no less than appropriate that traders assessed to VAT should be required to pay or deposit the tax in dispute, which they have or should have collected [23]. This logical connection is sufficient to justify the conclusion that VAT is different to those other taxes in this context regardless of the actual legislative reason for the imposition of the pay first requirement [24]. Lord Briggs considered what the position would have been had any of Income Tax, Capital Gains Tax or Stamp Duty Land Tax been a true comparator for the purposes of the principle of equivalence. The Proviso is not a free standing rule but part of the expression of the principle of equivalence and is directed to the standard of treatment which that principle imposes. Procedures should be broadly as favourable as that available for truly comparable domestic claims, rather than the very best available [45]. The Provisos purpose is to prevent member states from discriminating against claims based upon EU law by affording them inferior procedural treatment than comparable domestic claims [46]. While reaching no final decision on this point, Lord Briggs concluded that the Court of Appeals conclusion on this issue is therefore broadly correct [47].
The Appellant was arrested for driving with excess alcohol on 14 October 2008 and pleaded guilty to that offence on 5 November 2008. He was fined 50 and disqualified from driving for 12 months. A conviction for driving with excess alcohol is spent after five years. He has been an adult throughout. When the Appellant was arrested, the Police Service of Northern Ireland (PSNI) lawfully obtained from him: (i) fingerprints; (ii) a photograph; and (iii) a non intimate DNA sample by buccal swab. The fingerprints are held on a UK wide database and the photograph is held on a PSNI database to which only authorised PSNI personnel have access. A DNA profile was taken from the DNA sample. A DNA profile is digitised information in the form of a numerical sequence representing a very small part of the persons DNA. It indicates a persons gender and provides a means of identification. The profile is held on a Forensic Science Northern Ireland database. At present, the statutory position in Northern Ireland is that the PSNI may retain fingerprints, photographs, DNA samples and DNA profiles for an indefinite period after they have fulfilled the purpose for which they were taken, but they may only be used for specified policing purposes. The Criminal Justice Act (Northern Ireland) 2013, when it comes into force, will require the PSNI to destroy all DNA samples as soon as a DNA profile has been taken or within six months of the taking of the DNA sample, and will otherwise bring the position broadly into line with the current legislation applicable in England and Wales. The PSNI intends to destroy the Appellants DNA sample when that Act comes into force. This appeal therefore does not concern the retention of the DNA sample. The PSNI changed its practice following the decision in S and Marper v United Kingdom (2009) 48 EHHR 50 (S and Marper). The European Court of Human Rights (ECtHR) held that the UKs policy of indefinite retention of individuals fingerprints, cellular samples and DNA profiles after proceedings against them had led to acquittal or discontinuance was a disproportionate interference with their right to respect for private life under article 8 of the European Convention on Human Rights (ECHR). The PSNI now retains indefinitely biometric data only of those convicted of crimes. The Appellant says that the PSNIs retention of his data breaches article 8 ECHR. The Respondent accepts that there is an interference with the Appellants right to respect for his private life under article 8(1) and the Appellant accepts that the interference is in accordance with law and pursues a legitimate aim under article 8(2). The sole question is therefore whether the interference was proportionate. The Divisional Court held that it was. The Appellant appeals to the Supreme Court. The Supreme Court dismisses the appeal by a majority of 4:1. Lord Clarke, with whom Lord Neuberger, Lady Hale and Lord Sumption agree, gives the leading judgment. Lord Kerr dissents. The majority considers that in S and Marper the ECtHR was concerned only with the position of suspected but non convicted persons, not convicted persons; its criticism of the UKs blanket and indiscriminate data retention policy should be read with this focus in mind [30 32]. He recognises that it does not follow that the practice of Northern Ireland (and the UK) in relation to convicted persons is automatically compliant with article 8 and that the policy as it applies to convicted persons could be described as a blanket policy [33]. However, the policy is in fact proportionate: The ECtHR did recognise in S and Marper the importance of the use of DNA material in the solving of crime and that the interference in question is low [33]. It is also important to note that the present scheme is concerned only with the retention of the DNA profile and applies only to adults, whereas the scheme criticised by the ECtHR in S and Marper provided for the retention of the full sample and did not distinguish between children and adults [35]. Factors such as the threshold of offence, whether retention is permitted once a conviction has been spent and whether retention is permitted indefinitely or is subject to a time limit are potentially relevant but not decisive in the proportionality analysis [34, 36 39]. The potential benefit to the public of retaining the DNA profiles of those who are convicted is considerable and outweighs the interference with the right of the individual [40]. The retention may even benefit the individual by establishing that they did not commit an offence [41]. In S and Marper the ECtHR placed some reliance on the fact that the UK was almost alone among ECHR member states in indefinitely retaining biometric data of non convicted persons. In the case of convicted persons there is a much broader range of approaches, which broadens the margin of appreciation accorded to individual states [42 44]. Adopting a blanket measure is legitimate in some circumstances and it was legitimate here [45]. The retention policy is therefore within the UKs margin of appreciation, and the court has to decide for itself whether the policy is proportionate. Essentially on the basis of the factors already discussed and for the reasons given by the Divisional Court, the majority concludes that it is and dismisses the appeal [46 49]. Lord Kerr would have allowed the appeal. He explains that the critical questions on proportionality in this case are: (i) whether there is a rational connection between the legislative objective and the policy; and (ii) whether the policy goes no further than is necessary to fulfil the objective [61]. As to (i), it is important to recognise that the objective is not the creation of as large a DNA database as possible, but the actual detection of crime and identification of future offenders. There is a striking lack of hard evidence in this case to support the assumption that all persons who commit any recordable offence are potential suspects in any future crime [62 68]. As to (ii), it is clear in Strasbourg, CJEU and domestic case law that the question is whether a less intrusive measure could have been used without unacceptably compromising the attainment of the objective [73 77]. A far more nuanced and more sensibly targeted policy could easily be devised. In those circumstances it is impossible to say that the policy in its present form is the least intrusive means of achieving its stated aim [83 85]. As to whether a fair balance has been struck, the stigmatising application of the indefinite retention policy even to those whose convictions are spent frustrates the purpose of rehabilitation: this is an issue of first importance. It should not be relegated to the status of a single factor of no especial significance [90 96]. A domestic court should not be slow to condemn an ill thought out policy which does not address the essential issues of proportionality simply because a broad measure of discretion is available to an individual state [99 101].
The appellant is 22 years old and, prior to these proceedings, had no convictions. He was diagnosed with Autism Spectrum Disorder as a child and, in 2018, he was living with his mother. He began purchasing quantities of chemicals online. According to his account, this was because he had an obsessive interest in military matters, including bomb disposal. He claimed that he had acquired the chemicals because he wished to understand how explosives worked and to experiment with them. A search warrant was executed at the appellants home on 24 April 2018, where it was found that he had managed to make a small quantity (about 10 grams or less) of a sensitive primary explosive, Hexamethylene Triperoxide Diamine (HMTD). He also possessed manuals for making explosives, notes on making HMTD and a video on his mobile phone demonstrating the making of HMTD. When interviewed by the police, the appellant admitted his actions in acquiring chemicals and making explosives, and gave the explanation above. He was subsequently charged with certain offences, including two counts under section 4(1) of the Explosive Substances Act 1883 (1883 Act). As clarified in the course of the appeal, these counts are that the appellant knowingly had HMTD in his possession or under his control, in such circumstances as to give rise to a reasonable suspicion that he did not possess or control it for a lawful object. There is a defence if the accused can show that in fact he had the explosive substance in his possession or control for a lawful object which he identifies. In his defence statement, the appellant maintained that he had made the HMTD for a lawful object, namely interest, education and experimentation. The appellant said that he had made other explosives and used them to create small explosions in the garden of his house without causing harm, and he intended to do the same with the HMTD. At a preparatory hearing in the Crown Court at Birmingham, HH Judge Wall QC held that he was bound by Court of Appeal authority, R v Riding [2009] EWCA Crim 892, to find that experimentation and self education did not amount to a lawful object for the purpose of section 4(1). Accordingly, the judge ruled in advance of trial that the appellants proposed defence was bad in law. The appellant appealed unsuccessfully to the Court of Appeal, who considered themselves similarly bound. The court certified the following point of law of general public importance: for the purposes of section 4(1) of the 1883 Act, can personal experimentation or own private education, absent some ulterior unlawful purpose, be regarded as a lawful object? By a majority, the Supreme Court allows the appeal. Lord Sales gives the majority judgment, with which Lord Reed and Lord Carnwath agree. Lord Lloyd Jones and Lord Hamblen deliver a joint dissenting judgment. The majority examines the 1883 Act in light of other amendments to the criminal law around the same time, namely the Offences Against the Person Act 1861 (1861 Act) and the regulatory Explosive Substances Act 1875 (the 1875 Act) [14 15]. The 1883 Act was passed by Parliament with great speed due to fears around Irish nationalism and a perception that the 1861 Act, in particular, did not provide sufficiently for protection of the public [16]. The current regulatory regime is now primarily contained in the Explosives Regulations 2014, which like the 1875 Act before them, make clear that it is expected that private individuals, including hobbyists, may manufacture and keep explosives for their own private use [18]. In R v Fegan (1984) 78 Cr App R 189, the Court of Criminal Appeal in Northern Ireland explained that section 4(1) had been passed to address perceived deficiencies in other offences. These required proof of a specific mental element, and so were inadequate to guard against the risk of making or possessing explosives. The appellant, Fegan, acquired a firearm and ammunition (which qualified as explosive substances for the purpose of the 1883 Act) to protect himself and his family from threats against their safety and was convicted of an offence under section 4(1) of the 1883 Act. His appeal was successful, on the basis that although he had no licence for the possession of the firearm and ammunition, nonetheless he possessed them for a lawful purpose, i.e. to defend himself and his family [19 22]. Fegan was followed on similar facts in Attorney Generals Riding, the Criminal Division of the Court of Appeal of England and Wales held that in the particular circumstances of that case, curiosity did not qualify as a lawful object for the possession of a home made pipe bomb [26]. Under section 4(1), it is for a defendant to prove on the balance of probabilities that he had possession or control of an explosive substance for a lawful object. In English law, a purpose is lawful unless it is made unlawful by statute or the common law. A lawful object may, however, be tainted by an ulterior, unlawful purpose, including by knowledge or recklessness of a risk of injury or damage; but these would be matters to be explored on the evidence at trial [27 29]. The decision in Riding was correct on its facts, because the defence of the accused in that case was that he acted out of curiosity to see if he could construct a pipe bomb but he did not need to use real explosives for that; and it was no part of his defence that he had wanted to experiment by making it explode. The decision does not provide an answer in the different circumstances of the present case and was misinterpreted in the courts below. Experimentation and self education are objects within the ordinary meaning of that term and are capable of being lawful objects for the purposes of section 4(1). This view is reinforced by the background against which section 4(1) was enacted, including the 1875 Act, under which possession of explosive substances for private experimentation and use was regarded as lawful and legitimate [30 33], [35]. The Court of Appeal was wrong to conclude that the appellant was obliged to specify more precisely than he had done how the explosives would be used and that this would be lawful. The Court of Appeals reasoning was inconsistent with the Fegan and Attorney Generals was lawful [34]. As there is nothing unlawful about experimentation and self education as objects in themselves, they are capable of being lawful objects [37]. There is no requirement in law that a defence statement in relation to a charge under section 4(1) has to give a more detailed account of the proposed use of the explosive substance than that provided by the appellant [39]. The appellant ought to have been permitted to present his defence at trial [41], [43]. Lord Lloyd Jones and Lord Hamblen dissent from the majoritys reasoning and would dismiss the appeal. They take the view, in common with the courts below, that personal experimentation and private education cannot in law amount to lawful objects within the meaning of section 4(1) [51]. The word object refers to the reason for doing something, or the result you wish to achieve by doing it. As such, the Court of Appeal was correct to hold that, to make out the defence, a defendant is required to show the use to which the explosive substance is to be put, and to do so with sufficient particularity to demonstrate that that use is lawful [52]. education and personal experimentation is not enough, as the Court of Appeal previously held in Riding [54]. The defence is only made out if it is shown that the way in which the explosives were intended to be used is lawful. It is not enough to show that it may be lawful. A defence statement in response to a charge under section 4(1) should elaborate upon this and provide some details of the intended use. In the present case the appellant envisaged that experimentation would take the form of detonations of the explosives in his back garden, carrying an obvious risk of causing injury, damage to property, and a public nuisance. It was necessary to particularise how this would be carried out so as to avoid any such risk or would otherwise be lawful. Vague and general statements referring to personal experimentation or private education were insufficient and did not show how that was to be carried out lawfully [55]. Finally, Fegan and Attorney Generals was plausibly raised in each. In contrast, in the present case no lawful use is identified, and the claimed objects neither give sufficient indication of the use to which the explosives are to be put, nor do they permit assessment of the lawfulness of any such use [56].
Under the Rehabilitation of Offenders Act 1974, where a person is asked about his criminal record the question will be treated as not extending to spent convictions. Consequently, he is entitled not to disclose these and cannot be liable for a failure to do so. Equally, a prospective employer is not entitled to make any decision prejudicial to the individual by reference to spent convictions or to any failure to disclose them [6]. This applies to cautions, warnings or reprimands, which are spent as soon as they are given [76]. These appeals concern the Rehabilitation of Offenders Act 1974 (Exceptions) Order 1975 and sections 113A and 113B in Part V of the Police Act 1997. The 1975 Order makes certain questions exempt from the above provisions of the 1974 Act, including where they relate to specified professions and employments, and to working with children and vulnerable adults [78 79]. Part V of the 1997 Act deals with enhanced criminal record certificates (ECRCs). These are issued where an exempted question within the meaning of the 1975 Order is asked, including by a prospective employer. Disclosure is then made of every relevant matter recorded on the Police National Computer, including, at the relevant time, any spent conviction or caution [83 84]. In T, the police issued warnings in 2002 to an 11 year old boy in respect of the theft of two bicycles. The warnings were disclosed in 2008 under Part V of the 1997 Act when T applied for a part time job with a football club possibly involving contact with children. They were disclosed again in 2010 when he applied for a place on a sports studies course which again might have involved contact with children [117]. In JB, the police issued a caution to a 41 year old woman in 2001 in respect of the theft from a shop of a packet of false fingernails. In 2009 she completed a training course for employment in the care sector. She was required to obtain an ECRC, which disclosed the caution. The training organisation told JB that it felt unable to put her forward for employment in the care sector [118]. The respondents have no other criminal records. Both claim that the references in the ECRCs to their cautions violated their right to respect for private life under article 8 of the European Convention on Human Rights. T also asserts that his obligation to disclose the warnings violated the same right. T and JB were successful in the Court of Appeal, which made declarations that the relevant provisions of the 1997 Act were incompatible with article 8. The Court of Appeal in T held that the 1975 Order was also incompatible with article 8 and ultra vires (that is, that it went beyond the powers set out in) the 1974 Act. The Secretaries of State now appeal to this Court. While they have made amendment orders designed to eliminate the problems identified by the Court of Appeal, their appeals concern the 1975 Order and 1997 Act as they stood at the time [3]. The Court unanimously (1) dismisses the appeals against the declarations of incompatibility in respect of the 1997 Act; and (2) allows the appeal against the declaration that the 1975 Order was ultra vires [158]. The respondents cautions represent an aspect of their private lives, respect for which is guaranteed by article 8 [16]. Laws requiring a person to disclose his previous convictions or cautions to a potential employer constitute an interference with that right [138]. The disclosures in the ECRCs also constituted article 8 interferences, significantly jeopardising the respondents entry into their chosen fields of endeavour [20]. Lord Reed in line with 2012 decision of the European Court of Human Rights in MM v UK considers that sections 113A and 113B of the 1997 Act are incompatible with article 8 because they fail to meet the requirement of legality, that is, that the interference with the Convention right be in accordance with law. Legality requires safeguards which enable the proportionality of the interference to be adequately examined [108 119; 158]. Legislation like the present which requires the indiscriminate disclosure by the state of personal data which it has collected and stored does not contain adequate safeguards against arbitrary interferences with article 8 rights [113 119]. Lord Neuberger, Lord Clarke and Lady Hale agree with Lord Reeds conclusion on legality [158], while Lord Wilson disagrees [28 38], emphasising the importance of the distinction between the tests of legality and necessity in a democratic society. On this point he is critical of the European Courts approach in MM. The Court unanimously holds that the article 8 interferences under both the 1997 Act and the 1975 Order could not, in any event, be said to meet the requirement of being necessary in a democratic society [50; 121; 158]. Lord Wilson notes that it was the Home Secretary who identified a need to scale back the criminal records system to common sense levels [48]. Lord Reed points to a lack of a rational connection between dishonesty as a child and the question of whether, as an adult, the person might pose a threat to the safety of children with whom he comes into contact [142]. The Court upholds the declarations of incompatibility in relation to the 1997 Act. It is impossible to read and give effect to its provisions in a way which was compatible with the respondents Convention rights [53; 120]. The Court, however, allows the appeal in T against the decision that the 1975 Order was ultra vires. This was inconsistent with the declaration of incompatibility, which stated that it did not affect the validity or continuing operation of the 1997 Act, Part V of which in fact relied upon the validity of the terms of the Order [61 62]. No judicial remedy in relation to the Order is necessary. Lord Reed explains that it had no adverse consequences for T and he can be regarded for the purposes of the Convention as having obtained just satisfaction given the courts acceptance that his complaint is well founded and the resultant amendment of the Order [66;157 158].
The question arising in this appeal is whether it is lawful for a local housing authority to accommodate a homeless person a long way away from the authoritys own area where the homeless person was previously living. Local authorities have a statutory duty to provide accommodation in their own area so far as reasonably practicable under section 208(1) Housing Act 1996 (the 1996 Act). The accommodation must be suitable to the needs of the homeless person and each member of the household, and the location can be relevant to its suitability. Regard must be given to any guidance given by the Secretary of State for Communities and Local Government. While out of borough placements are not prohibited, the Homelessness (Suitability of Accommodation) (England) Order 2012 (the 2012 Order) requires authorities to take into account the distance of the accommodation being offered from its district and the disruption to caring responsibilities or the education of any member of the household. The obligation to secure accommodation as close as possible to where the household had previously been living was strengthened by Supplementary Guidance on the homelessness changes in the Localism Act 2011 and on the Homelessness (Suitability of Accommodation) (England) Order 2012 (the Supplementary Guidance), including the need to seek to retain established links with schools, doctors, social workers and other key services and support. The appellant is a single mother of five children aged between 8 and 14. She has serious health problems. In 2012 she was evicted from her privately rented home, in which she had been living since 2008, following the introduction of a cap on housing benefit, which left her unable to pay the rent. The respondent housing authority (Westminster) accepted that she was unintentionally homeless and that it owed a duty to provide her with suitable accommodation. It offered her temporary accommodation in a house in Bletchley, near Milton Keynes, with a brief explanation that due to a severe shortage of accommodation it was not reasonably practicable to offer her a home in Westminster, but that this house was suitable in view of her circumstances. The children were not of GCSE age so Westminster considered it suitable for them to move schools. The appellant refused the accommodation and Westminster served notice that its duty to house her had come to an end. Her application for a review of the decision was unsuccessful. Her appeals to the County Court and Court of Appeal were also dismissed. The Supreme Court unanimously allows the appeal and quashes Westminsters decision that it had discharged its duty to house the appellant because she had refused suitable accommodation. Lady Hale gives the only judgment. The 1996 Act and Guidance Local authorities have a statutory duty to accommodate persons within their area so far as this is reasonably practicable. Reasonable practicability imports a stronger duty than simply being reasonable. Where it is not reasonably practicable to accommodate in borough they must generally try to place the household as close as possible to where they were previously living. The combined effect of the 2012 Order and the Supplementary Guidance has changed the legal landscape when dealing with out of borough placement policies [19]. As an aspect of the suitability of the accommodation being offered, a decision to place an applicant out of borough falls within the grounds on which a review can be sought under section 202 of the 1996 Act [20]. The childrens welfare The exercise of the local authoritys functions under the 1996 Act is subject to section 11(2) of the Children Act 2004, which requires it to have regard to the need to safeguard and promote the welfare of children. Welfare encompasses physical, psychological, social, educational and economic welfare [23] and the duty applies both to the formulation of general policies and practices and to their application in an individual case [24]. The duty is clearly relevant to the question of the suitability of the accommodation being offered [27]. It does not, however, require that the childrens welfare should be the paramount or even a primary consideration [28]. There will almost always be children affected by decisions about where to accommodate households to which the main homelessness duty is owed, and invidious choices between them must sometimes be made, but this points towards the need to explain the choices made, preferably by reference to published policies [30]. Evidencing and explaining the authoritys decisions The Secretary of State intervened in the case to emphasise the duties on local authorities to evidence and explain their decisions [31]. It must be clear from the decision that proper consideration has been given to the relevant matters required by the 1996 Act and accompanying Code. The courts below were too ready to assume that Westminster had properly complied with its statutory obligations, which had the effect of immunising from judicial scrutiny automatic decisions to house people far from their home district [35]. This case The decision made in the appellants case suffers from these defects and more. No enquiries were made to assess the practicability of moving the family to Bletchley or as to the childrens needs, and no consideration seems to have been given to the duty to offer accommodation as close by as possible or explanation given [36]. It follows that Westminster still owes the appellant a duty to secure suitable accommodation. Guidance Ideally each local authority should have an up to date publically available policy for securing sufficient units of temporary accommodation to meet the anticipated demand for the coming year, reflecting its obligations under the 1996 Act and the Children Act 2004. It should also have a policy for the allocation of those units to individual homeless households, to which reference would be made in explaining any decisions to accommodate a household out of the area [39]. This way decisions will be properly evidenced and explained, and can be challenged if required [41].
This case concerns the scope of the new offence created by section 3ZB of the Road Traffic Act 1988 (the 1988 Act). This new section was added by section 21(1) of the Road Safety Act 2006 (the 2006 Act). It provides: A person is guilty of an offence under this section if he causes the death of another person by driving a motor vehicle on a road and, at the time when he is driving, the circumstances are such that he is committing an offence under (a) Section 87(1) of this Act (driving otherwise than in accordance with a licence); (b) Section 103(1)(b) of this Act (driving while disqualified), or (c) Section 143 of this Act (using a motor vehicle while uninsured or unsecured against third party risks). On conviction on indictment, this offence carries imprisonment for up to two years. On a Sunday afternoon in October 2009 the appellant was driving his family home in a campervan along the A69 towards Newcastle. Road conditions were normal and the appellants driving was faultless. The speed limit was 60 mph and the appellant was travelling at a steady speed of 45 55mph. At the same time Mr Dickinson was driving in the opposite direction. Mr Dickinson was driving erratically his car was veering all over the road, twice crossing into the wrong lane before smashing into the appellants campervan as it rounded a bend. The appellant and his family survived. However, Mr Dickinson suffered injuries as a result of the impact that proved to be fatal. Mr Dickinson was found to have had a significant quantity of heroin in his system and was a drug user. He was also overtired, having worked a series of 12 hour nightshifts in a power station in Largs, on the west coast of Scotland. He had already driven to Largs that day and had completed approximately 230 miles of his 400 mile return journey when the collision happened. At the time of the collision the appellant did not have a driving license and was not insured, both of which are offences under the Road Traffic Act 1988. Neither offence carries a sentence of imprisonment. It was accepted by the prosecution that the appellant was in no way at fault for the accident and could not have done anything to prevent it. The blame was entirely with the driving of Mr Dickinson, yet the appellant was prosecuted under section 3ZB of the 1988 Act for causing the death of Mr Dickinson whilst driving uninsured and without a license. At trial the judge directed the jury that they could only find the appellant guilty if they found he had contributed in a substantial way to Mr Dickinsons death i.e. in a way that was more than minimal. The prosecution appealed this ruling and the Court of Appeal, which felt itself bound by the decision in R v Williams [2010] EWCA Crim 2552, held that the prosecution did not have to prove any element of fault on the part of the appellant, his mere involvement in the fatal collision would be sufficient to commit the offence. The Supreme Court unanimously allows the appeal. Lord Hughes and Lord Toulson jointly give the judgment of the court. If the Court of Appeal were correct, then in this case the appellant would be criminally responsible for Mr Dickinsons death despite not being at fault at all for the collision. In addition, if any of the appellants family had died he would also be criminally responsible for their deaths despite the fact that if Mr Dickinson had survived he would have been guilty of causing death by, at the very least, careless driving when unfit to drive through drugs.[5 6]. It would plainly have been possible for Parliament to legislate in terms which left it beyond doubt that a driver was made guilty of causing death whenever a car which he was driving was involved in a fatal accident, if he were at the time uninsured, disqualified or unlicensed. It did not and instead used expression causesdeathby driving. This imports the concept of causation [19 20]. This is not a case where the concept of a deliberate intervening act applies to break the chain of causation. Mr Dickinson did not deliberately set out to kill himself. This is a case where there are potentially multiple causes of the death. The question is whether the appellants driving was in law a cause [22]. It was not; it was simply an event but for which the collision would not have happened. That would be much the same as saying, if the other driver had hit a tree rather than the defendants vehicle, that whoever planted the tree caused the death. The law draws a distinction between things which are but for circumstances which are just the background to an event, and things which truly cause that event. In R v Williams it was held that s.3ZB must catch cases that did not fall under s.2B (causing death by careless driving) but that case did not focus on the meaning of causesdeathby driving. It does not follow from the fact that section 3ZB contains no requirement that the defendant driver should have committed the offence of careless or inconsiderate driving that he is not required to have done or omitted to do something in the driving of the car which has contributed to the death, before he can be held to have caused it by his driving [24]. The gravity of a conviction for homicide, for which the sentence may be a term of imprisonment, is such that if Parliament wishes to displace the normal approach to causation recognised by the common law, and substitute a different rule, it must do so unambiguously [27]. There is no logical or satisfactory intermediate position between holding (a) that the law imposes guilt of homicide whenever the unlicensed motorist is involved in a fatal accident and (b) that he is guilty of causing death only when there is some additional feature of his driving which is causative on a common sense view, and the latter entails there being something in the manner of his driving which is open to proper criticism. The statutory expression cannot, the Court concludes, be given effect unless there is something properly to be criticised in the driving of the defendant, which contributed in some more than minimal way to the death. It is unwise to attempt to foresee every possible scenario in which this may be true but cases which might fall under s.3ZB but not s.2B (causing death by careless or inconsiderate driving) might, for example, include driving slightly in excess of a speed limit or breach of a construction and use regulation [32]. The trial judges ruling is reinstated. and the matter returned to Newcastle Crown Court.
In 1986, upon the flotation of the TSB Group plc, four Deeds were agreed and executed by which the appellant covenanted to provide four charitable foundations with payments totalling 1% of the TSB Groups pre tax profits. The respondent was one of those charitable foundations. The original Deed was executed in 1986, amended in 1993 and replaced in 1997. Under Clause 2 of the 1997 Deed, the appellant covenanted to pay the respondent the greater of either (a) an amount equal to 0.1946 per cent of the Pre Tax Profits for the relevant Accounting The term Pre Tax Profits was defined in Clause 1 of the Deed as in relation to any Accounting (as the case may be) shown in the Audited Accounts. At the time the Deed was entered into and at all times thereafter up until 2005, only realised profits were included in the consolidated income statement (the modern equivalent of a profit and loss account). This changed in 2005 as a result of the passage of the Regulation (EC) 1606/2002 which required that any gain on acquisition arising from a bargain purchase be recognised on the profit and loss account as of the acquisition date in line with International Financial Reporting Standards requirements. During the financial crisis in 2008, Lloyds TSB Group acquired HBOS. As a result of the acquisition, the appellants group Audited Accounts for 2009 included a figure for gain on acquisition of over 11 billion. This figure reflected the difference between the book value of HBOSs assets and the consideration given by Lloyds Bank of about half that amount. The inclusion of the gain on acquisition had the effect of converting a loss of over 10 billion into a profit before taxation of over 1 billion in the appellants Audited Accounts. The respondent asserts that this latter figure constitutes the group profit before taxation shown in the Audited Accounts, with the effect that they are due to receive a payment of 3,543,333 from the respondent pursuant to Clause 2 of the Deed. The appellant rejects this assertion and contends that it was unthinkable prior to 2005 that an unrealised gain on acquisition would be included in the consolidated income statement; indeed its inclusion would have been contrary to both the law and accounting practice. Accordingly, the appellant contends that the figure for gain on acquisition should be disregarded for the purposes of calculating the payments due to the respondent, with the effect that they should receive the fixed sum of 38,920. The Outer House of the Court of Session found for the appellant and the Lord Ordinary granted decree of absolvitor dismissing the claim. The Inner House of the Court of Session allowed the respondents appeal. The appellant appeals to the Supreme Court on the grounds that on its proper construction the figure for gain on acquisition should be disregarded when calculating the payments due under the Deed. The Supreme Court unanimously allows the appeal and restores the Lord Ordinarys decree of absolvitor dismissing the claim. The Deeds should be understood in the legal and accounting context at the dates when they were executed. In this respect, when the original Deed was made in 1986, amended in 1993 and replaced in 1997, two fundamental legal and accounting principles applied: (a) that a profit and loss account was concerned with ordinary activities before taxation and (b) that only profits realised at the balance sheet date could lawfully be included in the profit and loss account [7]. The Deed should be given a contextual and purposive interpretation [21]. Here the landscape, matrix and aim of the 1997 Deed were concerned with and aimed at realised profits or losses before taxation [22]. The change introduced in 2005 by Regulation (EC) 1606/2002 which required that negative goodwill be recorded in the profit and loss account as a gain on acquisition was wholly outside the parties original contemplation and is something they would not have accepted had they foreseen it [22]. Given that the 1997 Deed did not require an unrealised gain on acquisition to be taken into account in identifying the group profit before taxation, it is circular to try and draw any inference from the fact that the parties did not renegotiate or amend the Deed [14, 17]. Nor does the phrase group profit before taxation shown in the Audited Accounts have the effect of tying the appellant to any similarly phrased line which may be found in a future years Audited Accounts, no matter how different the basis on which that figure is arrived at from that which existed or was in mind when any of the Deeds were executed [20]. As the Deed has not been frustrated, it is necessary to determine how its language best operates in the fundamentally changed and entirely unforeseen circumstances in light of the parties original intentions and purposes; this is best achieved by ignoring the unrealised gain on acquisition in the 2009 accounts [23]. Ignoring the figure for gain on acquisition would not pose difficulties in later accounting periods [26 28]. Indeed, it is inconceivable that the parties could have intended the respondent to derive from an unrealised gain a benefit it could not derive from a realised profit yet this could occur were HBOS to be sold at a profit over and above its fair value as such a realised gain would be excluded from the calculation under the Deed [29]. The doctrine of equitable adjustment forms part of Scots law and resort may be made to it in cases where the contract has become impossible of performance or something essential to its performance has been totally or partially destroyed [46]. However, the Court cannot equitably adjust a contract on the basis that its performance, while not frustrated, is no longer that which was originally contemplated [44, 47].
On 4 March 2005, the appellant (the owners) bought a cruise ship called the New Flamenco (the vessel). The vessel had been chartered to the respondent (the charterers) by its previous owners by way of a time charterparty (the charterparty). By a novation agreement the appellant assumed the rights and liabilities of the previous owner under the charterparty effective as from 7 March 2005. In August 2005, the owners and the charterers concluded an agreement extending the charterparty for two years so that it was due to expire on 28 October 2007. At a meeting on 8 June 2007, the owners and charterers reached an oral agreement extending the charterparty for a further two years, expiring on 2 November 2009. The charterers disputed having made the agreement and maintained they were entitled to redeliver the vessel on 28 October 2007. The owners treated the charterers as in anticipatory repudiatory beach and accepted the breach as terminating the charterparty. The vessel was redelivered on 28 October 2007. Shortly before the delivery the owners agreed to sell the vessel to a third party for US$23,765,000. The owners commenced arbitration in London, as provided for by the charterparty, seeking damages for the charterers repudiatory breach [4]. The arbitrator found that an oral contract to extend the charterparty had been made, the charterers were in repudiatory breach of that contract and therefore the owners were entitled to terminate the charterparty. This finding is unchallenged [9 10]. However, there was a significant difference between the value of the vessel when the owners sold it, and its value in November 2009 (found by arbitrator to be US$7,000,000), when the vessel would have been redelivered to the owners had the charterers not been in breach [6]. The arbitrator declared that the charterers were entitled to a credit for this difference in value, amounting to 11,251,677 (the equivalent of US$16,765,000), which could be discounted from any damages payable by the charterer to the owners from the loss of profit claim. The credit was more than the owners loss of profit claim and would result in the owners recovering no damages [7, 12]. The owners appealed to the High Court pursuant to section 69 of the Arbitration Act 1996 on a question of law, namely whether when assessing the owners damages for loss of profits the charterers were entitled to take into account as diminishing the loss the drop in the capital value of the vessel [14]. Popplewell J held that they were not because the benefit accruing to the owners from the sale of the vessel in October 2007, instead of in November 2009, was not legally caused by the breach [15]. The charterers appealed to the Court of Appeal. The appeal was allowed on the basis that the owners took a decision to mitigate their loss by selling the vessel in October 2007 and there was no reason why the benefit secured by doing this should not be brought into account, in the same way that benefits secured by spot chartering a vessel during an unexpired term of charterparty would be [28]. The owners now appeal that judgment to the Supreme Court. The Supreme Court allows the owners appeal. The charterers are not entitled to a credit for the difference in the value of the vessel when sold in 2007, in comparison to its diminished value in 2009. Lord Clarke, with whom the other justices agree, gives the lead judgment. The fall in the value of the vessel is irrelevant because the owners interest in the capital value of the vessel had nothing to do with the interest injured by the charterers repudiation of the charterparty [29]. This is not because the benefit must be the same kind as the loss caused by the wrongdoer, but because the benefit was not caused either by the breach of the charterparty or by a successful act of mitigation [30]. The repudiation resulted in a prospective loss of income for a period of about two years. However, there was nothing about the premature termination of the charterparty which made it necessary to sell the vessel, at all or at any particular time. It could also have been sold during the term of the charterparty. When to sell the vessel was a commercial decision made at the owners own risk [32]. The owners would not have been able to claim the difference in the market value of the vessel if the market value had risen between the sale in 2007 and the time the charterparty would have terminated in 2009. The premature termination of the charterparty was at most the occasion for selling the vessel, but it was not the legal cause of it. There is equally no reason to assume that the relevant comparator is a sale in November 2009; there is no reason that a sale would necessarily have followed the lawful redelivery at the end of the charterparty term [33]. For the same reasons, the sale of the ship was not on the face of it an act of successful mitigation. If there had been an available charter market, the loss would have been the difference between the actual charterparty rate and the assumed substitute contract rate. Sale of the vessel would have been irrelevant. In the absence of an available market, the measure of the loss is the difference between the contract rate and what was or ought reasonably to have been earned from employment of the vessel under shorter charterparties. The relevant mitigation in that context is the acquisition of an alternative income stream to the income stream under the original charterparty. The sale of the vessel was not itself an act of mitigation because it was incapable of mitigating the loss of the income stream [34]. Popplewell J was therefore correct to hold that the arbitrator erred in principle. The Supreme Court prefers his conclusion to that of the Court of Appeals. His order, setting aside the part of the arbitral award that declared that the charterers were entitled to a credit of 11,251,677 in respect of the benefit that accrued to the owners when they sold the vessel in October 2007 as opposed to November 2009, is therefore restored [36].
In 2009, when the appellant was aged 14, the police charged him with two charges of lewd and libidinous practices at common law and one contravention of section 6 of the Criminal Law (Consolidation) (Scotland) Act 1995. One of the common law charges involved the allegation of showing online pornographic images to a young boy. The other common law charge and the statutory charge involved the allegation of exposing his penis to, and chasing after, three other children, who were girls aged 4, 12 and 13. The police reported the case to the Procurator Fiscal but a decision was made not to prosecute the appellant. In July 2015, when the appellant was aged 19, he was charged with having sexual intercourse with a girl who was aged 14 years and 11 months, contrary to sections 28 and 30 of the Sexual Offences (Scotland) Act 2009 (the 2009 Act). The appellant did not deny that sexual intercourse had taken place, but sought to rely on the defence in section 39(1)(a) of the 2009 Act that at the time he reasonably believed that the girl was aged 16. However, section 39(2)(a)(i) of the 2009 Act provides that the reasonable belief defence is not available to an individual who has previously been charged by the police with a relevant sexual offence. The relevant sexual offences are set out in Schedule 1 to the 2009 Act and cover a wide range of sexual offences, both consensual and non consensual, and include offences in which the age of the victim is not an essential component of the crime. The definition excludes consensual sexual activities between older children. The offences with which the appellant had been charged in 2009 fell within the meaning of a relevant sexual offence and the appellant could not therefore plead the reasonable belief defence. He argued that section 39(2)(a)(i) was not compatible with his rights as set out in the European Convention on Human Rights and therefore, in accordance with section 29 of the Scotland Act 1998, was outside the competence of the Scottish Parliament and was not law. He argued that that the provision breached both the presumption of innocence in Article 6(2) and his Article 8 right to privacy, and was unjustifiably discriminatory for the purposes of Article 14 read with Article 8. The Lord Advocate argued that any interference with the appellants Convention rights was justified in the interests of protecting older children from sexual exploitation. He argued that the prior charge acts as an official warning, alerting the person charged to the importance of a young persons age in relation to sexual behaviour, and therefore justifies depriving that person, if later charged with a sexual offence against an older child set out in section 28 to 37 of the 2009 Act, of the reasonable belief defence. The Supreme Court unanimously allows ABs appeal and remits the proceedings to the High Court of Justiciary. Lord Hodge gives the lead judgment, with which the other Justices agree. Lord Reed gives a concurring judgment, with which Lord Kerr, Lord Wilson and Lord Hughes agree. Article 6 Section 39(2)(a)(i) is not within the ambit of Article 6. Section 39(2)(a)(i) did not create an irrebuttable presumption that the appellant did not have a reasonable belief as to the age of the girl with whom he had sexual intercourse, thereby overriding the presumption of innocence in Article 6(2). The provision created what amounts to a strict liability offence, by treating as irrelevant the accused persons knowledge of the victims age. Such an offence does not violate Article 6(2), which is concerned with procedural guarantees and not with the substantive elements of a criminal offence [21]. Article 8 There was an interference with Article 8 which requires to be justified under Article 8(2) because the prosecutor relied on the earlier police charge in the criminal proceedings [23 24, 56 57]. Lord Hodge concludes that section 39(2)(a)(i) is incompatible with the appellants Article 8 rights [47]. The exclusion of the reasonable belief defence in this case is a disproportionate interference with the appellants Article 8 rights because the prior charges did not give the official warning or notice that consensual sexual activity with children between the ages of 13 and 16 is an offence [44]. Those prior charges were not charges of sexual activity with a child aged between 13 and 16 and therefore did not provide such a warning [29]. The list of relevant sexual offences includes charges in which the age of the victim is not an essential component, extends far beyond consensual sexual activity with an older child and excludes charges where the charged person was an older child at the time of the charge. This suggests that section 32(2)(a)(i) is likely in many other cases to give rise to infringements of article 8 because the prior charge does not objectively give the relevant warning [45, 47]. Lord Reed agrees that the interference with the Article 8 right is not proportionate where the necessary link between the prior charge and the supposed warning does not exist [66]. The difficulty arises from the width of the definition of relevant sexual offences. Since such offences are not confined to sexual conduct which is illegal because it is with children, prior charges of such offences cannot be taken to have alerted the accused to the importance of making sure that a person is over 16 before engaging in sexual activities. Further, since the definition includes non consensual offences, prior charges relating to those offences cannot be taken to have alerted the accused to the importance of age in the context of consensual sexual conduct [64]. The definition also excludes consensual sexual activities between older children, perhaps the clearest example of a situation where the charge alerts the person charged to the importance of the age of consent when engaging in consensual sexual behaviour [65]. Article 14 In light of the conclusions in relation to Article 8, it is unnecessary to discuss this challenge [46, 67]. Remedy It is not possible to interpret section 39(2)(a)(i) narrowly to bring it within the competence of the Parliament [48, 66]. Section 39(2)(a)(i) is therefore not law [66] and proceedings are remitted to the High Court of Justiciary to exercise the power to suspend or vary the effect of this decision [49 50].
In Cadder v HM Advocate [2010] UKSC 43, the Supreme Court held, having regard to the decision of the European Court of Human Rights in Salduz v Turkey (2008) 49 EHRR 421, that the Crowns reliance on admissions made by an accused who had no access to a lawyer while he was being questioned as a detainee at a police station was a violation of his rights under Article 6(3)(c), read with Article 6(1) of the European Convention on Human Rights. The issues in these cases are whether the right of access to a lawyer prior to police questioning, which was established by Salduz, applies only to questioning which takes place when the person has been taken into police custody; and, if the rule applies at some earlier stage, from what moment does it apply. The accused in the first case, John Paul Ambrose, was prosecuted on a charge of contravening section 5(1)(b) of the Road Traffic Act 1988 as being in charge of a motor vehicle while over the alcohol limit. He was questioned by two police officers by the roadside, who cautioned him but did not give him any specification about the offence he was suspected of having committed. In response to their questions, he confirmed that he was in possession of the car keys, and that he might be intending to drive the car. Breath tests indicated that he was substantially over the prescribed limit. In his trial the Crown led evidence of the questions and answers at the roadside. In M, the accused was charged with assault to severe injury, permanent disfigurement and permanent impairment. A few days after the incident, the police visited him at his home, cautioned him, and asked him a number of questions, in response to which he confirmed his attendance at the locus on the night in question and his involvement in the fight. He was detained the following day, and questioned further while he was in custody. At trial, he objected to the Crowns reliance upon the admissions he had made in his home, on the basis that he had not had access to legal advice prior to interview. In G, the accused was indicted with offences including the possession of controlled drugs under the Misuse of Drugs Act 1971 and possession of prohibited firearms and ammunition under the Firearms Act 1968. The police had obtained a warrant to search the accuseds flat. They forced entry and found him there. He struggled, and was handcuffed and cautioned. He admitted to having drugs in his pocket, and responded to a number of questions about items found in the flat. He was subsequently arrested and taken to a police station where he answered further questions. He objected to the Crowns reliance at trial on the statements he made during the course of the search of the flat. In each of the three cases, the Appeal Court of the High Court of Justiciary referred to this Court the question whether the act of the Lord Advocate in leading and relying on the evidence in question would be incompatible with the appellants rights under Article 6(1) and (3)(c) of the European Convention on Human Rights. The Supreme Court, by a majority of 4 to 1, finds that, in the cases of Ambrose and M, the act of the Lord Advocate in leading and relying at the trial on the evidence that was obtained from them in response to police questioning without having had access to legal advice was not incompatible with the Article 6(1) and (3)(c) right; and in the case of G that it was incompatible. In Ambrose and M, the question whether, taking all the circumstances into account, it would be fair to admit this evidence, is left open for decision by the Appeal Court and Sheriff Court respectively. Lord Hope gives the leading judgment. Lord Kerr gives a separate dissenting judgment finding in all three cases that the evidence would be inadmissible. In each of these three cases, the circumstances differ from those before the Supreme Court in Cadder and before the Grand Chamber in Salduz, in that the evidence in question was obtained through police questioning before the individuals were detained at a police station. The Supreme Court notes, firstly, that the jurisdiction of this court is limited to a consideration of the devolution issue which is raised by each of these references, and does not extend to ruling on how the circumstances referred to in each case would fall to be dealt with under domestic law. Secondly, it notes that a decision by this court that there is a rule that a person who is suspected of an offence but is not yet in custody has a right of access to a lawyer before being questioned by the police would have far reaching consequences for the investigation of crime by the authorities. Therefore, if Strasbourg has not yet spoken clearly on this issue, the court would be wise to wait until it has done so [14 15]. The duty of the domestic court in interpreting the Convention is to keep pace with the Strasbourg case law as it evolves over time. There is no obligation upon domestic courts to do more than that (R (Ullah) v Special Adjudicator [2004] UKHL 26, para 20 per Lord Bingham of Cornhill) [17]. The courts task in this case is to identify where the Strasbourg court stands on this issue. It is not for this court to expand the scope of the Convention right further than the jurisprudence of the Strasbourg court justifies [20]. In domestic law, where an individual has not yet been detained under section 14 of the Criminal Procedure (Scotland) Act 1995, the test for the admissibility of answers given to questions put by police is whether or not there was unfairness on the part of the police. The fact that the person did not have access to legal advice when being questioned is a circumstance to which the court may have regard in applying the test of fairness, but it carries no more weight than that. There is no rule in domestic law that provides that police questioning of a person without access to legal advice who is suspected of an offence but is not in police custody must always be regarded as unfair. The question is whether a rule to that effect is to be clearly found in the jurisprudence of the Strasbourg court [25]. The Grand Chamber in Salduz had in mind the need to protect an accused against abusive coercion while in custody. The judgment appears to have been concerned only with establishing a rule that there was a right of access to a lawyer where the person being interrogated was in police custody [33]. That assessment is supported by subsequent Strasbourg case law, in particular Zaichenko v Russia (Application no.39660/02), the only case to date in which the complaint was of lack of legal assistance during police questioning when the applicant was not in custody [46]. If the Salduz judgment were to apply to statements made by a person in response to police questioning before being taken into custody, the court would have had to have said so expressly. It did not do so [35]. The privilege against self incrimination is not an absolute right (Murray v United Kingdom (1996) 22 EHRR 29, para 47). It is primarily concerned with respecting the will of the person to remain silent (Saunders v United Kingdom (1996) 23 EHRR 313, para 68), and a person is free to confess if he is willing to do so. Police custody or its equivalent creates a need for protection of the accused against abusive coercion. The same is not the case for questioning at the locus or in a persons home [54]. In principle, the line as to when access to legal advice must be provided before the person is questioned should be drawn as from the moment that he has been taken into police custody, or his freedom of action has been significantly curtailed [55]. The correct starting point when considering whether the persons Convention rights have been breached is to identify the moment at which he is charged for the purposes of Article 6(1); that is whether his situation is substantially affected (Deweer v Belgium (1980) 2 EHRR 439, para 46; Eckle v Germany (1982) 5 EHRR 1, para 73). That will be the case as soon as the suspicion against him is being seriously investigated and the prosecution case compiled [62]. The fact that a person who has become a suspect and is not in custody is questioned without access to legal advice will be a relevant factor in the assessment whether the accused was deprived of a fair hearing, but it will be no more than that. In Ambrose and M, the question is whether the act of the Lord Advocate in leading and relying on evidence obtained in response to police questioning, conducted under common law at the roadside or at the accuseds home, without the accused having had access to legal advice, was incompatible with Article 6(1) and (3)(c). This is answered in the negative. Ambrose was charged for the purposes of Article 6 when he was cautioned. Suspicion that he was committing an offence fell on him as soon as he told the police that the keys were in his pocket [67]. M was charged when he was cautioned by the police officer at his home [69]. But it would be to go further than Strasbourg has gone to hold that the appellants are entitled to a finding that this evidence is inadmissible because, as a rule, access to a lawyer should have been provided to him when he was being subjected to questioning at the roadside [68 & 70]. The question whether, taking all the circumstances into account, it would be fair to admit this evidence, is left open for the Appeal Court and Sheriff Court respectively. In G, the question whether it is incompatible with his Convention rights for the Lord Advocate to lead evidence of his statements made during the course of the search is answered in the affirmative. He was charged for the purposes of Article 6 by the time the police began their search. The difference with this case was that there was a significant curtailment of Gs freedom of action. He was detained and had been handcuffed, and was, in effect, in police custody from that moment onwards. The circumstances were, therefore, sufficiently coercive for the incriminating answers that he gave to the questions that were put to him without access to legal advice to be inadmissible [71]. The same result need not, however, follow in every case where questions are put during a police search to a person who is to be take to have been charged for the purposes of Article 6; that, again, would be going further than Strasbourg has gone [72]. Lord Kerr would have found the evidence in question to be inadmissible in all three cases. It is not open to courts of this country to refrain from recognising a claim to a Convention right simply because Strasbourg has not spoken clearly on the matter [128]. In practice, it is inevitable that many claims to Convention rights will have to be determined by the UK courts without the benefit of unequivocal jurisprudence from Strasbourg. It is the duty of every domestic court to resolve the question of whether a claim to a Convention right is viable or not, even where the jurisprudence of the Strasbourg court does not disclose a clear current view [129]. As regards the right of access to a lawyer, the selection of the moment of being taken into custody as the first occasion on which legal representation becomes necessary is both arbitrary and illogical. The judgment in Salduz indicates that the need to have a lawyer is not to be determined on a geographical or temporal basis but according to the significance of what is taking place when the admissions in question are made [136]. The essential question is: when the questioning is taking place, is the suspect in a position where the advice of a lawyer is essential if a fair trial is to occur. If he is liable to incriminate himself at that time, a lawyers presence is required [145]. The judgment in Zaichenko is not clear, but does not indicate that formal arrest and interrogation in custody are essential prerequisites to the invocation of the right to legal assistance [158].
This matter centres around a married man and woman who, until 2015, had been living together in Australia with their two children. By the end of 2014 the marriage was in difficulties. The mother, who holds British citizenship, wanted to make a trip to England with the children before returning to work from maternity leave. The father agreed to an eight week stay. The mother and the children came to England on 4 May 2015 where they have since remained. Discussions between the mother and father resulted in the father agreeing to an extension of the eight week visit up to a year. Based on the extension, the mother gave notice to her employer and looked for work in England. In September 2015, the mother enrolled the older child at a local pre school. Without telling the father, on 2 November 2015, she applied for British citizenship for both children who had entered England on six month visitor visas. Her solicitors wrote a letter to the immigration authorities on her behalf indicating that she and the children could not return to Australia for fear of domestic abuse. In continuing correspondence, the father pressed the mother on the childrens expected date of return. The mother indicated that she did not know what her plans were but made clear that she would not be returning in May 2016. In June 2016, she expressed her intention to remain in the UK. The father made an application in the High Court under the Convention of 25 October 1980 on the Civil Aspects of International Child Abduction (the Abduction Convention). The issue of when the mother had decided not to return to Australia was in contention. The mothers own case was that by April 2016 she had felt she and the children would not be returning. The arguments before the Court meant that, on any view, there was a decision not to return to Australia before the expiry of the agreed year. The judge held that the children were habitually resident in England and Wales by the end of June 2016 so that mandatory summary return was unavailable under the Abduction Convention. But he accepted mothers evidence that she did not have the intention, in November 2015, or before April 2016, not to return to Australia. The mother now appeals against the Court of Appeals decision. The issues in the appeal are: (1) what is the effect on an application under the Abduction Convention if a child has become habitually resident in the destination state before the act relied on as a wrongful removal or retention occurs; and (2) if a child has been removed from their home state by agreement with the left behind parent for a limited period can there be a wrongful retention before the agreed period of absence expires (so called repudiatory retention)? The father cross appeals on the issue of habitual residence. The Supreme Court allows the appeal and dismisses the cross appeal. Lord Hughes gives the lead judgment with whom Lady Hale and Lord Carnwath agree. Lord Kerr and Lord Wilson each give judgments concurring on the two points of principle but dissenting on the outcome of this case on its facts. Issue 1 When considering the general scheme of the Abduction Convention, the construction that summary return is available if, by the time of the act relied on as a wrongful removal or retention, a child is habitually resident in the state where the application for return is made is unpersuasive. That construction is inconsistent with the operation of the Abduction Convention since 1980 and its treatment by subsequent international legal instruments. [19] The Abduction Convention is designed to provide a summary remedy which negates the pre emptive force of wrongful removal or retention and to defeat forum shopping. [21] The point of the scheme adopted by the Abduction Convention was to leave the merits to be decided by the courts of the place of the childs habitual residence. If the forum state is the habitual residence of the child, there can be no place for a summary return to somewhere else, without a merits based decision. This understanding of the scheme of the Abduction Convention is reflected in the provisions of both the Revised Brussels II Regulation and the 1996 Hague Convention on Recognition, Enforcement and Cooperation in respect of Parental Responsibility and Measures for the Protection of Children. [23] The Abduction Convention cannot be invoked if by the time of the alleged wrongful act, whether by removal or retention, the child is habitually resident in the state where the request for return is lodged. In such a case, that state has primary jurisdiction to decide on the merits, based on the childs habitual residence, and there is no room for a mandatory summary decision. [34] Issue 2 Repudiatory retention has been recognised in some jurisdictions, but no generally accepted international practice or authority exists on the point. [39] The desirability of inducing a prompt change of mind in the retaining parent is an argument for recognising a repudiatory retention when and if it occurs. The 12 month time limit for seeking mandatory summary return runs from the point a repudiatory retention occurs and that period may pass before an applicant is aware of the repudiatory retention. However, it is not a limitation period but a provision in the childs interest to limit mandatory summary return. Once elapsed it renders a summary return discretionary. The concern that repudiatory retention would make Abduction Convention applications longer and more complicated is a point well made. However, Family Division judges are used to managing applications actively and controlling any tendency to spill outside the relevant issues. Further, if repudiatory retention requires an overt act or statement, this lessens the danger of speculative applications. [46 48] Repudiatory retention is possible in law. The objections to it are insubstantial, whereas the arguments in favour are convincing and conform to the scheme of the Abduction Convention. It would be unwise to attempt an exhaustive definition of proof or evidence. An objectively identifiable act of repudiation is required, but it need not be communicated to the left behind parent nor does an exact date need to be identifiable. [50 51] On the present facts there could not have been a wrongful retention in April 2016 as the mothers internal thinking could not by itself amount to such. If she had such an intention in November 2015, the application to the immigration authorities could have amounted to a repudiatory retention. But it was open to the judge to believe the mothers evidence that she did not possess this intention in November. [55] There is no basis in law for criticising the judges decision as to habitual residence. [57] Lord Kerr dissents on the outcome of this case on its facts. He expresses misgivings about repudiatory retention requiring an overt act by the travelling parent. [63] The judges finding that wrongful retention did not arise in this case could not be reconciled with his statement that the mother had concluded by April 2016 that she and the children should remain in England. [68] Moreover, the judges conclusion that the mother had not formed any intention to retain the children in England in November 2015 is insupportable as he failed to address the question of what bearing the letter of November 2015 had on her intention. [72] Lord Wilson also dissents on the outcome of this case on its facts. The solicitors letter to the immigration authorities in November 2015 represented a major obstacle to any finding that the mother had not by then intended to keep the children in the UK indefinitely. The judges finding as to the mothers intention in November 2015 was flawed and the Court of Appeal were correct to order a fresh inquiry into her intention. [91 92]
This appeal concerns the legality under the Human Rights Act 1998 of an Enhanced Criminal Record Certificate (ECRC) issued in respect of the appellant (AR) under section 113B of the Police Act 1997. An ordinary Criminal Records Certificate is limited to the facts of convictions or cautions or their absence. By contrast, an ECRC includes information on the basis simply of the Chief Officers opinion as to its relevance, and whether it ought to be included in the certificate. In January 2011, AR was acquitted of rape by the Crown Court. He was a married man with children, of previous good character, and a qualified teacher, but was working at the time as a taxi driver. It was alleged that he had raped a woman who was a passenger in a taxi driven by him. His defence was that there had never been sexual contact with the victim. Following his acquittal, he applied for an ECRC in the course of an application for a job as a lecturer. The ECRC was issued with details of the rape charge for which he had been tried and acquitted. AR objected to this disclosure on the basis that there had been no actual conviction and it failed to give a full account of the evidence given and how the jury came to its conclusion. The judge and the Court of Appeal dismissed ARs appeal against the disclosure, holding that it was reasonable, proportionate and no more than necessary to secure the objective of protecting young and vulnerable persons. The main issue before the Supreme Court was whether the admitted interference with ARs rights under article 8 of the European Convention of Human Rights (ECHR) due to the disclosure was justified. There was also a question as to the proper role of an appellate court in reviewing the judges finding of proportionality under the ECHR. The Supreme Court unanimously dismisses the appeal. Lord Carnwath gives the judgment with which all the other justices agree. The leading authority on the operation of the ECRC regime is the Supreme Court decision in R (L) v Comr of Police of the Metropolis [2010] 1 AC 410 (Ls case). In Ls case, the ECRC disclosed details of alleged inadequate parental supervision by the applicant of her child. It was held that although article 8 was engaged, the essential issue was whether the disclosure was a proportionate interference with her private life, and that in the particular circumstances of the case, the significance of the information in respect of risk to children outweighed the prejudicial effect of the disclosure on the applicants employment prospects [22 29]. Following the initial hearing before the Supreme Court in the present case, the court sought more detailed information about the guidance available to chief officers and potential employers as to the operation of the ECRC system, and also any evidence about its impact on those affected. The resulting picture was not entirely clear or consistent [30 41]. On the issue of the proper role of the appellate court in approaching proportionality, Lord Carnwath noted that the purpose of the appeal is to enable the reasoning of the lower court to be reviewed and errors corrected, not to provide an opportunity for parties to reargue the same case [57]. The question in relation to the standard of review is whether the judge erred in principle or was wrong in reaching the conclusion which he did [61]. To limit intervention to a significant error of principle, as the Court of Appeal did, is too narrow an approach if it is taken as implying that the appellate court has to point to a specific principle which has been infringed by the judgment of the court below. The decision may be wrong because of an identifiable flaw in the judges reasoning which undermines the cogency of the conclusion. It is equally clear, however, that it is not enough that the appellate court might have arrived at a different evaluation. In the present case, it was sufficient for the Court of Appeal to consider whether there was any such error or flaw in the judges treatment of proportionality, and if there was not, there was no obligation for it to make its own assessment [64 65]. The procedural aspect of the complaint under article 8 is essentially that there was a lack of consultation, and this was rightly rejected for the reasons given by the judge as endorsed by the Court of Appeal. The officers were fully aware from the evidence at trial of the nature of ARs defence, and his personal circumstances, and they were aware and took account of the potential impact on his employment prospects. As the judge said, there was no indication of any further information he would have wished to advance [66]. On the substantive effect of article 8, Lord Carnwath rejected the submission that the interference involved in the disclosure could not be justified unless the officers were in a position to form a positive view of likely guilt, which could not be done without a full appraisal of the evidence at trial. He did not accept that, as a matter of domestic law or under article 8, it is necessary or appropriate for those responsible for an ECRC to conduct a detailed analysis of the evidence at the trial [67 68]. The judge went no further than to accept, as he was entitled to do, the Chief Constables view that the information was not lacking substance and that the allegations might be true. It was a matter for him to assess whether the information was of a sufficient wright in the article 8 balance. It should be borne in mind that the information about the charge and acquittal was a matter of public record, and might have come to the potential employers knowledge from other sources. The judge took full account of the possible employment difficulties for AR, but regarded them as no more than necessary to meet the pressing social need for which the ECRC process was enacted [69 70]. Lord Carnwath notes in a postscript to his judgment that although he has reached a clear conclusion on the limited issues raised by this appeal, it gives rise to more general concerns about the ECRC procedure in similar circumstances. There is no clear guidance as to what weight should be given to an acquittal in different circumstances, and there is a lack of information about how an ECRC is likely to be treated by a potential employer in such a case. Careful thought needs to be given to the value in practice of disclosing allegations which have been tested in court and have led to an acquittal [72 76].
Article 52(1) of the European Patent Convention (the EPC) provides that, in order to obtain a European patent, an invention must be susceptible of industrial application. Article 57 states that an invention is susceptible of industrial application if it can be made or used in any kind of industry. The primary issue in this case is the way in which the requirement of industrial applicability extends to a patent for biological material. The Appellant is the proprietor of European Patent (UK) 0,939,804 (the Patent). It describes the encoding nucleotide, the amino acid sequence, and certain antibodies of a new human protein called Neutrokine , and includes contentions as to its biological properties and therapeutic activities, as well as those of its antibodies. These contentions are predictions substantially based on the proposition that Neutrokine is a member of the THF ligand superfamily. The Patent was filed on 25 October 1996 and granted on 17 August 2005. The Respondent brought opposition proceedings in the Opposition Division of the European Patent Office (the EPO), following which the Patent was revoked. The Appellant appealed to the Technical Board of Appeal (the Board) of the EPO, which allowed the appeal and ordered that the Patent be maintained. Meanwhile, the Respondent brought parallel proceedings in the High Court for revocation of the Patent in the UK. The High Court revoked the Patent, on the basis that, in the light of the common general knowledge, the notional addressee of the Patent (a person skilled in the art) would have concluded that the functions of Neutrokine were, at best, a matter of expectation and then at far too high a level of generality to constitute a sound or concrete basis for anything except a research project. This decision was upheld by the Court of Appeal. The Supreme Court unanimously allows the appeal, dismisses the cross appeal, and remits the case to the Court of Appeal to deal with the outstanding issues. The leading judgments are given by Lord Neuberger and Lord Hope, with whom the other justices agreed. There is very little UK authority on the topic of industrial applicability, particularly as regards biological material [37] and [88], and the applicable principles are really to be found in the jurisprudence of the EPO and the Board [42]. While the reasoning in each decision of the Board is not binding upon national courts, the courts should normally follow the jurisprudence of the EPO, particularly where the Board has adopted a consistent approach to an issue in a number of decisions [84] and [87], as is the case with regard to the application of Article 57 to patents for biological material [88]. Further, there are strong policy reasons for seeking consistency of approach to patents in the biological field, as it is import for bioscience companies to be able to decide at what stage to file for patent protection, and to be able to obtain funding based on patent protection [96 102] and [141 143]. Despite the very wide ranging and generalised suggestions in the Patent as to the uses to which Neutrokine and its antibodies might be put, over and above revealing the existence and structure of the new protein and its encoding gene, the only relevant guidance in the Patent ultimately arises from its teaching as to the tissue distribution of Neutrokine , its expression in T cell and B cell lymphomas, and the fact that it is a member of the TNF ligant superfamily. The question is whether the Judge in the High Court was right, or at least entitled, to conclude that the inferences which would have been drawn from the Patent specification in 1996 would not have been enough to satisfy Article 57 [103]. That conclusion was based on the fact that the Patent neither revealed how Neutrokine could be used to solve any particular problem nor identified any disease or condition which it could be used to diagnose or treat [104] and [161]. That reasoning was not consistent with the approach adopted by the Board, from which a number of general and specific principles may be drawn [106 107]. In light of those principles, the disclosure of the existence and structure of Neutrokine and its gene, and its membership of the TNF ligand superfamily should have been sufficient, taking into account the common general knowledge, to satisfy the requirements of Article 57 [109]. This is because all known members of the TNF ligand family were expressed on T cells and were able to co stimulate T cell proliferation, and therefore Neutrokine would be expected to have a similar function [111]. The fact that the members of that superfamily were known to have pleiotropic effects is irrelevant where the value of the new member relates to the common features manifested by all known members [112 115]. Neither the Judge nor the Board considered that the unsatisfactory drafting of the Patent would actually have diverted the person skilled in the art from what their search of the literature, coupled with common general knowledge, would otherwise have led them to understand represented the teaching of the Patent [116 118]. The lower courts were wrong to focus on the speculative nature of some of the therapeutic uses of Neutrokine as disclosed in the Patent, and the degree of extra effort required to determine those uses, when the known activities of the superfamily were enough in themselves to justify patentability for the disclosure of a novel molecule (and its encoding gene) [119 121], [124 128] and [161]. For the same reason, the Respondents argument that the specification of the Patent is insufficient must fail [132 139]. The standard set by the Judge for susceptibility to industrial application was a more exacting one than that used by the Board. He was looking for a description that showed that a particular use for the product had actually been demonstrated, rather than that the product had plausibly been shown to be usable for the purposes of research work [151] and [154], which the Board must be taken to have regarded as an industrial activity in itself [155 156]. Notwithstanding the importance of deference to the findings of fact and value judgments of a court of first instance, especially where that decision is confirmed on appeal [94 95], [166], [168 170 and 172], in this case it is evident that the Judge and Court of Appeal failed to follow the principles of law clearly set out by the Board in this and previous cases. The appeal must therefore be allowed.
The appellant, Mr Carlyle, is a property developer. In 2007 he purchased a plot of land for development at Gleneagles, Perthshire from the Gleneagles Hotel. He had to complete the construction of a new house on the plot by 31 March 2011, before the Ryder Cup was due to be staged at Gleneagles golf course. The purchase was subject to a buy back clause entitling the vendor to re purchase the plot for the original price if the construction was not completed on time [9]. Mr Carlyle funded the purchase by taking a loan from the respondent, The Royal Bank of Scotland Plc (the bank). On 26 March 2007 Mr Carlyle met with representatives of the bank to discuss the proposed loan. The buy back clause was discussed and Mr Carlyle made it clear that he would need to borrow money to build the house as well as to purchase the plot [9]. In subsequent telephone calls he reiterated that the bank should not lend him the purchase money unless it was also committed to providing him with development funding [10]. On 14 June 2007 the banks representative told him by telephone that his proposal was all approved and Mr Carlyle accordingly paid a deposit to the vendor to secure the purchase [11]. In August 2008 the bank informed Mr Carlyle that it would not provide funding for construction and called in the loan. On 14 August 2008 the bank raised an action against Mr Carlyle for the payment of 1,449,660 plus interest. Mr Carlyle defended the action and counter claimed for his loss of profit on the development [3]. The central issue in the case was whether, on an objective assessment of what the parties said to each other, the bank intended to enter into a legally binding promise to lend Mr Carlyle money for not only the purchase but also the development of the plot [1]. On 10 May 2010 the Lord Ordinary declared the bank was in breach of a collateral warranty to make development funding of 700,000 available to Mr Carlyle [5]. He held that the telephone conversation of 14 June 2007, set in the context of the previous discussions, represented a commitment by the bank both to advance the purchase price and to provide a facility for the build cost [13]. The bank appealed. On 12 September 2013 the Second Division of the Inner House allowed the banks reclaiming motion [6], holding that: (i) in the conversation of 14 June 2007 the bank had simply informed Mr Carlyle of an internal decision to approve funding in principle; (ii) the bank was not under any legal obligation until there was a written loan agreement; and (iii) the alleged promise was legally ineffective because essential terms, including the maximum draw down, had not been agreed [18]. Mr Carlyle appealed to the Supreme Court. The Supreme Court unanimously allows the appeal, sets aside the interlocutor of the Second Division and remits the case to a commercial judge in the Court of Session to proceed accordingly [38]. Lord Hodge, with whom Lord Neuberger, Lord Kerr, Lord Clarke and Lord Reed agree, gives the judgment. Lord Hodge notes the limited power of an appellate court to reverse the findings of fact of the judge who has heard the evidence [2]. He comments that had he been deciding the matter at first instance, and if the findings of fact record all the material evidence, he might have shared the view of the Second Division that the bank had not entered into a legally binding obligation to provide the development funding [20]. However, when deciding that the trial judge has gone plainly wrong, the appeal court must be satisfied that the judge could not reasonably have reached the decision under appeal [21]. The rationale for this is both that the judge who has heard the evidence will have a deeper insight in reaching conclusions of fact and the different role assigned to the appellate court [22]. The Second Division disagreed with the Lord Ordinary on questions of fact without facing up to the restricted role of the appellate function on such questions [23]. The Lord Ordinary had a reasonable evidential basis for finding on an objective analysis that the bank made a legally binding promise in the telephone call of 14 June 2007 to provide development funding. He might have interpreted the evidence differently and concluded that there was no binding commitment, but he did not have to [25]. The fact that parties envisage that their agreement will be set out in a formal contract in the future does not, by itself, prevent that agreement from taking legal effect [25]. Although Mr Carlyle and the bank knew that the 14 June 2007 commitment would be superseded by more detailed loan agreements, this did not prevent it from having effect as a legally binding promise [26]. The fact that a previous loan transaction between Mr Carlyle and the bank had been conducted differently was not relevant, because in the earlier transaction there had been no buy back clause [30]. It was open to the Lord Ordinary to reach the conclusion he did despite the relatively ill defined nature of the obligation to provide the development funding [29]. The parties had proceeded on the basis that Mr Carlyle would need up to 700,000 for the development of the plot [27]. They were aware of the rates of interest applied to other loans, and the time constraints on the development of the plot [29]. Once the Lord Ordinary was satisfied that the bank had the intention to make a legally binding promise, he was entitled and indeed required to look for ways to give effect to that promise [29]. The pleading of a collateral warranty became a distraction in this case. It was not used as a term of article Either promise or unilateral undertaking would be a suitable description for the independent legal obligation under consideration [33]. In English contract law, the doctrine of consideration gives rise to the concept of a collateral contract, in which one partys promise or representation is given in exchange for the other party entering into the envisaged (separate) contract. In Scots law a unilateral undertaking that is intended to have legal effect, such as a promise, is binding without consideration passing from the recipient of that promise. The promise may be, but does not need to be, collateral to another contract. The issue is simply whether a legally binding obligation has been undertaken [35].
The issues in these appeals relate to the right to a fair trial. Alison McGarrigle had a son, Robert, by her former husband. Robert was subject to a residential supervision order requiring him to live with his father during the week but permitted him to visit his mother on Saturdays. On 14 June 1997 Robert did not return to his fathers address and instead he and his mother went to live with the appellants in a house in Largs. A drinking session took place there on or about the 20 June 1997 at which a number of people including the appellants, Robert and Mrs McGarrigle were present. The next morning she was gone and was never seen by Robert again. She was reported to police as missing on 16 February 1998. The investigation continued but in the meantime, on 17 June 1998, the appellants were convicted of sexual offences including offences against Robert McGarrigle and were sentenced to 6 and 8 years imprisonment respectively. Whilst serving their sentences the appellants were taken by police for questioning on suspicion of conspiracy to murder Alison McGarrigle. They were asked by the officers whether they were involved in her murder, but they both remained silent. Owing to a lack of evidence at that time, proceedings were not commenced against the pair. The appellants were eventually charged in 2005 for the murder of Mrs McGarrigle and remanded in custody. On 10 June 2010 the appellants were found guilty in the High Court of Justiciary at Glasgow of the murder of Mrs Allison McGarrigle between 21 June and 1 September 1997 and of a subsequent attempt to defeat the ends of justice by disposing of her body in the sea. In a separate trial held immediately before, the appellants were found guilty of a series of sexual offences relating to children. Both trials took place in front of the same judge, Lord Pentland, but with different juries. After the verdict in the first trial the Advocate Depute moved for sentence and handed the judge a list of the appellants previous convictions. The judge reserved sentencing for the sexual offences until after the trial for murder was complete. At the time of informing the appellants of this, the judge referred to their records and made comments to them that they were evil, determined, manipulative and predatory paedophiles of the worst sort. The two issues for the Supreme Court were: (1) when the appellants were charged for the purposes of their right to a trial within a reasonable time in terms of article 6(1) of the Convention (the appellants argued that time started to run when they were first questioned in 1998 and therefore there had been a breach of their right); and (2) whether the comments and conduct of the trial judge were such as to breach the appellants right to a fair trial by an impartial tribunal in terms of article 6(1) of the Convention and, if so, whether the act of the Lord Advocate in persevering with the trial was incompatible with the appellants rights under article 6(1). Both issues arose from the refusal of the Appeal Court to grant leave for the relevant grounds of appeal to be argued in the appeal in Scotland. The Appeal Court did however grant permission to appeal its refusal to the Supreme Court. The Supreme Court held that it had jurisdiction to consider the issues on the basis that they were compatibility issues in terms of the Criminal Procedure (Scotland) Act 1995 (as amended by the Scotland Act 2012), issue (1) being an appeal against a decision of the Appeal Court and issue (2) being a reference from the Appeal Court. The court determines the two compatibility issues as follows: (1) that the date when the reasonable time began for the purposes of the appellants article 6(1) Convention right was 5 April 2005; and (2) that the Lord Advocates act in proceeding with the trial on the murder charges was not incompatible with the appellants article 6(1) right to a trial before a tribunal that was independent and impartial. The proceedings will be remitted to the High Court of Justiciary [58]. Lord Hope gives the judgment of the court. The meaning of the word charged has been considered in a number of cases regarding article 6(1), which provides that in the determination of any criminal charge against him a person has the right to a fair trial within a reasonable time and article 6(3)(c) which provides a right to legal assistance for anyone charged with a criminal offence [25 32]. The focus of article 6(3)(c) is on the state of affairs when the suspect is first interrogated, as to wait until the stage is reached when there is sufficient evidence to charge before the suspect has the right of access to a lawyer could seriously prejudice his right to a fair trial. This is in contrast with the reasonable time guarantee of article 6(1): it relates to the running of time, not on what is needed to preserve the right to a fair trial. The rationale is the person should not remain too long in a state of uncertainty. Time runs from the date which the suspects position is substantially affected by the official notification. In the United Kingdom this could be some time after he is first questioned [33 34]. The date from which reasonable time begins is the subject of a separate guarantee from the guarantee that the trial will be fair and falls to be approached independently [36]. The appellants were certainly not at any stage of their interviews charged in the formal sense. They were both asked directly whether they killed Mrs McGarrigle. But, in the context in which these questions were being put, it cannot be said that this amounted to an official notification that they were likely to be prosecuted [37]. In the absence of any evidence to show where, when and how she had died, the police were in no position to initiate criminal proceedings. In August 2003 they received information that led to further enquiries and resulted in the appellants being charged with murder in 2005 [38]. On the issue of apparent bias, the test is contained in Porter v Magill [2001] UKHL 67 and considered in a number of authorities [47 52]. It would only be if the judge expressed outspoken opinions about the appellants character that were entirely gratuitous, and only if the occasion for making them was plainly outside the scope of the proper performance of his duties, that the fair minded and informed observer would doubt the judges ability to perform those duties with an objective judicial mind. The context indicates that nothing of the kind happened in this instance [53 54]. Furthermore, no objection was made by the defence at any point to the fact that Lord Pentland was to preside over the murder trial as well and there are no grounds for doubting his impartiality [55 56].
This appeal is concerned with the employment, by the Secretary of State for Children, Schools and Families, of teachers to work in the European Schools. These are schools set up to provide a distinctively European education principally for the children of officials and employees of the European Communities. The Staff Regulations, made by the Board of Governors pursuant to the Convention defining the Statute of the European Schools, limit the period for which teachers may be seconded to work in those schools to a total of nine years (or exceptionally ten). This is made up of an initial probationary period of two years, and a further period of three years, which is renewable for a further four years (the nine year rule). The principal question in the appeal is whether these arrangements can be objectively justified, as required by the Fixed term Employees (Prevention of Less Favourable Treatment) Regulations 2002 (SI 2002/2034) (the Fixed term Regulations). This was the measure chosen by the United Kingdom to implement Council Directive 1999/70/EC concerning the framework agreement on fixed term work (the Fixed term Directive). The effect of regulation 8 is that a successive fixed term contract is turned into a permanent employment unless the use of such a contract can be objectively justified. Mr Fletcher was employed by the Secretary of State and seconded to work in the European School in Culham, Oxfordshire, from 1 September 1998 until 31 August 2008. After his two year probationary period, he was employed for a further three year period, extended for a further four years, and then an additional one year. In 2007, he claimed that he was a permanent employee by virtue of regulation 8. Mr Duncombe was a teacher at the European School in Karlsruhe, Germany, from January 1996 until 31 August 2006. He too was employed under a series of fixed term contracts to reflect the nine year rule. He brought claims in the Employment Tribunal for wrongful dismissal or pay in lieu of notice, unfair dismissal and a declaration that he was a permanent employee. The Employment Tribunal, the Employment Appeal Tribunal and the Court of Appeal all held that the use of the successive fixed term contracts is not objectively justified. The Supreme Court unanimously allows the appeal, holding that it was objectively justified to employ these teachers on the current fixed term contracts and accordingly that these were not converted into permanent contracts by the operation of regulation 8 of the Fixed term Regulations. Lady Hale gives the leading judgment. The teachers complaint is not against the three or four periods comprised in the nine year rule but against the nine year rule itself. In other words, they are complaining about the fixed term nature of their employment rather than about the use of the successive fixed term contracts which make it up. But that is not the target against which either the Fixed term Directive or the Regulations are aimed. Employing people on single fixed term contracts does not offend against either the Directive or the Regulations. [23] The targets against which the Directive and Framework Agreement were directed were discrimination against workers on fixed term contracts and abuse of successive fixed term contracts in what was in reality an indefinite employment. It is not suggested that the terms and conditions on which the teachers were employed during their nine year terms were less favourable than those of comparable teachers on indefinite contracts. [24] It is not the nine year rule which requires to be justified, but the use of the latest fixed term contract bringing the total period up to nine years. And that can readily be justified by the existence of the nine year rule. The teachers were employed to do a particular job which could only last for nine years. The Secretary of State could not foist those teachers on the schools for a longer period, no matter how unjustifiable either he or the employment tribunals of this country thought the rule to be. The teachers were not employed to do any alternative work because there was none available for them to do. [25] It is not a question of whether the Staff Regulations trump the Directive. There is no inconsistency between them. The Staff Regulations are dealing with the duration of secondment, not with the duration of employment. [26]
Section 31 (2) of the Children Act 1989 imposes a threshold which must be satisfied before a care or supervision order can be made in respect of a child. First the child must have suffered or be likely to suffer significant harm; secondly, that harm must be attributable to the care given or likely to be given to the child. If the threshold is crossed then the court will treat the welfare of the child as its paramount consideration when deciding whether to make an order. The issue in this case is whether a child can be regarded as likely to suffer harm if another child has been harmed in the past and there is a possibility that the parent now caring for him or her was responsible for the harm to the other child. The local authority in this case brought care proceedings in respect of three children who are cared for by DJ and JJ. The two oldest are the children of DJ and his former partner, and have always lived with DJ. The youngest child is JJs daughter, her third child with her former partner, SW. The local authority submitted that the three children were likely to suffer significant harm because JJs first child with SW, T J, had died of non accidental injuries in 2004. In earlier care proceedings relating to JJ and SWs second child, who was subsequently adopted, a judge had found that either JJ or SW had caused the injuries to T J and the other had at the very least colluded to hide the truth. In the present proceedings the local authority sought to rely solely on the finding that JJ was a possible perpetrator of the injuries to T J. It submitted that this was a finding of fact sufficient as a matter of law to satisfy the s 31(2) threshold in respect of the three children now cared for by JJ and DJ. The High Court held on a preliminary issue that likelihood of significant harm can only be established by reference to past facts that are proved on the balance of probabilities. Mere possibility was insufficient. The Court of Appeal dismissed an appeal by the local authority but granted permission to appeal to the Supreme Court. The Supreme Court unanimously dismisses the local authoritys appeal. The main judgment is given by Lady Hale, with whom all the justices agree. Lord Wilson expresses disagreement on one point, which Lord Sumption shares. Lord Reed gives an additional judgment, with which Lord Clarke and Lord Carnwath agree. Lord Hope agrees with Lady Hale and Lord Reed. It is a serious matter for the state compulsorily to remove a child from his family of birth. The section 31(2) threshold is an important measure to protect a family from unwarranted intrusion while at the same time protecting children from harm [1] [75]. The wording of Section 31(2) has been the subject of six appeals to the House of Lords and Supreme Court. Those cases have consistently held that a prediction of future harm has to be founded on proven facts: suspicions or possibilities are not enough. Such facts have to be proved on the simple balance of probabilities [36]. This approach is supported by the legislative history of section 31(2) [45 46] [96]. It would be odd if the first limb (actual harm) had to be proved to the courts satisfaction but the basis of predicting future harm did not [47]. Care cases in which the only matter upon which the authority can rely is the possibility that the parent has harmed another child in the past are very rare. Usually there will be many readily provable facts upon which an authority can rely [5]. Even in cases where the perpetrator of injuries could not be identified there may be a multitude of established facts from which a likelihood that this parent will harm a child in the future could be shown. However, the real possibility that the parent caring for the child has harmed a child in the past is not by itself sufficient [54]. In this case there were many potentially relevant facts found in the earlier proceedings against JJ which might have been relevant to an assessment of whether JJ would harm children in the future, such as the collusion with SW which prevented the court from identifying the perpetrator, the failure to protect T J, and the deliberate failure to keep T J away from health professionals [56]. Other relevant matters for the assessment would have been consideration of the household circumstances at the time of T Js death and whether JJs new relationship with DJ looking after much older children was different [53]. As the local authority had chosen not to rely on these facts, however, it would not be fair to the whole family to allow these proceedings to go on. JJ has been looking after these three children and a new baby for some time without (so far as the court is aware) giving cause for concern and, should the local authority wish to make a case that any of these children is likely to suffer significant harm in the future, it will be open to it to bring new proceedings [57]. Lord Wilson, while agreeing with Lady Hale for the most part and in the disposal of the appeal, identified an issue on which he differed from the majority. In his view, since the consignment of a person to a pool of possible perpetrators of injuries to one child could not constitute a factual foundation for a prediction of likely significant harm to another child in his or her care, then as a matter of logic, it could not become part of the requisite foundation in combination with other facts and circumstances [80]. Lord Sumption agreed [92].
This appeal concerns the circumstances in which an asylum seeker should be sent back to the country where he or she first claimed asylum if it is claimed that such a return would expose the asylum seeker to the risk of inhuman or degrading treatment, which is forbidden by article 3 of the European Convention on Human Rights (ECHR). At this stage the appellants account of the risk that they face must be assumed to be true. They are an Iranian national (EH) and three Eritrean nationals (EM, AE, and MA) who have come to the United Kingdom via Italy. In each of their cases Italy is the country responsible for processing their asylum applications according to the relevant EU law, Council Regulation 343/2003 (commonly known as Dublin II). The basis of EHs asylum claim is that he was tortured as a political prisoner in Iran. He is now severely psychologically disturbed and needs treatment. He claims that if he were returned to Italy he would be homeless and without treatment. EM, AE, and MA were left homeless and destitute in Italy. AE and MA, who are women, claim that they were repeatedly raped there, despite having been recognised as refugees. MA has come to the UK with two of her children; a third was separated from the family during the attempt to make it here and has not been found. AEs experiences have traumatised her, and she is suicidal at the thought of being taken back to Italy. Italy is one of a list of countries which is presumed by the United Kingdom to be safe for returning asylum seekers. The Home Secretary therefore must be satisfied that the appellants claims that they will be subject to degrading and inhuman treatment are not clearly unfounded if they are to be allowed to stay in the United Kingdom while they pursue their asylum applications. That is important to the appellants because of the threats to their well being if they were returned to Italy. The Home Secretary certified all of the appellants claims as clearly unfounded because Italy was not in systemic breach of its international obligations to treat asylum seekers with dignity. The Court of Appeal considered that a systemic breach, rather than merely a breach, of those obligations was indeed required before the United Kingdom could decline to return an asylum seeker to Italy. The Court of Appeal reached that conclusion on the basis of a decision of the Court of Justice of the European Union (CJEU), NS (Afghanistan) v Secretary of State for the Home Department. The CJEU is responsible for interpreting EU law, including Dublin II. However, the Court of Appeal read the decisions of the European Court of Human Rights (ECtHR) as requiring only a breach, rather than a systemic breach, of a persons human rights. The ECtHR is responsible for interpreting the ECHR, and belongs to a separate legal system established by the Council of Europe. By virtue of legislation in the UK, decisions of the CJEU are binding on UK courts, while decisions of the ECtHR need only be taken into account. The Court of Appeal therefore felt bound to apply the CJEU case, as it understood it, over the ECtHR cases. Since it held that Italy was not in systemic breach of its duties, it found for the Home Secretary. The Supreme Court unanimously allows the asylum seekers appeals and remits all four cases to the administrative court to determine on the facts whether in each case it is established that there is a real possibility that, if returned to Italy, the claimant would be subject to treatment in violation of the Convention. The Court of Appeal was wrong to consider that only a systemic breach by the receiving country of its human rights obligations would justify not returning an asylum seeker to that country. The CJEUs judgment in NS had to be read according to the context in which it was given. While it did refer to a systemic breach, such a breach was well established on the cases facts. The CJEUs focus was therefore not on the sort of breach that had to be established, but rather on EU member states awareness of such a breach. There was therefore no warrant for concluding that CJEUs judgment was that there had to be a systemic breach; it only meant that a systemic breach would be enough. The CJEU was not calling into question the well established test applied in human rights law, which is that the removal of a person from a member state of the Council of Europe to another country is forbidden, if it is shown that there is a real risk that the person transferred will suffer treatment contrary to article 3 of the ECHR [5658]. Indeed, the EU requires its laws to be interpreted in accordance with fundamental rights, such as those guaranteed by the ECHR. And beyond that it is clear that the EU scheme of asylum law in general is to be applied in a way that respects the dignity of asylum seekers, and ensures a basic minimum standard of support. Council Directive 2003/9/EC (commonly known as the Reception Directive) requires that member states provide asylum seekers with at least enough to sustain their health and ability to subsist. And under Council Directive 2004/83/EC (the Qualification Directive), those granted refugee status are not to be discriminated against in terms of access to welfare support, accommodation, and so on [59 60]. These duties coalesce with the positive obligations on members of the Council of Europe who are also member states of the European Union. Article 4 of the EU Charter of Fundamental Rights contains a human rights protection in equivalent language to article 3 of ECHR. The UK, as an EU member state, is obliged to observe and promote the application of the Charter whenever implementing an instrument of EU law. There was no dispute before this Court that the positive obligations under article 3 of ECHR include the duty to protect asylum seekers from deliberate harm by being exposed to living conditions (for which the state bears responsibility) which cause ill treatment. And in R (Limbuela) v Secretary of State for the Home Department the House of Lords held that article 3 ECHR could be engaged where asylum seekers were by the deliberate action of the state, denied shelter, food or the most basic necessities of life [62]. Where, therefore, it can be shown that the conditions in which an asylum seeker will be required to live if returned under Dublin II are such that there is a real risk that he will be subjected to inhuman or degrading treatment, his or her removal to that state is forbidden. The evidence about breaches of a positive obligation is more likely to concern systemic failings, but a focus on such failings is only by way of establishing that there is a real risk of a breach of article 3, rather than a distinct hurdle to be surmounted [63].
A, a foreign national, arrived in the UK in 1991. He was later granted indefinite leave to remain, but in 1996 was sentenced to four years imprisonment for sexual offences against a child. In 1998, he was served by the Home Secretary with a notice to make a deportation order [4]. A appealed against the decision and protracted proceedings followed in which A cited risks due to his status as a known sex offender of death or ill treatment (contrary to articles 2 and 3 of the European Convention on Human Rights (ECHR)) should he be deported. As identity was withheld in the proceedings from 2001 onwards [5] [9]. In dismissing As appeal against the Home Secretarys refusal to revoke the deportation order, the First tier tribunal noted that the proceedings were anonymised, thus reducing the risk of As identification. In September 2012 A applied to the Court of Session for judicial review of the Upper Tribunals refusal of permission to appeal. The Secretary of State gave notice that she intended to remove A from the UK before the date fixed for the hearing of the judicial review application. On 7 November 2012 Lord Boyd heard As application for interim suspension of the removal decision and allowed him to amend his application for judicial review by substituting initials for his name and address. Lord Boyd also gave directions under section 11 of the Contempt of Court Act 1981 prohibiting the publication of As name or other identifying details and directing that no picture of him should be published or broadcast [2]; [10] [13]. He refused the application for interim suspension, concluding that A had not established a prima facie case for setting aside the Upper Tribunals decision [14]. The Inner House refused As appeal against that decision [16]. Media organisations had not been notified of and were not represented at the hearing before Lord Boyd. The BBC became aware of the section 11 order and applied for it to be set aside. In refusing the application, Lord Glennie noted that it was accepted that if the fact that A was being deported became known in his country of origin there would be a real risk of his article 3 rights being infringed. That was why an anonymity order had been made by the tribunal. It was necessary to withhold As identity to safeguard his article 3 rights and to preserve the integrity of the court proceedings, since publication of the information would give A grounds for a fresh application to the Home Secretary and frustrate the proceedings [17 18]. A was deported on 14 December 2012 [20]. The BBCs appeal against Lord Glennies decision was refused in May 2013. The Inner House considered that the material before the tribunal justified the conclusion that anonymity would be a significant protection of As article 3 rights and that setting aside the section 11 order would subvert the understanding on which As deportation had been authorised [21]. The appeal raises three issues [3]: (i) Whether the court possesses any common law power to protect the anonymity of a party where rights under the ECHR (given effect by the Human Rights Act 1998 (HRA)) are engaged; (ii) Whether the court acted compatibly with the BBCs rights under article 10 ECHR (which protects freedom of expression), both in terms of the substance of its decision and the procedure followed; and (iii) Whether the section 11 order fell within the scope of section 12 HRA, so that the BBC ought to have been notified and given an opportunity to make representations before it was made. In a unanimous judgment delivered by Lord Reed, the court dismisses the BBCs appeal. Lord Reed explains the importance of the general constitutional principle of open justice. Society depends on the courts to act as guardians of the rule of law, and this in turn necessitates the openness of the courts to public scrutiny. The principle has important consequences for the publishing of reports of court proceedings: open justice is inextricably linked to the freedom of the media to report on court proceedings [23] [26]. But there are exceptions. The courts have an inherent jurisdiction to determine how the principle of open justice should be applied [27] [37] and can permit the identity of a party or witness to be withheld from public disclosure where necessary in the interests of justice [38 41]. Central to the courts evaluation will be the purpose of the open justice principle, the potential value of the information in advancing that purpose, and any risk of harm that its disclosure may cause to the maintenance of an effective judicial process or to the legitimate interests of others [41]. The principle of open justice is protected and qualified by the ECHR as it is in domestic law [42] [54]. But the common law principle remains in vigour even where Convention rights also apply [55 57]. The purpose of section 11 is to support the exercise of the courts power to allow a name or other matter to be withheld in court proceedings, by conferring a statutory power to give ancillary directions prohibiting publication of the relevant information. The use of section 11 is not limited to protecting the public interest in the administration of justice [60], or to cases where members of the public are present in court [61]. Section 12 HRA does not apply to section 11 applications as they are not applications for relief made against any person [62 66]. Fairness nevertheless requires the media to have an opportunity to be heard, but both this and the medias right to an effective remedy are secured by enabling any person affected to seek recall of the order promptly at a hearing inter partes [67] [68]. Whilst article 10 ECHR was engaged in the present case, the arguments in favour of making the order were overwhelming. The tribunal had made a decision, the effect of which was to authorise As deportation, on the basis that anonymity would be a significant protection of his article 3 rights. The courts failure to make the order would have meant that the deportation might create all the risks that the tribunals directions as to anonymity had been intended to prevent. The order was justifiable under article 10, since it was both prescribed by law and necessary in a democratic society in order to protect the integrity of the legal proceedings and As article 3 rights [69] [76]. The order allowing A to withhold his identity was in accordance with the courts common law powers. The section 11 order was made in accordance with the power conferred by that provision. It was not incompatible with the BBCs Convention rights [75 76]. The BBC was able to apply promptly for recall of the order, and its application came before the court two days after the order was made [77].
The appellant ("the wife") used to be married to the respondent ("the husband"), a former solicitor. In 2002, the wife petitioned for a divorce. In response to her financial claims, the husband asserted that all of his ostensible wealth represented assets held on behalf of his clients [4 6]. In 2004, the wife's claims were settled at a Financial Dispute Resolution (FDR) meeting. The settlement order (the 2004 Order) provided that the husband should make a lump sum payment in final settlement of the wife's capital claims (which was eventually paid), and periodical annual payments (which the husband stopped paying in 2008). The 2004 Order included a recital that "the [wife] believes that the [husband] has not provided full and frank disclosure of his financial circumstances (although this is disputed by the [husband]), but is compromising her claims in the terms set out in this consent order despite this in order to achieve finality" (the Recital) [7 9]. In 2007, the wife applied by notice issued within the divorce proceedings to set aside the 2004 Order on the ground that the husband had fraudulently failed to disclose his assets. These proceedings were delayed, largely because in 2008 the husband was charged with serious money laundering offences dating from mid 2005. He was eventually convicted and committed to prison in 2011, and confiscation proceedings against him are ongoing [10 11]. In September 2012, after an eight day hearing, Moylan J set aside the 2004 Order [12 14]. His decision was made on the basis both that (a) there had been material non disclosure by the husband when the 2004 Order was made and, had he made full disclosure, the outcome would have been different, and (b) because the wifes evidence satisfied the criteria in Ladd v Marshall (which govern when fresh evidence may be adduced on appeal) it followed that her application should be allowed [24 25]. The Court of Appeal allowed the husbands appeal. It held that Moylan J had incorrectly applied the Ladd criteria and was wrong to allow the wifes application on that basis [24]. However, it held that the Ladd criteria were relevant in order to establish what evidence the wife could adduce in order to establish material non disclosure by the husband. Applying those criteria to the evidence before Moylan J, and discounting other inadmissible evidence, the Court of Appeal concluded that there was no admissible evidence to support Moylan Js conclusions on material non disclosure [26 29]. The Supreme Court unanimously allows Mrs Gohils appeal and reinstates Moylan Js order. Lord Wilson (with whom Lord Neuberger, Lady Hale, Lord Clarke, Lord Sumption, Lord Reed and Lord Hodge agree) gives the leading judgment. Lord Neuberger gives a short concurring judgment. The husband had suggested that, as a judge of the High Court, Moylan J did not have jurisdiction to set aside an order of the High Court. This argument was not pursued in the Court of Appeal, but the Supreme Court makes the following observations: (a) the Court of Appeal has long recognised that it is an inappropriate forum for inquiries into non disclosure issues raised in proceedings for the setting aside of a financial order; (b) this is shown by the present case, where an intensive fact finding hearing was necessary; (c) there is an urgent need for definitive confirmation of the High Courts jurisdiction to set aside a financial order made in that court; (d) the Supreme Court endorses the conclusion of the Family Procedure Rule Committee in relation to its Setting Aside Working Party, set out in the minutes of its meeting on 20 April 2015 [16 18]. The Recital Words such as those used in the Recital have no legal effect in a financial order in divorce proceedings. The husband owed a duty to the court to make full and frank disclosure of his resources, without which the court would be disabled from discharging its duty under section 25(2) of the Matrimonial Causes Act 1973. One spouse cannot exonerate the other from complying with this duty [19 22]. Criteria in Ladd v Marshall The Ladd criteria have no relevance to the determination of an application to set aside a financial order on grounds of fraudulent non disclosure [32]. The Court of Appeal was wrong to accept an argument that the criteria should apply to determine what evidence could be adduced because: (a) the Court of Appeal would not have conducted the necessary fact finding exercise, so the criteria for determining the admissibility of evidence in that court were irrelevant; (b) the first Ladd criterion presupposes that there has been a trial whereas, in this case, the wifes first opportunity to adduce the evidence was at the hearing before Moylan J; (c) the argument would not apply to an application to set aside a financial order made by a district judge and the evidential criteria should not depend on the level of court, and (d) the argument ignores the fact that, had the wifes claims proceeded to trial in 2004, the duty would have lain on the husband, not on her, to explain his resources [31]. In light of the erroneous approach to the admissibility of the wifes evidence, the dismissal of her set aside application cannot stand [33]. Consequences To decide whether Moylan Js order could be reinstated, it was necessary to consider what admissible evidence was before him and ask whether he would properly have found that the husband had been guilty of material non disclosure in 2004 [33 35]. Through no fault of his own, Moylan J had relied on evidence from the husbands criminal proceedings obtained from sources outside the UK (which had since been held inadmissible and had been discounted by the Court of Appeal) [13 15, 33]. However, even if Moylan J had referred only to the remaining admissible evidence [36 40], he would, in the light of his findings on it, still have concluded that the husband was guilty of material non disclosure [42]. Lord Neuberger agrees that Moylan Js order can be reinstated. Several factors make it clear that the material non disclosure issue should not be remitted, provided that there is no risk of injustice to the husband [49 55]. The court has to be satisfied that: (a) Moylan J would have decided that there had been material non disclosure even if he had not received the inadmissible evidence; or (b) looking at the totality of the admissible evidence, it could safely be concluded that there had been material non disclosure; or (c) if the issue was remitted, the judge could only realistically come to that conclusion in light of the totality of the admissible evidence [56 57]. For the reasons given by Lord Wilson, all three of these requirements were satisfied [58 61].
This appeal arises out of an action for damages for harassment and for an injunction to restrain its continuance. The question at issue is in what circumstances can such an action be defended on the ground that the alleged harasser was engaged in the prevention or detection of crime. Mr Willoughby was employed by one of Mr Hayess companies. In 2002, the two men fell out. In late 2003, Mr Willoughby embarked on a campaign against Mr Hayes centring on allegations of fraud, embezzlement and tax evasion in relation to Mr Hayess management of his companies. This took the form of sending numerous letters to the Official Receiver, the police and the Department of Trade and Industry. These bodies investigated and found no basis in the allegations, but Mr Willoughby continued to press these bodies and made a series of intrusions into Mr Hayess private life. The Protection from Harassment Act 1997 (the Act) makes harassment a civil wrong and a criminal offence, but under s.1(3) of the Act it is a defence for a person to show (a) that it was pursued for the purpose of preventing or detecting a crime; (b) that it was pursued under any enactment or rule of law, or (c) that in the particular circumstances, the pursuit of the course of conduct was reasonable. The trial judge found that Mr Willougbys conduct constituted harassment under s.1(1) of the Act but that he had a defence under s.1(3)(a) because he genuinely believed in the allegations involving Mr Hayes and wished to persist in investigating them. The Court of Appeal allowed Mr Hayess appeal on two main grounds: (1) only the purpose of the conduct not the purpose of the alleged harasser was relevant, and in this case it was not reasonably or rationally connected to the prevention of crime; and (2) the prevention of crime had to be the sole purpose of the alleged harasser, and the intrusions on Mr Hayess privacy were not related to that purpose. The Supreme Court dismisses the appeal by Mr Willoughby by a majority of four to one (Lord Reed Dissenting). Lord Sumption gives the judgment of the Court. There is no distinction between the purpose of the conduct and the purpose of the alleged harasser as such acts have no purpose other than that of their perpetrator. The issue is by what standard that persons purpose is to be assessed [10]. A wholly objective test (adopted by the Respondent) is not consistent with the wording or purpose of the Act. A test of reasonableness was not included in s.1(3)(a), as it was in other sections of the Act. It would also render the general defence of reasonableness in s.1(3)(c) otiose [11]. A wholly subjective test (adopted by the Appellant) is equally problematic [12]. Those who claim to be acting for the purpose of preventing or detecting crime may, at a purely subjective level, entertain views about what acts are crimes and what steps are calculated to prevent or detect them which have no relation to reality. Mere existence of belief, however absurd, in the mind of the harasser that he is detecting or preventing a crime, cannot justify him persisting in a course of conduct which the law recognises as oppressive. Some control mechanism is therefore required, even if it falls short of what is objectively reasonable [13]. The necessary control mechanism is to be found in the concept of rationality, familiar in public law but also increasingly significant in other areas, such as contractual discretions. Rationality is different to reasonableness. Reasonableness is an external, objective standard applied to the outcome of a persons thoughts or intentions. A test of rationality only applies a minimum objective standard to the relevant persons mental processes. It imports a notion of good faith in requiring some rational connection between the evidence and the ostensible reasons for the decision, and an absence of arbitrariness, capriciousness or reasoning so outrageous in its defiance of logical as to be perverse [14]. If the alleged harasser has rationally applied his mind to the material suggesting criminality and formed the view that the conduct said to constitute harassment was appropriate for its detection or prevention, the court will not test his conclusions by reference to what view a hypothetical reasonable man in his position would have formed. If he has not done so but proceeds anyway, he acts irrationally. He will not have a relevant purpose and there will be no causal connection between his purpose and the conduct constituting harassment. Such a test would in any event apply to public authorities. It is not a demanding test, and it is hard to imagine that Parliament could have intended anything less [15]. Applied to the facts, this tests means that after June 2007, Mr Willoughbys conduct against Mr Hayes was more than objectively unreasonable. It was irrational. He was no longer guided by any assessment of evidence, nor was there a rational connection between his supposed purpose and acts. By persisting in pressing his allegations on the Official Receiver and other investigatory bodies long after they refused to deal with him, he was acting in way that was incapable of furthering the alleged purpose [16]. Although not strictly necessary to decide the point, it was also held that for the purpose of s.1(3)(a) the prevention or detection of crime need not be the sole purpose of the alleged harasser, but only the dominant one [17]. Lord Reed (dissenting) agrees that reasonableness is not required under s.1(3)(a), but rejects the idea that Parliament intended to impose a rationality requirement for three reasons: (1) Parliament did not provide for any rationality test. (2) A statute should not be construed as extending criminal liability beyond the limits which Parliament itself enacted it. (3) Criminal liability would turn on the subtle distinction between irrationality and unreasonableness, which could create particular difficulties in giving clear directions to juries [24 28].
This appeal arises from the insolvency and administration of the Lehman Brothers group of companies. Lehman Brothers International (Europe) (LBIE) was the principal European trading company in the group and is incorporated in England as an unlimited company with its head office in London. LBIE is authorised and regulated by the Financial Services Authority (FSA). Its ultimate holding company is Lehman Brothers Holdings Inc (LBHI), incorporated in Delaware and based in New York. LBHI is now in Chapter 11 bankruptcy. LBIE was put into administration by order of the High Court made before the opening of business (at 07.56am) on Monday 15 September 2008: [24]. Many difficulties have arisen in the administration and the administrators have made several applications to the Companies Court for directions. The application which gives rise to this appeal relates to the provisions governing client money, namely chapter 7 (Client money: MiFID business) of the Clients Assets Sourcebook issued by the FSA (CASS 7), made under the Financial Services and Markets Act 2000 (FSMA): [25] [26]. CASS 7 provides for a normal approach and an alternative approach to discharging a firms client money segregation requirements. LBIE adopted the alternative approach. 7.4.16G of CASS 7, among other things, provides that Under the alternative approach, client money is received into and paid out of a firms own bank accounts A firm that adopts the alternative approach will segregate client money into a client bank account on a daily basis, after having performed a reconciliation of records and accounts of the entitlement of each client for whom the firm holds client money with the records and accounts of the entitlement of each client for whom the firm holds in client bank account and client transactions accounts to determine what the client money requirement was at the close of the previous business day. CASS 7 (7.7.2R) further provides that A firm receives and holds client money as trustee (or in Scotland as agent) Where the firm fails, this is known as a primary pooling event (PPE). 7.9.6R provides that If a primary pooling event occurs: (1) client money held in each client money account of the firm is treated as pooled; and (2) the firm must distribute that client money in accordance with CASS 7.7.2R, so that each client receives a sum which is rateable to the client money entitlement calculated in accordance with CASS 7.9.7R. A secondary pooling event occurs on the failure of a third party to which client money held by the firm has been transferred (7.9.14R). 7.9.13R provides that if both a primary pooling event and a secondary pooling event occur, the provisions relating to a primary pooling event are to apply: [44]. In an ideal world, the flawless operation of the scheme created by the CASS 7 rules would ensure that, upon a firms insolvency, the clients would receive back their money in full, free from the claims of the firms creditors. In the imperfect and highly complex real word occupied by LBIE and its numerous clients, there has been a falling short in the achievement of these objectives on a truly spectacular scale. This is a result of two prime causes: first, on the basis of the assumed facts, LBIE failed to identify as client money, and therefore also failed to segregate, vast sums received from or on behalf of a significant number of its clients; second is the failure of another LBIE affiliate, Lehman Brothers Bankhaus AG (Bankhaus) with which LBIE had deposited at least US$1 billion of segregated client money: [27]. Before the administration order, the last internal reconciliation of clients funds took place on the morning of Friday 12 September 2008. This appeal has proceeded on assumed facts and raises three issues concerning the true construction of CASS 7: (i) when does the statutory trust created by 7.7.2R arise; (ii) do the primary pooling arrangements apply to client money held in house accounts; and (iii) is participation in the notional client money pool (CMP) dependant on actual segregation of client money? [128] The Supreme Court dismisses the appeal by a majority (Lords Hope and Walker dissenting as to the second and third issues): the statutory trust under CASS7 arises on receipt of client money; the primary pooling arrangements apply to client money in house accounts; and, participation in the CMP is not dependant on the segregation of client money. As to when the statutory trust arises, the two competing answers are time of receipt and time of segregation of the funds. The Court unanimously holds that the trust arises at time of receipt. Where money is received from a client or from a third party on behalf of a client it would be unnatural, and contrary to the primary purpose of client protection, for the money to cease to be the clients property on receipt, and for it (or its substitute) to become his property again on segregation. It would also be contrary to the natural meaning of the comprehensive language of CASS 7.7.2R. Segregation without a trust would not achieve MiFIDs objective whereas under the alternative approach an immediate trust of identifiable client money does provide protection: [62] [63], [182] [183]. Lord Hope further supports this conclusion on the basis that the same result would be obtained using the law of agency, as it is in Scots law: [7] [14]. As to whether participation in the notional CMP is dependant on actual segregation of the clients money, the answer is to be found in the proper interpretation of CASS 7, in particular 7.9.6R, 7.9.7R and 7.7.2R: [139]. It does not depend on a consideration of any general principles of trust law: [145]. The general scheme of CASS 7 is that all client money is subject to a trust that arises upon receipt of the money. The client money rules are therefore intended to protect all the clients money received prior to a PPE. The distribution rules are intended to protect all the clients money in the event of a PPE, all client money is subject to a statutory trust and where there is a choice of possible interpretations, the court should adopt the one which affords a high degree of protection for all clients: [147]. The language of the relevant provisions of CASS 7 tends to support the claims basis for participation in the CMP (i.e. that no segregation is required). As the linguistic points are not conclusive, it is necessary to stand back from the detail and ask which interpretation better promotes the purpose of CASS 7. This clearly supports the claims basis: [159]. As to whether the primary pooling arrangements apply to client money held in house accounts, it is necessary to decide whether 7.9.6(1)R requires all identifiable client money to be treated as pooled, or only that client money which is held in the firms segregated client accounts. Since an examination of the text shows that there are two possible interpretations of the relevant CASS 7 rule, the correct interpretation is the one which best promotes the purpose of CASS 7 as a whole, namely to provide a high level of protection for client money. To exclude identifiable client money in house accounts from the distribution regime runs counter to this policy. The alternative would provide different levels of protection based on the happenstance of whether the firm has segregated money, which is arbitrary: [164] [165]. The primary pooling arrangements apply to client money in house accounts: [167].
These appeals concern the obligations of insurance companies under various contracts of employers liability (EL) insurance. In particular, the appeals concern the scope of the insurers obligations to indemnify employers against their liabilities towards employees who have contracted mesothelioma following exposure to asbestos. Mesothelioma has an unusually long gestation period, which can be in excess of 40 years between exposure to asbestos and manifestation of the disease. The insurers maintain that the EL policies only cover mesothelioma which manifested as a disease at some point during the relevant policy period. In contrast, the employers submit that the insurance policies respond to mesothelioma caused by exposure to asbestos during the relevant policy period but which develops and manifests itself sometime later. The usual rule in negligence cases is that the claimant must establish on the balance of probabilities that the defendants negligence caused his injury or disease. In Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22 and Barker v Corus UK Ltd [2006] UKHL 20 the House of Lords developed an exception to this general principle in cases involving mesothelioma caused by exposure to asbestos. The effect of this special rule is that an employer is liable where exposure to asbestos contributed to the risk that the employee would suffer mesothelioma and where the employee in fact develops the disease. The insurers submit that the special rule in Fairchild/Barker is not applicable when deciding, for the purposes of an EL insurance policy, whether an employees mesothelioma was caused by exposure to asbestos during a particular policy year. At first instance Burton J held that the policies should all be interpreted as having a causation wording. He therefore held that the liability trigger under the EL policy was when the employee inhaled the asbestos and not the date when the malignant lesion developed. A majority of the Court of Appeal (Rix and Stanley Burnton LJJ) upheld the judge in relation to some of the EL insurance policies (particularly those covering disease contracted during the relevant insurance period); however they concluded that other policies (particularly those covering disease sustained during the insurance period) responded only on an occurrence or manifestation basis. These appeals to the Supreme Court raise two issues: (i) On the correct construction of the EL policies, is mesothelioma sustained or contracted at the moment when the employee is wrongfully exposed to asbestos or at the moment when the disease subsequently manifests in the employees body? (ii) Does the special rule in Fairchild/Barker apply when determining whether, for the purposes of the EL policies, an employee sustained or contracted mesothelioma during a particular policy period? The Supreme Court dismisses the insurers appeal by a 4 1 majority; Lord Phillips dissenting on the second issue. Lord Mance gives the main judgment. To resolve the meaning of the EL policies it is necessary to avoid over concentration on the meaning of single words or phrases viewed in isolation, and to look at the insurance contracts more generally [19]. Several features point the way to the correct construction. First, the wordings of the policies on their face require the course of employment to be contemporaneous with the sustaining of the injury [20]. Second, the wordings demonstrate a close link between the actual employment undertaken during each period and the premium agreed by the parties for the risks undertaken by the insurers in respect of that period. Third, on the insurers case there is a potential gap in cover as regards employers breaches of duty towards employees in one period which only lead to disease or injury in another later period [24]. Fourth, on the insurers case employers would be vulnerable to any decision by the insurers not to renew the policy. A decision not to renew might arise from the employers complying with their duty to disclose past negligence upon any renewal. Employers who discovered that they had been negligent in the course of past activities in respects that had not yet led to any manifest disease would have such a duty. The insurers could then simply refuse any renewal or further cover [25]. Fifth, the way most of the policies deal with extra territorial issues throws doubt on any suggestion that the wordings are so carefully chosen that a court should stick literally to whatever might be perceived as their natural meaning [28]. Section 1 of the Employers Liability Compulsory Insurance Act 1969 also points the way to the correct interpretation. This states that every employer shall insure, and maintain insuranceagainst liability for bodily injury or disease sustained by his employees, and arising out of and in the course of their employment. In order to give proper effect to the protective purpose of that legislation, the Act requires insurance on a causation basis [47]. There is no difficulty in treating the word contracted as looking to the causation of a disease, rather than its development or manifestation. The word contracted used in conjunction with disease looks to the initiating or causative factor of the disease [49]. While the word sustained may initially appear to refer to the manifestation of an injury, the nature and underlying purpose of the EL insurances is one which looks to the initiation or causation of the accident or disease which injured the employee. Accordingly a disease may properly be said to have been sustained by an employee in the period when it was caused or initiated, even though it only developed or manifested itself later [50]. In relation to the second issue, the question is whether the EL policies cover employers liability for mesothelioma arising under the special rule in Fairchild/Barker [71]. Under that rule the law accepts a weak or broad causal link between the employers negligence and the employees mesothelioma. When construing the EL policies the concept of a disease being caused during the policy period must be interpreted sufficiently flexibly to embrace the role assigned to exposure by the Fairchild/Barker rule [74]. The purpose of the EL policies was to insure the employers against liability to their employees. Once it is held that the employers are liable to the employees, it would be remarkable if the insurers were not liable under the policies [88]. Accordingly, for the purposes of the EL policies, the negligent exposure of an employee to asbestos during the policy period has a sufficient causal link with subsequently arising mesothelioma to trigger the insurers obligation to indemnify the employer [74]. Lord Phillips dissents on the second issue. The special approach developed in Fairchild/Barker raises no implication or fictional assumption as to when mesothelioma is initiated. The consequence is that if claimants have to show that mesothelioma was initiated in a particular policy year in order to establish that insurers are liable they are unable to do so. This conclusion is not affected by section 3 of the Compensation Act 2009, which did not alter the jurisprudential basis of the Fairchild/Barker approach [132] [133].
This case concerns a pay protection provision in the 2002 version of NHS Terms and Conditions of Service for Hospital Medical and Dental Staff and Doctors in Public Health Medicine and the Community Health Service (England and Wales) (the NHS Terms). The Appellant is a doctor specialising in oral and maxillo facial surgery. She trained as a dentist in India but later qualified as a doctor and has been in the UK since 1996. From late 2006 to early 2007, the Appellant worked for the Luton and Dunstable Hospital NHS Foundation Trust as a trust grade doctor in oral surgery. She was subsequently employed by the Respondent as a Foundation Year 1 Pre Registration House Officer. That was a training position and was a necessary step to the Appellant qualifying as a consultant. The contract between the Appellant and the Respondent was governed by the NHS Terms. A document called the Pay Circular, to which reference was made in the NHS Terms, set out the basic rates of pay per annum of doctors and dentists in each of fifteen pay grades. Each pay grade had a number of levels of pay (referred to in the NHS Terms as incremental points). The Appellants position with the Luton and Dunstable Hospital NHS Foundation Trust was treated as falling within the lowest pay grade, namely Hospital Practitioner. The pay for that grade was determined, not on the basis of an annual salary, but on the basis of the number of sessions worked. A session consisted of a working period of 3.5 hours. The pay for the post was limited to a maximum of five sessions per week. The appellant in fact only worked two sessions per week. The NHS Terms included a pay protection provision which, in essence, enabled a doctor moving to a lower paid training post to retain his/her previous salary. Paragraph 132 of the NHS Terms provided: Where a practitioner takes an appointment in a lower grade which is recognised by the appropriate authority as being for the purpose of obtaining approved training the practitioner shall, while in the lower grade, continue to be paid on the incremental point the practitioner had reached in his or her previous appointment. Such a practitioner shall receive the benefit of any general pay awards. On reappointment to the higher grade or on appointment to another higher grade, the practitioner's starting salary should be assessed as if the period spent in the approved training post had been continuing service in the previous higher grade Paragraph 135 was an interpretative provision which provided: a. the rate of salary for a part time practitioner shall be taken to be the corresponding point in the salary scale, except for a practitioner employed as a part time medical or dental officer under paragraphs 94 or 105, for whom it shall be the maximum amount appropriate to nine notional half days The question in the present case concerns the meaning of those provisions. The Appellant brought a claim against the Respondent arguing, amongst other things, that there had been an unlawful deduction of her wages contrary to Part 2 of the Employment Rights Act 1996 (a concept that includes a failure to pay an employee what is contractually due to him or her). The Appellant argues that she is entitled to the basic hourly rate she had received whilst working for the Luton and Dunstable Hospital NHS Foundation Trust for each hour worked in her new post for the Respondent. By contrast, the Respondent argues that the Appellants pay protection should be limited to five sessions per week, i.e. the maximum period she could have worked in her previous post. The Employment Tribunal agreed with the Respondent. The Employment Appeal Tribunal reversed that decision. The Court of Appeal (Elias LJ dissenting) restored the decision of the Employment Tribunal. The Supreme Court unanimously allows the appeal and restores the decision of the Employment Appeal Tribunal. The case is remitted to the Employment Tribunal in order to determine the outstanding issues identified in the order of the Employment Appeal Tribunal. Lord Carnwath gives the only judgment. The different approaches adopted by the various judges who have considered this case demonstrate that paragraph 132 of the NHS Terms is not well drafted. It is disturbing that a provision designed to confer important rights on employees should be so obscure. The provision, however, is unlikely to cause problems in the future [21, 22]. The issue in this case has to be resolved by applying the ordinary principles of contractual interpretation according to which the object of the court is to ascertain the intention of the parties by examining the words they used and giving them their ordinary meaning in their contractual context [26]. The critical words in paragraph 132 of the NHS Terms are the practitioner shall continue to be paid on the incremental point the practitioner had reached in her previous appointment. The incremental point is a reference to the relevant point in the scale for the practitioners grade as shown in the Pay Circular. Since, for the Appellants grade (i.e. Hospital Practitioner), that point is expressed in terms of sessional rates, some means must be found to convert those rates into a form which can be applied to the different terms of her training post, in which her periods of work were measured in hours not sessions. [27]. The most obvious way of doing that is by converting the sessional rates to hourly rates [27]. It may be counter intuitive that those rates should not be limited in some way be reference to the number of sessions which were, or could have been, worked in the former post [27]. However, there is nothing in the wording of paragraphs 132 or 135(a) to support such a limitation [28]. The exceptions relating to part time medical and dental officers appointed under paragraphs 94(b) and 105 of the NHS Terms support that construction of paragraph 132. The existence of such a specific limitation makes it more difficult to imply some other unspoken limitation applicable to the Appellants case [28].
The present case concerns a tendering process carried out by the respondent in 2010 in respect of the provision of medical services to health authorities in Scotland. The appellant was the existing supplier of the services in question, but was unsuccessful in a tender competition for a replacement contract. The appellant challenged that decision on the ground that the respondent had breached certain of its duties under the Public Contracts (Scotland) Regulations 2006 (SSI 2006/1), which implemented certain EU Directives. In particular, the appellant complained that: (i) the criteria in the invitation to tender were insufficiently clear; and (ii) that the reasons given for the rejection of the tender bid were unclear and lacking in detail. [18] One of the EU Directives implemented by the Regulations is Directive 2004/18/EC of 31 March 2004, which concerns the coordination of procedures for the award of public works contracts, public supply contracts and public service contracts. The Directive seeks to ensure that the award of contracts by public authorities in the member states is subject to the principles of freedom of movement of goods, freedom of establishment and freedom to provide services, and to other principles derived from those, such as the principles of equal treatment, non discrimination, mutual recognition, proportionality and transparency. In particular, article 2 requires that contracting authorities shall treat economic operators equally and non discriminatorily and shall act in a transparent way. Article 41 entitles unsuccessful candidates to be informed of the reasons for the rejection of their applications. The Court of Justice of the European Union (the CJEU) has explained that the principle of transparency requires that that the award criteria for public contracts must be formulated in such a way as to allow all reasonably well informed and normally diligent tenderers [(RWIND tenders)] to interpret them in the same way. [57] The appellants case was rejected by the Outer House of the Court of Session. There, the Lord Ordinary, Lord Hodge, concluded that the award criteria met the required standard of clarity and that the reasons given by the respondent for rejecting the appellants tender were adequate. The appellants appeal to the Inner House of the Court of Session was refused. [1923] The issues before the Supreme Court on the appeal from the Inner House are: (i) in relation to the clarity of award criteria, whether the lower courts erred in treating the RWIND tenderer as a hypothetical construct, based on the courts objective assessment of the appropriate standard of clarity, rather than on the basis of the evidence of witnesses as to what an actual tenderer did or thought; and (ii) whether the lower courts had erred in concluding that the reasons given to the appellants for the rejection of their tender were adequate. The Supreme Court unanimously dismisses the appeal. Lord Reed gives the only judgment, with which the other Justices agree. The courts below applied the correct legal test to assess the clarity of the award criteria. In these circumstances it was not appropriate for the Supreme Court to interfere with the conclusion which they reached in the light of their evaluation of the evidence. Similarly, in assessing the adequacy of reasons given to the appellant, the lower courts applied the approach laid down by the CJEU, and it is not appropriate for the Supreme Court to interfere with their factual findings. Standard of clarity in award criteria When courts refer to the approach of a reasonable person or, in this case, an RWIND tenderer, they are describing an objective legal standard by reference to a hypothetical person. It follows that it would be misconceived for a party to seek to lead evidence from actual persons on how they would have acted in a given situation or what they would have perceived. [23] The decisions of the CJEU and the opinions of Advocates General in a series of cases also make clear that the RWIND tenderer standard is an objective one. The relevant question is not whether it had been proved that all actual or potential tenderers had in fact interpreted the criteria in the same way, but whether the court considered that the criteria were sufficiently clear to permit of uniform interpretation by all RWIND tenderers. An approach which depends on evidence of the actual or subjective understanding of tenderers would undermine the principle of legal certainty and the need that any review of a tender process be carried out as quickly as possible. [716] Reasons given to unsuccessful tenderers The scope of the duty to give reasons for an unsuccessful tender is described by the Court of First Instance in Strabag Benelux NV v Council of the European Union (Case T 183/00). The court stated that the obligation was fulfilled if tenderers were informed of the relative characteristics and advantages of the successful tenderer and the name of the successful tenderer. The lower courts followed this approach and each concluded that the reasons which had been given to the appellant were adequate since the appellant could have been left in no real doubt as to why it had been unsuccessful, and as to the relative characteristics and advantages of the successful tenderer. [19; 23; 30]
The appeal concerns a proposed development by Crisp Maltings Group Limited (CMGL) at a plant in the area of the North Norfolk District Council (the council). The development comprised two silos and a lorry park with associated facilities on a site close to the River Wensum. The appellant, Mr Champion, is a member of the Ryburgh Village Action Group, which opposed the development. The river is a Special Area of Conservation protected by the EU Habitats Directive (97/92/EC), given effect in the UK by the Conservation and Habitats Species Regulations 2010. Regulation 61, implementing article 6(3) of the Directive, requires that before giving consent for a project likely to have a significant effect on a European site, the competent authority must make an appropriate assessment of the implications for that site. It may agree to the project only after having ascertained that it will not adversely affect the integrity of the European site. Also relevant is the Environmental Impact Assessment (EIA) Directive, given effect by the Town and Country Planning (Environmental Impact Assessment) Regulations 2011. Under the regulations, a competent authority carries out screening to decide if a proposal is likely to have significant effects on the environment. If so, EIA is required. The EIA process involves an environmental statement and public consultation, which informs the decision whether to grant consent to an EIA development. CMGLs planning application of October 2009 included a Flood Risk Assessment (FRA) recognising a risk that surface water runoff from the site would pollute the river. There followed investigation of measures meant to prevent this pollution. From October 2009 June 2010 the council consulted with relevant statutory bodies. It issued a screening opinion on 23 April 2010 stating that EIA was not required. Between July 2010 and January 2011, two new FRAs and an ecological assessment were prepared, which led to the statutory bodies withdrawing their objections. The council decided on 20 January 2011 to give delegated powers to its officers to approve the development subject to conditions. This led to local complaints, including from the appellant, who argued that appropriate assessment and EIA were required. The council decided to refer the application back to committee and asked for further comments from the appellant, who did not respond. At the councils committee meeting of 8 September 2011, planning officers presented a detailed report concluding that appropriate assessment and EIA were not required. The committee resolved to approve the application subject to conditions, including monitoring the rivers water quality. The appellant challenged the consent successfully before the High Court for failure to comply with the EIA and Habitats legislation, but lost in the Court of Appeal. The Supreme Court unanimously dismisses the appeal. Lord Carnwath gives a judgment with which Lord Neuberger, Lord Mance, Lord Clarke and Lord Toulson agree. The two issues were the timing of the councils decisions that appropriate assessment or EIA was not required, and the relevance of measures meant to address adverse effects on the river from the site. On the first issue, there is nothing in the Habitats Directive or regulations to support a separate stage of screening in any formal sense. Case law of the Court of Justice of the European Union describes two stages under article 6(3) of the Directive: the appropriate assessment, and the decision in light of it. It used the word trigger to set the threshold for the first stage. The formal procedures in the EIA regulations, including screening, an environmental statement, and mandatory public consultation, have no counterpart in the habitats legislation. Where it is not obvious, the competent authority will consider whether the trigger for appropriate assessment is met, but this is not a screening in the EIA sense. All that is required is that, where there is found to be a risk of significant adverse effects to a protected site, there is an appropriate assessment. In this case, the planning authority and the expert consultees were satisfied that the material risk of significant effects on the river had been eliminated. Though the officers expressed this conclusion by saying that no appropriate assessment was required, there is no reason to think that the conclusion would have been different if they had decided from the outset that appropriate assessment was required. The mere failure to exercise the article 6(3) trigger at an earlier stage does not in itself undermine the legality of the final decision. [37 42] On timing of EIA screening, authorities should in principle adopt screening opinions early in the planning process. [43] Though a negative opinion, lawfully arrived at on the information then available, may need to be reviewed in light of subsequent information, this does not mean that a legally defective screening opinion not to require EIA, or a failure to conduct a screening opinion at all, can be cured by carrying out an assessment exercise outside the EIA regulations. In the present case it was accepted that the councils screening exercise in April 2010 was legally defective: the pollution prevention measures had not been fully identified at that point, so the council could not be satisfied then that mitigation measures would prevent a risk of pollutants entering the river. This was an archetypal case for EIA so that the risks and measures to address them could be set out in the environmental statement and subject to consultation and investigation. That defect was not remedied by what followed: it was not enough to say that the potential adverse effects had now been addressed in other ways. [45 47] On the second issue, the appellant disputed the legality of the councils reliance on mitigation measures, at the stage of granting planning permission, to dispense retrospectively with the requirement for EIA which should have been initiated at the outset. [48] There is nothing to rule out consideration of mitigation measures at the EIA screening stage, but the Directive and the regulations expressly envisage that they will where appropriate be included in the environmental statement. Cases of material doubt should generally be resolved in favour of EIA. [51]. The failure to treat this proposal as EIA development was a procedural irregularity, which was not cured by the final decision. [53] Despite the legal defect in the procedure leading to the grant of planning permission, the court retains a discretion to refuse relief if the applicant has been able in practice to enjoy the rights conferred by European legislation and there is no substantial prejudice. [54] There is nothing to suggest that the councils decision would have been different had the process taken place within the framework of the EIA regulations. There was only one issue of substance: measures to achieve adequate hydrological separation between the sites activities and the river. It is clear from the final report that the statutory agencies involved formed their own view of the measures effectiveness, and that the views of the public were taken into account. At the time the appellant was unable to raise specific concerns that had not been dealt with before the final decision, which remains the case. The appeal is dismissed. [59 62]
Three prisoners brought appeals concerning the circumstances in which the Parole Board is required to hold an oral hearing. Osborn was convicted in 2006 following an incident in which he was said to have brandished an imitation firearm at the home of his estranged wife. He was given a six year prison sentence and was released on licence in February 2009, the halfway point. He was recalled to prison later that day for breach of his licence conditions [18 29]. Booth and Reilly are indeterminate sentence prisoners who have served their minimum terms. In 1981, Booth [30 42] received a discretionary life sentence for attempted murder, with a minimum term of six and a half years. Reilly [43 53] was convicted in 2002 of robbery, attempted robbery and possession of an imitation firearm. He received an automatic life sentence with a minimum term of six years and eight months, which expired in September 2009. Both remain in custody. Each case was considered on paper by the boards single member panel. It decided not to direct the prisoners release or recommend their transfer to open prison conditions. Their solicitors made written representations to the board, disputing its findings and requesting an oral hearing in each case, but those requests were refused. All three sought judicial reviews of the decisions not to offer oral hearings. Only Reilly succeeded in the High Court, which found that the board had breached its common law duty of fairness, and had acted incompatibly with the appellants rights under article 5(4) of the European Convention on Human Rights1 by failing to offer him an oral hearing. This was overturned by the Northern Ireland Court of Appeal. The Supreme Court unanimously allows the appeals and declares that the board breached its common law duty of procedural fairness to the appellants, and article 5(4) of the European Convention, by failing to offer them oral hearings [116]. 1 Everyone who is deprived of his liberty by arrest or detention shall be entitled to take proceedings by which the lawfulness of his detention shall be decided speedily by a court and his release ordered if the detention is not lawful. The judgment, delivered by Lord Reed, emphasises that human rights protection is not a distinct area of the law based on the case law of the European Court, but permeates our legal system. Compliance with article 5(4) requires compliance with the relevant rules of domestic law [54 56]. The legal analysis of the problem does not begin and end with the Strasbourg case law [63]. Lord Reed sets out guidance (summarised at [2]) on complying with common law standards in this context. The board should hold an oral hearing whenever fairness to the prisoner requires one in the light of the facts of the case and the importance of what is at stake [81]. By doing so, it will act compatibly with article 5(4) [103]. It is impossible to define exhaustively the circumstances in which an oral hearing will be necessary, but these will often include: (a) where important facts are in dispute, or where a significant explanation or mitigation is advanced which needs to be heard orally in order fairly to determine its credibility [73 78; 85]; (b) where the board cannot otherwise properly or fairly make an independent assessment of risk, or of how it should be managed and addressed [79; 81; 86]; (c) where it is tenably maintained that a face to face encounter, or questioning of those who have dealt with the prisoner, is necessary to enable his case to be put effectively or to test the views of those who have dealt with him [82]; and (d) where, in the light of the prisoners representations, it would be unfair for a paper decision taken by a single member panel to become final without an oral hearing [96]. The purpose of the oral hearing is not only to assist in the boards decision making, but also to reflect the prisoners legitimate interest in being able to participate in a procedure with important implications for him, where he has something useful to contribute [82]. The likelihood of release or transfer is separate from the question of whether fairness requires an oral hearing [88 89]. When dealing with recalled prisoners cases, the board should bear in mind that they have been deprived of their freedom [83]. For indeterminate sentence prisoners, increased scrutiny should be afforded by the board in assessing whether the risk they present is unacceptable the longer they have spent in prison post tariff [83]. The board must be, and appear to be, independent and impartial [90 91] and guard against any temptation to refuse an oral hearing to save time, trouble and expense [91]. Lord Reed stresses that paper decisions are provisional; the right to request an oral hearing is not an appeal, and the prisoner need only persuade the board that an oral hearing is appropriate [94 95]. The common law duty to act fairly is influenced by the requirements of article 5(4); compliance with the former should ensure compliance with the latter [101 113]. Breach of article 5(4) will not normally result in an award of damages under the Human Rights Act unless the breach has resulted in the prisoner suffering a deprivation of liberty [114 115]. An oral hearing ought to have been offered to the appellants. Osborn and Reilly had advanced various explanations and mitigations [98] and their requests for an oral hearing were mistakenly characterised as appeals [99 100]. In Booths case, input from his psychiatrist at an oral hearing would have been helpful and it was relevant that he had spent so long in custody post tariff [99]. Reillys claim for damages failed it had not been argued that he had suffered any deprivation of liberty as a result of the article 5(4) breach [115].
Supplies of education to students in the United Kingdom are exempt from value added tax (VAT) if they are made by a college of a university within the meaning of Note 1(b) to Item 1, Group 6 of the Value Added Tax Act 1994 (the VAT Act). The appellant (SEL) contends that its supplies of education to students in the United Kingdom were and are exempt from VAT because it was and remains a college of Middlesex University (MU). SEL is a subsidiary of SAE Technology Group BV. Both are part of the SAE group of companies which trades around the world under the name SAE Institute (SAEI). MU is a United Kingdom university within the meaning of the VAT Act, Group 6, Item 1, Note 1(b). It has never had any financial interest in any SAE group company. Nevertheless, the relationship between MU and SAEI has been very close and is a reflection of a series of agreements addressing the nature of that relationship, the validation by MU of SAEI programmes of education and the accreditation of SAE group companies. SEL appealed against assessments raised by the Commissioners for Her Majestys Revenue and Customs (the Commissioners) in respect of its accounting periods 1 May 2009 to 29 February 2012. It has also appealed against subsequent assessments, but these have been stayed by agreement with the Commissioners pending the outcome of this appeal. SELs appeal was allowed by the First tier Tribunal (FTT). The Commissioners appealed that decision to the Upper Tribunal which allowed the appeal. SEL then appealed to the Court of Appeal, this appeal was dismissed. There are two issues for the Supreme Court: first, whether the Court of Appeal adopted the correct approach in determining whether SEL was a college of MU for the purposes of Note 1(b) to Item 1, Group 6 of the VAT Act; and secondly, if it did not, whether, upon application of the correct test, SEL was such a college. The Supreme Court unanimously allows the appeal. Lord Kitchin, with whom the rest of the Court agrees, delivers the judgment. The starting point for a consideration of the proper interpretation of Note 1(b) to Schedule 9, Group 6, Item 1 of the VAT Act must be articles 131 to 133 of the Principal VAT Directive. These make clear that member states must exempt transactions involving the provision of, among other things, university education by bodies governed by public law having such education as their aim. Member states must also exempt transactions by other organisations which they have recognised as having similar objects to those governed by public law and which also have education as their aim [41]. The general objective of the exemptions is to ensure that access to the higher educational services is not hindered by the increased costs that would result if those services were subject to VAT [43]. Parliament has chosen to exercise the discretion conferred upon it by exempting from VAT the provision of education by a United Kingdom university and any college of such a university. The term university is not defined in the VAT Act. However, the conditions under which a body in the United Kingdom is entitled to use the word university in its title are regulated by statute. Over 100 bodies are presently entitled to call themselves a university and they vary greatly in character. A small but nonetheless significant number of them are private and run for profit [46]. It is against the background of the range of possible arrangements between universities and their colleges that the meaning of the phrase college of such a university in Note (1)(b) falls to be determined [47]. In Lord Kitchins judgment the following points are material [47]. First, for its activities to fall within the scope of Item 1(a), any college of a university, as an eligible body, must provide education [48]. Secondly, the supply of educational services is exempt only if it is provided by bodies governed by public law or by other bodies recognised by the member state as having similar objects [49]. Thirdly, there is nothing in Note 1(b) or the broader context which would justify limiting the scope of the phrase any college of such a university to colleges which are a constituent part of a university in a constitutional or structural sense. To the contrary, if satisfaction of such a constituent part test were required, it would effectively exclude commercial providers such as SEL from the exemption for it is a test they will rarely if ever be able to satisfy [50]. Fourthly, it is necessary to examine the characteristics of those educational services and the context in which they are delivered rather than the precise nature of the legal and constitutional relationship between the body that provides them and its university [51]. Lord Kitchin recognises that the presence of a foundation or constitutional document or some other legal relationship establishing the college as a constituent part of the university in a constitutional or structural sense will be sufficient to prove that it is a college of the university within the meaning of Note 1(b), save in an exceptional case. However, that is not a necessary condition. In assessing whether a body is a college of a university the following five questions are also likely to be highly relevant: (i) whether they have a common understanding that the body is a college of the university; (ii) whether the body can enrol or matriculate students as students of the university; (iii) whether those students are generally treated as students of the university during the course of their period of study; (iv) whether the body provides courses of study which are approved by the university; and (v) whether the body can in due course present its students for examination for a degree from the university [53]. If a body can establish the presence of each of these five features, then it is highly likely to be a college of the university within the meaning of Note 1(b). This is not to suggest that that there may not be other cases where the degree of integration of the activities of the body and the university is such that it may properly be described as a college of the university. All will depend on the particular circumstances of the case [54]. Lord Kitchin concludes that the factual findings of the FTT were sufficient to justify its conclusion that SELs activities were integrated into those of MU and that it shared the objects of MU. The FTT was entitled to find that in May 2009 SEL became and thereafter remained a college of MU within the meaning of Schedule 9, Group 6, Item 1, Note (1)(b) of the VAT Act [73].
The appeal arises out of an unfortunate but isolated oversight in the offices of the West London Mental Health NHS Trust at the end of 2010. Mrs Modaresi, who suffers from schizophrenia, was detained under s.2 of the Mental Health Act 1983 (the Act) on 20 December 2010 for assessment. By s.66(1)(a) of the Act she had a right to apply to the First tier Tribunal within 14 days to review her detention. The tribunal would have been obliged to arrange a hearing within 7 days of receiving the application. On the afternoon of 31 December 2010 she gave a completed application form to a member of the hospital staff who faxed it to the appropriate office within the Trust. The administrator was out the office that day and the form was not seen by others in the office. The office was closed over the New Year holiday until 4 January 2011, when the form was found and faxed immediately to the tribunal. Officials in the tribunals office deemed it to be out of time. On 6 January 2011 Mrs Modaresi ceased to be detained under section 2 but became detained for treatment under s.3 of the Act. As such, she was entitled to make a separate application to the tribunal under s.66(1)(b) of the Act, which has no time limit for holding a hearing. Her solicitors wrote to the Secretary of State asking him, in the circumstances, to exercise his discretion under section 67(1) of the Act, which permits the Secretary of State, if he thinks fit, at any time to refer a patient detained under the Act to the tribunal. The Secretary of State noted that the reason for the delay was no fault of Mrs Modaresi, but having considered all the information, refused the application and suggested that she made a separate application as she was now detained under s.3. However, Mrs Modaresi did not pursue that course, and instead issued proceedings for judicial review against the Secretary of State for unlawfully declining to refer the case under s.67. The claim for judicial review was dismissed by the High Court and the Court of Appeal dismissed the appeal. The Supreme Court unanimously dismisses the appeal. Lord Carnwath gives the lead judgment and Lady Hale adds a concurring judgment. Mrs Modaresi was not deprived of her right of access to a court or tribunal to review her detention. She had such a right under s.3 of the Act. The issue was not the existence of the right but how speedily it might be exercised and whether it was as advantageous as might have been the case if her original application had been accepted [17]. S.67 did not enable the Secretary of State to insist on a hearing in 7 days, as would have been required under s.2. The timing was in the discretion of the tribunal, as it would be under s.3 and s.66(1)(b). There was no evidence to support the submission that the Secretary of State could have been more persuasive in that respect. An application could have been made under s.3 with a request for an urgent hearing in the circumstances. A direct approach to the Tribunal offered a much speedier resolution than the roundabout procedure actually adopted [18]. The practical advantage of a s.67 application was to avoid losing a right to make a second reference under s.3. However, Article 5(4) only required that a patient should have an entitlement to take proceedings to have the lawfulness of his or her detention decided speedily by the court. Mrs Modaresi had this entitlement under s.66(1) in respect of her detention under s.3. If there came a time when having unsuccessfully used up her s.3 application, Mrs Modaresi wanted to make a further application, she was entitled to ask the Secretary of State again under s.67, which he indicated that he would consider doing. The Secretary of State had a discretion under s.67 which it needed to exercise in accordance with normal public law principles and judicial review was available [19]. A reference under s.67 also had the advantages of convenience and accessibility over an alternative route [21, 35]. Lady Hale noted that Mrs Modaresi had undoubtedly been let down by the system through no fault of her own and there were some important lessons to be learnt [27]. The hospital failed to transmit her application to the tribunal on the day it was made [29]. The hospitals failure deprived the patient of the right of access to a tribunal which the law provides, and may well be a breach of the patients Convention rights, and the only safe course is to have a system which ensures this does not happen [31]. The tribunal also failed to accept her application when it arrived. The authorities show that when an Act of Parliament prescribes a period for doing an act which can only be done on a day when the court office is closed, the time is extended to the next day on which it is open [33].
The appellant, Fiona McDonald, is aged 45 and suffers from a personality disorder. In May 2005 her parents purchased 25 Broadway Close, Witney (the property), as a home for her, with the assistance of a loan from Capital Home Loans Ltd (CHL), which was secured by way of a registered legal charge over the property. From about June 2005, the respondents granted the appellant a series of assured shorthold tenancies (ASTs) of the property, the last of which was granted in July 2008 for a term of one year. The appellant continues to live in the property. Owing to financial difficulties with their business, the respondents failed to meet payments on the loan as they fell due. CHL accordingly appointed Andrew Hughes and Julian Smith (the Receivers) to act as receivers of the property. The rent due was regularly paid, but the arrears persisted. The Receivers subsequently served a notice, in the name of the appellants parents, on the appellant on 13 January 2012, indicating that they would be seeking possession of the property and, on the expiry of that notice, they issued proceedings in the name of the parents for possession of the property in the Oxford County Court. His Honour Judge Corrie heard the proceedings on 4 December 2012 and 7 March 2013. He gave judgment on 22 April 2013 and held that the court was not required to consider the proportionality of making an order for possession against a residential occupier where the person seeking possession was not a public authority, and as section 21(4) of the Housing Act 1988 (the 1988 Act) required him to make an order for possession against a person holding under an AST who had been served with an appropriate order, he had to make such an order. The judge added that, had he been entitled to consider proportionality, he would, on balance, have concluded that the claim for possession was disproportionate and dismissed the action. The Court of Appeal dismissed the appellants appeal. The appellant now appeals to the Supreme Court. The Supreme Court unanimously dismisses Fiona McDonalds appeal. Lord Neuberger and Lady Hale give the only judgment, with which the other Justices agree. This appeal raises three questions [1]: (i) whether a court, when entertaining a claim for possession by a private sector owner against a residential occupier, should, in light of section 6 of the Human Rights Act 1998 (the HRA) and article 8 of the European Convention on Human Rights (the ECHR) be required to consider the proportionality of evicting the occupier; (ii) whether, if the answer to question (i) is yes, the relevant legislation, in particular section 21(4) of the 1988 Act, can be read so as to comply with that conclusion; (iii) whether, if the answer to questions (i) and (ii) is yes, the trial judge would have been entitled to dismiss the claim for possession in this case, as he said he would have done. The appellants argument is that the judge should have taken into account the proportionality of making an order for possession for article 8 purposes and, on that basis, could have refused to make an order for possession despite the apparently mandatory terms of section 21(4) of the 1988 Act and section 89(1) of the Housing Action 1980 (the 1980 Act), which limits the period for which a court can postpone an order for possession taking effect [29 30]. It is well established that it is open to the occupier to raise the question of the proportionality of making an order for possession where the party seeking possession is a public authority within the meaning of section 6 of the HRA [34]. In deciding this issue in the case of Manchester City Council v Pinnock [2011] 2 AC 186, the Supreme Court made it clear that nothing in its judgment was intended to bear on cases where the person seeking possession was a private landowner [37]. The appellant contends that the same reasoning applies to a private sector landlord because the court which would grant the order for possession is a public authority for the purposes of the HRA [38 39]. The courts preliminary view is that it is not open to the tenant to contend that article 8 could justify a different order from that which is mandated by the contractual relationship between the parties, at least where there are legislative provisions through which the democratically elected legislature has balanced the competing interests of private sector landlords and residential tenants [40]. Were it otherwise, the ECHR could be said to be directly enforceable as between private citizens so as to alter their contractual rights and obligations [41]. As to the Strasbourg authorities, the admissibility decisions of Di Palma v United Kingdom (1988) 10 EHRR CD 149 and Wood v United Kingdom are inconsistent with the appellants case [48]. While subsequent authorities provide some support for the notion that article 8 is engaged on the making of the order for possession against a residential occupier such as the appellant, there is no support for the proposition that the judge could be required to consider the proportionality of the order which he would have made under legislation such as the 1980 and 1989 Acts [49 59]. The appeal is accordingly dismissed on the first issue [59 60]. As to the second issue, it would not be possible to read section 21(4) of the 1988 Act in the way contended for by the appellant [61 70]. Had the court been persuaded that the appellant was right on the first issue, a declaration of incompatibility under section 4 of the HRA would have been the only remedy [70]. As to the third issue, the judge did not consider whether, if he had found that the claim for possession were disproportionate, there might have been other solutions to the problem than dismissing the claim [71]. In those rare cases where a court is required to assess the proportionality of making a possession order, its powers to suspend or postpone the effect of that order are severely limited by section 89(1) of the 1980 Act [72]. The cases in which it would be justifiable to refuse, as opposed to postpone, a possession order must be very few and far between and could only be cases in which the landlords interest in regaining possession was heavily outweighed by the gravity of the interference in the occupiers right to respect for her home [73]. On the facts of this case, it seems likely that the most the appellant could hope for on a proportionality assessment would be an order for possession in six weeks time, the maximum permitted by section 89(1) of the 1980 Act [75].
The Institute of the Brothers of the Christian Schools (the Institute) was founded in 1680 with the mission to teach children, and its members are lay brothers of the Catholic Church. The question arising in this appeal is whether the Institute is responsible in law (vicariously liable) for alleged acts of sexual and physical abuse of children by its members between 1952 and 1992 at St Williams, a residential institution at Market Weighton for boys in need of care (the School). The Institute did not own the School, which was founded in 1865 by a group of Catholic benefactors and run locally as a reformatory school for boys. In 1933 it became an approved school for boys convicted of custodial offences under a group of managers. Brothers from the Institute taught at the School alongside lay teachers and a brother always acted as headmaster of the School. In 1973 the School became an assisted community home for children in the care of the local authority, managed by the Middlesbrough Diocesan Rescue Society until 1982, and thereafter by the Catholic Child Welfare Society (Diocese of Middlesbrough). In 1990 the headmaster of the School, Brother James, was expelled from the Institute after it was discovered he was guilty of systematic sexual abuse of boys in his care. In 1993 and 2004 he was convicted of numerous counts of serious sexual offences against boys over a period of 20 years. The School was closed in 1994. Claims have been brought by 170 men in respect of abuse to which they allege they were subjected at the School, by Brother James and by other brothers. The claims are against two groups of defendants. The first group consists of the managers of the school from 1973, who inherited the statutory liabilities of the former managers and entered into contracts of employment with the brother teachers (the Middlesbrough Defendants). The second group consisted of members of the Institute (the Institute Defendants). As a preliminary issue, the High Court held that the Institute Defendants were not vicariously liable for the acts of abuse committed by brothers at the School. The Court of Appeal upheld that ruling. The Middlesbrough Defendants appealed to the Supreme Court. The Supreme Court unanimously allows the appeal. It holds that it is fair, just and reasonable for the Institute Defendants to share with the Middlesbrough Defendants vicarious liability for sexual abuse committed by the brothers. The judgment is given by Lord Phillips, with whom the other Justices agree. The law of vicarious liability had developed recently to establish a number of important propositions: It is possible for unincorporated associations (such as the Institute) to be vicariously liable for the wrongful acts of its members It is possible to be vicariously liable even if the wrongdoers act is in breach of the duty he owes to the person liable and even if the act was the criminal offence of sexual assault; and It is possible for two or more different defendants each to be vicariously liable for a single wrongful act [20] The criteria to be demonstrated to establish vicarious liability involved a synthesis of two stages: first, whether the relationship between the member and the Institute was one which was capable of giving rise to vicarious liability; and secondly examination of the connection that linked the relationship between them with the members wrongful act or omission [21]. Both were in issue in this case. The Institute relied on the fact that the Middlesbrough Defendants entered into contracts of employment with the brothers and managed and controlled both them and the School, and contended that the absence of these critical features meant that the relationship between the brothers and the Institute could not give rise to vicarious liability. The Institute was an unincorporated association but because of the manner in which it carried on its affairs Lord Phillips considered that it was appropriate to approach the case as if it was a corporate body existing to perform the function of providing a Christian education, able to own property and other assets through charitable trusts [33]. When two sets of defendants are potentially vicariously liable for the act of a wrongdoer it was necessary to give independent consideration to the relationship with each set in order to decide whether they should be liable [45]. The relationship between the Institute and the teaching brothers at the School had all of the essential elements of that between employer and employee. The teaching activity was undertaken because the brothers were so directed by the Institute; it was in furtherance of the mission of the Institute and the manner in which the brother teachers were obliged to conduct themselves was dictated by the Institutes rules. The fact that they were bound to the Institute by their vows rather than contract, and transferred all their earnings to the Institute, did not make a material difference [56 58]. Thus the first stage of the test for vicarious liability was satisfied. Stage 2 of the test is usually satisfied when a wrongdoer does something he has been required or requested to do pursuant to his relationship with the defendant in a manner that is negligent. But sexual abuse can never be a negligent way of performing such a requirement. Where abusers have been members of a church or religious order, what has weighed with the courts has been the fact that the relationship has facilitated the commission of the abuse by placing the abusers in a position where they enjoyed both physical proximity to their victims and the influence of authority over them as teachers and men of god [84]. The necessary close connection between the relationship between the defendant and the wrongdoer is established where a defendant, whose relationship with the abuser put it in the position to use the abuser to carry on its business, did so in a way which created or significantly enhanced the risk of abuse [86]. In this case the Institute placed the brothers in teaching positions and in particular the position of headmaster, responsible for running the School. The boys living on the premises were particularly vulnerable, not just as children in a school but because they were virtually prisoners and would have difficulty making credible allegations of abuse because of their personal histories [92]. The status of a brother was no doubt treated by the managers as an assurance that children could safely be entrusted to his care. The placement of brother teachers in a residential school thus greatly enhanced the risk of abuse by them if they had a propensity for such misconduct. This was not a borderline case and it was fair just and reasonable for the Institute to share vicarious liability in this case with the Middlesbrough Defendants [94].
In domestic law, the polices power to retain data is controlled by the Data Protection Act 1998 and by a mandatory Code of Practice and accompanying Guidance issued under the Police Act 1995. Individuals also have a right to respect for their private lives under Article 8 of the European Convention on Human Rights (ECHR). The Code of Practice limits the handling of police information to police purposes, limits the circumstances under which data can be shared between police forces, and requires that information originally recorded for police purposes must be reviewed for deletion at prescribed intervals. The Guidance says that the object of such reviews is to ensure that there is a continuing policing purpose for holding the record, the record is accurate, up to date and not excessive, the Data Protection Act has been complied with, and the assessment of the risk level presented by the data subject is correct. Mr Catt, a 91 year old man from Brighton, participates in political protests, including with a group called Smash EDO. Mr Catt is a peaceful protestor, but some members of Smash EDO commit violent offences. The police overtly collect information from Smash EDO public demonstrations. Because Smash EDO has associations with violent crime, information is retained even where no crime has been committed. Events are recorded in Information Reports and some individuals are the subject of a nominal record. These records are stored on a Domestic Extremism Database. At one point there was a nominal record and a photograph for Mr Catt, but both were deleted in separate reviews before these proceedings began. However, information about Mr Catt, including his presence, date of birth, and address, is contained in 107 Information Reports primarily directed to the activities of other people (including at mainstream non Smash EDO protests). Ms T is alleged to have said a homophobic insult to her neighbours friend in July 2010. The police made a Crime Reporting Information System (CRIS) record about the incident and sent her a Prevention of Harassment Letter notifying her that she may be liable for arrest and prosecution should she commit any act or acts amounting to harassment. The practice of the Metropolitan Police is to retain a copy of the letter in their electronic records for seven years, and the corresponding CRIS for 12 years. The police deleted the materials in January 2013 in the course of preparing for this appeal. Mr Catt and Ms T accept that it was lawful for the police to make records of the events as they occurred. However, they contend that the Metropolitan Polices policy in thereafter retaining the data on a searchable database is unlawful because it is contrary to their rights under Article 8 ECHR. Both of their claims failed at first instance. Their claims were heard together in the Court of Appeal, which allowed both appeals. In the case of Mr Catt, the Supreme Court allows the appeal by a majority of 4 1 and restores the first instance judgment. Lord Sumption (with whom Lord Neuberger agrees) gives the leading judgment. Lady Hale delivers a concurring judgment, agreeing with Lord Sumption, and Lord Mance agrees with both Lady Hale and Lord Sumption. Lord Toulson would have dismissed the appeal. In the case of Ms T, the Supreme Court unanimously allows the appeal and restores the first instance judgment. Lady Hale and Lord Toulson (with whom Lord Mance agrees) say that the policy was lawful. Lord Sumption (with whom Lord Neuberger agrees) says that the policy was not originally lawful but became so in this case. Lord Sumption explains that the states systematic collection and storage in retrievable form even of public information about an individual is clearly an interference with private life under Article 8(1) ECHR [3 5]. These appeals therefore turn on whether the retention of the data can be justified under Article 8(2), and in particular whether the retention is (i) in accordance with the law and (ii) proportionate to its objective of securing public safety or preventing of disorder and crime [6]. The in accordance with the law condition under Article 8(2) requires that the applicable rules not be so wide or indefinite as to permit interference with the right on an arbitrary or abusive basis and that their application be reasonably predictable [11]. The retention of data in police information systems in the United Kingdom is in accordance with the law: there are some discretionary elements in the scheme, but this is inevitable, and the space of discretionary judgment is limited and subject to judicial review; further, future disclosure is limited by comprehensive restrictions [13 17]. Lady Hale [47 49], Lord Mance [58 59] and Lord Toulson [60] all agree that the real issue in these appeals is proportionality. Proportionality: Mr Catt Lord Sumption holds that the interference with Mr Catts private life is minor: the information stored is personal but not intimate or sensitive; the primary facts recorded have always been in the public domain, and it is known that the police records them; there is no stigma attached to the inclusion of his information in the database as part of reports primarily directed to the activities of other people; the material is usable and disclosable only for police purposes and in response to requests made by Mr Catt himself under the Data Protection Act; and the material is regularly reviewed for deletion according to rational and proportionate criteria contained in the publicly available Code of Conduct and accompanying Guidance [26 28]. There are numerous proper policing purposes to which the retention of evidence of this kind makes a significant contribution. The longer term consequences of restricting the availability of this method of intelligence gathering to the police would potentially be very serious, and the amount of labour required to excise information relating to persons such as Mr Catt from the database would be disproportionate [29 31]. Lady Hale agrees with Lord Sumptions analysis of the case of Mr Catt [56], though adds that it would have been disproportionate to keep a nominal record about Mr Catt since he has not been and is not likely to be involved in criminal activity himself and the keeping of such records has a potentially chilling effect on the right to engage in peaceful public protest [50 52]. Lord Mance agrees with both Lord Sumption and Lady Hale [58]. Lord Toulson would have dismissed the appeal in the case of Mr Catt. He does not think that the evidence given by the police explains why it is necessary to retain for many years after the event information about someone about whom they have concluded that he was not known to have acted violently. He notes in particular that information was retained about Mr Catts attendance at mainstream political protest events and does not see how this could be thought necessary and proportionate [65 66]. The suggestion that it would be over burdensome for the police to have to review information about individuals such as Mr Catt was not supported by the evidence, especially since the police already conduct regular reviews [67 68]. Proportionality: Ms T Lady Hale [54 56] and Lord Toulson [76] both say that retaining information about previous harassment complaints serves a vital purpose, particularly in domestic abuse cases, and it is not unlawful for the police to adopt a standard practice of retaining such information for several years, provided that the policy is flexible enough to allow it to be deleted when retention no longer serves any useful policing purposesas in fact happened in this case [76]. Lady Hale notes that the Information Commissioner could not have secured the withdrawal of the Prevention of Harassment Letter and that is presumably why these proceedings were launched [53]. Lord Mance agrees with Lady Hale and Lord Toulson, but adds that even if the policy were originally inflexible, he would still have allowed the appeal for the reasons given by Lord Sumption [59]. Lord Sumption says that the Prevention of Harassment Letter, while in this case unnecessarily accusatorial, clearly serves a legitimate policing purpose, but the standard period of retention applied by the Metropolitan Police is wholly disproportionate in light of the trivial nature of the incident in this case. However, Ms Ts Article 8 rights have not been violated because the material was in fact retained for only two and a half years, a period at the far end of the spectrum but not disproportionate [42 44]. The dispute could have been more appropriately resolved by applying to the Information Commissioner [45].
It can be a condition of release from prison of certain medium or high risk prisoners that they must live at Approved Premises (APs). APs are single sex establishments. There are 94 APs for men, distributed around England and Wales including several in London. There are only 6 APs for women, who constitute 5% of the prison population, and none of them is in London or in Wales. This means that women are much more likely than men to be placed in an AP which is far from their homes and communities. In 2004 the appellant was sentenced to life imprisonment with a tariff of 11 years and 3 months, which was due to expire in November 2015. She anticipated that on her release she would be required to live at an AP, which would necessarily be at a considerable distance from her family in London. In 2013 the appellant brought proceedings seeking a declaration that the current provision of APs amounted to unlawful sex discrimination, contrary to the Equality Act 2010 (the EA) and her rights protected by articles 8 and 14 of the European Convention on Human Rights, and that the Secretary of State had acted in breach of the public sector equality duty (PSED) under section 49 EA by failing to have due regard for the need to eliminate discrimination against women in the provision of APs. In the High Court, Cranston J dismissed the discrimination claim but granted a declaration that the Secretary of State had failed to discharge the PSED. The Secretary of State has not appealed that declaration. The appellants appeal against the finding of no discrimination was dismissed by the Court of Appeal. Since that time, she has been released from prison and required to live in an AP in Bedford. The appellant has brought a discrimination claim in the county court which has been stayed pending the outcome of these proceedings. The Supreme Court unanimously allows the appellants appeal to the extent of granting a declaration that provision of APs constitutes direct discrimination against women which is unlawful unless justified, and that the Secretary of State has yet to show such justification. Lady Hale, with whom all the other Justices agree, gives the only judgment. APs are commissioned rather than directly provided by the Secretary of State. However, under s 29(6) EA, a person exercising a public function such as this must not do anything that constitutes discrimination, harassment or victimisation. It is not necessary for the appellant to show that every female prisoner required to live at an AP has suffered the detriment of being placed at an AP far from her home in order to establish a case of direct discrimination on grounds of sex [29 31]. Nor are there differences between the circumstances of male and female prisoners required to live in APs, which are material to the issue of accommodating them close to home. In this respect their circumstances are comparable, and the risk of being placed far from home is much greater for women than for men [32]. The reason for this is not any deliberate desire to treat the women less favourably than the men but a function of the much smaller numbers of female offenders and the policy decision that the particular vulnerability of women required to live in an AP means that all APs should be single sex [33]. The appellants case is one of direct rather than indirect discrimination [43]. Direct discrimination can only be justified in certain limited and defined circumstances, including those set out in paragraph 26 of Schedule 3 to the EA. Paragraph 26(2) relates to the provision of separate and different services for men and women, and provides that these will not contravene s 29 if (a) a joint service for persons of both sexes would be less effective, (b) it is not reasonably practicable to provide a service which is not different, and (c) the limited provision is a proportionate means of achieving a legitimate aim [24, 34]. There is no dispute that in relation to APs, providing a joint service would be less effective. The appellant also accepts that the much lesser extent to which women require APs makes it not reasonably practicable to provide the same number of APs for each sex [38]. The crucial question is whether the limited provision for women is a proportionate means of achieving a legitimate aim [39]. Saving cost is a legitimate objective of public policy but if a benefit is to be limited to save costs it must be limited in a non discriminatory way [40]. The appellant accepts that in principle the different provision for men and women might be justified but the Ministry of Justice has never properly addressed its mind to the problem of providing sufficient and suitable places in APs for women which achieve, so far as is practicable, the policy of placing them as close to home as possible. There are other options which could have been considered, such as replacing large womens APs with smaller units more widely spread, or replacing one or more of the existing womens APs with APs closer to the areas where many women offenders have their homes, or considering alternative forms of accommodation for women released on licence [41]. It is for the Secretary of State to show that the discrimination is justified and, in the light of the breach of the PSED, she has so far failed to do this [42]. The Supreme Court therefore allows the appeal to the extent that it makes a declaration that the provision of APs in England and Wales constitutes direct discrimination against women contrary to s 13(1) EA, which is unlawful unless justified under paragraph 26 of Schedule 3 to the EA. No such justification has yet been shown by the Secretary of State. Individual women who are less favourably treated as a result of the provision of APs may bring sex discrimination claims in the county court but it will be open to the Secretary of State to resist such claims on the ground that the provision is justified under paragraph 26 [45].
On 9 September 1988, the Appellant (SerVaas) entered into an agreement with the Iraqi Ministry of Industry for the supply of equipment, machinery and related services required for the commissioning of a state owned copper and brass processing factory in Iraq. On 2 August 1990 Iraq invaded Kuwait and on 4 August 1990, the assets of Rafidain Bank (Rafidain) in the UK were frozen in accordance with a United Nations sanctions regime. On 13 August 1990 SerVaas terminated the agreement and subsequently commenced proceedings in the Paris Commercial Court against the Ministry to recover money due under the agreement. It gave judgment in favour of SerVaas for US$14,152,800 (the Judgment). The Judgment was recognised in the Netherlands and SerVaas recovered US$966,515 by partial enforcement there against Iraqs assets. In July 2002, SerVaas received US$6,736,285 from the UN Claims Commission by way of compensation for losses caused by Iraq as a result of the invasion of Kuwait. In May 2003, the regime of Saddam Hussein in Iraq fell. On 22 May 2003 the UN Security Council passed Resolution 1483 establishing the Development Fund for Iraq (DFI). On 21 November 2004, Iraq made a debt cancellation agreement with government creditors comprising the Paris Club. In December 2004, Iraq began a process of debt restructuring with its commercial creditors and the creditors of other specified Iraqi entities, including Rafidain under the auspices of the Iraq Debt Reconciliation Office (the IDRO Scheme). Rafidain, in the meantime, had had a winding up petition presented in respect of it by the Bank of England, in relation to which Provisional Liquidators had been appointed in respect of its UK assets, but which petition had been adjourned generally. On 26 July 2005, Iraq announced an offer to repurchase claims for the commercial creditors of specified Iraqi debtors, including Rafidain, where claims arose before 6 August 1990. In May 2006, Iraq issued an invitation to tender claims for cash purchase and for exchange. Thereafter Iraq took assignments of certain debts owed to Rafidains creditors by Rafidain in accordance with the IDRO Scheme. On 3 April 2008, a scheme of arrangement for the distribution of assets held by the Provisional Liquidators to Rafidains creditors was sanctioned (the Scheme). By 19 August 2009, Iraq had submitted claims in the Scheme which were admitted in the sum of US$253.8 million (the Admitted Claims). The original commercial debts constituting the Admitted Claims were acquired by Iraq by way of assignment from existing creditors of Rafidain. On 4 November 2009, SerVaas obtained an order registering the Judgment in England and Wales against the Ministry under the Civil Jurisdiction and Judgments Act 1982 (the Registration Order). It was served on Iraq on 2 May 2010 and became enforceable against the Ministry and Iraq in England and Wales on 2 September 2010. On 11 October 2010 Iraqs US lawyers responded to a request from the Scheme Administrators by stating that the dividend payment on the Admitted Claims should be paid to the account in the name of the DFI with the Federal Reserve Bank in New York. As at November 2010, the debt due in respect of the Judgment is said to have amounted to US$34,481,200.49. In the meantime on 7 October 2010 Mann J granted an application by SerVaas lifting the stay on proceedings against Rafidain and made an order preventing Rafidain, the Provisional Liquidators and the Scheme Administrators from making any payment to Iraq under the Scheme in respect of the Admitted Claims or recognising or giving effect to any assignment or transfer of the Admitted Claimant to a third party which would have the effect of reducing the amount payable to Iraq to an amount less than the Judgment debt. On 13 October 2010 SerVaas issued an application for a Third Party Debt Order, that is, an order that the debts payable to Iraq by Rafidain by way of dividend under the Scheme be instead paid to SerVaas insofar as necessary to satisfy the Judgment. On 30 November 2010, the Charg dAffaires and Head of Mission of the Embassy of Iraq in London signed a certificate (the Certificate) that the Admitted Scheme Claims have never been used, are not in use and are not intended to be for use for any commercial purpose. Iraq applied to discharge the injunction on the ground that monies due to Iraq by Rafidain were immune from execution by virtue of section 13(2)(b) of the State Immunity Act 1978. In the High Court, Arnold J held that the Admitted Claims were immune from execution by reason of s.13(2)(b) and (4) because they were not property which was for the time being in use or intended for use for commercial purposes. By a majority, the Court of Appeal dismissed SerVaas appeal. SerVaas appealed to the Supreme Court. The Supreme Court unanimously dismisses the appeal. Whether property is for the time being in use or intended for use for commercial purposes within the meaning of s.13(4) of the State Immunity Act 1978 does not depend on whether that property has in the past been used for commercial purposes. Lord Clarke gives the leading judgment with which Lord Phillips, Lady Hale, Lord Sumption and Lord Reed agree. It was common ground that (a) the monies payable under the Scheme are a debt and a chose in action and as such are property within the meaning of s.13(2)(b); (b) that Iraqs state intention is to transfer the proceeds of the Admitted Claims to the DFI; (c) that, the Certificate creates a rebuttable presumption that the Admitted Claims are not in use or intended for use for commercial purposes; (d) that the onus lies on SerVaas to show a real prospect that it can rebut that presumption; and (e) that the debts were intended for use for sovereign and not commercial purposes. As these are summary proceedings, the issue is whether there is any real prospect of SerVaas rebutting the presumption. The central question in this appeal is therefore whether the nature of the origin of the debts is relevant to the question whether the property in question was in use for commercial purposes. It is not. This conclusion is based on the language of s.13(4) and on previously decided domestic and comparative authority. As to language, s.13(4) should be given its ordinal and natural meaning having regard to its context and it would not be an ordinary use of language to say that a debt arising from a transaction is in use for that transaction. Parliament did not intend a retrospective analysis of all the circumstances which gave rise to property but an assessment of the use to which the state had chosen to put the property. The language of s.13(4) can also be contrasted with other sections of the Act. As to authority, Lord Diplock in Alstom v Republic of Columbia [1984] AC 580 distinguished between the origin of the funds on the one hand and the use of them on the other. Various decisions of the American Federal courts of appeals and of the Court of Appeal in Hong Kong also support this distinction.
The question in this case is whether the respondent local authority were entitled to be satisfied that the appellant, Ms Haile, became homeless intentionally. If the authority were not satisfied that she became homeless intentionally (section 193(1) of the Housing Act 1996), then they were under a duty to secure that accommodation was available for her occupation (section 193(2)). By section 191(1) of the 1996 Act: a person becomes homeless intentionally if he deliberately does or fails to do anything in consequence of which he ceases to occupy accommodation which is available for his occupation and which it would have been reasonable for him to continue to occupy. The appellant surrendered her tenancy of a bedsit in a hostel on 25 October 2011. She moved to temporary accommodation, which ended in November 2011 when she was asked to leave because of overcrowding. She then applied to the respondent authority for accommodation as a homeless person. On 15 February 2012 she had a baby daughter. Had she still been living in the hostel, she would have had to leave, as only single persons were allowed to reside there. On 1 August 2012 the authority decided that the appellant was homeless, eligible for assistance, and had a priority need, but that she became homeless intentionally. On 31 January 2013, a decision to the same effect was made by a review officer. The basis of the finding was that she had surrendered her tenancy of the room in the hostel and in consequence had ceased to occupy accommodation which was available for her occupation, and which it would have been reasonable for her to continue to occupy until she gave birth. Her contention that she would have had to leave the hostel in any event when she gave birth was regarded as irrelevant. Her appeal was dismissed by the County Court and the Court of Appeal. Before the Supreme Court, she argued that the birth of her baby broke the chain of causation between her intentionally leaving the hostel, and her state of homelessness when her application was considered. The appeal invited the court, if necessary, to depart from the House of Lords decision in Din v Wandsworth London Borough Council [1983] 1 AC 657. The Supreme Court allows the appeal by a majority of 4 1 (Lord Carnwath dissenting). Lord Reed gives the lead judgment, with which Lord Neuberger, Lady Hale, and Lord Clarke agree. Lord Neuberger adds a concurring judgment. Lord Reed reasons that the requirement in section 193(1) is meant to prevent queue jumping by persons who, by intentionally rendering themselves homeless, would obtain a priority in the provision of housing to which they would not otherwise be entitled. It is in relation to the current state of being homeless that one asks, did the applicant become homeless intentionally? [22 24] Section 193(1) must therefore be understood as being concerned with whether the applicants current homelessness was caused by intentional conduct on his part. This depends, first, on whether he deliberately did or failed to do anything in consequence of which he ceased to occupy accommodation meeting the requirements of section 191(1). If yes, the further question arises under section 193(1) whether the appellants current homelessness was caused by that intentional conduct. [25, 28] Thus, section 193(1) is read as meaning the local authority are not satisfied that [the applicant is homeless because] he became homeless intentionally. [27] Din concerned the interpretation of the definition of becoming homeless intentionally in section 17(1) of the Housing (Homeless Persons) Act 1977. The decision that the elements of that definition were to be considered as at the time when the applicant ceased to occupy accommodation meeting the requirements of the definition is still correct. It also remains true that if the definition is satisfied at that point in time, subsequent hypothetical events are immaterial. Finally, the conclusion in Din that there must be a continuing causal connection between the deliberate act satisfying the definition of intentional homelessness, and the homelessness existing at the date of the local authoritys inquiry, remains good law. [38, 40, 59 60] Later authorities applied that principle and provide examples of events interrupting the causal connection, such as marital breakdown. [44, 62] In the present case, the review officer did not consider whether the cause of Ms Hailes current state of homelessness was her surrender of her tenancy of the room in the hostel. The birth of the baby meant that she would be homeless, at the time her case was considered, whether or not she had surrendered the tenancy. She had not jumped the queue as a result of surrendering the tenancy. [66 67] Lord Neuberger agrees with Lord Reeds analysis, [69] and adds some reasons of his own. Lord Carnwath, in his dissenting judgment, would have dismissed the appeal. In his view the reasoning of the review officer was an orthodox reflection of the majority approach in Din. [89]
In December 2001 the Appellant, Paul McInnes, was convicted at the High Court of the Justiciary in Glasgow in respect of an assault outside a hotel in Duntocher, Dunbartonshire. He was sentenced to eight years imprisonment. The crucial issue at the trial was the identification of the persons who participated in the assault. The prosecution evidence in the case included the statement of Mr Brian Pearce, a steward at the nearby hotel who described how he witnessed the Appellant assault the victim. At the trial, Mr Pearce described how he witnessed the Appellant administer a kick to the head of the victim of the assault. Prior to the Appellants criminal trial, Mr Pearce failed to correctly identify the Appellant at two separate identity verification parades. On the first occasion (when the Appellant did not participate in the parade), Mr Pearce identified a person other than the Appellant as the perpetrator of the assault. On the second occasion (when the Appellant did participate), Mr Pearce identified a police stand in as the perpetrator. The prosecution failed to disclose either failed identification to the Appellant in advance of his trial. The Appellants defence were of the view that if they had been in possession of the information relating to the failed identification parades it would have enabled them to more effectively attack the reliability of Mr Pearces evidence at trial and that this may have affected the outcome of the proceedings. Following discovery of this omission, the Appellants case was referred to the High Court of Justiciary. The Appellant lodged a Devolution Minute contending that the failure of the prosecution to disclose the relevant information had rendered his trial unfair contrary to Article 6 of the European Convention on Human Rights and constituted a miscarriage of Justice. The High Court of Justiciary dismissed the Devolution Minute, holding that it could not be said that the non disclosure gave rise to a real risk of prejudice. The Appellant appealed to the Supreme Court on the basis that the High Court of Justiciary had erred in applying the wrong legal test. Rather than enquiring as to whether there was a real risk of prejudice it was contended that the court should simply have considered whether disclosure could have made a difference to the outcome of the prosecution. The Supreme Court unanimously dismisses the appeal, with Lord Hope delivering the leading judgment of the Court. The law on disclosure is reasonably well settled. The prosecution must disclose any material which might materially weaken its case or strengthen the defence. Accordingly, all police statements as a class must be disclosed [para [1] per Lord Hope]. The Courts jurisdiction in this case was confined to analysing whether the High Court of Justiciary had applied the correct legal test. The application of the test to the facts of the case was exclusively within the jurisdiction of the High Court of Justiciary [para [18] per Lord Hope]. Two questions arise for determination in this type of case. Firstly, whether the information is of a type that must be disclosed. In respect of police statements the answer to this question is clearly affirmative [para [19] per Lord Hope]. Secondly, whether, taking all the circumstances of the trial into account, there was a real possibility that the jury would have arrived at a different verdict if the relevant information had been disclosed [para [20] per Lord Hope]. The question which the appeal court must ask itself is whether after taking account of all the circumstances of the trial, including the non disclosure in breach of the appellants Convention right, the jurys verdict should be allowed to stand. The question will be answered in the negative if there was a real possibility at a different outcome if the jury might reasonably have come to a different view on the issue to which it is directed its verdict if the withheld material had been disclosed to the defence [para [24] per Lord Hope; see also paras [30] [31] per Lord Rodger and paras [35] and [38] per Lord Brown]. The test to determine whether there has been a fair trial in terms of Article 6 is the same that is to be applied to determine whether there has been a miscarriage of justice [para [23] per Lord Hope]. It is clear from the judgment of the High Court of Justiciary that it applied the correct legal test [para [25] per Lord Hope].
The question raised on this appeal is whether a person released from prison on a home detention curfew, and then recalled to prison under section 255 of the Criminal Justice Act 2003, has rights pursuant to article 5(4) of the European Convention of Human Rights. Article 5 protects the right to liberty, and article 5(4) confers on an individual who has been deprived of their liberty an associated right to challenge that deprivation before a judicial body. On 5 October 2010, the appellant, Stuart Whiston, was sentenced to 18 months in prison for robbery. He was entitled to automatic release on licence after serving half his sentence on 5 July 2011. However, on 21 February 2011, he was released on licence under a so called home detention curfew pursuant to section 246 of the 2003 Act. On 7 April 2011, the Secretary of State decided to revoke the licence under section 255 of the 2003 Act, because the appellants whereabouts could no longer be monitored in the community, and he was recalled to prison. The decision of the Secretary of State was not subject to any statutory judicial control or review. The appellant contends that, as a result of the licence granted on 21 February 2011, he regained his liberty, and the subsequent revocation of his licence and his consequent recall to prison on 7 April 2011 therefore constituted a deprivation of his liberty which infringed article 5(4). The Secretary of State argues that, at least where the sentence in question is determinate, in any case where a prisoner who has been released on licence is recalled to prison during the currency of his requisite custodial period, the requirements of article 5(4) are satisfied by the original sentence lawfully passed by the court by which he was originally imprisoned. The Supreme Court unanimously dismisses the appeal. Lord Neuberger, with whom Lord Kerr, Lord Carnwath and Lord Hughes agree, gives the main judgment. Lady Hale gives a concurring judgment. Under Strasbourg jurisprudence, where a person is lawfully sentenced to a determinate term of imprisonment by a competent court, he is not, at least in the absence of unusual circumstances, able to challenge his loss of liberty during that term on the ground that it infringes article 5(4). Where the Secretary of State exercises her discretion to release a prisoner before the end of the requisite custodial period of their sentence, article 5(4) is not infringed if that licence is subsequently revoked. All the statutory provisions relevant to this appeal are in the 2003 Act, as amended most recently by the Legal Aid, Sentencing and Punishment of Offenders Act 2012. Where a person has been convicted and given a determinate prison sentence of twelve months or more (a sentence period), section 244(1) provides that, subject to certain specified exceptions, once he has served half his sentence it is the duty of the Secretary of State to release him on licence. A prisoner may also be released on licence during the requisite custodial period under section 246(1). A licence, whether under section 244 or 246, remains in place until the end of the sentence period, unless the licence is revoked and the person subject to the licence (the licensee) is recalled. The Secretary of State has the power to revoke a licence and recall a licensee back to prison pursuant to two different statutory provisions [3 8]. First, section 254(1) of the 2003 Act gives the Secretary of State a general power to revoke any licence and to recall the licensee to prison. Where the power of revocation is exercised under section 254(1), the licensee is entitled to be told the reasons for his recall and to make representations to the Secretary of State, and, ultimately, to the Parole Board. Secondly, section 255(1) confers a specific power on the Secretary of State to revoke a section 246 licence. This power of recall can only be exercised until the end of the requisite custodial period, when the licensee would have been entitled to be let out on licence as of right. Unlike the position in relation to the section 254 power of recall, there is no provision for review by the Parole Board of the exercise of the Secretary of States section 255 power of recall [9 10]. Under Strasbourg jurisprudence, where a person is lawfully sentenced to a determinate term of imprisonment by a competent court, there is (at least in the absence of unusual circumstances) no question of his being able to challenge his loss of liberty during that term on the ground that it infringes article 5(4). This is because, for the duration of the sentence period, the lawfulness of his detention has been decidedby a court, namely the court which sentenced him to the term of imprisonment [38]. On this approach, article 5(4) could not normally be invoked in a case where, in relation to those serving determinate terms, domestic discretionary early release provisions are operated by the executive. The notion that article 5(4) is satisfied by the original sentence appears entirely principled, and the consequence that a person under such a regime has to rely on his domestic remedies, at least unless other Convention rights are engaged, is not unreasonable in practice [40]. The common law should be well able to afford appropriate protection to the rights of people in the position of Mr Whiston without recourse to the Convention [45]. Consequently, in so far as it held that article 5(4) was engaged by the revocation of a mandatory licence, the House of Lords in Smith and West [2005] 1 WLR 350 were incorrect and the observations of Lord Brown in R (Black) v Secretary of State for Justice [2009] 1 AC 949 are wrong in so far as they suggest that the law of the United Kingdom in relation to article 5(4) differs from the Strasbourg jurisprudence [46]. Lady Hale agrees that the revocation of a discretionary licence does not infringe article 5(4). However, Lady Hale holds that the present law draws a principled distinction between those determinate prisoners who have reached the point in their sentence at which they are entitled to be released on licence and those who have not. If the former are recalled from their licence, and their representations to the Secretary of State are unsuccessful, they are entitled to have their case referred to the Parole Board. The latter, whose release on licence are discretionary, are not [50]. Once a prisoner has passed the point of mandatory release on licence, the basis for any later recall and detention is the risk of reoffending rather than the original order of the court, and, therefore, article 5(4) applies [52].
These appeals concern the obligations of insurance companies under various contracts of employers liability (EL) insurance. In particular, the appeals concern the scope of the insurers obligations to indemnify employers against their liabilities towards employees who have contracted mesothelioma following exposure to asbestos. Mesothelioma has an unusually long gestation period, which can be in excess of 40 years between exposure to asbestos and manifestation of the disease. The insurers maintain that the EL policies only cover mesothelioma which manifested as a disease at some point during the relevant policy period. In contrast, the employers submit that the insurance policies respond to mesothelioma caused by exposure to asbestos during the relevant policy period but which develops and manifests itself sometime later. The usual rule in negligence cases is that the claimant must establish on the balance of probabilities that the defendants negligence caused his injury or disease. In Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22 and Barker v Corus UK Ltd [2006] UKHL 20 the House of Lords developed an exception to this general principle in cases involving mesothelioma caused by exposure to asbestos. The effect of this special rule is that an employer is liable where exposure to asbestos contributed to the risk that the employee would suffer mesothelioma and where the employee in fact develops the disease. The insurers submit that the special rule in Fairchild/Barker is not applicable when deciding, for the purposes of an EL insurance policy, whether an employees mesothelioma was caused by exposure to asbestos during a particular policy year. At first instance Burton J held that the policies should all be interpreted as having a causation wording. He therefore held that the liability trigger under the EL policy was when the employee inhaled the asbestos and not the date when the malignant lesion developed. A majority of the Court of Appeal (Rix and Stanley Burnton LJJ) upheld the judge in relation to some of the EL insurance policies (particularly those covering disease contracted during the relevant insurance period); however they concluded that other policies (particularly those covering disease sustained during the insurance period) responded only on an occurrence or manifestation basis. These appeals to the Supreme Court raise two issues: (i) On the correct construction of the EL policies, is mesothelioma sustained or contracted at the moment when the employee is wrongfully exposed to asbestos or at the moment when the disease subsequently manifests in the employees body? (ii) Does the special rule in Fairchild/Barker apply when determining whether, for the purposes of the EL policies, an employee sustained or contracted mesothelioma during a particular policy period? The Supreme Court dismisses the insurers appeal by a 4 1 majority; Lord Phillips dissenting on the second issue. Lord Mance gives the main judgment. To resolve the meaning of the EL policies it is necessary to avoid over concentration on the meaning of single words or phrases viewed in isolation, and to look at the insurance contracts more generally [19]. Several features point the way to the correct construction. First, the wordings of the policies on their face require the course of employment to be contemporaneous with the sustaining of the injury [20]. Second, the wordings demonstrate a close link between the actual employment undertaken during each period and the premium agreed by the parties for the risks undertaken by the insurers in respect of that period. Third, on the insurers case there is a potential gap in cover as regards employers breaches of duty towards employees in one period which only lead to disease or injury in another later period [24]. Fourth, on the insurers case employers would be vulnerable to any decision by the insurers not to renew the policy. A decision not to renew might arise from the employers complying with their duty to disclose past negligence upon any renewal. Employers who discovered that they had been negligent in the course of past activities in respects that had not yet led to any manifest disease would have such a duty. The insurers could then simply refuse any renewal or further cover [25]. Fifth, the way most of the policies deal with extra territorial issues throws doubt on any suggestion that the wordings are so carefully chosen that a court should stick literally to whatever might be perceived as their natural meaning [28]. Section 1 of the Employers Liability Compulsory Insurance Act 1969 also points the way to the correct interpretation. This states that every employer shall insure, and maintain insuranceagainst liability for bodily injury or disease sustained by his employees, and arising out of and in the course of their employment. In order to give proper effect to the protective purpose of that legislation, the Act requires insurance on a causation basis [47]. There is no difficulty in treating the word contracted as looking to the causation of a disease, rather than its development or manifestation. The word contracted used in conjunction with disease looks to the initiating or causative factor of the disease [49]. While the word sustained may initially appear to refer to the manifestation of an injury, the nature and underlying purpose of the EL insurances is one which looks to the initiation or causation of the accident or disease which injured the employee. Accordingly a disease may properly be said to have been sustained by an employee in the period when it was caused or initiated, even though it only developed or manifested itself later [50]. In relation to the second issue, the question is whether the EL policies cover employers liability for mesothelioma arising under the special rule in Fairchild/Barker [71]. Under that rule the law accepts a weak or broad causal link between the employers negligence and the employees mesothelioma. When construing the EL policies the concept of a disease being caused during the policy period must be interpreted sufficiently flexibly to embrace the role assigned to exposure by the Fairchild/Barker rule [74]. The purpose of the EL policies was to insure the employers against liability to their employees. Once it is held that the employers are liable to the employees, it would be remarkable if the insurers were not liable under the policies [88]. Accordingly, for the purposes of the EL policies, the negligent exposure of an employee to asbestos during the policy period has a sufficient causal link with subsequently arising mesothelioma to trigger the insurers obligation to indemnify the employer [74]. Lord Phillips dissents on the second issue. The special approach developed in Fairchild/Barker raises no implication or fictional assumption as to when mesothelioma is initiated. The consequence is that if claimants have to show that mesothelioma was initiated in a particular policy year in order to establish that insurers are liable they are unable to do so. This conclusion is not affected by section 3 of the Compensation Act 2009, which did not alter the jurisprudential basis of the Fairchild/Barker approach [132] [133].
This appeal raises the question of how the concepts of sufficiency and infringement are to be applied to a Swiss form patent relating to a specified medical use of a known pharmaceutical compound. The Appellant (Warner Lambert) is part of the Pfizer group of companies. It is the proprietor of European Patent No 0641330 for Isobutylgaba. This is used for the treatment of seizure disorders, including epilepsy. Pregabalin, a derivative compound of Isobutylgaba, is marketed by Warner Lambert under the Lyrica brand. Patent No 0641330 expired on 17 May 2013. This appeal concerns a second European Patent No EP(UK) 0934061 entitled Isobutylgaba and its derivatives for the treatment of pain, with a priority date of 24 July 1996 (the Patent). The claims of the Patent (which define the scope of the patent protection) are all purpose limited. Most relevant are Claims 1 3 on the use of pregabalin for treating (1) pain, (2) inflammatory pain and (3) neuropathic pain. Lyrica has marketing authorisation in the EU for treatment of peripheral and central neuropathic pain, epilepsy and generalised anxiety disorder. It is one of Pfizers most successful drugs in the UK. The First Respondent (Mylan) and the Second Respondent, Actavis Group PTC EHF (Actavis), are pharmaceutical companies mainly engaged in marketing generic pharmaceutical products. Actavis markets a generic pregabalin product under the brand name Lecaent, launched in 2015. In these proceedings, Mylan and Actavis claimed the revocation of the Patent on the grounds of lack of inventive step and insufficiency. Warner Lambert claim that Actavis infringes Claims 1 and 3 above. At first instance, Arnold J rejected the arguments based on lack of inventive step. These are no longer in issue. Further, he held that Claim 1 (pain) and Claim 3 (neuropathic pain) were invalid because he construed Claim 1 as extending to all pain and Claim 3 as extending to all neuropathic pain. He found that there was sufficient disclosure in the specification to support the claim that pregabalin was efficacious in the treatment of inflammatory and peripheral neuropathic pain, but not central neuropathic pain. Both claims therefore failed for insufficiency. The result of the judges decision was to remove patent protection for the manufacture of pregabalin for the treatment of both peripheral and central neuropathic pain. Arnold J also rejected as an abuse of process an application concerning an amendment to narrow the Patent. The Court of Appeal (Floyd, Kitchin and Patten LJJ) upheld the judges findings, so far as relevant to this appeal, and his decision on the amendment application. The judge and Court of Appeal differed in their approach to infringement in patent cases confined to manufacture for a particular use. On appeal to the Supreme Court, Warner Lambert contend that all the claims of the Patent were valid. Their main aim is to establish the validity of their claims relating to neuropathic pain or, at least, peripheral neuropathic pain. Actavis and Mylan cross appeal, arguing that none of the claims as to neuropathic pain are valid. They only accept as valid the claims limited to inflammatory pain, for which there is no marketing authorisation. This gives rise to four issues on appeal: (i) the construction of the claims (in particular, Claim 3 as to neuropathic pain); (ii) the sufficiency of the disclosure in the specification; (iii) amendment and abuse of process; and (iv) the test for infringement of a patent in relation to manufacturing for a limited use. The Supreme Court dismisses the appeal and allows the cross appeal (Lord Mance and Lord Hodge dissenting in part on whether there was sufficient disclosure in the specification for Claims 1 and 3). Lord Sumption gives the leading judgment, with which Lord Reed, Lord Hodge and Lord Briggs agree, save on some issues specified in the separate judgments of Lord Briggs, Lord Hodge and Lord Mance. Issues (i) and (iii) Construction of the claims and amendment/abuse of process: The court unanimously affirms (for reasons given by Lord Briggs): (1) the view of both courts that Claim 1 extends to all pain and Claim 3 to all neuropathic pain, whether peripheral or central, and (2) Arnold Js decision rejecting Warner Lamberts application to amend the Patent to narrow it [15(1), 16 (Lord Sumption); 99 106, 118 120 (Lord Briggs); 181 (Lord Hodge); 195 196 (Lord Mance)]. Issue (ii) Sufficiency of disclosure in specification for Claims 1 and 3: The court holds, by a majority (Lord Sumption, Lord Reed and Lord Briggs), that the disclosure in the specification supports the claims in relation to inflammatory pain, but not neuropathic pain, whether peripheral or central. Claims 1 and 3 therefore fail for insufficiency. Thus, the appeal is dismissed and the cross appeal allowed [15(2), 43 54]. The majoritys approach requires the patentee to demonstrate that the specification discloses some scientific reason why the implied assertion of efficacy in the patent claim may well be true [36 37]. More than a bare assertion or mere possibility of therapeutic efficiency is required, though a priori reasoning (not necessarily only experimental data) may suffice [37]. This respects the principle that the patentee cannot claim a monopoly of new use for an existing compound without real disclosure [35]. Lord Hodge (dissenting) proposes an alternative approach to sufficiency, preferring a lower standard of plausibility, and would have dismissed the cross appeal [186 190]. Lord Mance agrees with Lord Hodge on this issue, concluding that the majoritys approach imposes too high a threshold [198 201]. Issue (iv) Correct test for infringement of patent manufactured for a limited use: The court unanimously holds that if Claims 1 and 3 had been valid, they would not have been infringed by Actavis [15(3)]. The reasons for arriving at this agreed result differ substantially. Lord Sumption and Lord Reed consider that the intention of the alleged infringer, whether subjective or objective, is irrelevant and that the sole criterion of infringement is whether the product as it emerges from the manufacturing process, including any labelling or accompanying leaflet, is presented as suitable for the uses which enjoy patent protection the outward presentation test [15(3), 71 86]. On the facts of this case, it is not disputed that Lecaent was sold with labels and patient information to the effect that it was for the treatment of seizure disorders and general anxiety disorder [8, 15(3)]. Lord Mance agrees that the test depends on the objective appearance and characteristics of the product as it is prepared, presented and put on the market, but considers that in rare cases the context may make it obvious that these are not to be taken at face value [15(3); 218 223]. Lord Briggs and Lord Hodge prefer the view of Arnold J that the test is whether the alleged infringer subjectively intended to target the patent protected market (Arnold J found they had not so intended) [15(3); 170 177 (Lord Briggs); 193 (Lord Hodge)].
In 2002 Mr Khan was registered as a pharmacist. Between 2010 and 2012 Mr Khan pleaded guilty to three incidents of domestic violence. In 2012 the General Pharmaceutical Council referred to its Fitness to Practise Committee (original committee) an allegation that Mr Khans fitness to practise as a pharmacist was impaired by reason of his misconduct. On 27 June 2013 the original committee found that the impairment of his fitness to practise was established. When it turned to identify the sanction in article 54(2)(d) of the Pharmacy Order 2010 (the Order) which would properly reflect the gravity of Mr Khans misconduct, the original committee rejected the option of suspending his right to practise for 12 months on the basis that this sanction would be insufficient to mark the degree of gravity of Mr Khans misconduct. Instead it directed that his entry in the register of pharmacists be altogether removed. The Extra Division of the Court of Session allowed Mr Khans appeal against the direction for removal and remitted the case to the original committee for it to determine the appropriate sanction in light of its Opinion. It found that the original committee had made no mention of its power under article 54(3)(a)(ii) of the Order to conduct a review following a direction for suspension and to direct that the suspension of the entry be extended for such further period not exceeding 12 months as may be specified in the direction. In its view there was therefore a middle way between suspension for 12 months, which the original committee has considered to be insufficient, and removal, which the original committee had acknowledged perhaps appeared harsh. In light of the original committees power to conduct later reviews, it had been reasonably incidental to its power of suspension for 12 months for it to indicate that it considered that the suspension should be extended thereafter, for a further 12 months or longer. Although this indication would not bind the review committee, the Extra Division reasoned that it must be assumed that the later committee will be obliged to respect the indication and if it departs from it will be expected to give reasons for doing so. The General Pharmaceutical Council appealed the decision of the Extra Division. Its appeal concerns whether a review committee may impose a further suspension to reflect the original committees conclusion that the gravity of the registrants misconduct demanded a longer period of suspension than the 12 months it was permitted to imposed. Mr Khan cross appealed against the Extra Divisions implicit rejection of his argument that in any event his removal from the register was disproportionate. The Supreme Court unanimously allows both the General Pharmaceutical Councils appeal and Mr Khans cross appeal. Lord Wilson gives the judgment, with which the other Justices agree. The Appeal The powers of the review committee following the original committees direction for a registrants suspension are set out in article 54(3)(a) of the 2010 Order. That article does not indicate how the powers should be exercised, and only limited assistance may be derived from the General Pharmaceutical Council (Fitness to Practise and Disqualification etc) Rules 2010. Rule 34(4) requires the representative of the council to inform the review committee of the background to the case and sanction previously imposed and to direct its attention to any relevant evidence. Rule 34(4) and (5) permit both parties to adduce evidence in relation to the person concerns fitness to practice. Rule 34(6) provides that, following a direction for suspension, the review committee must receive further evidence although the subject of it is not identified. It certainly seems that the reference to the registrants fitness to practise relates to his fitness at the time of the review hearing [24]. Greater assistance is collected from the Indicative Sanctions Guidance which makes clear that the focus of the review is upon the current fitness of the registrant to resume practice, judged in the light of what he has, or has not, achieved since the date of suspension. The review committee asks: does his fitness to practise remain impaired [27]. The recent work of the three UK Law Commissions in this area proposes that the review committee should address changes relevant to impairment which have or have not occurred since the date of the original committees direction [28]. It is also noteworthy that in the fifth report of the Shipman Inquiry, the Chairman stated that review hearings should focus the doctors mind on the need to undertake any necessary remediation [29]. The Extra Divisions conception is alien to the generally accepted conception of a review as a vehicle for monitoring the steps taken by the registrant towards securing professional rehabilitation [31]. Taylor v General Medical Council [1990] 2 AC 539, not cited to the Extra Division, expressly holds that the conception favoured by the Extra Division is misplaced [32]. In Taylor the court held that it can never be a proper ground for the exercise of the power to extend the period of suspension that the period originally directed was insufficient to reflect the gravity of the original offence [33]. The Extra Division was too ingenious. There was no middle way. It was wrong to remit the case to the committee on that basis [35]. The Cross Appeal The original committee itself acknowledged that its direction for removal might appear harsh. Serious though Mr Khans misconduct certainly was, the sanction appropriate to the disrepute into which Mr Khans conduct had brought, or was likely to bring, the profession of pharmacy was suspension of his registration, which, at the time of the committees determination, should no doubt have been for a period of a year [40]. A direction for suspension should be substituted in place of the original committees direction for removal from the register. In light of Mr Khans interim suspension since the date of the original committees direction, the period of his suspension should be four months and a review committee should conduct a review prior to the expiry of this period [41].
Ewa Michalak was a doctor employed by the Mid Yorkshire Hospitals NHS Trust from April 2002 until she was dismissed in July 2008. Following her dismissal, Dr Michalak brought an unfair dismissal claim against the Trust in the Employment Tribunal. The tribunal found that her dismissal had been unfair due to sex and race discrimination and victimisation. Dr Michalak received a compensation award and a public apology from the Trust. Before the tribunal had issued its determination, the Trust had reported Dr Michalak to the General Medical Council to consider whether she should continue to be registered as a medical practitioner. The Trust later accepted that there had not been proper grounds to refer her and she remains registered as a medical practitioner. In the meantime, however, the GMC had begun fitness to practice proceedings against Dr Michalak. She claims that the GMC discriminated against her in the way in which it pursued those proceedings, including a failure to investigate complaints she had made against other doctors employed by the Trust. In August 2013, Dr Michalak brought a claim in the Employment Tribunal against the GMC, its chief executive and one of its investigation officers in relation to these complaints. The GMC argued that section 120(7) of the Equality Act 2010 meant that the Employment Tribunal did not have jurisdiction to hear the claim, as judicial review already provides for an appeal in these matters. The issue in this appeal was whether the availability of judicial review proceedings in respect of decisions or actions of the GMC can properly be described as proceedings in the nature of an appeal and, on that account, the jurisdiction of the Employment Tribunal is excluded by section 120(7) of the 2010 Act. The Supreme Court dismisses the appeal. Lord Kerr gives the judgment with which all other members of the panel agree. Lord Mance provides an additional short judgment with further explanation of his reasons. It is accepted that Dr Michalak could seek judicial review of the decisions that are alleged to constitute discrimination. The issue is whether the availability of judicial review comes within section 120(7) of the 2010 Act. This depends on two requirements being satisfied: (i) whether judicial review can be described as a proceeding in the nature of an appeal and (ii) whether it is available by virtue of an enactment [13]. The Employment Tribunal was designed to be a specialised forum for the resolution of disputes between the employee and employer, with the power to award a comprehensive range of remedies. Where Parliament has provided an alternative route of challenge to a decision through an appeal or an appeal like procedure, however, it makes sense for the appeal procedure to be confined to that statutory route. This avoids the risk of expensive and time consuming satellite proceedings and is convenient for both the appellant and respondent. Employment tribunals should be prepared to examine critically whether statutory appeals are available, and where they are, should strike out proceedings before them. This rationale only applies where the alternative route is capable of providing an equivalent means of redress, however. [16 18]. Conventionally, an appeal is a procedure which entails a review of an original decision in all its aspects an appeal body may thus examine the basis on which the original decision was made, assess the merits of the conclusions reached and, if it disagrees, substitute its own view. Judicial review, by contrast, is a proceeding in which the legality of or procedure by which a decision is reached is challenged. It cannot partake of the nature of an appeal the remedy available on a judicial review application in circumstances such as the present is a declaration that the decision is unlawful or that the decision be quashed. The court cannot substitute its own decision for that of the decision maker and, in that sense, the decision of the GMC could not be reversed. An appeal in a discrimination case must confront directly the question whether discrimination has taken place, not whether the GMC had taken a decision which was legally open to it. [20 22]. The origins of judicial review lie within the common law and it is not a procedure which arises by virtue of any statutory source section 31 of the Senior Courts Act 1981 did not establish judicial review as a procedure, but rather regulated it. If Parliament had intended that judicial review was within the scope of the procedures contemplated by section 120(7), one would expect that it would have provided for it expressly [32 33, 35]. Lord Mance agrees with Lord Kerr but adds that he would not necessarily limit the ability of judicial review to cater for a close examination of a claim on its merits in appropriate circumstances, judicial review may lead the court to a conclusion that there exists only one possible outcome of a properly conducted legislative or executive decision making process. In this situation, however, the Employment Tribunal offers the natural means of recourse and there is no need to strain the ordinary understanding of the concept of appeal to embrace judicial review [37 38].
This appeal relates to bonds issued by the respondent, the Republic of Argentina (Argentina), in respect of which, together with all its other debt, Argentina declared a moratorium in December 2001. The appellant NML Capital Limited (NML) is an affiliate of a New York based hedge fund, which purchased the bonds at little over half their face value between June 2001 and September 2003 and then pursued the respondent for the return of their full principal value and interest in the New York courts. On 11 May 2006 NML obtained summary judgment on the bonds from a Federal Court in New York for over $284m. It then sought to enforce the judgment against assets held by Argentina in England by bringing a common law action on the judgment in London. NML applied for permission to serve the claim form out of the jurisdiction on Argentina, initially alleging two reasons why Argentina was not entitled to state immunity. The first was that Argentina had waived immunity as a term of its agreement with Bankers Trust to issue the bonds, and the second that the claim constituted proceedings relating to a commercial transaction and so fell within the exception in section 3(1)(a) of the State Immunity Act 1978 (the 1978 Act). Permission was granted and the proceedings were served. Argentina then applied to have the order for service set aside. NML resisted this application by reliance on two different grounds: the provisions of section 31 of the Civil Jurisdiction and Judgments Act 1982 (the 1982 Act) and terms as to waiver and jurisdiction in the bonds themselves. The High Court dismissed Argentinas application but this decision was reversed by the Court of Appeal, which held that Argentina was protected by state immunity from the claim. The Supreme Court unanimously allows the appeal. Lord Phillips and Lord Clarke find that the claim falls within the scope of section 3(1)(a) of the 1978 Act. Lord Mance, Lord Collins and Lord Walker disagree on this point but all agree that the appeal should in any event succeed by reason of the provisions of s 31 of the 1982 Act and by Argentinas submission and waiver of immunity in the bonds. The scope of s 3(1)(a) of the 1978 Act Sections 1 11 of the 1978 Act are a comprehensive statement of the scope of state immunity under the law of the United Kingdom. During the twentieth century there was a growing recognition around the world of the restrictive doctrine of state immunity, under which immunity was given to governmental acts in the exercise of sovereign authority but not to commercial activities carried on by the state [10]. S 3(1)(a) made it clear that the UK was adopting the restrictive doctrine. The context for interpreting the phrase proceedings relating to a commercial transaction in this case was the enforcement of a judgment which absent state immunity would be permitted by Civil Procedure Rule 6.20(9). Lord Phillips and Lord Clarke considered that the words relating to should be given a broad rather than a narrow meaning. The proceedings related both to the foreign judgment and to the transaction underlying that judgment [26]. Although Parliament was unlikely to have thought the 1978 Act to apply to a class of foreign judgments at a time when there was no procedural machinery to serve a defendant out of the jurisdiction, s 3(1)(a) should be given an updated meaning consistent with the statutory purpose of the act [152]. Lord Mance did not think it was justified to treat the wording of s 3(1)(a) as applying to a foreign judgment against a foreign state, which had long been recognised as a special area of private international law [80]. S 31 of the 1982 Act was the means by which Parliament had achieved for the first time a comprehensive treatment of the issue of state immunity in respect of foreign judgments [98][118]. Lord Collins (with whom Lord Walker agreed) pointed to the almost invariable use in international loan agreements and bond issues since the 1970s of clauses providing for submission to national jurisdiction and waivers of immunity [103]. He preferred a narrow interpretation of s 3(1)(a). There was no policy reason to give it a wider meaning in the light of s 31 of the 1982 Act and the widespread use of express waivers [116]. Effect of s 31 of the 1982 Act S 31 of the 1982 Act reflected and in part replaced the categories of exemption from state immunity set out in the 1978 Act as far as foreign judgments were concerned. It was an alternative scheme rather than an additional hurdle [47]. State immunity could not be raised as a bar to the recognition and enforcement of a foreign judgment if, under the principles of international law recognised in this jurisdiction, the state against whom the judgment was given was not entitled to immunity in respect of the claim [49]. Did the bonds contain a submission to the jurisdiction of the English court? The High Court had correctly held that the agreement in the bond was more than a mere waiver and amounted to a submission to jurisdiction [59]. It was the only meaning the provision could sensibly bear [62] [128]. Could NML rely on new issues at the High Court hearing? Lord Phillips considered that the rule in Parker v Schuller (1901) should no longer be applied. Allowing a party to amend a pleading where no prejudice was caused to the other party that could not be dealt with by an appropriate order for costs accorded with the overriding objective [75]. The other justices did not think that the rule applied on the facts of this case, but agreed with his observations.
On 13 May 2012 China Gateway International Limited (CGI) submitted an application for planning permission to the local planning authority, Dover District Council (DDC). It sought permission for a large residential development within an area of outstanding natural beauty (AONB). The proposal was controversial. A planning officers report was circulated to the Planning Committee on 7 June 2013. It recommended the grant of permission with amendments to CGIs proposal, including a reduction in the number of planned houses at one site from 521 to 365. They also recommended ensuring, through an agreement with CGI (the Section 106 Agreement), various economic benefits including a planned hotel and conference centre. The planning officers report had regarded the level of harm to the AONB as significant but concluded that the suggested amendments created a finely balanced public interest. The Planning Committee met on 13 June 2013. Three members of the Committee expressly stated that harm to the AONB could be minimised by effective screening. The planning officers report had nonetheless expected that screening would be largely ineffective. After discussion, the Committee carried a motion approving the planning officers recommendation, but without the proposed reduction in the number of houses. On 18 December 2013 the application for planning permission returned to the Planning Committee with an updated planning officers report. The updated report confirmed that, contrary to the officers earlier recommendation, the Section 106 Agreement did not require CGI to provide the hotel but instead served to create an opportunity for a hotel. The Section 106 Agreement was executed on 1 April 2015. Planning permission was granted on the same day. Campaign to Protect Rural England Kent (CPRE Kent) sought a judicial review of that decision. Although it was unsuccessful at first instance, the Court of Appeal allowed the subsequent appeal and quashed the decision to grant permission. In this appeal to the Supreme Court it was not in dispute that the DDC was in breach of a specific requirement under the Town and County Planning (Environmental Impact Assessment) Regulations 2011 (the EIA Regulations) to provide a statement of the main reasons and considerations on which the decision was based. The issue is whether the Court of Appeal was right to quash the decision on that basis. The Supreme Court unanimously dismisses the appeal. Lord Carnwath gives the judgment, with which Lady Hale, Lord Wilson, Lady Black and Lord Lloyd Jones agree. The Court reviewed various statutory rules relating to the provision of reasons for planning decisions, observing that these rules are to be found in subordinate legislation and that it is hard to detect a coherent approach to their development. The three main categories of planning decision are: (i) decisions of Secretaries of State and inspectors, (ii) decisions by local planning authorities in connection with planning permission, and (iii) decisions, at any level, on applications for EIA development [21 23]. Special duties arise under the EIA Regulations where an application (as in this case) involves a development which is likely to have significant effects on the environment by virtue of factors such as its nature, size or location (an EIA development). Regulation 3(4) provides that decision makers shall not grant planning permission, where the application involves an EIA development, without first taking the environmental information into consideration, and that they must state in their decision that they have done so. Article 6.9 of the Aarhus Convention (Convention on Access to Information, Public Participation in Decision making and Access to Justice in Environmental Matters), to which the United Kingdom is a party, also requires each party to make accessible to the public the text of certain decisions involving an EIA, along with reasons and the considerations on which it is based [31 34]. Where there is a legal requirement to give reasons, what is needed is an adequate explanation of the ultimate decision. The content of that duty should not in principle turn on differences in the procedures by which the decision is arrived at. The essence of the duty, and the central issue, is whether the information so provided by the authority leaves room for genuine doubt as to what it has decided and why [35 42]. The Court rejects DDCs argument that a breach of the EIA duty alone should be remedied by a mere declaration of the breach. DDC relied on R (Richardson) v North Yorkshire County Council [2004] 1 WLR 1920 in which the Court of Appeal remedied a failure to provide a statement of reasons without quashing the decision, by ordering only that the statement be provided. However, in that case it was possible to take the planning committee as adopting the reasoning in the officers report which had recommended granting permission [46 49]. In view of the specific duty to give reasons under the EIA regulations, it is unnecessary to address the common law position. However, the particular circumstances of this case would, if necessary, have justified the imposition of a common law duty to provide reasons for the grant of permission. Planning law is a creature of statute, but the proper interpretation of the statute is underpinned by general common law principles, including fairness and transparency. It is appropriate for the common law to fill the gaps in the present system of rules, but its intervention should be limited to circumstances where legal policy reasons for it are strong [50 60]. The meeting on 13 June 2013 occurred only days after receipt of the planning officers detailed report, which proposed new and controversial amendments. A decision maker must not only ask himself the right question, but must take reasonable steps to acquaint himself with the relevant information to enable him to answer it correctly. Even if there was pressure to reach a decision in this case, it seems unfortunate that the committee members did not apparently consider deferring detailed discussion of the proposed amendments [62 63]. A mere declaration of the breach of the EIA duty is not an appropriate or sufficient remedy. In the three years since the permission was issued, no attempt has been made to formulate the reasons so as to make good the admitted breach, perhaps underlining the difficulty of reconstructing the reasons of the committee on the basis of its minutes alone. The recorded views of those members who supported the proposal do not indicate whether those views were shared by the majority, nor why the members felt able to reject the view of their own advisers without further investigation. Their omission of any legal mechanism to secure the proposed economic benefits, in particular the hotel and conference centre, required explanation. Furthermore, it was critical to understand the basis of the members belief that the harm to the AONB could be minimised, which conflicted with the planning officers view that screening would be largely ineffective. The quashing order of the Court of Appeal is consequently affirmed and the appeal is dismissed [61 69].
This appeal concerns the application of planning law to a dwelling house disguised as a hay barn. The first issue is whether the building is within the provisions of the Town and Country Planning Act 1990 which impose a time limit for taking enforcement action against breaches of planning control. The second issue is whether the owners dishonest scheme disentitles him from benefitting from those provisions. In 2001 Mr Beesley, the Second Respondent, applied for and obtained planning permission to construct a hay barn for grazing and haymaking on open land which he owned in the Green Belt. In 2002 he constructed a building which was to all external appearances the permitted barn, but internally was a fully fitted out dwelling house with garage, living room, study, bedrooms, bathrooms and gym. In August 2002 he moved in with his wife and lived there continuously for four years. Welywn Hatfield Borough Council, the Appellant, in whose area the property lies, remained unaware that the building was constructed as, or was being used as, a dwelling house. In August 2006 Mr Beesley made an application for a certificate of lawfulness for use of the building as a dwelling house. He contended that the four year time limit for taking enforcement action in section 171B(2) of the Town and Country Planning Act 1990 (the 1990 Act) was applicable and had elapsed. The section provides that where there has been a breach of planning control consisting in the change of use of any building to use as a single dwelling house, no enforcement action may be taken after the end of four years beginning with the date of the breach. The certificate was granted and subsequently upheld by the Court of Appeal, which decided that there had been a change of use within section 171B(2) such that immunity from enforcement was established. The Council appealed to the Supreme Court on two grounds. First, it challenged the Court of Appeals decision that there had been a relevant change of use. Secondly, it argued that even if there had been such a change, the principle of public policy that no one should be allowed to profit from his own wrong precluded Mr Beesley from relying on section 171B(2). The Supreme Court unanimously allows the appeal. It holds that: (i) there had been no change of use within section 171B(2); (ii) in any event, Mr Beesleys dishonest conduct meant that he could not rely on the section. Lord Mance gives the lead judgment. Lords Rodger and Brown deliver additional concurring judgments. On the first issue, the question was whether there had been any relevant change of use such as to bring the building within section 171B(2). The Supreme Court held first that the building which Mr Beesley constructed was not the permitted barn: it was a dwelling house. Therefore there could not have been a change of use within section 171B(2) from the use permitted by the planning permission. In any event, it was doubtful whether change of use under section 171B(2) could consist of a simple departure from permitted use. The word use in the section is directed to real or material use, not permitted use. [13] [14] Nor was there a relevant change of use on the basis that in the short period between completion of the building in July 2002 and its residential occupation on 9 August 2002 the building had no use, such that there was a change of use from no use to use as a dwelling house on and after 9 August. It is artificial to say that a building has no use when its owner has just built it to live in and is about to move in a few days time. The question of whether it is right to describe a building as having no use is not one which can sensibly be answered on a day by day basis, but rather calls for a broader and longer term view. For all these reasons, section 171B(2) is simply not apt to encompass the use of a newly built house as a dwelling house. [27] [30]; [68] In light of this conclusion on the first issue, it was not strictly necessary to address the second issue, but given its importance the Court went on to do so. The issue involved consideration of the scope and application of the principle that, unless the contrary intention appears, statutes are to be construed to the effect that no one should be allowed to profit from his own wrong. The Court noted that Mr Beesley intended to deceive the Council from the outset by his statements in the planning application. This was positive deception in matters integral to the planning process and directly intended to undermine that process. His conduct was not identifiably criminal but the principle is not only relevant where there has been the commission of a crime. The Court further considered the rationale of the statutory provision: the four year period in section 171B(2) must have been conceived as a period during which a planning authority would normally be expected to discover an unlawful use and after which the general interest in proper planning control should yield and the status quo prevail. Positive and deliberately misleading false statements by an owner which prevent discovery take a case outside that rationale. It would in fact frustrate the policy of the section if the time limit for enforcement was to apply on the facts of the present case. It would also damage the publics confidence in planning law: any law abiding citizen would be astonished to suppose that Mr Beesleys dishonest scheme, once being discovered, would not be enforced against but rather crowned with success. It is unthinkable that Parliament intended such an outcome. Even if, therefore, Mr Beesley had come within the literal wording of section 171B(2), his conduct took him outside its scope. [31]; [53] [58]; [67]; [80]
Interest bearing loan notes (the notes) to the value of 660m were issued to certain companies (the Noteholders) by a special purpose vehicle formed by the Lehman Brothers group, Eurosail UK 2007 3BL (the Issuer). The Issuer used the issue of the notes to fund the purchase of a portfolio of mortgage loans, to the value of 650m, secured on residential property in the United Kingdom. The notes were issued in 5 principal classes in order of priority for repayment. Those classes run from A through to E, and comprise a total of 14 sub classes. The A notes hold highest priority, are of the highest value, and are designated either A1, A2 or A3. The final redemption date of the lowest priority notes is in 2045. The terms governing the issue of the notes (the Conditions) provide that in the event of an Event of Default, an Enforcement Notice may be served by the trustee of the Noteholders rights, namely BNY Corporate Trustee Services Ltd (the Trustee). If the Issuer becomes unable to pay its debts under the terms of section 123 of the Insolvency Act 1986 (the 1986 Act), that would constitute an Event of Default. That section provides that a company is deemed unable to pay its debts, first, if it is unable to pay those debts as they fall due or, secondly, if the value of the companys assets is less than the amount of its liabilities, taking into account its contingent and prospective liabilities. The former is often referred to as the cash flow test, and the latter as the balance sheet test. The effect of section 123 was incorporated into the Conditions. Further, on the date on which the notes were issued, on behalf of the Noteholders the Trustee entered into an agreement (PECO) with another company (OptionCo). OptionCo was granted the option to purchase all of the notes from the Noteholders, plus the accrued interest, for a nominal consideration in the event that the Trustee determines that the Issuer is unable to pay its debts under section 123. The intended effect of this was that, in the event that the assets of the Issuer were exhausted, the remaining claims of the Noteholders against the Issuer would be assigned to OptionCo, and the Issuer would not be regarded as unable to pay its debts. The Issuer had entered into swap agreements with two of Lehman Brothers companies, with the consequence that when the latter became insolvent the Issuer suffered a significant deficiency in its net asset position, though it continued to pay its debts. The holders of the A1 and certain other of the notes had been repaid by this time. The A2 Noteholders were to have priority over A3 Noteholders in receiving repayments of principal out of sums raised by the Issuer from the redemption of mortgages in the portfolio, though those two groups would rank equally for repayment of interest. However, a finding that the Issuer was unable to pay its debts, and the consequent issuing of an Enforcement Notice, would alter this position significantly: all notes would become immediately due and payable and, importantly, A2 and A3 Noteholders would rank equally for repayment of principal. Against that background, though adopting a neutral position, the Trustee commenced these proceedings to seek a determination of whether the difficulties suffered by the Issuer constituted an Event of Default on the basis that it was unable to pay its debts within the meaning of section 123 of the 1986 Act. This appeal is therefore concerned with the construction of section 123. The Issuer and certain of the A2 Noteholders successfully argued in both the High Court and the Court of Appeal that the Issuer was not unable to pay its debts within the meaning of section 123. The Appellants, who are A3 Noteholders, argue to the contrary, and seek a stricter construction of section 123 than that which was applied by the lower courts. By way of cross appeal the Issuer renews its argument, rejected by the Court of Appeal, that in the event that the Issuer was otherwise deemed unable to pay its debts under section 123, the effect of the PECO should serve to alter that conclusion. The Supreme Court unanimously dismisses the appeals and the Issuers cross appeal. Lord Walker, with whom Lord Mance, Lord Sumption and Lord Carnwath agree, gives the lead judgment. Lord Hope gives a concurring judgment. Having regard to previous relevant legislation, to the authorities pertaining to those provisions and to section 123 of the 1986 Act itself, the enactment of section 123 should be seen as having made little significant change in the law. The changes in form therein emphasise that the cash flow test is concerned with debts falling due from time to time in the reasonably near future, in addition to those debts presently due. What is to be regarded as the reasonably near future will depend on the circumstances at hand, but especially the nature of the companys business [37]. However, once one moves beyond the reasonably near future, any attempt to apply the cash flow test will become completely speculative. In that situation, a comparison of present assets with present and future liabilities, the latter having been discounted to account for contingencies and deferment of payments, becomes the only sensible test. That is the reason for the inclusion of the balance sheet test in section 123, though it is still very far from an exact test. It is for the party asserting balance sheet insolvency to establish insolvency of that nature [37]. Whether or not the balance sheet test of insolvency is satisfied must depend on the available evidence as to the circumstances of the particular case. In that regard, the Issuer is not engaged in normal, on going trading activities, and therefore its present assets should be a better guide to its ability to meet its long term liabilities. Against that, the impact of factors relevant to its business in the period until the final redemption rate in 2045, such as currency movements, interest rates and the economy and housing market of the United Kingdom, must be considered. However, they are a matter of speculation rather than calculation or prediction on a scientific basis [38, 49]. As the Issuers liabilities can, as matters stand, be deferred until 2045, and as it is currently paying its debts as they fall due, the Court should proceed with the greatest caution in deciding that it is in a state of balance sheet insolvency [42]. Its ability to pay all its debts, present or future, may not be finally determined until much closer to 2045. The Conditions contain several mechanisms to ensure that liabilities in respect of principal can be deferred until that date. That being so, the Court cannot be satisfied that there will eventually be an inability on the part of the Issuer to pay its debts [49]. Though it is not required to decide the point because the appeal is dismissed, PECO agreements are of importance to the securitisation market. So the Court gives reasons for its decision to dismiss the cross appeal [51]. In that regard, the intended legal and commercial effects of the PECO, having regard to the wording of the documents pertaining to the transaction as a whole, point in the same direction: they do not affect the quantification of the Issuers liabilities. The meaning to be given to the language used by the parties on this point is not open to doubt. It would not be consistent with commercial good sense to depart from it [64].
These proceedings challenge a deportation order made in respect of the appellant, an Iraqi national who has lived unlawfully in the UK since 2000. He made an asylum claim in 2002, which was rejected, and his subsequent appeal was dismissed. In November 2005 he was convicted of Class A and C drug possession and was fined. On 4 December 2006, he was convicted of two counts of Class A drug possession with intent to supply, and sentenced to four years' imprisonment. Upon completion of his sentence in January 2011 he was considered to present a low risk of re offending. He has been in a relationship with his fiance, a British citizen, since 2005. The appellant has two children who probably reside in the UK, and with whom he has no contact. He has no remaining family in Iraq. On 5 October 2010 the Secretary of State made an automatic deportation order under s.32(5) of the UK Borders Act 2007. S.32(5) requires deportation orders to be made in respect of foreign criminals unless one of the exceptions in s.33 applies, which include breach of ECHR rights. A foreign criminal is defined in s.32(1) as a person who is not a British citizen, who is convicted in the UK of an offence and is sentenced to a period of imprisonment of at least 12 months. The Secretary of State found that the appellant did not fall within any of the exceptions in s.33: she accepted that deportation might interfere with the appellants ECHR article 8 rights to private and family life, but considered that this was proportionate to the aim of preventing disorder or crime and the maintenance of effective immigration control. The appellant appealed the Secretary of States decision, but the First tier Tribunal dismissed his appeal. The appellant then appealed to the Upper Tribunal, which set aside the First tier Tribunals decision and re heard the appeal, allowing it on the ground that a s.33 exception applied: the appellants removal would be incompatible with his rights under article 8. The Court of Appeal allowed the Secretary of States appeal on the ground that the Upper Tribunal had failed, in its assessment of proportionality, to take into account the new Immigration Rules which had come into force in July 2012, and had failed to recognise the importance of the public interest in deporting foreign criminals. The Court of Appeal remitted the appeal for re consideration by a differently constituted Upper Tribunal. This is the appeal against the decision of the Court of Appeal to remit. The Supreme Court dismisses Mr Alis appeal by a majority of 6 to 1. Lord Reed gives the lead judgment (with which Lord Neuberger, Lady Hale, Lord Wilson, Lord Hughes and Lord Thomas agree). Lord Wilson and Lord Thomas each add a concurring judgment. Lord Kerr gives a dissenting judgment. The Immigration Rules (the Rules) were a relevant and important consideration which the Upper Tribunal ought to have taken into account when assessing the proportionality of the interference with the appellants article 8 rights. It should also have taken into account that his relationship with his partner was formed at a time when his immigration status was such that the persistence of family life within the UK was uncertain [60]. The European Court of Human Rights has provided guidance to the factors which should be taken into account in the balancing exercise (for example in Boultif v Switzerland (2001) 33 EHRR 50, Maslov v Austria [2009] INLR 47, Jeunesse v Netherlands (2014) 60 EHRR 17). These factors involve wide ranging consideration of the appellants circumstances including the nature of his private and family life in the UK, his links to the destination country, and the likelihood of him re offending [26 33]. The weight to be attached to each factor in the balancing exercise falls within the margin of appreciation of the national authorities [35]. The Rules set out the Secretary of States assessment of the weight generally to be afforded to some of these factors. In particular, the Rules prescribe a presumption that the deportation of foreign criminals is in the public interest, except where specified factors are present which the Rules accept outweigh that interest. Outside of those specified factors (for example in every case where a custodial sentence of 4 years of more has been imposed, as here), the Rules state that exceptional circumstances that is, compelling reasons are required to outweigh the public interest in deportation. The Rules are not law, but do have a statutory basis and require the approval of Parliament. It is within the margin of appreciation to adopt rules reflecting the assessment of the general public interest made by the Secretary of State and endorsed by Parliament. [15 23, 36 39]. As an appellate body, the Upper Tribunals decision making process is not governed by the Rules, but should nevertheless involve their consideration. The Upper Tribunal must make its own assessment of the proportionality of deportation, on the basis of its own consideration of the factors relevant to the particular case, and application of the relevant law. But in doing so, it must not disregard the decision under appeal. Where the Secretary of State has adopted a policy in relation to the assessment of proportionality, set out in the Rules and endorsed by Parliament, the Upper Tribunal should give considerable weight to that policy. In this case that policy was that a custodial sentence of four years or more represents such a serious level of offending that the public interest in the offenders deportation almost always outweighs countervailing considerations of private or family life [39 50, 60 64]. Lord Wilson adds that public concern (as reflected in the Rules endorsed by Parliament) can assist a courts objective analysis of where the public interest lies [65 81]. Lord Thomas emphasises the importance of clear reasoning at first instance through a structured balance sheet approach [82 84]. In a dissenting judgment, Lord Kerr would have allowed the appeal and upheld the decision of the Upper Tribunal. He concluded that the application of the Rules, and their prescription of the weight to be given to the public interest in the deportation of foreign criminals, were not compatible with the balancing exercise that had to be undertaken in considering the relevant factors arising under article 8 in a particular case. It had been sufficient for the Upper Tribunal to take into account those relevant factors. Undue or unique reliance on the Rules, at the expense of a comprehensive survey of the pertinent article 8 factors was not appropriate.
These appeals concern the obligations of insurance companies under various contracts of employers liability (EL) insurance. In particular, the appeals concern the scope of the insurers obligations to indemnify employers against their liabilities towards employees who have contracted mesothelioma following exposure to asbestos. Mesothelioma has an unusually long gestation period, which can be in excess of 40 years between exposure to asbestos and manifestation of the disease. The insurers maintain that the EL policies only cover mesothelioma which manifested as a disease at some point during the relevant policy period. In contrast, the employers submit that the insurance policies respond to mesothelioma caused by exposure to asbestos during the relevant policy period but which develops and manifests itself sometime later. The usual rule in negligence cases is that the claimant must establish on the balance of probabilities that the defendants negligence caused his injury or disease. In Fairchild v Glenhaven Funeral Services Ltd [2002] UKHL 22 and Barker v Corus UK Ltd [2006] UKHL 20 the House of Lords developed an exception to this general principle in cases involving mesothelioma caused by exposure to asbestos. The effect of this special rule is that an employer is liable where exposure to asbestos contributed to the risk that the employee would suffer mesothelioma and where the employee in fact develops the disease. The insurers submit that the special rule in Fairchild/Barker is not applicable when deciding, for the purposes of an EL insurance policy, whether an employees mesothelioma was caused by exposure to asbestos during a particular policy year. At first instance Burton J held that the policies should all be interpreted as having a causation wording. He therefore held that the liability trigger under the EL policy was when the employee inhaled the asbestos and not the date when the malignant lesion developed. A majority of the Court of Appeal (Rix and Stanley Burnton LJJ) upheld the judge in relation to some of the EL insurance policies (particularly those covering disease contracted during the relevant insurance period); however they concluded that other policies (particularly those covering disease sustained during the insurance period) responded only on an occurrence or manifestation basis. These appeals to the Supreme Court raise two issues: (i) On the correct construction of the EL policies, is mesothelioma sustained or contracted at the moment when the employee is wrongfully exposed to asbestos or at the moment when the disease subsequently manifests in the employees body? (ii) Does the special rule in Fairchild/Barker apply when determining whether, for the purposes of the EL policies, an employee sustained or contracted mesothelioma during a particular policy period? The Supreme Court dismisses the insurers appeal by a 4 1 majority; Lord Phillips dissenting on the second issue. Lord Mance gives the main judgment. To resolve the meaning of the EL policies it is necessary to avoid over concentration on the meaning of single words or phrases viewed in isolation, and to look at the insurance contracts more generally [19]. Several features point the way to the correct construction. First, the wordings of the policies on their face require the course of employment to be contemporaneous with the sustaining of the injury [20]. Second, the wordings demonstrate a close link between the actual employment undertaken during each period and the premium agreed by the parties for the risks undertaken by the insurers in respect of that period. Third, on the insurers case there is a potential gap in cover as regards employers breaches of duty towards employees in one period which only lead to disease or injury in another later period [24]. Fourth, on the insurers case employers would be vulnerable to any decision by the insurers not to renew the policy. A decision not to renew might arise from the employers complying with their duty to disclose past negligence upon any renewal. Employers who discovered that they had been negligent in the course of past activities in respects that had not yet led to any manifest disease would have such a duty. The insurers could then simply refuse any renewal or further cover [25]. Fifth, the way most of the policies deal with extra territorial issues throws doubt on any suggestion that the wordings are so carefully chosen that a court should stick literally to whatever might be perceived as their natural meaning [28]. Section 1 of the Employers Liability Compulsory Insurance Act 1969 also points the way to the correct interpretation. This states that every employer shall insure, and maintain insuranceagainst liability for bodily injury or disease sustained by his employees, and arising out of and in the course of their employment. In order to give proper effect to the protective purpose of that legislation, the Act requires insurance on a causation basis [47]. There is no difficulty in treating the word contracted as looking to the causation of a disease, rather than its development or manifestation. The word contracted used in conjunction with disease looks to the initiating or causative factor of the disease [49]. While the word sustained may initially appear to refer to the manifestation of an injury, the nature and underlying purpose of the EL insurances is one which looks to the initiation or causation of the accident or disease which injured the employee. Accordingly a disease may properly be said to have been sustained by an employee in the period when it was caused or initiated, even though it only developed or manifested itself later [50]. In relation to the second issue, the question is whether the EL policies cover employers liability for mesothelioma arising under the special rule in Fairchild/Barker [71]. Under that rule the law accepts a weak or broad causal link between the employers negligence and the employees mesothelioma. When construing the EL policies the concept of a disease being caused during the policy period must be interpreted sufficiently flexibly to embrace the role assigned to exposure by the Fairchild/Barker rule [74]. The purpose of the EL policies was to insure the employers against liability to their employees. Once it is held that the employers are liable to the employees, it would be remarkable if the insurers were not liable under the policies [88]. Accordingly, for the purposes of the EL policies, the negligent exposure of an employee to asbestos during the policy period has a sufficient causal link with subsequently arising mesothelioma to trigger the insurers obligation to indemnify the employer [74]. Lord Phillips dissents on the second issue. The special approach developed in Fairchild/Barker raises no implication or fictional assumption as to when mesothelioma is initiated. The consequence is that if claimants have to show that mesothelioma was initiated in a particular policy year in order to establish that insurers are liable they are unable to do so. This conclusion is not affected by section 3 of the Compensation Act 2009, which did not alter the jurisprudential basis of the Fairchild/Barker approach [132] [133].
Mr Thevarajah entered into an agreement with the Appellants, Mr Riordan and Eugene and Barrington Burke, to buy the shares that they owned in Prestige Property Developer UK Ltd (the Company). Having paid 1.572m to the Appellants, Mr Thevarajah sought specific performance of the agreement in proceedings issued in March 2013. On 17 May 2013, Mr Thevarajah obtained a freezing order (the freezing order) which required the Appellants to provide by 24 May 2013 information and documents relating to all their assets, including assets held by the Company, as well as details of bank accounts. The Appellants did not provide the disclosure required by the freezing order by 24 May 2013. Mr Thevarajah subsequently applied for and obtained an unless order from Henderson J, which provided that: (i) the Appellants were required to disclose certain identified assets that they had failed to disclose; and (ii) in default of compliance by 1 July 2013, the Appellants would be debarred from defending the claim. The Appellants failed to comply fully with the unless order. Mr Thevarajah subsequently applied to the Chancery Division of the High Court for an order debarring the Appellants from defending their claim; the Appellants applied for a determination that they had complied with the unless order or, if they had not, for relief from sanctions. On 9 August 2013, Hildyard J heard the applications, made the debarring order sought by Mr Thevarajah and dismissed the Appellants application for relief from sanctions. There was no appeal against Hildyard Js order. The trial of the action was due to start on 3 October 2013. Having instructed fresh solicitors, the Appellants issued a second application for relief from sanctions (the second relief application) on 2 October, accompanied by a lengthy affidavit which provided what the Appellants considered to be full disclosure as required by the freezing order. Mr Andrew Sutcliffe QC, sitting as a Deputy High Court Judge, heard the second relief application and granted the defendants relief against the debarring order, and fixed a fresh date for the trial. Mr Thevarajah appealed to the Court of Appeal. The Court of Appeal allowed the appeal and restored the debarring order imposed by Hildyard J. The Appellants now appeal to the Supreme Court. The Supreme Court unanimously dismisses the appeal. Lord Neuberger gives the only judgment, with which the other Justices agree. The Appellants contended that the Court of Appeal had erred in two respects: (i) in holding that the Appellants needed to establish a material change of circumstances in order to succeed on the second relief application; or, in the alternative (ii) in holding that the Appellants had failed to establish such a material change. The Appellants first ground is rejected. The effect of Henderson Js unless order, coupled with Hildyard Js finding that the Appellants had failed to comply with the disclosure requirements in that order, was that the Appellants were debarred from defending the claim unless they were granted relief from sanctions under CPR 3.9. [11] Hildyard Js reasoning in refusing relief from sanctions is consistent with authoritative guidance subsequently set out by the Court of Appeal in the cases of Mitchell v News Group Newspapers Ltd [2014] 1 WLR 795 and Denton v TH White Ltd [2014] 1 WLR 3926. [13] The Court of Appeals conclusion that the Deputy Judge should not have considered the second relief application on its merits because CPR 3.1(7) required the Appellants to show that there had been a material change of circumstances since the hearing of the first relief application, was correct. [14 19] Even if CPR 3.1(7) did not apply, this was the position as a matter of ordinary principle. [18] The Appellants second ground is also rejected. Where a party has had imposed on it a debarring order for failing to comply with an unless order, its subsequent compliance with that unless order cannot without more amount to a material change of circumstances. [21] In refusing relief from sanctions, a court is effectively saying that it was now too late for that party to comply with the unless order and obtain relief from sanctions. [21] However, that does not mean that late compliance cannot, in certain circumstances, give rise to a successful second application for relief from sanctions, at least where it occurs in the context of some other relevant change in circumstances. [22] On the facts, there were no grounds which justified the Deputy Judge entertaining the second relief application on the merits. [23] The Deputy Judge was not entitled to come to a different conclusion on what where essentially the same facts as were before Hildyard J. [24] Further, the evidence before the Deputy Judge was insufficient to justify his finding that the Appellants former solicitors were partly to blame for the Appellants failure to comply with the unless order. [25] The issue of delay is also relevant. There is considerable force in the Court of Appeals view that the Appellants should have been in difficulties on the second relief application because of their delay in bringing it. The second relief application was made eight weeks after Hildyard J made his order and one day before the trial was due to begin, without any satisfactory explanation. [26]
London suffered from serious rioting from 6 to 9 August 2011. In one incident at 11:40pm on 8 August 2011, a gang of youths broke into the Sony DADC distribution warehouse at the business park on Solar Way in Enfield. They stole goods from the warehouse and threw petrol bombs, starting a fire which destroyed the warehouse and the stock, plant and equipment within. The insurers of Sony DADC (which were the lessees of the warehouse), the insurers of the freehold owner of the warehouse, and companies (which were customers of Sony DADC and whose uninsured stock in the warehouse had been destroyed) made claims for compensation from the appellant, the Mayors Office for Policing and Crime (the MOPC) under s.2 of the Riot (Damages) Act 1886 (the 1886 Act). The MOPC contested those claims on both its liability to pay compensation and the quantification of loss. The High Court and the Court of Appeal decided that the MOPC was liable and that finding is no longer in issue. The remaining issue before the Supreme Court is the quantification of the claims, which raises a question of statutory construction: whether persons who suffer loss when rioters destroy their property can in principle obtain compensation for consequential losses, including loss of profits and loss of rent, under s.2 of the 1886 Act. The High Court held that s.2 provided compensation only for physical damage and not for consequential losses, but the Court of Appeal reversed that decision, holding that s.2 provided a right to compensation for all heads of loss, including consequential loss, proximately caused by physical damage to property for which the trespassing rioter is liable at common law. The Supreme Court unanimously allows the MOPCs appeal. Lord Hodge gives the lead judgment, with which the other Justices agree. The wording of the 1886 Act by itself does not provide a clear cut answer to the issue. In particular, s.2(1) does not clarify whether the loss for which a claimant may claim compensation is simply the physical damage to his property, or extends to consequential losses [14 15, 17]. The 1886 Act does not expressly provide compensation for either (a) personal injury caused by rioters, or (b) damage to property in the streets (e.g. a parked car) and there is no jurisprudence to support the view that such losses could be claimed. On any view, therefore, the Act provides only partial compensation for damage caused by rioters [16]. Legislative history The 1886 Act must be construed in the light of the prior legislative history [13]. Parliament first provided for compensation for riot damage in the Riot Act 1714 (the 1714 Act). Section 6 provided that when rioters had demolished certain buildings, the inhabitants of the hundred (an historical administrative subdivision of a county or shire) were liable to pay damages to the injured party [18 19]. The 1714 Act did not specify the scope of the damages to be paid by the hundred, so guidance was provided by the courts, which extended the right to compensation for loss occasioned to furniture and household goods [20 23]. However, this case law does not support a general principle that the hundred stood as sureties for the wrongdoer [24]. This is because: (1) While the 1714 Act imposed on the hundred the obligation to compensate only for loss occasioned by the destruction of, or damage to, buildings, the prior laws governing the hundreds liability for a failure to raise hue and cry imposed no such restriction [25]. (2) While under the 1714 Act the hundred incurred strict liability for the riot, the prior hue and cry legislation allowed the community to escape liability if hue and cry were raised and the offenders caught: the hundred were not sureties for the offender unless they failed to apprehend him [26]. (3) Most importantly, the legislative history after the 1714 Act shows there was no broad principle of compensation. In particular, the wording of the Remedies against the Hundred (England) Act 1827 (which amended and consolidated the prior legislation and remained in force until the 1886 Act) makes it clear that the statutory compensation was confined to physical damage to property [27 30]. The 1886 Act made certain changes to the statutory scheme, including transferring the liability to pay compensation from the hundred to the local police authority, and directing the police authority to have regard to the claimants conduct when deciding what compensation was due. None of these changes suggest an intention to extend the measure of compensation beyond physical damage to property [31 32]. When regard is had to the words of the 1886 Act in the context of its legislative history, there is no reason to think that Parliament ever intended that the statutory compensation scheme should mirror the rioters liability in tort, or should develop as the law of damages for tort developed [33]. The Act, like its predecessors, sets out a self contained statutory compensation scheme which does not extend to cover consequential losses [34].
This appeal concerns the circumstances in which an employer is vicariously liable for wrongs committed by its employees, and also whether vicarious liability may arise for breaches by an employee of duties imposed by the Data Protection Act 1998 (DPA). The appellant operates a chain of supermarkets and employed Andrew Skelton on its internal audit team. In July 2013, Skelton received a verbal warning after disciplinary proceedings for minor misconduct and bore a grievance against the appellant thereafter. In November 2013, Skelton was tasked with transmitting payroll data for the appellants entire workforce to its external auditors, as he had done the previous year. Skelton did so, but also made and kept a personal copy of the data. In early 2014, he used this to upload a file containing the data to a publicly accessible filesharing website. Skelton later also sent the file anonymously to three UK newspapers, purporting to be a concerned member of the public who had found it online. The newspapers did not publish the information. Instead, one alerted the appellant, which took immediate steps to have the data removed from the internet and to protect its employees, including by alerting police. Skelton was soon arrested and has since been prosecuted and imprisoned. The respondents, some of the affected employees, brought proceedings against the appellant personally and on the basis of its vicarious liability for Skeltons acts. Their claims were for breach of statutory duty under the DPA, misuse of private information, and breach of confidence. At trial, the judge concluded that the appellant bore no primary responsibility but was vicariously liable on each basis claimed. The judge rejected the appellants argument that vicarious liability was inapplicable given the DPAs content and its foundation in an EU Directive. The judge also held that Skelton had acted in the course of his employment, on the basis of Lord Toulsons judgment in Mohamud v WM Morrison Supermarkets plc [2016] UKSC 11 (Mohamud). The appellants subsequent appeal to the Court of Appeal was dismissed. The Supreme Court unanimously allows the appeal. Lord Reed gives the only judgment, with which Lady Hale, Lord Kerr, Lord Hodge and Lord Lloyd Jones agree. The primary issue before the Court is whether the appellant is vicariously liable for Skeltons conduct. The starting point is Lord Toulsons judgment in Mohamud, which was not intended to change the law of vicarious liability but rather to follow existing precedents [16 21]. One such authority was the House of Lords decision in Dubai Aluminium Co Ltd v Salaam [2003] 2 AC 366 (Dubai Aluminium), where Lord Nicholls explained the existing close connection test of whether the wrongful conduct was so closely connected with acts the employee was authorised to do that for the purposes of the liability of the employer to third parties, it may fairly and properly be regarded as done by the employee while acting in the ordinary course of his employment. The test had to be applied having regard to the circumstances of the case and previous court decisions, following Dubai Aluminium [22 23]. Having explained the close connection test, Lord Toulson summarised the law in the simplest terms. The first question was what functions or field of activities the employer had entrusted to the employee. Next, the court must decide whether there was sufficient connection between the position in which he was employed and his wrongful conduct to make it right for the employer to be held liable under the principle of social justice which goes back to Holt CJ. This had been more fully explained in Dubai Aluminium by Lord Nicholls as set out above [25]. Lord Toulson was not suggesting any departure from Lord Nicholls approach [26]. Further, read in context, Lord Toulsons comments that on the facts of Mohamud there was an unbroken sequence of events and a seamless episode referred to the capacity in which the employee had been purporting to act when the wrongful conduct took place, namely about his employers business [28]. Lord Toulsons comment, in relation to the facts of Mohamud, that motive is irrelevant should not be taken out of context: whether the employee was acting on his employers business or for personal reasons was important, but, on the facts of Mohamud, the reason why he had committed the tort could not make a material difference to the outcome [29 30]. The Court concludes that the judge and the Court of Appeal misunderstood the principles governing vicarious liability in a number of respects. First, the online disclosure of the data was not part of Skeltons field of activities, as it was not an act which he was authorised to do. Secondly, the satisfaction of the factors referred to by Lord Phillips in Various Claimants v Catholic Child Welfare Society [2012] UKSC 56 was not to the point: those factors were relevant to whether, where the wrongdoer was not an employee, the relationship between wrongdoer and defendant was sufficiently akin to employment for vicarious liability to subsist. They were not concerned with whether employees wrongdoing was so closely connected with their employment that vicarious liability ought to be imposed. Thirdly, a temporal or causal connection alone does not satisfy the close connection test. Finally, it was highly material whether Skelton was acting on his employers business or for purely personal reasons [31]. Considering the question afresh, no vicarious liability arises in the present case. Skelton was authorised to transmit the payroll data to the auditors. His wrongful disclosure of the data was not so closely connected with that task that it can fairly and properly be regarded as made by Skelton while acting in the ordinary course of his employment. On long established principles, the fact that his employment gave him the opportunity to commit the wrongful act is not sufficient to warrant the imposition of vicarious liability. An employer is not normally vicariously liable where the employee was not engaged in furthering his employers business, but rather was pursuing a personal vendetta. The close connection test elucidated by Lord Nicholls in Dubai Aluminium, in light of the cases that have applied it and on the particular facts of the present appeal, is not satisfied [32 47]. The second major issue before the Court is whether the DPA excludes imposition of vicarious liability for either statutory or common law wrongs. It is not strictly necessary to consider this in light of the above conclusion, but as full argument was heard, it is desirable that the Court expresses a view [48]. Ultimately the Court finds the appellants argument that liability is excluded unpersuasive. Imposing statutory liability on a data controller like Skelton is not inconsistent with the co existence of vicarious liability at common law, whether for breach of the DPA or for a common law or equitable wrong, as the DPA says nothing about a data controllers employer. It is irrelevant that a data controllers statutory liability under the DPA is based on a lack of reasonable care, while vicarious liability for an employees conduct requires no proof of fault. The same contrast exists at common law between, for example, an employees liability in negligence and an employers vicarious liability. It makes no difference that an employees liability may arise under statute instead [54 55]. The appeal is therefore allowed [56].
This appeal concerns the assessment of claimants for personal independence payment (PIP), a non means tested allowance paid to certain people with long term health problems or disability. The appeals focus is on one of the markers used to determine the extent to which the ability of claimants to carry out daily living activities is limited by their physical or mental condition. The particular activity in question is engaging with other people face to face and the issue is the interpretation of descriptor 9c found in Part 2 of Schedule 1 to the Social Security (Personal Independence Payment) Regulations 2013 (the Regulations). The Regulations are made under Part 4 of the Welfare Reform Act 2012 (the Act). Descriptor 9c reads: Needs social support to be able to engage with other people. The respondent is a man in his forties. He made a claim for PIP in February 2015. His entitlement to the daily living component of the allowance at the standard rate depended on whether he satisfied descriptor 9c, which would give him 4 points towards the required score of at least 8 points overall under regulation 5. His claim was rejected on the ground that his ability to engage with other people face to face only satisfied descriptor 9b, Needs prompting to be able to engage with other people, which gave him a score of 2 points. The respondent appealed unsuccessfully to the First tier Tribunal (FTT). His appeal to the Upper Tribunal was allowed on the ground that the FTT had given an inadequate explanation of why he satisfied descriptor 9b rather than 9c. The case was remitted to the FTT for rehearing and directions were given as to the interpretation of descriptor 9c. The appellant Secretary of State appealed to the Inner House of the Court of Session in relation to the directions. The Inner House refused the appeal while modifying some of the directions. The Secretary of State appealed to the Supreme Court. Before the Supreme Court the Secretary of State accepted that the social support required for descriptor 9c may consist of prompting, as with 9b, but for this descriptor the support had to be from a person trained or experienced in assisting people to engage in social situations. Whilst he accepted that a friend or family member who knows the claimant well could have the relevant training or experience, he argued that a need for help simply from someone familiar or trusted was not sufficient (the qualitative issue). He also argued that the social support needed to be contemporaneous with the face to face engagement, ie that the person offering the social support had to be physically present (the timing issue). The Supreme Court unanimously allows the appeal in the limited sense of interpreting the relevant legal provisions differently from the Inner House. The respondents claim will now return to the FTT for determination in accordance with this interpretation. The judgment is given by Lady Black. The qualitative issue The activity of engaging with people face to face can take many differing forms, as can the form of the assistance that is needed for the claimants engagement to occur [29 30]. A narrow and technical approach to the words social support in descriptor 9c is unwarranted; it is inconsistent with the governments objective of creating a benefit which is easier to understand and reaches those who need extra support to live independently and participate in everyday life [32]. What brings the claimant within descriptor 9c rather than 9b is that, to be able to engage with others, he or she needs the support to come from someone trained or experienced in assisting people to engage in social situations i.e. the support will only be effective if delivered by someone who is not just familiar with the claimant, but also trained/experienced in assisting engagement in social situations [34 35]. Careful scrutiny of the facts will sometimes be necessary in order to determine whether descriptor 9c applies, including probing the information provided by sensitive questions [38] and, where support is already being provided by family/friends, exploring how they have come to know what to do, whether that help could come from any well meaning friend or family member, and what additional help (if any) is required [37]. The timing issue It is helpful to consider examples of practical ways in which a person can be helped to engage face to face with others. For instance, preparation prior to the engagement might avoid overwhelming psychological distress, and, during the engagement, a supporter might be able to give the claimant reminders, direct the conversation away from topics that trigger anxiety, give private signs of reassurance, or recognise the need to remove the claimant from the meeting [40]. The Secretary of States insistence on it being necessary for the supporter to be present at the engagement would stand in the way of means of support which do not involve physical presence and would be likely to impede attempts to improve the claimants abilities to handle matters in future with less support [41]. It would be undesirable to construe descriptor 9c in a way that runs counter to these considerations, and there is nothing in the wording of the descriptor to require that. The word needs indicates a continuing need, to be found as respects every time over the 12 month period made relevant by s 81 of the Act, but it does not, of itself, exclude the possibility of assistance outside the confines of the engagement [43] and nor is there anything else to dictate such an interpretation [45]. Given that social support is likely to take many different forms, depending on the individual needs of the claimant, it is undesirable to attempt to prescribe in the abstract which other forms of support will be sufficient. It will be a question of fact and degree [46], addressed with close attention to the words of the descriptor and the required period condition [48]. The Inner Houses acceptance that a temporal or causal link was required between the help given and the activity should not be adopted. It is difficult to see how support which is linked in time to a face to face engagement but has no causal link to what occurs could have any relevance [47].
When she was 13 Ms Jackson, the pursuer, was hit by a car driven by the defender. She appeals to the Supreme Court from the Inner House of Sessions assessment of her contributory negligence at 70%. Section 1(1) of the Law Reform (Contributory Negligence) Act 1945 states: Where any person suffers damage as the result partly of his own fault and partly of the fault of any other person or persons, a claim in respect of that damage shall not be defeated by reason of the fault of the person suffering the damage, but the damages recoverable in respect thereof shall be reduced to such extent as the court thinks just and equitable having regard to the claimants share in the responsibility for the damage. The collision took place on 12 January 2004 on the A98 road between Banff and Fraserburgh, near its junction with a private road leading to the farm where the pursuer lived with her family. There was a 60mph speed limit, and no street lighting. The pursuer alighted from her school bus, which had just parked on the opposite side of the road from the entrance to the farm road. It was about 40 minutes after sunset and light was fading. The respondent was driving home in the opposite direction, travelling at about 50mph. His lights were on. The pursuer passed between the rear of the bus, which was still stationary, and the car behind it. She paused briefly at the offside rear of the bus, took one or two steps into the road into the path of the defenders oncoming car, before breaking into a run. While running she was struck by the defenders car, still travelling at about 50mph. She was projected into the air, the car passed beneath her and she landed on the road. The pursuer sustained serious injuries. The Lord Ordinary found that the defender had failed to drive with reasonable care and was negligent. If he had been travelling at a reasonable speed the pursuer would have made it safely past him, so the accident would not have occurred. The Lord Ordinary also considered that the principal cause of the accident was the reckless folly of the pursuer: either she did not look to the left before crossing or, having looked, she failed to identify and react sensibly to the presence of the car in close proximity. On either scenario the greater cause of the accident was her movement into the path of the defenders car at a time when it was impossible for him to avoid a collision. The Lord Ordinary assessed the pursuers contributory negligence at 90%. On appeal the Extra Division of the Inner House allowed the pursuers appeal and assessed her contributory negligence instead at 70%. The Supreme Court allows Ms Jacksons appeal by a majority of 3 2 (Lord Hodge and Lord Wilson dissenting) and awards her 50% of the agreed damages. Lord Reed (with whom Lady Hale and Lord Carnwath agree) gives the lead judgment allowing the appeal. Lord Hodge (with whom Lord Wilson agrees) would have dismissed the appeal. Lord Reed did not accept the appellants contention that there was no basis for a finding of contributory negligence at all on the findings made by the Lord Ordinary. [17 18] Section 1(1) of the 1945 Act does not specify how responsibility is to be apportioned. Decided cases show two aspects to apportionment: the respective causative potency of the parties acts and their respective blameworthiness. The court consistently imposed a high burden on drivers to reflect the potentially dangerous nature of driving. [20 26] There is no demonstrably correct apportionment. Since different judges may legitimately take different views of what is just and equitable in particular circumstances, those differing views should be respected, within the limits of reasonable disagreement. [27 28] The lower court must have gone wrong: in the absence of an identifiable error, only a difference of view as to apportionment that exceeds the ambit of reasonable disagreement will warrant that conclusion. [35] Apportionments are not altered because of disagreement as to the precise figure. However, appellate courts have intervened on the basis of disagreement as to whether one party bore much greater responsibility than the other: there is a qualitative difference between a finding of 60% contribution and a finding of 40%. [38] The Extra Division provided only a very brief explanation of their apportionment of 70%. Given their conclusion that the causative potency of the defenders conduct was greater than that of the pursuer, the result can only be explained on the basis that they considered the pursuer far more blameworthy. They rightly considered that she did not take reasonable care for her own safety, but regard has to be had to her circumstances. She was only 13. An assessment of the defenders speed in the circumstances was far from easy. Attempting to cross a relatively major road with a 60mph speed limit, after dusk and without street lighting, is not straightforward, even for an adult. The Extra Division considered that the defenders behaviour was culpable to a substantial degree, with which Lord Reed agrees. Overall the Extra Divisions reasoning does not provide a satisfactory explanation of their conclusion that the pursuer bore the major share of responsibility. Lord Reed considered the defenders conduct played at least an equal role to that of the pursuer in causing the damage and was at least equally blameworthy. He therefore allows the appeal and awards 50% of the agreed damages to the pursuer. [39 44] Lord Hodge would have dismissed the appeal. He agrees on the facts and the legal principles set out in Lord Reeds judgment. [45] The Lord Ordinarys assessment of 90% appears to have been influenced by eyewitnesses impressions, however the defenders prior failure to reduce his speed was a potent cause of the accident, which put a different perspective on the matter. The Extra Division was entitled to conclude the Lord Ordinary had gone wrong. The Extra Divisions assessment is not open to the same criticism, given the Lord Ordinarys findings. [47 49] Not to look or to knowingly run into the path of the car displayed a very high degree of carelessness. The Extra Division were entitled, because of the extent of her blameworthiness, to attribute to the pursuer the major share of responsibility. [57]
The issue in this appeal is whether the conditions of entitlement to State Pension Credit (SPC) are compatible with a rule of EU law which prohibits discrimination between nationals of different Member States. State Pension Credit is a means tested non contributory benefit. Section 1(2)(a) of the State Pension Credit Act 2002 provides that a claimant is entitled to SPC if he or she is in Great Britain. The State Pension Credit Regulations 2002 (the Regulations) provide when someone is or is not to be treated as being in Great Britain. This is not just a matter of where the person is. In outline, a person is treated as not in Great Britain if he or she is not habitually resident in the United Kingdom, the Channel Islands, the Isle of Man or the Republic of Ireland (the Common Travel Area). And no person is to be treated as habitually resident in the Common Travel Area if that person does not have a right to reside in the Common Travel Area. (Cumulatively, these are the conditions.) Regulation 1408/71 (EC) (the EU Regulation) provided for the application of social security schemes to employed persons and their families moving within the European Community. (It has now been replaced by other EU legislation.) The Regulation applies to SPC. One of the categories of person to which it applies is employed person. This includes EU nationals who have retired from employment but who remain insured by the social security system of a Member State because of contributions paid during their working life. The dispute in this case relates to the effect of Regulation 3. It provides that persons to whom this Regulation applies shall be subject to the same obligations and enjoy the same benefits under the legislation of any Member State as nationals of that State. This prohibits both direct discrimination (in broad terms, where people are treated differently because of their nationality) and indirect discrimination (broadly, where an apparently neutral test is applied to nationals and non nationals and places non nationals at a particular disadvantage). Mrs Patmalniece is Latvian pensioner, now 72 years of age and in receipt of a Latvian retirement pension. She came to the UK in June 2000. Her asylum claim was unsuccessful, but she was not removed from the UK. She has never worked in the UK. Latvia joined the EU on 1 May 2004. In August 2005, Mrs Patmalniece claimed SPC. But her claim was refused on the ground that she was not in Great Britain because she did not have a right to reside in the UK. She appealed against the refusal of SPC, arguing that the requirement that she have a right to reside in the UK was directly discriminatory on the grounds of her nationality, in breach of Regulation 3 of the EU Regulation. The Social Security Appeal Tribunal allowed her appeal. However, the Social Security Commissioner allowed the Secretary of States appeal and held the conditions to be indirectly discriminatory but justified. The Court of Appeal upheld that decision. Mrs Patmalniece appealed to the Supreme Court. The Supreme Court, by a majority, dismisses the appeal. It holds unanimously that the conditions are indirectly discriminatory. But the majority (Lord Walker dissenting) hold that this discrimination is justified because the Regulations are a proportionate response to the legitimate aim of protecting the UK public purse and that this justification is independent of the claimants nationality. The court also holds unanimously that the different treatment afforded to Irish nationals is protected by the Protocol on the Common Travel Area. Lord Hope gives the main judgment, with which Lord Rodger agrees. Lady Hale gives a separate judgment. Lord Brown agrees with both Lord Hopes and Lady Hales judgments. Lord Walker gives a dissenting judgment. The court considers three issues: (1) do the conditions of entitlement for SPC give rise to direct discrimination? (2) If they give rise only to indirect discrimination, is that discrimination justified? (3) Is that conclusion undermined by the favourable treatment that the Regulation gives to Irish nationals? Direct/indirect discrimination All UK nationals would automatically satisfy the right to reside element of the test, whereas nationals of other Member States would not automatically do so. However, UK nationals still had to satisfy the requirement of habitual residence. The result is that the in Great Britain test would be satisfied by some, but not all, UK nationals, and some, but not all, nationals of other Member States. It was more likely to be satisfied by UK nationals than nationals of other member states: [25] [28]. The court applies the decision of the Grand Chamber of the European Court of Justice in Bressol v Gouvernement de la Communaut Franaise (Case C 73/08). In Bressol the ECJ had considered a Belgian law which set down eligibility criteria to study in Belgium which were similarly structured to the entitlement conditions for SPC. Although Advocate General Sharpston in Bressol had proposed that the ECJ treat the provisions of the Belgian law as directly discriminatory on the grounds of nationality, the ECJ did not follow her approach. Although the reasons for the ECJs position were not fully explained in its judgment, the court has decided that it should follow its conclusion and hold that the entitlement conditions for SPC were only indirectly discriminatory: [30] [35], [73], [89] [92]. Justification A difference in treatment which amounts to indirect discrimination can be justified only if it is based on objective considerations independent of the nationality of the persons concerned and is proportionate to a legitimate aim: [36], [94]. The parties were agreed that the measures here were proportionate. The issues were whether the conditions pursued a legitimate aim and whether it was independent of the nationality of the persons affected. The majority holds that both tests are satisfied. The aim was to ensure that claimants were economically or socially integrated in the UK, or elsewhere in the Common Travel Area, thereby protecting the social security system against the risk of benefit or social tourism: [38], [46], [51] [52], [96] This justification was independent of nationality. Lady Hale notes, additionally, that the Governments aims in introducing the right to reside test were consistent with the aims of Regulation 1408/72 and that it is logical that if a person does not have a right under EU law to reside in a particular state, that state should not have the responsibility under EU law for ensuring their minimum level of subsistence: [103]. Lord Walker dissents on the issue of justification. He would have held that the provisions were probably aimed at discriminating against economically inactive foreign nationals on the grounds of their nationality: [79]. The relevance of the treatment of Irish nationals The Appellant argued that, as entitlement to SPC was extended to Irish nationals, it was discriminatory not to extend it to nationals of other Member States. The court rejects that argument. The provision for Irish nationals in the conditions is protected by Article 2 of the Protocol on the Common Travel Area, which provides that the UK and Ireland may continue to make arrangements between themselves relating to the movement of persons between their territories: [54] [60], [80].
Mr Renford Braganza disappeared between 1am and 7am on 11 May 2009 while working as the Chief Engineer on an oil tanker in the mid North Atlantic managed by the respondents (collectively BP). BP formed the opinion that the most likely explanation for his disappearance was that he had committed suicide by throwing himself overboard. As a result his widow was not entitled to death benefits under his contract of employment, which provided that compensation would not be payable if in the opinion of the Company or its insurers, the deathresulted fromthe Officers wilful act, default or misconduct. The question arising in this appeal is the proper test for the court to apply when deciding whether BP was entitled to reach the opinion it did. BP set up its own inquiry team into Mr Braganzas disappearance, to examine whether its systems could be improved, and it reported on 17 September 2009. The report identified six factors supportive of suicide and concluded that the most likely scenario was that Mr Braganza had jumped overboard deliberately. The report was forwarded to Mr Sullivan, the General Manager of the BP company which employed the officers on board the vessel. Mr Sullivan made no further inquiries of his own and on the basis of the report concluded that there had been wilful default within the meaning of Mr Braganzas employment contract so that death in service benefits were not payable to his widow. Mrs Braganza brought a claim in contract against BP for death benefits and damages for negligence under the Fatal Accidents Act 1976 and the Law Reform (Miscellaneous Provisions) Act 1934. In the High Court the judge was unable to make a finding as to the cause of Mr Braganzas death on the evidence. He upheld the contractual claim, finding that the opinion formed by BP was not reasonable because Mr Sullivan did not direct himself that cogent evidence commensurate with the seriousness of a finding of suicide was necessary and he had failed to take into account the real possibility that Mr Braganza had suffered an accident. The judge rejected the claim for negligence and Mrs Braganza did not appeal against this. BPs appeal in relation to the contractual claim was allowed by the Court of Appeal, which held that the employer did not have to approach the matter the way required by the judge, and that the conclusion of suicide was a reasonable one for BP to reach in all the circumstances. The Supreme Court by a majority of 3 to 2 allows Mrs Braganzas appeal. In the majority, Lady Hale gives the lead judgment and Lord Hodge gives a concurring judgment. Lord Kerr agrees with Lady Hale and Lord Hodge. Lord Neuberger gives a dissenting judgment, with which Lord Wilson agrees. The appeal raises two inter linked questions of principle: (i) the meaning of the general requirement that the decision of a contractual fact finder must be a reasonable one and (ii) the proper approach of a contractual fact finder who is considering whether a person may have committed suicide [17]. The court is not the primary decision maker but will seek to ensure that where there is a conflict of interest for the party charged with making a decision under the contract (which is heightened where there is a significant imbalance of power between the contracting parties), such contractual powers are not abused. The standard of review of contractual decisions is akin to that adopted for judicial review of administrative action [19]. That test involves two limbs: the first focusing on the decision making process and the second focused on its outcome [24]. The court will imply a term into the contract that the decision making process be lawful and rational in the public law sense, ie that the decision is made rationally, in good faith and consistently with its contractual purpose [30], but much depends on the context of the particular contract involved. This case involved an employment contract which has an implied obligation of trust and confidence, in accordance with which any fact finding function entrusted to the employer concerning whether a person has committed suicide must be exercised [32]. It would have been open to BP to conclude that it was unable to form an opinion as to the cause of Mr Braganzas death but instead it made a positive finding of suicide and the question was what was required for this conclusion [33]. As to this, it is not the consequences of a finding of suicide which demands that there be cogent evidence to support it but its inherent improbability [35]. A decision that an employee has committed suicide is not a rational or reasonable decision unless the employer has had it clearly in mind that suicide is such an improbability that cogent evidence is required to form a positive opinion that it has taken place [36]. On the facts of this case, Mr Sullivan should not simply have accepted the view of the inquiry, which was conducted for a different purpose, that suicide was the most likely explanation for Mr Braganzas disappearance. In order to make a positive finding of suicide he had to direct himself that cogent evidence was required sufficient to overcome its inherent improbability [39]. In this case there were no positive indications of suicide and the six factors relied on in the report were straws in the wind [40]. They should have been set against the evidence of his normal behaviour immediately before his disappearance, his concern about the weather and the fact that he was a Roman Catholic, which increased the inherent improbability of suicide in his case [41]. The lack of evidence supporting the hypothesis of an accident was still consistent with Mr Braganza having sustained an accident through carelessness [59]. Thus the judge had been right to find that the decision was unreasonable in the public law sense of having been formed without taking relevant matters into account [42, 63]. Lord Neuberger agrees with the majority that where a contract allocates power to a party to make decisions which have an effect on both parties the court should review the decision in the same way as it reviews administrative decisions [103]. BP had to carry out the investigation with honesty, good faith and genuineness, and avoid arbitrariness, capriciousness, perversity and irrationality. The courts approach when reviewing this decision should be similar to that of an appellate court reviewing a trial judges decision. In the present case, Lord Neuberger would have held that there was a combination of reasons which could fairly be said to be sufficiently cogent to justify the finding that Mr Braganza had taken the unusual and tragic course of committing suicide [114 125].
This appeal arises out of a joint venture between Apex Global Management Ltd (Apex), a Seychelles company owned by Mr Almhairat, and Global Torch Ltd (Global), a British Virgin Islands company owned by Prince Abdulaziz (the Prince), Mr Abu Ayshih and Mr Sabha. Apex and Global set up an English company Fi Call Ltd (Fi Call), and then fell out badly. In December 2011, Global issued a petition under sections 994 996 of the Companies Act 2006 against Apex, Mr Almhairat and Fi Call seeking share purchase orders, and pecuniary and declaratory relief. Ten days later, Apex issued a very similar cross petition against the Prince, the Princes father Prince Mishal, Global, Mr Abu Ayshih, and Fi Call. Allegations and counter allegations of seriously unlawful misconduct are involved, including money laundering, financial misappropriation, and funding of terrorism. The two petitions were ordered to be heard together. The relief sought by Apex included a claim for just under US$6 million plus interest, which it contended was owing to Apex by the Prince. The Prince denied that the $6m was owing on the ground that he had paid it into the bank account of certain companies. In July 2013 Vos J made a number of directions, including that each party file and serve a disclosure statement certified by a Statement of Truth signed personally. The Prince, who had objected to the order, failed to comply. This was on the basis that, as a member of the Saudi Royal Family, he was bound by a protocol which prevented him from taking part in litigation personally or from signing court documents. Apex applied to Norris J for, and obtained, an order that unless the Prince complied with the order, and in particular signed a Statement of Truth, his Defence be struck out and judgment be entered against him (an Unless Order). The Prince maintained his position and Apex applied to Norris J for, and obtained, judgment in its favour under Civil Procedure Rules (CPR) r.3.5(2). The Prince applied under CPR 3.1(7) for a variation of the Vos Js order and for relief from sanctions. Mann J refused to vary Vos Js order and rejected the application for relief from sanctions under CPR 3.9. In July 2014, Hildyard J refused an application for summary judgment in relation to the question of whether the $6m had in fact been repaid. The Prince unsuccessfully appealed the decisions of Vos J, Norris J and Mann J to the Court of Appeal, and was given permission to appeal to the Supreme Court on terms that he paid $6m to his solicitors to abide the order of the Court. The issue in this appeal is therefore whether The Prince is entitled to the relief he seeks. The Supreme Court dismisses the appeal by a majority of 4 1. Lord Neuberger (with whom Lord Sumption, Lord Hughes and Lord Hodge agree) gives the main judgment. Lord Clarke gives a dissenting judgment. The language of the CPR and of the relevant Practice Direction suggests that the standard form of disclosure by a party does require personal signing by the party and such an order reflected the normal practice [12 13]. Vos Js decision to make the order was well within the margin accorded to case management decisions [15]. Similarly, the approaches taken by Norris J to making an Unless Order and of Mann J to refusing relief from sanctions each represented a correct approach in principle and a careful consideration of the competing arguments, and Norris Js second decision was almost administrative in nature [18]. The decisions of Vos J, Norris J and Mann J are individually unassailable [20 21]. The contention that the consequence of these orders is disproportionate is difficult to maintain; the importance of litigants obeying court orders is self evident and if the disobedience continues notwithstanding the imposition of a sanction, the enforcement of the sanction is almost inevitable [23]. There are no special factors which justify reconsidering the original orders, and the Prince had two very clear opportunities to comply with the simple obligation to give disclosure in an appropriate fashion. [24 25]. The strength of a partys case on the ultimate merits of the proceedings is generally irrelevant when it comes to case management issues of this sort, though there may be an exception where a party has a case the strength of which would entitle him to summary judgment. A trial involves directions and case management decisions, and it is hard to see why the strength of either partys case should, at least normally, affect the nature or the enforcement of those directions and decisions [29 31]. The Prince would have a good prospect of establishing that the $6m was paid as he contends in his defence, but his prospects cannot be said to be any higher [33]. It is true that the question of whether the Prince has paid may be determined in the very proceedings which he would have been debarred from defending. However, it is inherent in orders such as default judgment that the claimants will obtain judgment for relief to which it may subsequently be shown they were not entitled. [36 37]. The Supreme Court should be very diffident about interfering with the guidance given or principles laid down by the Court of Appeal when it comes to case management and application the CPR [39]. Lord Clarke would have allowed the appeal on the basis that justice requires that the Prince should be allowed to challenge the claim against him, and all parties would be protected because the court would be able to resolve all the issues between the parties [46]. Lord Clarke would not limit the relevance of the merits to a case where the strength of a partys case would entitle him to summary judgment. [75]. Nobody had suggested that it will not be possible to have a fair trial because of the Princes breach of the orders which led to judgment being entered against him [77]. Lord Clarke agrees with Lord Neubergers comments on the role of the Supreme Court in relation to case management and the CPR [79]. Postscript After the oral argument on this appeal had been concluded and the Court had notified the parties of its conclusion, but before judgment was handed down, the Court was advised of recent judgments of Hildyard J in the principal action, given on 3 and 5 November, when he reluctantly adjourned the trial to 2015 on the application of Mr Almhairat. It would not be right for this Court to address the question whether to reconsider its decision to dismiss the Princes appeal in the light of these developments, and in particular in the light of any breaches of the CPR or any orders by any of the Apex parties. If, in the light of events which have occurred since the Court heard and decided the Princes appeal, reconsideration, revocation or modification of any of the orders is appropriate then that is a matter which should be raised before a Judge of the Chancery Division, who should also be responsible for deciding how the $6m should be dealt with [42 44].
Mr and Mrs Hickin became the joint tenants of a three bedroom terraced house in Chelmsley Wood, Solihull in 1967 [2]. The Appellant, Elaine Hickin, is their daughter who has lived in the house since the beginning of the tenancy. The Respondent, Solihull Metropolitan Borough Council, became the freehold owner and landlord in 1980 [2]. On 3 October 1980, the tenancy became a secure tenancy pursuant to Part II of the Housing Act 1980 [2]. The Housing Act 1980 was later consolidated into the Housing Act 1985. Mr and Mrs Hickin both lived in the house until some time after 1980 when Mr Hickin left. The tenancy remained a joint tenancy in the names of Mr and Mrs Hickin [2]. Mrs Hickin continued to live there with the Appellant until her death on 8 August 2007 [2]. Shortly after Mrs Hickins death, the Respondent served noticed on Mr Hickin to quit the property as it considered that he had become the sole tenant and since he no longer resided there the tenancy was no longer secure. It also commenced proceedings against the Appellant for possession of the house [3]. The Appellant resisted the possession proceedings on the basis that on her mothers death the secure tenancy had vested in her, rather than her father, as a result of section 89 of Housing Act 1985 [3]. At the trial, on agreed facts, Deputy District Judge Hammersley ordered possession. HHJ Oliver Jones QC, sitting in the High Court, allowed the appeal and declared that the tenancy vested in the Appellant. The Court of Appeal allowed the Respondent councils appeal and restored the order of the Deputy District Judge [3]. The Appellant appealed to the Supreme Court. The issue in the appeal is whether the common law rights of Mr Hickin as joint tenant of the secure tenancy had been displaced by the Housing Act 1985 statutory scheme in favour of the Appellant upon the death of Mrs Hickin [3]. The Supreme Court dismisses the appeal by a 3 2 majority, Lord Mance and Lord Clarke dissenting. Lord Sumption gives the leading judgment (with whom Lord Walker agrees) restoring the order of the Deputy District Judge. Lord Hope, Deputy President, gives a short concurring judgment. A secure tenancy under the Housing Act 1985 is not just a personal right of occupation, but is also an estate in land [6]. At common law, upon the death of a joint tenant, the tenancy is vested in the surviving joint tenant or in all of the survivors if there is more than one [1]. There is no transmission of the tenancy upon death, rather the interest of the deceased person is extinguished [8]. Sections 87 to 91 of the Housing Act 1985 operate to determine the transmission of a secure tenancy [6]. Subject to limited exceptions, a secure tenancy cannot be passed on to a third person with the benefit of the statutory security, either during the lifetime of the tenant or in the course of the administration of their estate after their death, except if that person is qualified to succeed under section 87 [6]. A person is qualified to succeed if he or she is the deceased tenants spouse or civil partner or any other member of the deceaseds family, within a broad definition contained in section 113 [6]. Section 89 of the Housing Act 1985 vests a tenancy in a qualified person if (i) a secure tenant has died; (ii) the tenancy was a periodic tenancy; (iii) the qualified person occupies the house as her only or principal home for the period of twelve months proceeding the death and (iv) the tenant was not herself a successor within the meaning of Section 88 [5]. The Housing Act 1985 does not, however, wholly displace the common law. The Act necessarily operates by reference to basic principles of the law of property and does not modify the common law governing the transmission of tenancies; rather it merely affects the statutory security of the tenure available when the tenancy has been transferred [7]. At common law and by virtue of section 8 of the Housing Act 1985, Mr and Mrs Hickin were joint secure tenants for as long as one of them occupied the property as an only or principal home [8]. Upon Mrs Hickins death, Mr Hickin remained the sole tenant under the agreement with the Respondent, to which he remained party, but since he was not occupying the property the tenancy ceased to be secure [8]. Mr Hickin could have made the tenancy secure again by moving back to the property at any time before the local authority served a notice to quit [8]. The provisions of the Housing Act 1985 do not affect this result. For the purposes of section 89(1), a secure tenant dies only when a sole tenant dies; if the tenancy is a joint tenancy a secure tenant has not died if there remains at least one living joint tenant [11]. The provisions of the Housing Act 1985 concern the transmission of the tenancy to a person other than the previous tenant on account of the latters death. Where there is a surviving joint tenant, the whole statutory basis for disposing of the succession to the tenancy is absent [11]. It is only necessary to provide for the transmission of a tenancy on death if there is a vacancy but where a joint tenant remains living there is none [11, 25]. The surviving tenant has the same contractual rights as he always did [11]. If Parliament had intended the section to operate to exclude the common law rights of a joint tenant it would have done so expressly [12]. Lord Hope notes that such express provision was made in the Housing (Scotland) Act 2001 [21 23] and whilst that Act could not be an aid to the construction of the relevant provisions of the Housing Act 1985, it indicates the kind of statutory language that can be used if the policy is to override the common law right of survivorship [23]. Lord Mance would have allowed the appeal on the grounds that where the surviving joint tenant is not in occupation, the secure tenancy cannot continue in the surviving tenant who cannot be a secure tenant [47]. In this situation, nothing in the Housing Act 1985 recognises or permits any right of survivorship to oust the mandatory statutory provisions contained in section 89 [47]. The tenancy vested in the Appellant upon Mrs Hickins death [47]. Lord Clarke would also have allowed the appeal on the ground that transmission under the Housing Act 1985 to a qualified person occurs when any individual joint tenant dies, and in this case the person qualified to succeed Mrs Hickin was the Appellant [60].
The appellant, Michael Mark Junior Darnley, was struck on the head on 17 May 2010. A friend, Robert Tubman, drove the appellant to the Accident and Emergency (A&E) Department at Mayday Hospital, Croydon which was managed by the respondent, NHS Trust. He attended at 20:26. The trial judge found that at the A&E reception, the appellant informed the receptionist that he thought he had a head injury and that he was feeling very unwell. The appellant and Mr Tubman both told the receptionist that the appellant was really unwell and needed urgent attention. The receptionist told the appellant that he would have to wait up to four to five hours before he could be seen by a clinician. The appellant told the receptionist he could not wait that long as he felt he was about to collapse. The receptionist replied that if he did collapse then he would be treated as an emergency. The identity of the A&E receptionist is unknown, save that it must have been one of the two receptionists on duty, neither of which had any recollection of the conversation. However, each described her usual practice when a person with a head injury asked about waiting times. One would say that they could expect to be seen by a triage nurse within 30 minutes of arrival. The other would say that the triage nurse would be informed and that they would be seen as soon as possible. The appellant left after 19 minutes because he felt too unwell to remain and went to his mothers home. The appellant became distressed at 21:30 and an ambulance was called. He was taken back to Mayday Hospital and a CT scan identified a large extradural haematoma with a marked midline shift. He was transferred to St Georges Hospital and underwent an operation at 01:00. Unfortunately, the appellant suffered permanent brain damage in the form of a severe and very disabling left hemiplegia. The appellant brought proceedings against the respondent alleging a breach of duty by the reception staff concerning the information he was given about the time he would have to wait and the failure to assess the appellant for priority triage. The High Court dismissed the claim. The appellant appealed to the Court of Appeal. The appeal was dismissed by a majority on the grounds that neither the receptionist nor the health trust acting by the receptionist owed any duty to advise about waiting times, the damage was outside the scope of any duty owed, and there was no causal link between any breach of duty and the injury. The appellant appealed to the Supreme Court. The Supreme Court unanimously allows the appeal and remits the case to the Queens Bench Division for assessment of damages. Lord Lloyd Jones gives the sole judgment with which the other Justices agree. Duty of care First, the present case falls squarely within an established category of duty of care: it has long been established that such a duty is owed by those who provide and run a casualty department to persons presenting themselves complaining of illness or injury and before they are treated or received into care in the hospitals wards. The duty is to take reasonable care not to cause physical injury to the patient. In the present case, as soon as the appellant was booked in at reception he entered into a relationship with the respondent of patient and health care provider. The scope of this duty of care extends to a duty to take reasonable care not to provide misleading information which may foreseeably cause physical injury [16]. Secondly, the duty of care is owed by the respondent and it is not appropriate to distinguish, in this regard, between medical and non medical staff. The respondent had charged its non medically qualified staff with the role of being the first point of contact for persons seeking medical assistance and, as a result, with the responsibility for providing accurate information as to its availability [17]. Thirdly, the judgments of the majority in the Court of Appeal elide issues of the existence of a duty of care and negligent breach of duty. [21] Fourthly, observations on the social cost of imposing such a duty of care are misplaced as this is not a new head of liability for NHS health trusts and, in any event, the undesirable consequences of imposing the duty in question were considerably overstated. The Court did, however, acknowledge that the very difficult circumstances under which A&E departments operate may well prove highly influential in many cases when assessing whether there has been a negligent breach of duty [22]. Negligent breach of duty A receptionist in an A&E department is expected to take reasonable care not to provide misleading advice as to the availability of medical assistance. The standard required is that of an averagely competent and well informed person performing the function of a receptionist at a department providing emergency medical care [25]. Moreover, responding to requests for information as to the usual system of operation of the A&E department is well within the area of responsibility of receptionists [26]. The two receptionists on duty were aware of the standard procedure, but the appellant was told to sit down to wait for up to four to five hours. That information was incomplete and misleading. The trial judge made the finding that it was reasonably foreseeable that a person who believes it may be four to five hours before he will be seen may decide to leave. In light of that finding, the provision of such misleading information by a receptionist as to the time within which medical assistance might be available was negligent [27]. Causation The appellants decision to leave was reasonably foreseeable and was made, at least in part, on the basis of the misleading information [29]. The trial judge made further findings of fact that, (1) had the appellant been told he would be seen within 30 minutes he would have waited, been seen by a doctor and admitted, and (2) had the appellant suffered the collapse at 21:30 whilst at the Mayday Hospital, he would have undergone surgery earlier and he would have made a nearly full recovery [30]. Thus, the appellants departure did not break the chain of causation.
During her marriage, Mrs Staveley and her husband set up a company. She had a pension fund with the companys occupational scheme, and this fund was transferred to another pension scheme (the pension scheme) for her when she and her husband divorced. In December 2006, Mrs Staveley died. Shortly before her death, she transferred funds from the pension scheme into a personal pension plan (PPP). The transfer was motivated by Mrs Staveleys desire to ensure that her ex husband did not benefit from the return to the company of any surplus in the fund. She did not take any pension benefits during her life and, in those circumstances, death benefit was payable under the PPP. Mrs Staveley nominated her two sons as beneficiaries of the death benefit, subject to the discretion of the pension scheme administrator. After her death, the death benefit was paid to them. Her Majestys Revenue and Customs (HMRC) determined that inheritance tax was due on the death benefit, on the basis that both the transfer of funds from the pension scheme into the PPP, and Mrs Staveleys omission to draw any benefits from the plan before her death, were lifetime transfers of value within section 3 of the Inheritance Tax Act 1984 (IHTA). The issue in this appeal is whether HMRC were right to take that view. The appellants are the three executors of Mrs Staveleys estate (her two sons and a solicitor, Mr Parry). They argue that inheritance tax is not payable, either on the transfer of funds into the PPP (the transfer) or on Mrs Staveleys omission to draw pension benefits from the PPP (the omission). The issue has divided the courts below. The First tier Tribunal (Tax Chamber) held that inheritance tax was due on the omission but not the transfer. The Upper Tribunal (Tax and Chancery Chamber) disagreed, finding that no inheritance tax was payable on either transaction. The Court of Appeal held that both the transfer and the omission gave rise to a charge to inheritance tax. By a majority, the Supreme Court partially allows the appeal, holding that the omission gave rise to a charge to inheritance tax, but the transfer did not. Lady Black gives the lead judgment, with which Lord Reed and Lord Kitchin agree. Lord Hodge gives a separate judgment, dissenting in part, with which Lord Sales agrees. Inheritance tax is chargeable on the value transferred by a disposition which is a transfer of value under the IHTA. Section 3(3) IHTA extends the meaning of disposition to include deliberate omissions by which the disponors estate is diminished and the value of another persons estate is increased. Section 10(1) IHTA provides that a disposition that was not intended, and was not made in a transaction intended, to confer any gratuitous benefit on any person is not a transfer of value and so does not give rise to a charge to inheritance tax. Section 10(3) IHTA provides that a transaction for section 10(1) purposes includes a series of transactions and any associated operations [6 13]. In the lead judgment, Lady Black sets out the three issues that arise on the appeal: (1) whether section 10 IHTA applies to the transfer as a disposition viewed on its own, i.e. whether the transfer was not intended to confer any gratuitous benefit on any person (Transfer Issue 1); (2) whether section 10 IHTA applies to the transfer looked at in its wider context, i.e. whether the transfer was not made in a transaction intended to confer any gratuitous benefit on any person (Transfer Issue 2); and (3) whether the terms of section 3(3) are satisfied in respect of the omission, i.e. whether, by her omission to draw pension benefits from the PPP, the value of Mrs Staveleys estate was diminished and the values of her sons estates were increased (Omission Issue) [17 20]. Transfer Issue 1 (transfer on its own) Section 10 IHTA applies where the overall effect of the disposition is not intended by the disponor to confer a gratuitous benefit. In the present case, the transfer had not been motivated by any intention to improve the sons position. Mrs Staveleys sole intention in transferring the funds was to eliminate any risk that any part of the funds might be returned to her ex husband. The mere fact that the sons inheritance was intended to be enjoyed in a different legal form after the transfer did not mean that Mrs Staveley intended to confer a gratuitous benefit her sons. When viewed on its own, section 10 IHTA applied to the transfer [65, 66]. Transfer Issue 2 (transfer and omission as associated operations) The parties accepted that, in omitting to draw lifetime benefits under the PPP, part of Mrs Staveleys intention was to benefit her sons. The question for the Supreme Court was whether this intention coloured the transfer with an intention to confer gratuitous benefit which the transfer would not have on its own. The majority of the Supreme Court did not accept, as the Court of Appeal had found, that the omission and the transfer together formed part of a scheme (a series of transactions and any associated operations) intended to confer a gratuitous benefit. The present case could be distinguished from Inland Revenue Comrs v Macpherson [1989] AC 159, where the House of Lords had found that a series of operations formed part of a such a scheme and so gave rise to tax. In Macpherson, the operations under consideration had been clearly linked by a common intention. In contrast, in the present case, Mrs Staveleys essential scheme had been her omission to draw a lifetime pension in order to benefit her sons by leaving them her death benefits. That could have been achieved without the transfer. Accordingly, the transfer and the omission were not linked by a common intention and did not form part of a single scheme intended confer a gratuitous benefit on any person. Section 10 IHTA therefore applied to the transfer, even when viewed in its wider context. It followed that the transfer did not give rise to a charge to inheritance tax [67, 88]. The Omission Issue The limited discretion of the pension scheme administrator did not break the chain connecting the omission and the payment of the death benefit. The omission was the operative cause of the increase in the sons estates and therefore gave rise to a charge to inheritance tax under section 3(3) IHTA [94]. Lord Hodge would have dismissed the appeal in relation to the transfer and the omission. He agrees with Lady Black on Transfer Issue 1 and the Omission Issue but reaches a different conclusion on Transfer Issue 2. The transfer, when taken with the omission and the nomination of the sons as beneficiaries of the death benefit, formed a contributory part of a scheme to confer a gratuitous benefit. Mrs Staveleys intention to benefit her sons pre dated the transfer and was reflected in her nomination in the PPP. Therefore, he concludes, section 10 does not apply and tax ought to be payable [104, 109].
These appeals concern the legality of the Secretary of States Employment, Skills and Enterprise Scheme (ESES), which was designed to assist claimants of job seekers allowance (JSA) to obtain employment or self employment. The Jobseekers Act 1995 (the 1995 Act) provides for JSA to be paid to certain categories of unemployed persons. Section 17A of the 1995 Act provided that the Secretary of State could make regulations requiring JSA claimants in prescribed circumstances to participate in work or work related schemes of a prescribed description for a prescribed period. By section 35, prescribed means specified in or determined in accordance with regulations. Purportedly acting under section 17A of the 1995 Act, the Secretary of State made the Jobseekers Allowance (Employment, Skills and Enterprise Scheme) Regulations 2011 (the 2011 Regulations:). These set up the ESES, defined by Regulation 2 as a scheme under section 17A to assist JSA claimants to obtain employment, which could include work related activity. By Regulation 3, the Secretary of State could select a JSA claimant for participation in the Scheme. Regulations 4 provided that such a claimant was required to participate once the Secretary of State had provided a notice in writing specifying (among other things) details of what participation involved, and Regulations 4 8 provided that failing to participate without good cause would lead to benefits sanctions. A number of work for your benefit programmes were created under ESES including the sector based work academy scheme (SBWA scheme), a short term scheme aimed at clearly employable individuals, and the Community Action Programme (CAP) aimed at the very long term unemployed. The first Respondent unwillingly participated for four weeks in the SBWA scheme having been informed, wrongly, that her participation was mandatory. She received no written notice. The second Respondent was selected to participate in the CAP. He was informed orally that he would be required to work for 30 hours/week for 26 weeks or until he found employment. He repeatedly refused to participate, and was subject to benefits sanctions with the effect that he received no JSA for 6 months. The Respondents brought judicial review claims. They argued that (i) the 2011 Regulations are unlawful, since they did not fulfil the requirements of section 17A of the 1995 Act in prescribing the programmes, the circumstances by which individuals are selected, or the period of participation (lawfulness), (ii) the Respondents did not receive the information required by Regulation 4 of the 2011 Regulations (notification), (iii) the Government was required to have a published policy setting out the details of the relevant schemes (publication), and (iv) that the first Respondent had been subject to forced or compulsory labour contrary to Article 4 ECHR (forced labour). The High Court found for the Respondents on ground (ii) only: the Secretary of State had accepted that the first Respondents notice did not satisfy Regulation 4, and the Court held that the second Respondents notice also failed to comply. The Court of Appeal upheld the High Courts decision on ground (ii), but also allowed the appeal on ground (i), and thereby quashed the 2011 Regulations. The Secretary of State appeals to the Supreme Court against the Court of Appeals decision on grounds (i) and (ii). The Respondents cross appeal against the Court of Appeals decision on grounds (iii) and (iv). Following the Court of Appeals decision, the Government passed the Jobseekers Allowance (Schemes for Assisting Persons to Obtain Employment) Regulations 2013 (the 2013 Regulations) and the Jobseekers (Back to Work Schemes) Act 2013 (the 2013 Act). The effect of these was retrospectively to validate the 2011 Regulations and to set out fuller details of seven schemes, including the SBWA scheme and the successors of the CAP, pursuant to section 17A of the 1995 Act. Lord Neuberger and Lord Toulson give the unanimous judgment of the Court. On ground (i) lawfulness, the Supreme Court dismisses the Secretary of States appeal, holding that the 2011 Regulations are invalid, since they did not contain a sufficiently detailed prescribed description of the SBWA or CAP schemes. On ground (ii) notification, the Court dismisses the Secretary of States appeal, holding that the notice given to the second Respondent was insufficiently detailed. On ground (iii) publication, the Supreme Court holds that the Secretary of State had failed to provide sufficient information about the schemes to the Respondents. On ground, (iv) forced labour, the Court dismisses the Respondents cross appeal: the Regulations do not constitute forced or compulsory labour. Given the existence of the 2013 Act and 2013 Regulations, however, the appropriate form of the order would require submissions from counsel. (i) Lawfulness: The SBWA and CAP are schemes falling within the 2011 Regulations. However, Regulation 2 contains no prescribed description of the ESES, SBWA scheme or CAP [45]. Even taking into account the need for flexibility in the detail of schemes, where a statute allows the making of regulations with a significant impact on peoples lives, the need for legal certainty is of crucial importance [46 47]. To be meaningful, the prescribed description must add something to what is said in the 1995 Act, and the description of ESES in the 2011 Regulations added nothing to the words of section 17A [48 50]. Therefore the 2011 Regulations were unlawful. However, the prescribed circumstances were sufficiently set out in Regulations 3 and 4 together, given the obvious need for flexibility [51]. Likewise, it was legitimate for the prescribed period to be an open ended one [52]. (ii) Notification: The notice served on the second Respondent simply informed him that he had to perform any activities requested by the private company operating the CAP, without any indication of the nature of the likely tasks, hours or places of work. This was insufficient to satisfy Regulation 4(2)(c), which required that the notice give the second Respondent details of what [the second Respondent] is required to do by way of participation in the Scheme [54 55]. However, the letter was sufficiently detailed with regard to the consequences of failure to participate: while there might have been imperfections, the second Respondent was not significantly prejudiced or misled [56 57]. (iii) Publication: The Regulations invoked a statutory power which involved a requirement to work on pain of loss of benefits. Therefore fairness required that the claimants should have sufficient information about the scheme to be able to make freely informed representations before a decision was made [64 66], which the Secretary of State failed to do [67 73, 76]. (iv) Forced labour: Article 4 ECHR requires that no one shall be required to perform forced or compulsory labour. However, this does not include work forming part of normal civic obligations. The latter provision delimits the ambit of the former [78, 81 82]. Therefore it was wrong to say that any work done under threat of penalty constituted forced labour unless it was required by lawfully imposed civic obligations [79 80]. JSA is a benefit for work seekers, and the 2011 Regulations impose a condition on that benefit directly linked to its purpose. This comes nowhere close to the type of exploitative conduct at which article 4 is aimed [83, 90]. The fact that, as a matter of domestic law, the first Respondents notice was unlawful made no difference [91].
The Information Commissioner ordered the disclosure of information held by Ofcom concerning the precise location of mobile phone masts. On appeal, the Information Tribunal found that the public interest in public security, and in the protection of intellectual property rights, were both engaged but that under each separate exception the public interest in disclosure outweighed the interest alleged by Ofcom. It dismissed the argument of Ofcom that under the Environmental Information Regulations 2004 the Tribunal should conduct a third balancing test weighing all the interests in favour of disclosure against all the public interests in refusing disclosure. The High Court upheld the Information Tribunal. On appeal, the Court of Appeal overturned the Tribunal. It held that the Regulations must be construed in the light of European Directive 2003/4/EC, which they implement. The language of both documents supported an aggregate weighing exercise to assess the overall public interest. The Supreme Court unanimously holds that the appeal raises an issue of general principle and that the answer is not obvious. Different members of the Court hold different views on the correct construction of Environmental Information Regulations 2004, and Directive 2003/4/EC which they implement. Consequently, the Supreme Court is under a duty to refer the question in the appeal to the European Court of Justice (paras [3], [10], [14]). The question referred to the European Court under Article 267 of the Treaty on the Functioning of the European Union is: Under Council Directive 2003/4/EC, where a public authority holds environmental information, disclosure of which would have some adverse effects on the separate interests served by more than one exception (in casu, the interests of public security served by article 4(2(b) and those of intellectual property rights served by article 4(2)(e)), but it would not do so, in the case of either exception viewed separately, to any extent sufficient to outweigh the public interest in disclosure, does the Directive require a further exercise involving the cumulation of the separate interests served by the two exceptions and their weighing together against the public interest in disclosure? (para [15]) A majority of the Court would have upheld the judgment of the Court of Appeal. The majority consider that there are certain linguistic clues in the Directive which favour an aggregate weighing exercise which considers the overall public interest. The diversity of reasons is a positive reason to accumulate them, and certain heads already involve more than one public interest (paras [10], [12]). The minority of the Court also finds linguistic clues in the Directive to suggest that no cumulation of factors is possible given the disparate public interests involved which considered together would produce incongruities and be impractical (para [13]).
The respondent, Mr Perry, is a retired miner. By the time he stopped working, he was suffering from a condition known as Vibration White Finger (VWF). Common symptoms include a reduction in grip strength and manual dexterity, often leading to an inability to carry out routine domestic tasks unaided. In the late 1990s, a group of test cases established that the National Coal Board (later British Coal) had been negligent in failing to take reasonable steps to limit the exposure of its miners to VWF from the excessive use of vibratory tools. In 1999, the Department for Trade and Industry (DTI) set up a scheme (the Scheme) to provide tariff based compensation (i.e. based on the severity of the injury) to miners suffering from VWF following exposure to excessive vibration. The Scheme was administered under a Claims Handling Arrangement dated 22 January 1999 made between the DTI and solicitors firms representing miners. The Scheme contemplated the making of two main types of compensatory award to such miners, which broadly reflected general and special damages for personal injuries. Pursuant to a Services Agreement dated 9 May 2000, special damages could include a Services Award to qualifying miners. This depended on establishing what became known as the factual matrix. In summary: (1) prior ability to undertake one or more of six defined routine domestic tasks (the six tasks) without assistance; (2) current inability to undertake those tasks without assistance because of VWF; and (3) current receipt of the necessary assistance with those tasks from others. The six tasks were gardening, window cleaning, DIY, decorating, car washing, and car maintenance. Qualification for a general damages award required affected miners to undertake a medical interview and examination designed to assess the severity of their VWF. Sufferers at certain high levels of severity also became entitled to a rebuttable presumption that they qualified for a Services Award. The Scheme provided for relatively light touch checks of Services Award claims. Compensation was payable to qualifying claimants according to an index linked tariff. Proportionate deductions could be made if a further medical examination showed that there were other contributing medical conditions. Mr Perry engaged the appellant law firm, Raleys, to pursue a VWF claim in October 1996. His claim ultimately fell within the Scheme. In October 1997, he was given medical ratings (stagings) sufficient both for him to obtain general damages and for a Services Award to be presumed. However, Mr Perry settled his claim in November 1999 for the payment of general damages only (11,600) and made no claim for a Services Award within the specified time. He made a professional negligence claim against Raleys in February 2009, claiming that the firms negligent failure to give him competent legal advice deprived him of the chance to claim a Services Award. His estimated loss was 17,300.17 plus interest. At trial in the County Court, Raleys ultimately admitted breach of duty, but denied causation of loss. It also alleged that his claim was time barred. The trial judge, Judge Saffman, rejected the limitation defence, but held that Mr Perry had not proved that Raleys breach of duty had caused him any loss. This conclusion was based on the finding that Mr Perrys VWF had not caused him any significant disability in performing any of the six tasks without assistance, such that he could not have been able to make an honest claim for a Services Award. The judge dismissed the claim, but nonetheless proceeded to make findings on the assessment of damages. The Court of Appeal reversed the finding on causation and concluded that the alternative findings on quantum meant a re trial was unnecessary. It granted Mr Perry loss of chance damages of 14,556.15 plus interest. Raleys appealed to the Supreme Court, seeking restoration of Judge Saffmans order. The Supreme Court allows the appeal and restores the order of the County Court judge. Lord Briggs gives the sole judgment, with which all members of the Court agree. Loss of chance damages have been developed by the courts to deal with the difficulties arising from the assessment of counter factual and future events [16]. In both types of situation, the courts at times depart from the ordinary burden on a claimant to prove the facts required for a successful claim on the balance of probabilities (i.e. more likely than not) standard [17 18]. However, this does not mean that the basic requirement that a negligence claim requires proof that loss has been caused by the breach of duty is abandoned [19]. The correct approach, following Allied Maples Group Ltd v Simmons & Simmons [1995] 1 WLR 1602 (CA), is to require a claimant to prove what he or she would have done on the balance of probabilities, while what others would have done (if relevant) depends on a loss of chance evaluation [20 21]. These principles apply equally to negligence claims based on loss of the opportunity to achieve a better outcome in a negotiated transaction and ones, as in this case, based on loss of the chance to bring a legal claim [22]. It is not unfair to subject medical and oral evidence as to facts within the claimants own knowledge to forensic analysis on the balance of probabilities standard [30]. The case law only establishes that, where the question for the court is one which turns on the assessment of a lost chance, it is generally inappropriate to conduct a trial within a trial [31]. It does not establish a principle that it is always wrong to try an issue relevant to causation in a professional negligence case, merely because that issue would have fallen for determination in the underlying claim (lost due to alleged negligence) [35 37]. Whether an issue should be tried to the usual standard depends on whether it concerns the claimants conduct (where it should be) or third party conduct (only requiring a real and substantial chance) [37]. Applying this approach, Mr Perry needed to prove that, properly advised by Raleys, he would have made a claim to a Services Award under the Scheme within time [25]. Further, the judge was correct to impose the additional requirement of the claim having to be an honest claim [25]. A concession in the courts below had been rightly made as to the honest claim requirement, because: (1) a claimant giving an honest description of his or her condition to a solicitor would not be advised to bring a claim if the facts were insufficient; (2) a court may fairly presume that the client would only make honest claims; and (3) it is not the proper role of the courts to reward dishonest claimants [25 27]. On the facts, Mr Perry had to believe the following to bring an honest claim: (1) before developing VWF, he had carried out all or some of the six tasks without assistance; (2) after developing VWF, he needed assistance in carrying out all or some of those tasks; and (3) the need for assistance was due to complications from VWF [28]. Question (3) might require expert medical opinion, but all the other necessary elements fell within his own knowledge [29]. Such facts do not raise issues of counter factuality or futurity which engage loss of chance principles [30]. Accordingly, Judge Saffman had made no legal error in conducting a trial of the issue whether Mr Perry would (or could) have brought an honest claim for a Services Award [41]. Further, the judge did not (wrongly) apply a second causation hurdle requiring Mr Perry to prove that his claim would have been successful (not merely honest) [42 48]. In addition, the Court of Appeal wrongly interfered with the judges factual determination the very stringent test for appellate court interference was not met in this case [49 66].
The Respondent (B) has been in the UK since 1993. He was originally detained under section 21 of the Anti Terrorism, Crime and Security Act 2001 and was subsequently subject to a control order under the Prevention of Terrorism Act 2005. On 11 August 2005, he was notified of the Secretary of States intention to make a deportation order against him on national security grounds. He was detained under Schedule 3 of the Immigration Act 1971 (the 1971 Act) pending deportation. He appealed, using a false identity, to the Special Immigration Appeals Commission (SIAC) against his deportation. The UK Government sought assurances from the Algerian authorities that, if returned to Algeria, B would not be subject to treatment incompatible with Article 3 of the European Convention on Human Rights (ECHR). On 10 July 2006, the Algerian authorities confirmed that the details of his identity given by B were those of an individual present in Algeria. On 19 July 2007, SIAC ordered B to provide details of his true identity. On 30 July 2008, SIAC held that the Secretary of States case against B on the risk to national security had been made out. On 26 November 2010, SIAC held that B had disobeyed its earlier order of 19 July 2007 and imposed a prison sentence on B of four months. Following his eventual release from prison, bail conditions were imposed on B. On 13 February 2014, SIAC held that there was no reasonable prospect of removing B to Algeria and the ordinary legal basis for justified detention under the Immigration Acts therefore fell away. Thereafter, the Secretary of State did not authorise Bs further detention and his bail conditions were relaxed. Bs appeal against the notice of decision to deport him was struck out by SIAC in light of his continuing contempt of court. SIAC also rejected Bs submission that, following SIACs findings of 13 February 2014, it no longer had jurisdiction to grant bail to B or to impose bail conditions. This decision was upheld by the High Court. B appealed to the Court of Appeal, which allowed his appeal on the ground that SIAC had no jurisdiction to impose bail conditions on B if his detention would be unlawful. On 12 December 2016, SIAC allowed Bs substantive deportation appeal. As a result, Bs bail fell away and it is common ground that the immigration power is now unavailable. The Supreme Court granted the Secretary of State permission to appeal against the decision of the Court of Appeal on the issue of SIACs bail jurisdiction. The Supreme Court unanimously dismisses the appeal. Lord Lloyd Jones gives the lead judgment with which the other Justices agree. The so called Hardial Singh principles concerning the operation of the detention power contained in paragraph 2 of Schedule 3 to the 1971 Act form an important part of the background to this appeal. These principles are that (i) the Secretary of State must intend to deport the person and can only use the detention power for that purpose; (ii) the deportee may only be detained for a reasonable period; (iii) if it becomes apparent that the Secretary of State will not be able to effect deportation within a reasonable period then he should not seek to exercise the power of detention; and (iv) the Secretary of State should act with reasonable diligence and expedition to effect removal [24 25]. At the heart of the case is a dispute about what the correct approach to the availability of immigration bail is when the Hardial Singh limit on actual detention is reached. The Secretary of State suggested that a purposive interpretation of the legislation should apply so that bail is available regardless of whether the individual is lawfully detained or would hypothetically be lawfully detained [28]. The Court saw no basis for such an approach. It is a fundamental principle of the common law that Parliament is presumed not to intend to interfere with the liberty of the subject without making such an intention clear. This was a situation where the principle of legality was in play. Accordingly, the Court was required to interpret the statutory provisions strictly and restrictively [29]. It was common ground that being detained was a condition precedent to the exercise of the power to grant bail conferred by paragraphs 22 and 29 of Schedule 2 to the 1971 Act. Applying the strict approach to interpretation, the Court found that the reference to detained means lawful detention [30 31]. Furthermore, detained does not only refer to the state of affairs which must exist at the time when the power is first exercised. Unless there is a continuing power to detain, the system of bail would encounter substantial difficulties in operation [32]. Where it ceases to be lawful to detain a person pending deportation there is no longer a power of detention under paragraph 16 of Schedule 2, and there is therefore no longer a power to grant bail under paragraphs 22 or 29 [33]. The Secretary of State submitted that as both bail and temporary admission or temporary release are ameliorating possibilities of alternatives to detention, it is sensible for both powers to persist for some duration beyond the point at which actual detention can no longer continue. Temporary admission or release is covered by paragraph 21 of Schedule 2. However, unlike paragraphs 22 and 29, it refers to a person liable to be detained and not detention which is a material difference. Accordingly, the comparison does not assist the Secretary of State [34 39]. The Court did not agree with the Secretary of States submission that the interpretation of paragraphs 22 and 29 favoured by the Court of Appeal would lead to impracticability in their application. In any event, if administrative inconvenience is a consequence the remedy lies with Parliament [40 45]. The Court found considerable modern authority which supported the Court of Appeals statement of principle that the power to grant bail presupposes the existence of and the ability to exercise the power to detain lawfully. [47 51]. However, this is not necessarily a principle of universal application. While the clearest possible words would be required to achieve a contrary result, Parliament could do so [53]. In the circumstances it was not necessary to address the arguments under Article 5 ECHR which added nothing to the resolution of the issues on appeal [56].
This appeal concerns the defence of illegality. The Supreme Court is asked to decide whether a firm of solicitors, Stoffel & Co, can escape liability to Ms Grondona for their negligent failure to register documents effecting a transfer of property because the transfer formed part of an illegal mortgage fraud. Ms Grondona had a business relationship with Mr Mitchell. In or about July 2002, Mr Mitchell purchased a 125 year lease of the rear ground floor flat at 73b Beulah Rd, Thornton Heath (the property). Shortly afterwards, he borrowed 45,000 from BM Samuels Finance Group Plc. The loan was secured by a legal charge over the property (the BM Samuels charge). In October 2002, Ms Grondona bought the property from Mr Mitchell, with the assistance of a mortgage advance of 76,475 from Birmingham Midshires. The plan was that the mortgage advance would be secured by a charge over the property, which Ms Grondona entered into on 31 October 2002 (the Birmingham Midshires charge). Ms Grondona procured the mortgage advance by fraud. According to the trial judge, the purpose of the fraud was to raise capital for Mr Mitchell from a high street lender, which he would not otherwise have been able to obtain. Ms Grondona and Mr Mitchell had previously entered into an agreement which confirmed that Mr Mitchell would be responsible for the mortgage payments. Stoffel & Co solicitors acted for Ms Grondona, Mr Mitchell and Birmingham Midshires in connection with the transaction. Stoffel & Co negligently failed to register at the Land Registry the form TR1 transferring the property from Mr Mitchell to Ms Grondona, the form DS1 releasing the BM Samuels charge, and the Birmingham Midshires charge. This meant that Mr Mitchell remained the registered owner of the property, which continued to be subject to the BM Samuels charge. Mr Mitchell subsequently received further advances from BM Samuels on the basis of that charge. In 2006, Ms Grondona defaulted on payments under the Birmingham Midshires charge. Birmingham Midshires brought proceedings against her. Ms Grondona, in turn, sought damages from Stoffel & Co. Stoffel & Co admitted that their failure to register the forms with the Land Registry constituted negligence or a breach of retainer. However, they argued that they were entitled to rely on the illegality defence, because Ms Grondona had only instructed them to further the illegal mortgage fraud. The trial judge held that Ms Grondonas claim was not barred by the illegality defence. The Court of Appeal dismissed Stoffel & Cos appeal. Stoffel & Co appealed to the Supreme Court. The Supreme Court unanimously dismisses Stoffel & Cos appeal, and holds that Ms Grondonas claim is not barred by the illegality defence. Lord Lloyd Jones gives the judgment, with which all members of the Court agree. The Supreme Court decision in Patel v Mirza [2016] UKSC 42 set out a new policy based approach to the illegality defence at common law. In that case, the majority held that, when a claim is tainted by illegality, the court should ask itself whether enforcing the claim would lead to inconsistency that is damaging to the integrity of the legal system. In making this assessment, the court should consider: (a) the underlying purpose of the illegality in question, and whether that purpose would be enhanced by denying the claim; (b) any other relevant public policy on which denying the claim may have an impact; and (c) whether denying the claim would be a proportionate response to the illegality [22 23]. The application of this trio of considerations should not be a mechanistic process. Accordingly, the court will identify the policy considerations at stages (a) and (b) of the trio at a relatively general level. The courts task is to establish whether enforcing a claim that is tainted with illegality would be inconsistent with the policies to which the law gives effect or, where the policies compete, to decide where the balance lies. The court is not required to evaluate the underlying policies themselves. In contrast, when considering proportionality at stage (c), it is likely that the court will need to look closely at the case before it. However, it is not necessary for the court to consider proportionality in every case. If, after it has examined the policy considerations at stages (a) and (b), the court determines that the claim should not be barred by the illegality defence, there will be no need for it to go on to consider proportionality. This is because the claim will be allowed, so there is no risk of disproportionate harm to the claimant by refusing relief to which he or she would otherwise be entitled [26]. In considering stage (a), Lord Lloyd Jones observes that Ms Grondona was engaged in mortgage fraud, which is a serious criminal offence. However, denying her claim would not enhance the underlying purpose of the prohibition on mortgage fraud. Fraudsters are unlikely to be deterred by the risk that they will be left without a civil remedy if their solicitors prove to be negligent [29]. On the other hand, allowing Ms Grondonas claim would enhance the protection that the law provides to mortgagees and other members of the public, which is a further underlying purpose of the prohibition on mortgage fraud. By the time of Stoffel & Cos negligence, it was in the interests of both Ms Grondona and Birmingham Midshires that the transfer should be registered [30 31]. Turning to stage (b), denying Ms Grondonas claim would run counter to a number of important public policies. In particular, it would be inconsistent with the policy that solicitors should perform their duties to their clients diligently and without negligence, as well as with the policy that the victims of solicitors negligence should be compensated for the loss they have suffered [32]. Denying the claim would also result in an incoherent contradiction in the law, because the law accepts that an equitable interest in the property passed to Ms Grondona, even though she was engaged in mortgage fraud [33 34]. The balancing of the policy considerations at stages (a) and (b) indicate that Ms Grondonas claim should not be barred by the illegality defence. There is, consequently, no need to consider proportionality, but Lord Lloyd Jones does so nevertheless [35]. He concludes that it would not be proportionate to deny Ms Grondonas claim because it is conceptually entirely separate from the mortgage fraud [43]. Enforcing the claim would not allow Ms Grondona to profit from her wrongdoing [44 45]. In any case, following Patel v Mirza, the courts focus should be on the need to avoid inconsistency that is damaging to the integrity of the legal system. The question of whether the claimant will profit from the illegality remains a relevant consideration, but it is no longer the true focus of the courts inquiry [46].
In Cadder v HM Advocate [2010] UKSC 43, the Supreme Court held, having regard to the decision of the European Court of Human Rights in Salduz v Turkey (2008) 49 EHRR 421, that the Crowns reliance on admissions made by an accused who had no access to a lawyer while he was being questioned as a detainee at a police station was a violation of his rights under Article 6(3)(c), read with Article 6(1) of the European Convention on Human Rights. The issues in these cases are whether the right of access to a lawyer prior to police questioning, which was established by Salduz, applies only to questioning which takes place when the person has been taken into police custody; and, if the rule applies at some earlier stage, from what moment does it apply. The accused in the first case, John Paul Ambrose, was prosecuted on a charge of contravening section 5(1)(b) of the Road Traffic Act 1988 as being in charge of a motor vehicle while over the alcohol limit. He was questioned by two police officers by the roadside, who cautioned him but did not give him any specification about the offence he was suspected of having committed. In response to their questions, he confirmed that he was in possession of the car keys, and that he might be intending to drive the car. Breath tests indicated that he was substantially over the prescribed limit. In his trial the Crown led evidence of the questions and answers at the roadside. In M, the accused was charged with assault to severe injury, permanent disfigurement and permanent impairment. A few days after the incident, the police visited him at his home, cautioned him, and asked him a number of questions, in response to which he confirmed his attendance at the locus on the night in question and his involvement in the fight. He was detained the following day, and questioned further while he was in custody. At trial, he objected to the Crowns reliance upon the admissions he had made in his home, on the basis that he had not had access to legal advice prior to interview. In G, the accused was indicted with offences including the possession of controlled drugs under the Misuse of Drugs Act 1971 and possession of prohibited firearms and ammunition under the Firearms Act 1968. The police had obtained a warrant to search the accuseds flat. They forced entry and found him there. He struggled, and was handcuffed and cautioned. He admitted to having drugs in his pocket, and responded to a number of questions about items found in the flat. He was subsequently arrested and taken to a police station where he answered further questions. He objected to the Crowns reliance at trial on the statements he made during the course of the search of the flat. In each of the three cases, the Appeal Court of the High Court of Justiciary referred to this Court the question whether the act of the Lord Advocate in leading and relying on the evidence in question would be incompatible with the appellants rights under Article 6(1) and (3)(c) of the European Convention on Human Rights. The Supreme Court, by a majority of 4 to 1, finds that, in the cases of Ambrose and M, the act of the Lord Advocate in leading and relying at the trial on the evidence that was obtained from them in response to police questioning without having had access to legal advice was not incompatible with the Article 6(1) and (3)(c) right; and in the case of G that it was incompatible. In Ambrose and M, the question whether, taking all the circumstances into account, it would be fair to admit this evidence, is left open for decision by the Appeal Court and Sheriff Court respectively. Lord Hope gives the leading judgment. Lord Kerr gives a separate dissenting judgment finding in all three cases that the evidence would be inadmissible. In each of these three cases, the circumstances differ from those before the Supreme Court in Cadder and before the Grand Chamber in Salduz, in that the evidence in question was obtained through police questioning before the individuals were detained at a police station. The Supreme Court notes, firstly, that the jurisdiction of this court is limited to a consideration of the devolution issue which is raised by each of these references, and does not extend to ruling on how the circumstances referred to in each case would fall to be dealt with under domestic law. Secondly, it notes that a decision by this court that there is a rule that a person who is suspected of an offence but is not yet in custody has a right of access to a lawyer before being questioned by the police would have far reaching consequences for the investigation of crime by the authorities. Therefore, if Strasbourg has not yet spoken clearly on this issue, the court would be wise to wait until it has done so [14 15]. The duty of the domestic court in interpreting the Convention is to keep pace with the Strasbourg case law as it evolves over time. There is no obligation upon domestic courts to do more than that (R (Ullah) v Special Adjudicator [2004] UKHL 26, para 20 per Lord Bingham of Cornhill) [17]. The courts task in this case is to identify where the Strasbourg court stands on this issue. It is not for this court to expand the scope of the Convention right further than the jurisprudence of the Strasbourg court justifies [20]. In domestic law, where an individual has not yet been detained under section 14 of the Criminal Procedure (Scotland) Act 1995, the test for the admissibility of answers given to questions put by police is whether or not there was unfairness on the part of the police. The fact that the person did not have access to legal advice when being questioned is a circumstance to which the court may have regard in applying the test of fairness, but it carries no more weight than that. There is no rule in domestic law that provides that police questioning of a person without access to legal advice who is suspected of an offence but is not in police custody must always be regarded as unfair. The question is whether a rule to that effect is to be clearly found in the jurisprudence of the Strasbourg court [25]. The Grand Chamber in Salduz had in mind the need to protect an accused against abusive coercion while in custody. The judgment appears to have been concerned only with establishing a rule that there was a right of access to a lawyer where the person being interrogated was in police custody [33]. That assessment is supported by subsequent Strasbourg case law, in particular Zaichenko v Russia (Application no.39660/02), the only case to date in which the complaint was of lack of legal assistance during police questioning when the applicant was not in custody [46]. If the Salduz judgment were to apply to statements made by a person in response to police questioning before being taken into custody, the court would have had to have said so expressly. It did not do so [35]. The privilege against self incrimination is not an absolute right (Murray v United Kingdom (1996) 22 EHRR 29, para 47). It is primarily concerned with respecting the will of the person to remain silent (Saunders v United Kingdom (1996) 23 EHRR 313, para 68), and a person is free to confess if he is willing to do so. Police custody or its equivalent creates a need for protection of the accused against abusive coercion. The same is not the case for questioning at the locus or in a persons home [54]. In principle, the line as to when access to legal advice must be provided before the person is questioned should be drawn as from the moment that he has been taken into police custody, or his freedom of action has been significantly curtailed [55]. The correct starting point when considering whether the persons Convention rights have been breached is to identify the moment at which he is charged for the purposes of Article 6(1); that is whether his situation is substantially affected (Deweer v Belgium (1980) 2 EHRR 439, para 46; Eckle v Germany (1982) 5 EHRR 1, para 73). That will be the case as soon as the suspicion against him is being seriously investigated and the prosecution case compiled [62]. The fact that a person who has become a suspect and is not in custody is questioned without access to legal advice will be a relevant factor in the assessment whether the accused was deprived of a fair hearing, but it will be no more than that. In Ambrose and M, the question is whether the act of the Lord Advocate in leading and relying on evidence obtained in response to police questioning, conducted under common law at the roadside or at the accuseds home, without the accused having had access to legal advice, was incompatible with Article 6(1) and (3)(c). This is answered in the negative. Ambrose was charged for the purposes of Article 6 when he was cautioned. Suspicion that he was committing an offence fell on him as soon as he told the police that the keys were in his pocket [67]. M was charged when he was cautioned by the police officer at his home [69]. But it would be to go further than Strasbourg has gone to hold that the appellants are entitled to a finding that this evidence is inadmissible because, as a rule, access to a lawyer should have been provided to him when he was being subjected to questioning at the roadside [68 & 70]. The question whether, taking all the circumstances into account, it would be fair to admit this evidence, is left open for the Appeal Court and Sheriff Court respectively. In G, the question whether it is incompatible with his Convention rights for the Lord Advocate to lead evidence of his statements made during the course of the search is answered in the affirmative. He was charged for the purposes of Article 6 by the time the police began their search. The difference with this case was that there was a significant curtailment of Gs freedom of action. He was detained and had been handcuffed, and was, in effect, in police custody from that moment onwards. The circumstances were, therefore, sufficiently coercive for the incriminating answers that he gave to the questions that were put to him without access to legal advice to be inadmissible [71]. The same result need not, however, follow in every case where questions are put during a police search to a person who is to be take to have been charged for the purposes of Article 6; that, again, would be going further than Strasbourg has gone [72]. Lord Kerr would have found the evidence in question to be inadmissible in all three cases. It is not open to courts of this country to refrain from recognising a claim to a Convention right simply because Strasbourg has not spoken clearly on the matter [128]. In practice, it is inevitable that many claims to Convention rights will have to be determined by the UK courts without the benefit of unequivocal jurisprudence from Strasbourg. It is the duty of every domestic court to resolve the question of whether a claim to a Convention right is viable or not, even where the jurisprudence of the Strasbourg court does not disclose a clear current view [129]. As regards the right of access to a lawyer, the selection of the moment of being taken into custody as the first occasion on which legal representation becomes necessary is both arbitrary and illogical. The judgment in Salduz indicates that the need to have a lawyer is not to be determined on a geographical or temporal basis but according to the significance of what is taking place when the admissions in question are made [136]. The essential question is: when the questioning is taking place, is the suspect in a position where the advice of a lawyer is essential if a fair trial is to occur. If he is liable to incriminate himself at that time, a lawyers presence is required [145]. The judgment in Zaichenko is not clear, but does not indicate that formal arrest and interrogation in custody are essential prerequisites to the invocation of the right to legal assistance [158].
Mrs Hewage was born in Sri Lanka, and has been a British citizen since 1998. On 1 December 1993 she commenced employment with Grampian Health Board (the Board) at Aberdeen Royal Infirmary as a consultant orthodontist. In 1996 she became Head of Service for the Orthodontic Department. She resigned from that position on 30 November 2003. On 24 December 2004 she resigned from her employment with the Board with effect from 31 March 2005. In September 2005 she commenced proceedings against the Board in which she claimed under section 94(1) of the Employment Rights Act 1996 that she had been unfairly dismissed from that employment. She also claimed under the Sex Discrimination Act 1975 and the Race Relations Act 1976 that she had been discriminated against on the grounds of her sex and race. At a hearing before an employment tribunal, the Board conceded that Mrs Hewage had been constructively and unfairly dismissed. The tribunal held on 4 December 2007 that she had been unlawfully discriminated against on a number of grounds of both sex and race. On 15 April 2009 the Employment Appeal Tribunal (the EAT) upheld an appeal by the Board and dismissed Mrs Hewages claims of discrimination. She appealed to the Inner House of the Court of Session, and on 14 January 2011 the Second Division allowed her appeal and quashed the decision of the EAT. It remitted the case to the employment tribunal to decide whether, if it had had regard to the only issues which the court considered to be relevant to the claims of discrimination, it would have come to the same or a different conclusion. The Board appeals against that decision. The complaints have their source in allegations by Mrs Hewage that she was bullied and harassed by employees of the Board. In her position as Head of Service she attended monthly management meetings of the department. She claimed that at one of the meetings, two colleagues (Mrs Helen Strachan, the service manager for surgical specialities, and Mrs Edith Munro, the clinical nurse manager) were verbally abusive, hostile and aggressive towards her. An occupational health doctor wrote on her behalf to the Boards Chief Executive, and Mrs Hewage met with the Chief Executive to discuss the matter. But she was not satisfied by his response and resigned from her position as Head of the Department. The conduct of Mrs Strachan had previously been brought to the attention of the Boards senior management by Professor John Forrester, Head of Service for the Department of Opthalmology, who had resigned as a result of her behaviour towards him. In response to his resignation, the department was reorganised, Mrs Strachan was removed from the position of service manager, and Professor Forrester was re appointed. When Mrs Hewage resigned as Head of Service in November 2009, Mr Colin Larmour, a consultant orthodontist, took over from her, initially on a temporary basis. On his appointment to a permanent role, the Boards General Manager and Associate Medical Director assured him of their support, especially in relation to Mrs Strachan. In meetings regarding the appointment of dental nurses, Mrs Edith Munro and Sister Moira Munro willingly agreed to a proposal by Mr Larmour which they had fiercely resisted when it was proposed by Mrs Hewage. Mrs Hewage complained that the formal investigation undertaken by a panel under the Boards Dignity at Work Policy resulted in a report that was full of inaccuracies and omissions, and did not reach any conclusions or make any recommendations. She repeatedly explained her concerns about the inadequacy of the report to the Boards Medical Director, but he took no further action. So she raised discrimination proceedings against the Board on the basis that other white male consultants, such as Professor Forrester and Mr Larmour, were not subject to the same bullying and harassing treatment that she suffered and that she would not have been treated in the way in which she was were it not for her sex and race. The Supreme Court unanimously dismisses the Boards appeal, and affirms that part of the Second Divisions interlocutor in which it allowed the appeal to the Inner House and quashed the decision of the Employment Appeal Tribunal. The judgment is given by Lord Hope, with whom the other Justices agree. The employment tribunal was entitled to hold that Professor Forrester and Mr Larmour were appropriate comparators, despite the fact that the situations which were being compared in each case were not precisely the same. The question whether the situations were comparable is a question of fact and degree, and there was a good deal of evidence to indicate that they were indeed comparable [21 22]. Previous case law is clear on how cases should be approached under section 63A(2) of the Sex Discrimination Act 1975 and section 54A(2) of the Race Relations Act 1976. The employment tribunals approach to the two stage test set out in those provisions was correct. At stage one, the complainant must prove facts from which the tribunal could conclude, in the absence of an adequate explanation, that the respondent has committed an act of discrimination against the complainant which is unlawful. So the prima facie case must be proved, and it is for the claimant to discharge that burden. In considering at that stage what inferences or conclusions can be drawn from the primary facts, the tribunal must assume that there is no adequate explanation for those facts. The purpose of that assumption is to shift the burden of proof onto the respondent at the second stage. It does not diminish in any way the burden of proof at the first stage, when the tribunal is looking at the primary facts that must be established. But it is important not to make too much of the role of the burden of proof provisions. They will require careful attention where there is room for doubt as to the facts necessary to establish discrimination. They have nothing to offer where the tribunal is in a position to make positive findings on the evidence one way or the other. That was the position in this case [25 & 32]. It is clear that the tribunal addressed whether the situations of Professor Forrester and Mr Larmour were like for like comparisons. Having done that, it found that difference of treatment had been proved for which, in its judgment, there appeared to be no adequate explanation. It was entitled in these circumstances to draw a prima facie inference of sex and race discrimination in Mrs Hewages favour, which it was for the Board to rebut and it failed to do. There is no substance in the suggestion that the tribunal misdirected itself or that it considered only part of the evidence that it was required to examine at the first stage [26]. It was not necessary for the question remitted to the employment tribunal by the Inner House to be remitted to a differently constituted tribunal. There was an obvious advantage in remitting the matter to the original tribunal as it had already heard and been able to assess the evidence. This was pre eminently a matter for the Inner House, and there are no grounds for thinking that it made the wrong choice [33].
The respondents to these appeals (Mrs Gallagher, P, G and W) have all been convicted or received cautions or reprimands in respect of relatively minor offending. The disclosure of their criminal records to potential employers has made, or may in future make, it more difficult for them to obtain employment. In each case, the relevant convictions and cautions were spent under the legislation designed for the rehabilitation of ex offenders, set out below. Nonetheless, criminal records had to be disclosed if they applied for employment involving contact with children or vulnerable adults. In 1996, Mrs Gallagher was convicted of one count of driving without wearing a seatbelt, for which she was fined 10, and three counts of carrying a child under fourteen years old without a seatbelt, for which she was fined 25 on each count. In 1998, she was again convicted of two counts of the latter offence and fined 40 on each count. Mrs Gallagher has no other convictions. In 2013, having qualified as a social carer, she was admitted to the Northern Ireland Social Care Council Register of Social Care Workers. In 2014, she applied for a permanent position at a day centre for adults with learning difficulties and received a conditional offer of employment. On a disclosure request, she only disclosed the 1996 convictions regarding her children, but not the 1996 conviction as to herself, nor the 1998 convictions. Her job offer was withdrawn after the Enhanced Criminal Record Certificate disclosed all her previous convictions. In 1999, P received a caution for the theft of a sandwich from a shop. In the same year, P was convicted of the theft of a book worth 99p and of failing to surrender to the bail granted to her after her arrest for that offence. She received a conditional discharge for both offences. At the time of the offences, P was 28 years old, homeless and suffering from undiagnosed schizophrenia which is now under control. She has committed no further offences. P is qualified to work as a teaching assistant but has not been able to find employment. She believes this is the result of her disclosure obligations. In 1982, W was convicted of assault occasioning actual bodily harm. He was 16 years old at the time when the assault occurred during a fight between a number of boys on their way home from school. He received a conditional discharge, and has not offended since. In 2013, aged 47, he began a course to obtain a certificate in teaching English to adults. He believes that his chances of obtaining teaching employment will be prejudiced by the need to obtain a criminal record certificate for a job as a teacher. In 2006, G, aged 13, was arrested for sexually assaulting two younger boys. The offences involved sexual touching and attempted anal intercourse. There was exceptional mitigation. The police record indicates that the sexual activity was consensual and seems to have been in the form of dares and is believed to have been a case of sexual curiosity and experimentation on the part of all three boys. The Crown Prosecution Service decided it was not in the public interest to prosecute but suggested a reprimand. G received two police reprimands in September 2006. He has not offended since. In 2011, when working as a library assistant in a local college, he was required to apply for an enhanced criminal record check because his work involved contact with children. The police proposed to disclose the reprimand, with an account of the mitigation. As a result, G withdrew the application and lost his job. He has since felt unable to apply for any job requiring an enhanced criminal record check. In all four of the appeals, the respondents challenge two related statutory disclosure schemes as being incompatible with Article 8 of the European Convention on Human Rights 1950 (ECHR), protecting the right to respect for private and family life. This raises two separate questions, namely whether any interference with Article 8 ECHR is: (1) in accordance with the law (the legality test) and (2) necessary in a democratic society (the proportionality test). The first scheme, governing disclosure by the ex offender, is that under the Rehabilitation of Offenders Act 1974 (the 1974 Act) in England and Wales and the corresponding provisions of the Rehabilitation of Offenders (Northern Ireland) Order 1978 (SI 1978/1908) in Northern Ireland, which are materially the same. By sections 4(2) (3) of the 1974 Act, where a question is put to an ex offender about previous convictions, offences, conduct or circumstances, there is no duty of disclosure. However, for any of thirteen specified purposes in the Rehabilitation of Offenders Act 1974 (Exceptions) Order (SI 1975/1023) (1975 Order) and the Rehabilitation of Offenders (Exceptions) Order (Northern Ireland) (SR(NI) 1979/195) (1979 Order), there is a duty of disclosure. The second scheme, governing disclosure by the Disclosure and Barring Service in England and Wales or Access NI in Northern Ireland, is governed by Part V of the Police Act 1997, as amended (the 1997 Act). Sections 113A and 113B deal with Criminal Record Certificates and Enhanced Criminal Record Certificates. These provisions create a system of mandatory disclosure of all convictions and cautions on a persons record if the conditions for the issue of a certificate were satisfied. In 2014, a more selective system for disclosure was introduced under the second scheme by the Disclosure and Barring Service the Police Act 1997 (Criminal Record Certificates: Relevant Matters) (Amendment) (England and Wales) Order (SI 2013/1200) and the Police Act 1997 (Criminal Record Certificates: Relevant Matters) (Amendment) (Northern Ireland) Order (SI 2014/100). Broadly corresponding limitations were imposed in relation to the first scheme by the Rehabilitation Act 1974 (Exceptions) Order 1975 (Amendment) (England and Wales) Order (SI 2013/1198) and the Rehabilitation Act 1974 (Exceptions) (Amendment) Order (Northern Ireland) Order (SI 2014/27). The Court of Appeal in England and in Northern Ireland (EWCA and NICA), affirming the decisions of the Divisional Court or High Court (except in Ws case), upheld the respondents case. First, the statutory schemes were considered incompatible with Article 8 ECHR for failing the legality test because of the breadth of the categories in the legislation. Secondly, the statutory schemes were considered disproportionate for failing to sufficiently distinguish between convictions and cautions of varying degrees of relevance. The appellants now appeal to the Supreme Court. Ps cross appeal concerns the refusal to quash article 2A(3)(c) of the 1975 Order for breach of Article 8 ECHR. The Supreme Court dismisses the appeals (except in Ws case), but varies parts of the orders below. A majority of the Court (Lord Sumption, Lord Carnwath, Lord Hughes and Lady Hale) reach that result based on a partial breach of the proportionality test. Lord Sumption (with whom Lord Carnwath and Lord Hughes agree) gives the lead judgment. Lady Hale (with whom Lord Carnwath also agrees) gives a concurring judgment. On the cross appeal, the Court varies the order of the Divisional Court by adding a declaration that article 2A(3)(c) of the 1975 Order is incompatible with Article 8 ECHR. Lord Kerr gives a separate judgment, disagreeing with the majoritys approach to the legality test and its application of the proportionality test. Lord Kerr would have dismissed the appeals (including in Ws case) and affirmed the declarations of incompatibility made by the EWCA and NICA. All members of the Supreme Court agree that Article 8 ECHR is engaged and that two conditions thus apply, namely satisfaction of: (1) the legality test and (2) the proportionality test [12], [73], [153]. They also all agree that the legality test requires, at least, accessibility and foreseeability [16], [73], [182]. Majority judgments (Lord Sumption and Lady Hale): Lord Sumption considers that the legality test, whether under Article 8 ECHR or otherwise, does not involve questions of degree [14]. For him, accessibility requires that it must be possible to discover what the provisions of a legal measure are, while foreseeability requires that a measure does not confer an unconstrained discretion [17], [31]. However, if the issue is how much discretion is too much (i.e. a question of degree), only the proportionality test can be used for review [17]. In the ECHR case law, in particular MM v United Kingdom (App. no. 24029/07), the Strasbourg Court has treated the need for safeguards as part of the foreseeability requirement and applied it as part of the legality test in cases where a discretionary power would otherwise be unconstrained and lack certainty of application [24]. There must be sufficient safeguards, exercised on known legal principles, against the arbitrary exercise of a discretion, so as to make its application reasonably foreseeable [31]. Lord Sumption disagrees with the EWCA and NICA as to the effect of the Supreme Courts decision in R (T) v Chief Constable of Greater Manchester Police [2014] UKSC 35, concerning the regime governing disclosure of criminal records in England before the changes introduced in March 2014 [15], [35 41]. He does not accept that R (T) decided that a measure may breach the legality test even where there is no relevant discretion and the relevant rules are precise and entirely clear [37]. For Lord Sumption, the rules governing the disclosure of criminal records under both the 1974 Act and the 1997 Act are highly prescriptive, mandatory and leave no discretion [42]. There is thus no real difficulty in assessing the proportionality of the two statutory schemes, so the legality test is satisfied both schemes are in accordance with the law for the purposes of Article 8 ECHR [42 45]. As to proportionality, Lord Sumption considers that two questions arise: (1) whether the legislation can legitimately require disclosure by reference to pre defined categories at all and (2), if so, whether the current boundaries of these categories are acceptable [46]. As to the first question, Lord Sumption considers that legislation by reference to pre defined categories is justified [50]. This is because: (1) the final decision about the relevance of a conviction should be that of the employer, who is best placed to assess the individual circumstances; (2) there is limited evidence that employers cannot be trusted to take an objective view; (3) the 1997 Act scheme is carefully aligned with the disclosure scheme under the 1974 Act, necessitating a category based approach; and (4) it would be impracticable to require a system of individual assessment [51 54]. On the second question, Lord Sumption considers that, with two exceptions, the carefully drawn categories in the legislation are not disproportionate [61 62]. The first exception is the multiple convictions rule, which does not achieve its purpose of indicating propensity as it applies irrespective of the nature, similarity, number or time intervals of offences [63]. The second exception concerns warnings and reprimands for younger offenders, the purpose of which is instructive and specifically designed to avoid damaging effects later in life through disclosure [64]. In Ps case, the disclosure was based on the multiple convictions rule under the 1997 Act, so the appeal against the declaration of incompatibility falls to be dismissed on that limited ground [65]. However, as to Ps cross appeal, article 2A(3)(c) of the 1975 Order is only to be declared incompatible with Article 8 ECHR (rather than quashed) [66]. As Mrs Gallaghers case also concerns the multiple convictions rule, she is also entitled to a declaration of incompatibility both as to the 1997 Act and the 1979 Order [67]. In Gs case, concerning a reprimand against a younger offender, the declaration of incompatibility as to the mandatory disclosure requirement under the 1997 Act is affirmed [68]. In Ws case, the High Courts order is restored since assault occasioning actual bodily harm may be a serious offence and it was appropriate to include it within the category of offences requiring disclosure [69]. Lady Hale agrees with Lord Sumption that, given the changes to the statutory schemes in 2014, the legality test is satisfied [72 73]. She considers that the law in question does not have to contain an individual review mechanism in every case. The requirement is only that it is possible to test, both the law itself and the decisions made under it, for proportionality [73]. The present schemes are not indiscriminate in nature and have been carefully devised to balance the competing public interests in (1) rehabilitation, (2) safeguarding and (3) practicability [75]. Given the need for a practicable and proportionate scheme, bright line rules are necessary [76 77]. She agrees with Lord Sumption that the categories used are proportionate, save as to the two exceptions above, and accordingly agrees with him on the disposal of each appeal and the cross appeal [78 79]. Lord Kerrs minority judgment: Lord Kerr would have dismissed the appeals (including in Ws case) and affirmed the declarations of incompatibility made by the EWCA and NICA. Lord Kerr disagrees with the majority on compliance with the legality test and the proportionality test. He illustrates the issues with the current statutory schemes by reference to a fuller account of the circumstances of each of the respondents [80 100]. He also reviews in detail the operation of the two statutory schemes before and after the 2014 amendments [101 146]. Lord Kerr considers that two important points follow from the Supreme Courts decision in R (T). These are: (1) that there must be adequate safeguards built into a disclosure scheme which allow for a proper evaluation of proportionality and (2) that the provisions then in force were condemned for the lack of any mechanism for independent review [149]. Lord Kerr identifies five central precepts that are relevant to the legality test [153, 158], but adds that not all of these must necessarily be satisfied [159]. He considers that the fundamental requirement is that the operation of the safeguards must permit a proper assessment of the proportionality of the interference with the Article 8 ECHR right [159]. He also clarifies that his approach to the legality test goes beyond only satisfying the two requirements of accessibility and foreseeability, contrary to Lord Sumptions approach [182 187]. Lord Kerr would have found the scheme in England and Wales to fail the legality test since the cases show that there is at least the potential for widespread disproportionate outcomes in disclosure [162]. Therefore, it cannot be said that there are safeguards adequately to examine proportionality [162]. He suggests two potential modifications: (1) a provision which linked the relevance of the data to be disclosed to the nature of the employment sought [165 173] and (2) an individual review mechanism in some cases, such as that introduced in Northern Ireland in 2016 [174 175]. Further, Lord Kerr would have found the scheme disproportionate [188 190].
This appeal arises from an unsuccessful management buyout of Evo Medical Solutions (Evo) made through Evo Medical Solutions Ltd (EMSL) in 2006. The buyout was funded by an interest bearing loan of 15m to EMSL by Swynson Ltd (Swynson), a company owned and controlled by Mr Hunt, a wealthy investor. Prior to the buyout, Swynson instructed an accountancy firm formerly known as Hurst, Morrison Thomson, now Lowick Rose LLP (in liquidation) (HMT), to carry out due diligence on Evo. It is common ground that HMT was negligent in failing to draw attention to fundamental problems about Evos finances, and that the transaction would not have gone ahead but for that failure. By July 2007, Evo was at risk of financial collapse. As a result, Mr Hunt caused Swynson to lend EMSL a further 1.75m in July 2007 and 3m in June 2008. At or about the same time, Mr Hunt acquired the majority beneficial ownership of EMSL. In December 2008, the 2006 and 2007 loans were refinanced. Mr Hunt and EMSL entered into a loan agreement under which Mr Hunt personally made a short term non interest bearing loan of 18.663m to EMSL. This was for the specific purpose of enabling EMSL to repay Swynson the original loan, with the aim of cleaning up Swynsons balance sheet and reducing its liability to tax. EMSL duly repaid the loan, but eventually ceased business and was unable to meet its liabilities. Swynson and Mr Hunt brought proceedings against HMT seeking to recover damages for losses resulting from the buyout and the making of all three loans in 2006, 2007 and 2008. HMT contends that they have no liability for damages on the basis that Swynson has suffered no loss, because EMSL repaid Swynson the whole of the original loan in December 2008. The Court of Appeal by a majority upheld the judges award of damages of 15m. This was because they held that the judge had been right to regard the December 2008 refinancing as res inter alios acta. It did not therefore affect the amount of Swynsons recoverable loss. The Supreme Court unanimously allows HMT (Lowick Rose LLP)s appeal. Lord Sumption gives the lead judgment, with which Lord Neuberger, Lord Clarke and Lord Hodge agree. Lord Neuberger and Lord Mance give concurring judgments. Res inter alios acta The general rule that loss which has been avoided is not recoverable as damages is subject to an exception in respect of collateral payments (res inter alios acta), where these are received independently of the circumstances giving rise to the loss [11]. The payments made by Mr Hunt to EMSL and by EMSL to Swynson to pay off the 2006 and 2007 loans cannot be regarded as collateral. First, the transaction discharged the very liability whose existence represented Swynsons loss [13]. Secondly, the money Mr Hunt lent to EMSL in December 2008 was not an indirect payment to Swynson, even though it ultimately reached them. Mr Hunts agreement to make that loan and the earlier agreements of Swynson to lend money to EMSL were distinct transactions between different parties, each made for valuable consideration [13]. Thirdly, the consequences of refinancing could not be recoverable as the cost of mitigation, because the loan to EMSL was not an act of Swynson and was not attributable to HMTs breach of duty [13; 45; 97]. Transferred loss The judge and Court of Appeal were correct to reject Swynsons second argument, based on the principle of transferred loss. This principle is a limited exception to the general rule that a claimant can recover only loss which he has himself suffered [14 15; 52 53; 102 105]. It does not arise here because it was no part of the object of the engagement of HMT, or any other aspect of the 2006 transaction, to benefit Mr Hunt [17; 54; 108]. Unjust enrichment HMT was not unjustly enriched by Mr Hunts provision of funds to EMSL to repay Swynson, with the result that Mr Hunt may not be subrogated to Swynsons claims against them: i. Lord Sumption is prepared to assume for the sake of argument that HMT was enriched [20], while Lord Neuberger and Lord Mance consider that HMT has undoubtedly been enriched in economic terms as a result of the discharge by EMSL of the loan due to Swynson [113; 57]. ii. Lord Sumption is again prepared to assume that if HMT was enriched, it was at Mr Hunts expense [20]. Lord Neuberger considers that HMTs enrichment was not sufficiently directly effected by Mr Hunts advance of the new loan [114 115], while Lord Mance notes that the questions whether a benefit was obtained at the expense of the claimant and whether it would be unjust for the defendant to retain it are difficult to separate in the present case [58 68]. iii. Mr Hunts case is that the enrichment of HMT was unjust because he made a mistake in assuming that the December 2008 refinancing would not affect the claim he and/or Swynson had against HMT [21; 78]. But the purpose of the law of unjust enrichment is to correct normatively defective transfers of value [22; 117]. The role of equitable subrogation in this context is to replicate as far as possible the element of the transaction whose absence made it defective [31; 86]. The December 2008 refinancing was not a defective transaction: Mr Hunt got precisely what he intended to get, namely the discharge of EMSLs debt to Swynson and a right to recover the new loan from EMSL [32 35; 119; 87]. As Lord Neuberger and Lord Mance also note, the fact that HMT received a benefit as an unforeseen and incidental consequence of Mr Hunts pursuit of those objectives does not establish any normative or basic defect in the arrangements made [117; 87 89].
This case concerns jurisdiction over claims relating to thefts from two cargo containers. The first container was allegedly hi jacked in Belgium in September 2011. 756 of an original 1,386 cigarette cartons disappeared from the second container while it was parked overnight near Copenhagen. The consignors of these containers, and respondents in both appeals, are two companies in the British American Tobacco Group. The consignors are claiming against the English main contractors, Exel Europe Ltd (Exel), who undertook responsibility for the carriage and also against the sub contractors, the appellants, in whose hands the containers were when the alleged losses occurred. Exel has played no part in the appeals. In the appeal concerning the first container the appellants are H Essers Security Logistics BV and H Essers Transport Company Nederland BV (collectively, Essers) and in the appeal concerning the second container the appellant is Kazemier Transport BV (Kazemier). In both cases the carriage of containers was subject to the Convention on the Contract for the International Carriage of Goods by Road 1956 (CMR), which is given the force of law in the United Kingdom by the Carriage of Goods by Road Act 1965. Article 31 CMR sets out certain bases on which courts and tribunals can have jurisdiction over disputes arising from contracts of carriage. Article 34 CMR provides that where a contract of carriage is performed by successive carriers, each is responsible for the performance of the whole operation. Under article 36 CMR, claims under the contract may be brought against the first and last carriers and the carrier in possession of the goods when the loss, damage or delay occurred. The issue in the appeals is whether the consignors can found jurisdiction in England against Essers and Kazemeir as successive carriers within the meaning of CMR by relying on the presence in England of, and the proceedings brought against, Exel and/or upon a provision in the main contract for English jurisdiction. The High Court set aside the service of claim forms on Essers and Kazemeir. The Court of Appeal reached the opposite conclusion. Essers and Kazemeir appealed to the Supreme Court. The Supreme Court allows the appeals by Essers and Kazemeir and restores the High Courts order setting aside the service of claim forms. Lord Mance writes the lead judgment, with which Lord Neuberger and Lord Reed agree. Lord Sumption and Lord Clarke write concurring judgments. Does Article 31 apply to successive carriers? Article 31 CMR applies to disputes where a successive carrier is involved [19 20, 61]. Article 31 CMR confers jurisdiction only on the courts or tribunals of the jurisdiction (i) as agreed by the parties, (ii) where the defendant is ordinarily resident, or has his principal place of business, or the branch through which the contract was made, or (iii) the place of consignment or delivery [33, 62]. In order to found jurisdiction in England, the respondents must bring their claim under one of these heads. It is common ground that England was not the place of consignment or delivery for either container [31, 64]. Did Essers and Kazemeir agree to English jurisdiction? The appellants became party to the respective contracts made between the respondents and Exel but article 34 CMR qualifies this position by adding that a successive carrier becomes party to the contract of carriage under the terms of the consignment note [23]. It would be contrary to the general principle that contract depends on agreement to hold a successive carrier bound by a choice of court clause, or any other clause not evidenced by the consignment note and of which he had no express notice. On that basis neither Essers nor Kazemeir were bound by an English jurisdiction clause in the original contracts [26]. Can a claimant bring subsequent carriers into the jurisdiction where it is already suing one carrier? The first, last and performing carriers under article 36 CMR hold joint and several liability, rather than alternative liability [34 42]. CMR reflects a considered decision as to the balance of interests between all potential claimants and defendants and ought not to be interpreted to include an additional head of jurisdiction allowing for a defendant domiciled in one member state to be sued in the courts of the place where a co defendant was domiciled, unlike the article 6.1 of the Regulation on Civil Jurisdiction and Judgments, Council Regulation (EC) No 44/2001 of 22 December 2000 (the Brussels Regulation) [46 47]. Was England the location of the branch or agency through which the relevant contract of carriage was made? Article 31 CMR refers to the original contract between the consignor and the primary carrier, rather than that through which any successive carrier acceded. Therefore England is not the relevant location of such branch or agency [32, 66]. Does the Brussels Regulation provide any other basis for jurisdiction or aid the interpretation of the CMR? The Brussels Regulation does not provide any other basis for jurisdiction over Essers or Kazemeir or otherwise act as an aid to the interpretation of the CMR. The CMR represents a balanced regime adopted across 55 states, only half of which are in the EU. It does not impinge on any of the principles of EU law which the Court of Justice has, in other cases, held to prevail over such international agreements [48 57]. Lord Clarke and Lord Sumption both consider that the commercial logic of articles 34 and 36 points towards the recognition of a jurisdiction to receive claims against all three carriers in one set of proceedings but agree with Lord Mance that the language of the CMR clearly provides otherwise [60, 69].
Gaming duty is an excise duty that is charged in accordance with the provisions of the Finance Act 1997 (the FA) on any premises where dutiable gaming takes place. Non negotiable gaming chips and free bet vouchers (collectively referred to as Non Negs for the purposes of this appeal) are promotional tools provided free of charge by some casino operators to selected gamblers to encourage them to gamble in their casinos. Non Negs may typically only be used to place bets at the gaming tables for their face value and cannot be used to buy goods or services, nor encashed. This appeal is about whether Non Negs should be taken into account as part of the bankers profits for the purposes of computing gaming duty in accordance with section 11(8)(b) read with section 11(10) of the FA. Bankers profits are defined by section 11(10) as the value, in money or moneys worth, of the stakes staked with the banker in any such gaming (section 11(10)(a)) less the value of the prizes provided by the banker to those taking part in such gaming otherwise than on behalf of a provider of the premises (section 11(10)(b)). The valuation of prizes for the purposes of section 11(10)(b) is governed by section 11(10A) of the FA which incorporates by reference, with any necessary modifications, certain provisions of the Betting and Gaming Duties Act 1981 (the BGDA). From October 2008 until September 2012, London Clubs Management (LCM) included the face value of all the Non Negs played by gamblers and retained by its casinos in the calculation of its bankers profits for the purposes of computing its liability for gaming duty. It subsequently considered that this approach was incorrect and that it had overpaid gaming duty by over 1.97 million. Her Majestys Revenue and Customs (HMRC) rejected LCMs claim for repayment of the alleged overpayment and LCM appealed that decision. The First tier Tribunal dismissed LCMs appeal. The Upper Tribunal allowed LCMs appeal. The Court of Appeal dismissed HMRCs further appeal. HMRC appealed to the Supreme Court. The Supreme Court unanimously dismisses the appeal. The leading judgment is given by Lord Kitchin, with whom Lord Carnwath and Lady Black agree. Lady Arden gives a separate judgment, agreeing that the appeal should be dismissed but for materially different reasons. Lord Sales gives a separate judgment agreeing with the majority on the decisive issues in the appeal but reaching a different conclusion on a related issue. Lord Kitchin holds that Non Negs are neither stakes staked, nor have any value in money or moneys worth within the meaning of section 11(10)(a) of the FA [31 48]. First, the assessment of the gross gaming yield from any premises requires a focus on the activity of gaming and not the provision of other goods or services [36]. Second, the assessment required by section 11(10) must be carried out from the bankers perspective, for it is the bankers profits which must be brought into account in calculating the gross gaming yield [37]. Third, the reference to money or moneys worth in section 11(10)(a) emphasises that it is the real world value of the stakes in the hands of the banker which matters [38]. A gambler who plays with cash chips in a casino is not staking the chips as such but the money those chips represent, which the gambler has deposited with the casino [31 35]. That is not the case when a gambler places a bet using a Non Neg, which essentially amounts to a free bet. Therefore, a Non Neg holds no real world value to the casino when a gambler loses it in a bet, save that it eliminates the chance that the casino may have to pay out the winnings corresponding to that bet. However, that does not impart a value, in money or moneys worth to the Non Neg, nor mean that it is a stake staked within the meaning of section 11(10)(a) of the FA [44]. Lord Sales agrees with Lord Kitchins reasoning in this regard and provides further reasons in support of the conclusion [88 92]. Lady Arden adopts different reasoning for dismissing the appeal [61 78]. She holds that the value of a stake for the purposes of section 11(10)(a) is what a person would pay for it in the open market [62]. The objective valuation of the stake means that it can be taken into account at an appropriate value if it has generated gaming activity, which is what gaming duty is charged upon [70]. A Non Neg is a stake for the purposes of assessing the bankers profits under section 11(10)(a) of the FA [75]. However, the appeal fails on the facts of this case as no evidence was adduced to support any objective valuation [68]. The Supreme Court also considers the related issue of what value, if any, should be given to Non Negs for the purposes of section 11(10)(b) of the FA. Lord Kitchin expresses the view that, as with section 11(10)(a), it is the real world cost to the banker of providing the prizes that must be brought into account for the purposes of assessing the value of the prizes provided by the banker, subject to the relevant provisions of the BDGA [38]. Under the BGDA, the cost to the banker of awarding a voucher as a prize is taken to be its face value if, among other things, it may be used in place of money as whole or partial payment for benefits of a specified kind obtained from a specified person. Non Negs do not satisfy this condition and must therefore be treated as having no value for the purposes of section 11(10)(b) of the FA [54]. Lord Kitchin is confirmed in his views by the consideration that the result is a coherent scheme for the treatment of Non Negs, whether used by gamblers to place bets or when returned to gamblers as prizes [56]. Lord Sales reaches a different conclusion from the majority in this regard. When a Non Neg is awarded as a prize, it represents a real cost to the banker which ought to be brought into account [93]. A Non Neg given as a prize satisfies the relevant conditions under the BDGA 1981, such that it should normally be treated as having its face value for the purposes of section 11(10)(b) of the FA [95 104]. Lady Arden agrees with Lord Saless conclusion in this regard [82 84].
The appellant, Ms Samuels, was an assured shorthold tenant of a property in West Bromwich, Birmingham, where she lived with four children. In July 2011, having fallen into rent arrears, she was given notice to leave. She later applied to the respondent council to be treated as homeless under Part VII of the Housing Act 1996 (the 1996 Act). A local housing authority becomes under a duty to secure accommodation to a person found homeless if certain conditions are satisfied. One condition is that they are not satisfied that the person became homeless intentionally. That depends on whether she deliberately did or failed to do anything which caused her to leave accommodation that was available and would have been reasonable for her to continue to occupy. Article 2 of the Homelessness (Suitability of Accommodation) Order 1996 provided that, in determining whether it would be reasonable for a person to continue to occupy accommodation, the local authority will take into account whether that accommodation is affordable. That includes consideration of the financial resources available to that person, including social security benefits, and consideration of the persons other reasonable living expenses. The local authority is required to have regard to guidance given by the Secretary of State, which at the time was the Homelessness Code of Guidance for Local Authorities (the Code) issued in 2006. Paragraph 17.40 of the Code stated: In considering an applicants residual income after meeting the costs of the accommodation, the Secretary of State recommends that housing authorities regard accommodation as not being affordable if the applicant would be left with a residual income which would be less than the level of income support or income based jobseekers allowance that is applicable in respect of the applicant, or would be applicable if he or she was entitled to claim such benefit. [] The council decided that Ms Samuels was intentionally homeless, on the grounds that the accommodation in West Bromwich was affordable and reasonable for her to continue to occupy, and that its loss was the result of her deliberate act in failing to pay the rent. In concluding that the accommodation was affordable, the council found that the shortfall in rent could have been met by greater flexibility in the household budgeting. Ms Samuelss appeal to the County Court against the councils decision was dismissed and her further appeal was dismissed by the Court of Appeal. The central issue in her appeal to the Supreme Court is whether the council adopted the correct approach in determining that the accommodation was affordable for the purposes of the 1996 Act. The Supreme Court unanimously allows the appeal and quashes the councils decision. Lord Carnwath gives the judgment of the court. The 1996 Order requires the authority to take into account all sources of income, including all social security benefits. There is nothing in it to require or justify the exclusion of non housing benefits of any kind. It also requires consideration of the applicants reasonable living expenses, which necessitates an objective assessment, not simply the subjective view of the case officer [34]. Even if the recommendation in paragraph 17.40 of the Code in respect of income support is not interpreted as extending to benefits for children, the lack of a specific reference does not make the level of those benefits irrelevant. Benefit levels are not generally designed to provide a surplus above subsistence needs for the family. If comparison with relevant benefit levels is material to the assessment of the applicant, it should not be any less material in assessing what is reasonable by way of living expenses in relation to other members of the household. The duty to promote and safeguard the welfare of children under the Children Act 1989 is also relevant [35]. As one would expect, the guidance makes clear that the amount of an applicable benefit will vary according to the circumstances and composition of the applicants household. It also refers to the current tariffin respect of such benefits (plural), implying that the tariff may be looked at in respect of benefits other than income support, and is at least a good starting point for assessing reasonable living expenses [35]. The review officer in Ms Samuels case asked whether there was sufficient flexibility to enable her to cope with the shortfall between her rent and her housing benefit. But the question ought to have been what her reasonable living expenses were (other than rent), to be determined having regard to both her needs and those of the children. The total expenses shown in the schedule provided by her solicitors (1,234.99) was well within the amount regarded as appropriate by way of welfare benefits (1,349.33). It is difficult to see by what standard those expenses could be regarded as unreasonable [36]. The appeal is therefore allowed, and the review decision quashed. In light of the information available to the Court, Lord Carnwath finds it hard to see on what basis the finding of intentional homelessness could be properly upheld. He therefore hopes that on reconsideration the council will be able to accept full responsibility under Part VII of the 1996 Act for Ms Samuels and her family [37].
This appeal is concerned with the roles of the case investigator and the case manager when handling concerns about a doctors performance under the disciplinary procedures introduced over eight years ago for doctors and dentists in the National Health Service. The national policy framework is known as Maintaining High Professional Standards in the Modern NHS (MHPS), which the Trust has implemented through its own policies D4 and D4A. Dr Chhabra was employed by the Trust as a consultant forensic psychiatrist at Broadmoor Hospital, which is a high security unit, on 3 September 2009. Following concerns about Dr Chhabras performance, Dr Nicholas Broughton, the Trusts medical director and case manager for these concerns, appointed Dr Amanda Taylor, a consultant forensic psychiatrist from another trust, as case investigator on 15 December 2010. He instructed Dr Taylor to investigate the following: (1) an allegation that Dr Chhabra, travelling on a busy train, discussed an incident involving a patient and read a medical report on a patient whose name and personal details could be clearly seen; (2) an allegation that Dr Chhabra had dictated patient reports when travelling on a train; (3) concerns about Dr Chhabras working relationship with her clinical team; and (4) a complaint from one of Dr Chhabras patients, made through a solicitor. Dr Taylor found, in her report of June 2011, that Dr Chhabra had breached, and admitted breaching, patient confidentiality (1) by having patient documents clearly visible and (2) by dictating sensitive reports on the train. She also found that there were difficulties within Dr Chhabras clinical team which were issues of capability that needed to be addressed; and that the solicitors complaint did not have merit. Her report also recorded Dr Chhabras unchallenged account that she had not appreciated at the time that her practice compromised patient confidentiality and that she believed that she had ensured that no other passengers were close by when she dictated the reports. Dr Taylor also reported on an allegation by Dr Chhabras former secretary, which had not been expressly included in her terms of reference, that she had made telephone calls when travelling by train to work in which she had discussed patient information. Dr Taylor did not make any finding on the veracity of this allegation. In response to a concern raised by Dr Chhabra, the Trust had undertaken that Mr Wishart, its associate human resources director, could take no part in the investigation. But, unknown to Dr Chhabra, Dr Taylor had communicated with Mr Wishart during the investigation. Most significantly, Dr Taylor had sent Mr Wishart a draft of her report and Mr Wishart prepared suggested amendments to the draft. The extensive amendments had stiffened the criticism of Dr Chhabra. Dr Taylor had accepted some of the suggested amendments but not others. Among those she had accepted was the characterisation as serious of breaches of confidentiality she had described in her report. On 12 August 2011 Dr Broughton wrote two letters to Dr Chhabras solicitors. One informed her that he regarded the concerns about her team working to be matters of capability. The other stated that he proposed to put to a disciplinary panel not only the admitted breaches of confidentiality but also (i) the allegation, on which Dr Taylor had noted there had been a conflict of opinion, that Dr Chhabra, while travelling by train, had telephoned her secretary to discuss patient related information, and (ii) an allegation, which was not within Dr Taylors remit and on which she had not reported, that Dr Chhabra had breached patient confidentiality by disclosing information via email to her medical protection society and legal advisers. Dr Broughton expressed the view that the charges were potential gross misconduct and that dismissal was a possible outcome of the hearing before the disciplinary panel. Dr Chhabras solicitors objected to the charge of breach of patient confidentiality by disclosing information to her protection society and her legal advisers, which had not been the subject of Dr Taylors investigation. At their request the Trust agreed to instruct Dr Taylor to investigate that allegation. Dr Taylor carried out this further investigation and reported that there was no complaint to answer. As a result, on 17 January 2012 Dr Broughton informed Dr Chhabra by letter that that charge would not be pursued at the disciplinary hearing. On 1 June 2012 Judge McMullen QC granted Dr Chhabra a declaration and injunctive relief preventing the disciplinary panel from investigating the confidentiality concerns, including those not grounded in Dr Taylors report, as matters of gross misconduct. On 25 January 2013 the Court of Appeal upheld the Trusts appeal. The case manager was not confined to the findings of fact of the case investigator but could consider complaints supported by evidence reported by the case investigator, even if denied by the practitioner. The conduct panel would resolve issues of disputed fact. Dr Broughton was entitled to regard the breach of confidentiality as a potentially serious offence and as a result was justified in deciding to convene the conduct panel. Dr Chhabra appeals to this court. The Supreme Court unanimously allows Dr Chhabras appeal and orders the Trust not to pursue any of the confidentiality concerns contained in the Trusts letter of 12 August 2011 as matters of gross misconduct; and not to pursue any confidentiality concerns without first re starting and completing an investigation under its policy D4A. The first and most significant issue is the roles of the case investigator and the case manager. The procedures do not allow the case investigator to determine the facts. Their aim is to have someone, who can act in an objective and impartial way, investigate the complaints identified by the case manager to discover if there is a prima facie case of a capability issue or misconduct. It would introduce an unhelpful inflexibility into the procedures if (i) the case investigator were not able to report evidence of misconduct which was closely related to but not precisely within the terms of reference (as in the former secretarys allegations) or (ii) the case manager were to be limited to considering only the case investigators findings of fact when deciding on further procedure. Similarly, it would be unduly restrictive to require the case manager to formulate the complaint for consideration by a conduct panel precisely in the terms of the case investigators report. Neither MHPS or the Trusts policies in D4 and D4A are so inflexible or restrictive. But the procedure does not envisage that the case manager can send to a conduct panel complaints not considered by the case investigator or for which the case investigator has gathered no evidence. The Trust was therefore correct in acceding to Dr Chhabras request for a second report from Dr Taylor in relation to the new allegation of breach of confidentiality in her communications with the protection society and her solicitors. There were number of irregularities in the proceedings against Dr Chhabra that cumulatively render the convening of the conduct panel unlawful as a material breach of her contract of employment. First, Dr Taylors findings were not capable, taken at their highest, of supporting a charge of gross misconduct, defined in the policy as so serious as to potentially make any further relationship and trust between the Trust and the employee impossible. The breaches of confidentiality she recorded, including the former secretarys allegations, were qualitatively different from a deliberate breach of confidentiality such as speaking to the media about a patient. Secondly, in reaching the view that Dr Chhabras behaviour could amount to gross misconduct, Dr Broughton founded on the words added to para 13.4.1 with effect from 28 March 2011, after the incidents in this case. The list of misconduct in para 13.4.1 comprised only typical examples of what the Trust saw as amounting to gross misconduct and was not a comprehensive statement of the concept. But Dr Broughton relied on the amended provision in support of his view that the complaints might amount to gross misconduct and quoted it in his letter of 12 August 2011 relating to the disciplinary procedure. Thirdly, the Trust breached its contract with Dr Chhabra when Mr Wishart continued to take part in the investigatory process in breach of the undertaking the Trust had given. In particular, when Mr Wishart proposed extensive amendments to Dr Taylors draft report and Dr Taylor accepted some of them, which strengthened her criticism of Dr Chhabra, the Trust went outside the agreed procedures which had contractual effect in ways going beyond clarifying its conclusions. The report had to be the product of the case investigator. It was not. Further, the disregard for the undertaking amounted to a breach of the obligation of good faith in the contract of employment. It was also contrary to policy D4s principle that managers act in a way that an objective observer would consider reasonable: Dr Chhabra had an implied contractual right to a fair process, which Mr Wisharts involvement undermined. Fourthly, Dr Broughton did not re assess the decision in his letter of 12 August 2011 that the matters were considered as potential gross misconduct after he departed from the additional complaint once he had received Dr Taylors second report. He was obliged to do so under para 3.1 of policy D4: an objective observer would not consider it reasonable to fail to do so. The cumulative effect of those irregularities is that it would be unlawful for the Trust to proceed with the disciplinary procedure and that the Court should grant relief. The categorisation of Dr Chhabras conduct as gross misconduct is itself a sufficient ground for injunction.
This case concerns an aspect of the Supreme Court's jurisdiction to hear appeals in Scottish civil cases. Mr and Mrs Politakis are the directors and the only shareholders of Apollo Engineering Ltd (Apollo). They wish to appeal against two orders that were made in a case stated for the opinion of the Inner House of the Court of Session under section 3 of the Administration of Justice (Scotland) Act 1972. The stated case arose from an arbitration between Apollo and James Scott Ltd in relation to a contractual dispute [2]. Apollo ran out of funds and could no longer afford legal representation. On 18 January 2012 the Inner House made an order refusing Mr Politakis request that he represent Apollo, on the basis that as a matter of Scots law, a company requires to be legally represented. On 27 November 2012 the Inner House made a further order in which, among other things, it (1) refused Mr Politakis leave to appeal to the Supreme Court against the order of 18 January 2012; (2) refused to allow Mr Politakis to be joined as a party either to replace or in addition to Apollo; and (3) dismissed the stated case [3, 4]. Section 40 of the Court of Session Act 1988 (the 1988 Act) regulates appeals to the Supreme Court in Scottish civil cases. It provides that it is competent to appeal from the Inner House to the Supreme Court without the leave of the Inner House, against: (1) a judgment on the whole merits of the cause; (2) an interlocutory judgment where there is a difference of opinion among the judges; or (3) where the interlocutory judgment is one sustaining a dilatory defence and dismissing the action. It is also competent to appeal to the Supreme Court with the leave of the Inner House against any other type of interlocutory judgment of the Inner House [6]. The House of Lords had decided in the case of John G McGregor (Contractors) Ltd v Grampian Regional Council 1991 SC (HL) 1 that an opinion of the court in a stated case did not constitute a judgment within the meaning of section 40 of the 1988 Act. In the present case, an opinion on the legal issues in the stated case had not been given [7]. Two issues are before the Supreme Court: (1) whether the McGregor principle applies in the present case so that an appeal against the order of 27 November 2012 is incompetent; and (2) if not, whether the part of the order of 27 November 2012 which dismissed the stated case can competently be appealed to the Supreme Court under section 40 of the 1988 Act without the leave of the Inner House. The Supreme Court directed that these two issues should be the subject of an oral hearing [8]. There is no self standing right of appeal to the Supreme Court against the order of 18 January 2012, because it was an interlocutory judgment under section 40 of the 1988 Act and the Inner House has refused leave to appeal against it [9]. Apollo can competently appeal to the Supreme Court without the leave of the Inner House against the part of the order of 27 November 2012 which dismissed the stated case, as long as the appeal raises a question which can be responsibly be certified by counsel as reasonable [16, 28, 29]. Lord Hope gives the judgment of the Court. On issue (1), the Court holds that none of the cases in this area, including the McGregor case, offer direct assistance on the question that is to be resolved, and none deals with the situation where the court has declined to do what the statute provides for, which is to give an opinion [10 14]. The ordinary use of language indicates that an appeal to the Supreme Court against an opinion of the Inner House under section 3 of the 1972 Act is excluded by necessary implication because it is for the opinion of that court only that the case has been stated. But the 1972 Act makes no provision for the course of action that the Inner House felt obliged to take in this case: dismissing the stated case without giving its opinion on the questions that were before it at all [15]. On issue (2), the Court holds that the order dismissing the stated case cannot be regarded as an interlocutory judgment of the kind which is appealable only with leave under section 40 of the 1988 Act. All the issues that were in controversy before the Inner House were disposed of when the stated case was dismissed. In dismissing the stated case, the court exhausted its functions under the statute, save as to resolving any outstanding issues about expenses. The effect of the order was to end the proceedings completely, in just the same way as if it had encompassed the courts opinion on the questions that were before it [22, 23]. It is not easy to characterise the order dismissing the stated case as one sustaining a dilatory defence and dismissing the action. The order gave effect to a motion by James Scott Ltd based on Apollos inability to fulfil the courts rules of practice about representation. It would be stretching the language of the statute to say that this objection was a defence, especially as the procedure under section 3 was not one that could, in the ordinary sense of the word, be defended [27]. It may not matter much whether the order is to be regarded as a judgment on the whole merits of the cause or as one sustaining a dilatory defence, as both are appealable without the leave of the Inner House. On balance, however, it would be more correct to regard it as a judgment on the whole merits of the cause within the meaning of section 40 of the 1988 Act, even though the Inner House did not address itself to the issues raised in the stated case [27]. As is the case with all other orders that are appealable without leave however, Apollos petition of appeal must be certified by two counsel as reasonable the test for which is whether the appeal raises arguable points of law which are of general public importance. The only question which the Supreme Court can consider is whether the order of the Inner House to dismiss the stated case was one which was open to it to make under the jurisdiction given to it by the statute. Unless something has gone seriously wrong, however, this was an exercise of judgment on a matter of procedure with which this court would not normally wish to interfere [29]. The question whether there was any way in which Apollos interests could have been represented which might have avoided the situation in which the Inner House felt obliged to dismiss the stated case is not before the Court. But it is a troublesome aspect of this case, and there may be grounds for thinking that the rule which disables a company from being represented other than by counsel or a solicitor with a right of audience needs to be re examined [30].
The Appellant is a practising barrister and is black. She alleges that the Respondent discriminated against her on grounds of her race by bringing disciplinary proceedings which ended in her acquittal on appeal. On 9 June 2010, the Respondents Complaints Committee brought 6 disciplinary charges against the Appellant. On 23 May 2011, the Disciplinary Tribunal found 5 of these charges proved. The Appellant appealed to the Visitors of the Inns of Court (the Visitors). On 17 August 2012, her appeal was allowed on the basis that none of the alleged conduct involved any breach of the Bar Code of Conduct. On 21 February 2013, the Appellant issued the present proceedings, which included an allegation of violation of Article 14 of the European Convention on Human Rights (ECHR) read in conjunction with Article 6 ECHR, contrary to section 6 of the Human Rights Act 1998 (the 1998 Act). In its defence, the Respondent maintained that this claim was time barred under section 7(5)(a) of the 1998 Act which provides that proceedings must be brought before the end of the period of one year beginning with the date on which the act complained of took place. On 3 January 2014, the Respondent issued an application seeking an order that the statement of case be struck out on the basis that none of the Appellants claims had a real prospect of success and, in any event, there was a complete defence under section 7(5)(a). On 2 April 2014, the Respondents application for strike out was granted. The Appellant appealed. On 18 December 2014, Warby J held that there was a sufficiently pleaded case that the Respondent indirectly discriminated against the Appellant. However, he also held that the claim was time barred under section 7(5)(a) of the 1998 Act. The Appellant appealed to the Court of Appeal. The Court of Appeal held that the limitation period under section 7(5)(a) had started to run when the Disciplinary Tribunal had found the charges against the Appellant proved and so had expired before she had issued her claim. The Appellant was granted permission to appeal to the Supreme Court on the time limit issue under section 7(5)(a). The issues arising before the Supreme Court were: (i) whether the disciplinary proceedings against the Appellant were to be considered a series of discrete acts or a single continuing act and (ii) if the latter, did that act end with the verdict of the Disciplinary Tribunal or with the verdict of the Visitors? The Supreme Court unanimously allows the appeal. Lord Lloyd Jones gives the lead judgment with which the other Justices agree. As a preliminary issue, the Court was required to determine the precise nature of the discrimination claim which the Appellant wished to make [15]. In this regard, the Court concluded that the Appellants challenge was to the conduct of the Respondent in bringing and pursuing disciplinary proceedings against her, not to an alleged state of affairs in which BME lawyers were more likely to be the subject of such proceedings. Therefore, the bringing and pursuit of the disciplinary proceedings must be the focus of the investigation in terms of section 7(5)(a) of the 1998 Act [16 21]. The question which then arose in relation to the application of section 7(5)(a) was whether the bringing of disciplinary proceedings by the Respondent was to be considered a series of discrete acts or a single continuous act [22]. Section 7(5)(a) should not be read narrowly and must be capable of providing an effective and workable rule for situations where the infringement of a Convention right arises from a course of conduct. Leaving a claimant to have recourse only to the discretionary remedy in section 7(5)(b) is inappropriate [23]. The alleged infringement of Convention rights in this case arose from a single continuous course of conduct. The essence of the complaint made by the Appellant was the initiation and pursuit of the proceedings to their conclusion. It cannot have been the intention of Parliament that each step should be an act to which the one year limitation period should apply [29]. Under section 7(5)(a) time begins to run from the date when the continuing act ceased, not when it began [30]. In determining when the continuing act ceased, it was necessary to consider whether the Respondents conduct in proceedings before the Visitors should be considered as forming part of the same continuing act as its conduct in proceedings before the Disciplinary Tribunal. In order to answer this question, it was necessary to consider the nature of the regulatory scheme and the precise features of the Respondents conduct [32 34]. Several features of the regulatory scheme and the Visitors jurisdiction, as applicable to the disciplinary proceedings against the Appellant, led to the conclusion that the Respondents part in the proceedings before the Disciplinary Tribunal and those before the Visitors should be regarded as part of a single continuing act [35]. Therefore, the single continuing act in this case continued until the Visitors allowed the Appellants appeal on 17 August 2012. The Appellant commenced the present proceedings on 21 February 2013, within the period of one year beginning with the date on which the act complained of took place, as required by section 7(5)(a) and the appeal should accordingly be allowed [39]. The Respondent asked the Court to uphold the Court of Appeals decision on the alternative ground that Warby J was wrong to hold that the Article 14 ECHR claim of indirect discrimination had real prospects of success. The Respondent argued that this claim could have no real prospect of success without statistics sufficient to raise a potential case of discrimination, general statements of disproportionate impact being unlikely to be sufficient [41]. The Court observed that it was adventitious that this point was before the Court [42]. However, the Appellant was entitled to rely upon a 2013 report into the Respondents complaint system which analysed data from 2007 11, in conjunction with the unhappy history of the proceedings against her, as supporting her case that she had been the victim of indirect discrimination. The European Court of Human Rights had made clear that indirect discrimination can be proved without statistical evidence [43].
This appeal concerns the application of Article 3 of the Hague Convention on the Civil Aspects of International Child Abduction (the Convention). Under Article 3 it is unlawful to remove or retain a child in breach of rights of custody attributed to a person under the law of the state in which the child was habitually resident immediately before removal or retention. This case concerns two small children, born and raised in France, who were brought to Scotland by their mother in July 2013 with the consent of their father, who remained in France. The mother and children were to live in Scotland for the period of about a year. In November 2013 the relationship between the parents ended. On 20 November 2013 the mother commenced proceedings in which she sought a residence order in respect of the children and an interdict against the father removing them from Scotland. The father argued that the initiation of those proceedings was a wrongful retention within the meaning of the Convention on the basis that the children were habitually resident in France immediately before proceedings commenced. The Outer House of the Court of Session concluded that the children were still habitually resident in France on 20 November 2013. This judgment was based on the fact that the move to Scotland had not been intended by both parents to be permanent. The Inner House of the Court of Session reversed the Outer Houses decision on the basis that shared parental intention to move permanently to Scotland was not an essential element in any alteration of the childrens habitual residence. The Inner House concluded that the children were habitually resident in Scotland at the material time. The father appealed to the Supreme Court on the basis that the Outer House had been correct, and that the Inner House had in any event erred in its approach. The mother argued that there had in any event been no wrongful retention. The Supreme Court unanimously dismisses the appeal. The Court considers that, for the purposes of habitual residence, the stability of residence, rather than its degree of permanence, is important. There is no requirement that the child should have been resident in the country in question for a particular period of time or that one or both parents intend to reside there permanently or indefinitely. As the Court has previously held in a series of cases, habitual residence is a question of fact which requires an evaluation of all relevant circumstances [16]. In determining habitual residence, the focus is upon the situation of the child, with the intentions of the parents being merely one of the relevant factors. It is necessary to assess the degree of the integration of the child (or, in the case of an infant or young child, the degree of integration of those on whom the child is dependent) into a social and family environment in the country in question. There is no rule that one parent cannot unilaterally change the habitual residence of a child [17]. In the present case, the children were habitually resident in Scotland within the meaning of the Convention. The absence of a joint parental intention to live permanently in Scotland was not decisive, nor was an intention to live in a country for a limited period inconsistent with becoming habitually resident there. The important question is whether the residence has the necessary quality of stability, not whether it is necessarily intended to be permanent [21]. Following the childrens move with their mother to Scotland, their life there had the necessary quality of stability. Their home was Scotland for the time being, their social life and much of their family life was there. The longer time went on, the more integrated they became into their environment in Scotland [23]. Given this conclusion, the question of wrongful retention did not arise [25].
The appellant, KM, is a profoundly disabled man aged 26, who lives with his mother, brother and sister [1]. He was born without eyes and has a range of serious mental and physical medical conditions [8]. He is intelligent and articulate with many achievements including GCSE passes in French and music [8]. He needs substantial support in feeding, self care and other aspects of daily living [9]. The respondent, Cambridgeshire County Council, is the appellants local authority. Under section 2 of the Chronically Sick and Disabled Persons Act 1970 [the Act] a local authority owes a duty to a disabled person, if it is necessary in order to meet his needs, to make arrangements for the provision of a number of specified services including practical assistance in the home, recreational facilities inside and outside the home and assistance with adaptations to the home [11 13]. In fulfilling its duty, a local authority must act according to the guidance of the Secretary of State for Health, but in place of directly providing services it can, and in some cases must, make a direct payment to the disabled person to enable him to arrange his own support [22]. To determine the appropriate direct payment, the respondent applied its Resource Allocation System, which calculates the payment due based on the average funding for people with specific needs in the local authority area; with additional sums calculated in severe cases by reference to an Upper Banding Calculator. An independent social worker had produced a report apparently estimating the total annual cost of supporting the appellant as c. 157,000. The respondent did not expressly contest this assessment and classified his needs as being critical, but disputed the proposed required level of funding. The appellant challenges by way of judicial review a decision by the respondent, communicated at the latest by a letter dated 3 June 2010, to pay him roughly 85,000 annually. The sum reflected the maximum 61,000 calculated by reference to the Resource Allocation System and an additional amount calculated by reference to the Upper Banding Calculator. The appellant contends that the decision is unlawful either because it was not adequately supported by reasons or because it was irrational [2]. As part of his challenge, the appellant invited the Supreme Court to reconsider the decision of the House of Lords in R v Gloucestershire County Council, Ex parte Barry [1997] AC 584 on the extent to which the resources of a local authority may be taken into account in making decisions under section 2 of the Act. Consequently, a court of seven justices was convened; and four charities and the Secretary of State for Health intervened. At the hearing, however, it became clear that the issue did not arise on the facts since the respondent did not rely on resource constraints to justify its decision. The Court therefore did not hear full argument on Barry, which is not reconsidered in its judgments [4 7; 40 41]. The Supreme Court unanimously dismisses the appeal. Lord Wilson gives the leading judgment, with which the other justices agree. Lady Hale gives a separate concurring judgment. When a local authority is required to consider whether it is necessary, in order to meet the needs of a disabled person, for that authority to make arrangements for the provision of any of the relevant services, then it must ask itself three questions: (i) what are the needs of the disabled person; (ii) in order to meet these needs is it necessary for the authority to make arrangements for the provision of any of the listed services; and (iii) if so, what are the nature and extent of the services for which it is necessary for the local authority to make arrangements? [15]. The guidance produced by the Secretary of State for Health in respect of the duties under section 2 of the Act reflects these stages of inquiry [16 18]. In considering the question of eligibility, at the second stage, the local authority can ask whether the needs of the disabled person can reasonably be met by family or friends, by institutions such as the NHS or charities, or out of the persons own resources [19]. The decision in R v Gloucestershire County Council, Ex parte Barry [1997] AC 584 established that the availability of its resources is also relevant [19]. If the needs of a disabled person are deemed eligible, the duty of a local authority to meet those needs is then absolute, and the Court of Appeal erred in holding otherwise [19]. Under regulations made pursuant to the Health and Social Care Act 2001, if a local authority is satisfied that a disabled persons need for the relevant service can be met by the provision of a direct payment to the disabled person then it can, and in some cases must, with the disabled persons consent, make a direct payment to enable him to purchase the relevant service [22]. In that case, a fourth stage of enquiry arises: (iv) what is the reasonable cost of securing provision of the services for which it is necessary for the authority to make arrangements? In answering that question, it is unduly laborious for a local authority at first to cost each service for every disabled person [24]. Local authorities therefore use resource allocation systems to provide a ball park figure of the appropriate payment, subject to adjustment. The systems generally work by allocating points to eligible needs and then ascribing a cost to each point. A realistic connection between needs and points, and then between points and costs, is crucial [25]. The systems are a lawful tool to provide an approximate indicative sum [26]. Once the indicative sum has been identified, the requisite services should be costed in a reasonable degree of detail to permit a judgement on whether the sum is correct [28]. Adequate reasons can be achieved with reasonable brevity and it will often be sufficient for a local authority to list the required services, and the suggested timings and hourly costs [37]. By reference to its resource allocation system the respondent ascribed points to the appellant in excess of the maximum, which equated to 61,000 [25]. It had developed a second indicative tool, an upper banding calculator, for persons who scored points in excess of the maximum: this brought the figure up to 85,000 [27]. The costing of individual services should then have been carried out, preferably in conjunction with the disabled person, by the making of a support plan [28]. The respondent accepted all of the appellants presenting needs as eligible [29]. It did not accept his mothers representations that the family would not offer support but, unfortunately, it did not expressly say so [29 30]. Had an adjustment been made for a reasonable amount of such support, the indicative figure would have been only c. 46,000 [30]. The jointly instructed social worker uncritically reported the appellants familys wishes, failing to make an expert assessment of the costs [32 33]. The respondent considered the report to be manifestly excessive but, again, failed expressly to say so in proposing the annual sum of c. 85,000 [34]. The respondent broadly explained how this annual sum might be deployed, but not how it had been computed [35]. The appellants solicitors pressed for an explanation and the respondent accepts that it should have provided it earlier than its letter of 3 June 2010 [35]. The challenge to the rationality of the decision fails, since it was entirely rational for the respondent to use its resource allocation system and upper banding calculator and it appears that any flaw in the computation is likely to have been in the appellants favour [38]. The respondent should have made a more detailed presentation of its assessment of the reasonable cost of the necessary services in the appellants case. Nevertheless, in the light of the subsequent amplification of its reasoning during the litigation, it would be a pointless exercise of discretion to quash the decision so that his entitlement might be considered again, perhaps even to his disadvantage [38].
This appeal considers the scope of what is known in public procurement law as the Teckal exemption. It considers whether a local authority was entitled to enter into contracts of insurance with a mutual insurer, established in co operation with other local authorities, without first putting those contracts out to tender in accordance with the Public Contracts Regulations 2006 (the UK Regulations) In 2006 and 2007 various London local authorities co operated and entered arrangements for mutual insurance. The aim was to reduce the cost of insurance premiums, by removing the element of profit built in to an ordinary commercial insurance premium, and to increase the standard of risk management. London Authorities Mutual Limited (LAML) was established in order to provide insurance to participating London authorities and their affiliates. Brent London Borough Council (Brent) was one of ten authorities involved. It became a member of LAML, made a payment in order to capitalise LAML and also provided a guarantee, undertaking to pay further sums to LAML on demand. Once LAML was established, it contracted to provide Brent with insurance. Brent did not conduct a tendering process for the award of the contract of insurance which it entered into with LAML. Risk Management Partners (RMP), a commercial insurer, claimed that there should have been a tendering process complying with the UK Regulations and that, had one been carried out, it might have obtained the insurance which was placed with LAML. RMP claimed damages from Brent. Although that claim has now been settled, the Courts decision will determine other damages claims against various other London authorities and will clarify for the future whether the UK Regulations apply to contracts which a public body proposes to award to an organisation such as LAML. The UK Regulations apply whenever a contracting authority seeks offers in relation to the award of certain public contracts. They give effect in domestic law to an EU Directive, Council Directive 2004/18/EC on the co ordination of procedures for the award of public work contracts, public supply contracts and public service contracts (the Directive). It applies to the award of public contracts. Case law of the European Court of Justice has developed an exception, known as the Teckal exemption. It provides that, in certain circumstances, the award of a contract by one public body to another separate legal person will not fall within the definition of public contract in the Directive, with the result that the Directive will not apply to it and EU law will not require the contract to be put out to tender. The Teckal exemption comprises both a control test and a function test. (1) The local authority must exercise over the person to whom the contract is proposed to be awarded a control which is similar to that which it exercises over its own departments, and (2) that person must carry out the essential part of its activities with the controlling local authority or authorities. This appeal concerned three questions of principle. (1) Does the Teckal exemption apply to the UK Regulations at all? (2) Does the Teckal exemption apply to contracts of insurance? (3) In order for the Teckal exemption to apply, must the control which the contracting authority exercises over the contractor be exercised by that authority individually or is it sufficient that it could be exercised collectively, together with other local authorities? The High Court and the Court of Appeal held that the Teckal exemption did apply to the UK Regulations and that it was available in respect of contracts of insurance. However, they concluded that the control test was not satisfied because LAML was too independent from the local authorities which made up its membership. The claim between RMP and Brent having settled, Harrow was given permission to continue the appeal so as to have the issues of principle decided. The Supreme Court unanimously allows the appeal. It holds that the Teckal exemption does apply to the UK Regulations, that it is available in respect of insurance contracts and that it is sufficient for it to apply that the co operating public authorities together exercise collective control over the party to whom contracts are awarded. The requirements of the Teckal exemption were satisfied. Lord Hope and Lord Rodger both give judgments; Lord Walker, Lord Brown and Lord Dyson agree with both. The Teckal exemption applies to the UK Regulations. Although the definitions in the UK Regulations differ in some respects from the Directive, the purpose of the UK Regulations was to give effect in domestic law to the Directive. There is nothing to indicate that the UK Regulations intended to depart from the European Court of Justices case law: [22] [26], [92]. The Teckal exemption is available in respect of contracts of insurance. It did not matter that insurance was not a service that the local authority could provide for itself. What matters is whether the arrangement satisfies the control test: [27] [30]. Both Lord Hope and Lord Rodger review the ECJ case law to consider the purpose and scope of the Teckal exemption. The Directive is not intended to protect the commercial sector by forcing public authorities to obtain the services which they need on the commercial market. Rather, its purpose is to ensure that, if public authorities do decide to obtain the services which they need from outside bodies, proper procedures are followed to ensure that potential providers have an opportunity to compete for the work. The Directive therefore does not apply where a public authority obtains the product or services from its own resources. Nor, in light of Teckal, does it apply where an authority obtains services from a separate body which is so closely connected with it that the authority should still be regarded as, in substance, obtaining the services in house. There is no reason in principle to distinguish between a situation where the body from which services are obtained exists to serve the interests of a single local authority and where it exists to serve the interests of several local authorities: [67] [73]. The Teckal control test requires that the public authority exercise a power of decisive influence over both the strategic objectives and significant decisions of the other body: [40]. That need not be exercised individually. It is sufficient that the public authority could exercise control over the contractor alone or together with other public authorities: [41], [45], [47] [49], [52], [80] [85]. There must be no private investment in the contractor: [53] & [75]. No injury will be caused to the policy objectives of the Directive if public authorities are allowed to participate in the collective procurement of goods and services, so long as no private interests are involved and they are acting solely in the public interest in the carrying out of their public service tasks: [53]. The argument that individual control is necessary is simply inconsistent with European Court of Justices thinking: [80]. Here, the participating local authorities did exercise collective control over LAML. Meetings of LAMLs board were not validly constituted unless a majority of those present were directors representing a participating member. Participating local authorities each had one vote at general meetings and retained a power to direct the board by special resolution. The fact that a director could not participate in a board meeting which considered a claim by a member which he or she represented was a matter of detail: [57]. No private interests were involved. The function test was also satisfied. LAML existed only in order to serve the insurance needs of its members: [59]
In September 2006, Mr Davidson (the Second Respondent) entered into a trust deed for the benefit of his creditors. It was a protected trust deed to which provisions of the Bankruptcy (Scotland) Act 1985 (1985 Act) applied. Clause 11 of the deed provided for the deeds termination on the occurrence of one of three events, one of which was a final distribution of the estate. Before he entered into the deed, Mr Davidson had been mis sold payment protection insurance (PPI), for which the bank agreed to pay him compensation of around 56,000 in April 2015. Dooneen Ltd (the First Respondent) was Mr Davidsons agent for the purpose of making the claim and Mr Davidson had assigned 30% of any compensation received to Dooneen. The dispute between the parties was about whether Mr Mond (the Appellant), as trustee, or the Respondents were entitled to the compensation. This turned on whether Mr Mond had made a final distribution when he distributed what he called a first and final dividend of 22.41 pence in the pound to the creditors in November 2010 and was discharged accordingly in circumstances where he did not know at the time that Mr Davidson had been mis sold PPI in respect of which he was entitled to compensation. The Lord Ordinary, Lord Jones, found in favour of the Respondents, and that decision was upheld by the Inner House. Mr Mond now appeals to the Supreme Court on the ground that the courts below had misinterpreted final distribution. The Supreme Court unanimously dismisses the appeal with the result that Dooneen and Mr Davidson are entitled to the payment of compensation. Lord Reed, with whom the rest of the Court agrees, delivers the judgment. Mr Mond argued that, regardless of whether or not the trustee knew of all of the assets, a final distribution can only occur when either all assets are distributed or enough assets are distributed so as to pay all creditors in full. This construction is rejected because it would have consequences which the debtor cannot have intended when granting the deed [12]. First, one could never be certain whether any distribution was in fact final so that the deed would potentially be of indeterminate duration. This would be particularly difficult to reconcile with other parts of the deed that vest in the trustee assets and income acquired by the debtor during the currency of the trust deed [13]. Second, it would make it impossible for the debtor or anyone doing business with him to know whether or not the debtor has been finally discharged [14]. Third, it would undermine the purpose of the public Register of Insolvencies, where certificates are registered under the 1985 Act signifying that a final distribution has been made, as it could no longer be relied on as accurate [15]. Lord Reed observes that the outcome of the case is scarcely satisfactory, and notes that the Court raised with the parties the question whether the relevant acts of the trustee might be reduced (set aside) if they were the result of an error as to the extent of the trust estate. Although the parties were invited to make submissions on this, they declined to do so, and it would accordingly not have been appropriate for the Court to consider these matters on this occasion [23].
The appellants company hired out items of machinery. Following an arson attack orchestrated by the appellant on a competitors premises, the police raided the premises of the appellants company and discovered that a significant proportion of the machinery present had been stolen. The appellant was convicted of handling stolen goods, and sentenced to 15 months imprisonment. Following this conviction, there was a hearing pursuant to section 6 of the Proceeds of Crime Act 2002 (POCA). The appellant conceded he had a criminal lifestyle and thus the judge had to decide whether and to what extent he had benefited from this. Not all of the machinery hired out by the appellants company had been stolen. The judge assessed the benefit obtained by the appellant at 2,275,454.40, comprising 1,960,754.40 from general criminal conduct and a further 314,700. Of this, the 1,960,754.40 was calculated on the basis that the proportion of stolen items to the total stock over the relevant period was 38%, and the companys aggregate turnover for the relevant period was 5,159,880 (inclusive of VAT). A confiscation order was made in the sum of 2,275,454.40. The appellant was given six months (later extended to 12 months) to pay, and was ordered to serve ten years (reduced to eight years by the Court of Appeal) in default of payment. This appeal considers whether the judge was right to make the confiscation order on the basis that the VAT had been obtained by the appellant for the purposes of POCA, in circumstances where the appellants company had already accounted for the VAT to HMRC. Before the Court of Appeal, the appellant argued that to include VAT in the amount of the confiscation order would involve an unacceptable degree of double counting. The Crown argued that a benefit is obtained for the purpose of POCA if it has been received by a defendant, even if he has subsequently had to account to a third party for some, or even all, of it. The Court of Appeal accepted the Crowns case. The Supreme Court allows Mr Harveys appeal by a majority of 3:2. Lord Neuberger and Lord Reed give the leading judgment. Lord Mance writes a concurring judgment. Lord Hughes and Lord Toulson each write dissenting judgments. Lord Neuberger and Lord Reed find that the VAT for which a defendant has accounted to HMRC is in a different category from either income or corporation tax, and from expenses incurred in connection with acquiring money or an asset [24]. First, income and corporation tax are computed on a taxpayers overall or aggregate income. They cannot be invoked to reduce the value of the property or money obtained from criminal activity when assessing what has been obtained for the purposes of POCA. VAT liability on the other hand arises on each taxable supply, and can be directly and precisely related to the obtaining of the property in question. In a case where the VAT on a transaction has been accounted for to HMRC, then the Court of Appeals approach would lead to the UK government enjoying double recovery of the VAT: once under POCA and once through the Value Added Tax Act 1994 [25 26]. Secondly, VAT is intended to be neutral in its impact on taxable persons: where money is paid to a defendant as a result of a transaction which is liable to VAT, the defendant is regarded as collecting the VAT element on behalf of HMRC. It is difficult to regard VAT which has been collected and accounted for to HMRC as forming part of the economic advantage derived from criminal offences [27]. Thirdly, it would be particularly harsh, where a defendant has accounted to HMRC for all the VAT for which he is liable, not to allow him credit for that sum, but this would be the effect of his being rendered liable to a confiscation order in respect of the output tax on his transactions. He would then be treated in the same way regardless of whether he had paid the tax or not [28]. Fourthly, the possibility of double recovery has been recognised, and avoided by extra statutory means, in the context of excise duty: HMRC does not seek to recover the excise duty due in respect of smuggled goods where a confiscation order has been made in the same sum [29]. Lord Neuberger and Lord Reed acknowledge that these factors give rise to a powerful argument that when VAT has been accounted for to HMRC, it has not been obtained by the defendant. However, they reject the appellants submission that this conclusion follows from the wording of POCA, because of the principle enunciated in R v Waya [2012] UKSC 51 that property obtained as a result of or in connection with crime remains the defendants benefit whether or not he retains it [30]. Nonetheless, these factors are relevant to consideration of whether the effect of the Crowns interpretation of POCA breaches Article 1 Protocol 1 of the European Convention on Human Rights (A1P1), the right to peaceful enjoyment of possessions [31]. Although a provision effecting double recovery is not forbidden by A1P1, it is at risk of being found disproportionate, given that sums payable pursuant to POCA are intended to be deterrent and not punitive [32]. Waya made clear that where the proceeds of crime are returned to the loser it would be disproportionate under A1P1 to treat such proceeds as the benefit obtained. That situation is similar to the collection of VAT, and the policy underlying the principle is in part that a defendant who makes good a liability to pay or restore should not be worse off than one who does not [33]. In R v Ahmad [2014] UKSC 36 it was held that it would be disproportionate for the same sum to be recovered from two co conspirators in respect of the same property which had been jointly obtained. The observations made in these cases are applicable in relation to VAT which has been accounted for to HMRC [34]. Although the burden may be on the Crown to establish the gross value of the benefit obtained by the defendant, the burden of establishing any sum which should be deducted to reflect the VAT accounted for to HMRC lies on the defendant. There is nothing disproportionate about a judge taking a broad brush approach where the evidence is confusing, unreliable or incomplete [35]. Thus, where VAT has been accounted for to HMRC, it would be disproportionate under A1P1 to make a confiscation order calculated on the basis that the VAT, or a sum equivalent, was obtained by the defendant for the purposes of POCA. The position in relation to VAT for which a defendant is liable but has not accounted to HMRC is left open [36]. Lord Mance agrees with Lord Neuberger and Lord Reed and writes a concurring judgment [38 48]. Lord Hughes would have dismissed the appeal. He states that POCA is not designed to restore money to the state, which in most cases is not the loser, but is designed to deprive the offender [55]. When the defendant was paid a VAT inclusive sum by his customers, he obtained the VAT element, and this is not affected by his obligation to declare it [66]. It is not disproportionate to confiscate the gross proceeds of offending without giving credit for taxes paid to the state [76]. Lord Toulson would also have dismissed the appeal [102]. He finds that it would not be disproportionate under A1P1 to treat the entirety of the companys receipts from its criminal conduct, ignoring associated outgoings such as tax liabilities, as having been obtained by the appellant [125].
The appellants made false representations in their applications for United Kingdom citizenship. The issue in these appeals is whether those misrepresentations made the subsequent grant of citizenship to them a nullity rather than rendering them liable to be deprived of that citizenship under sections 40 and 40A of the British Nationality Act 1981 (BNA). Mr Hysaj is an Albanian citizen. He claimed asylum in the UK in July 1998, falsely stating he was born in Kosovo, was therefore a citizen of the Federal Republic of Yugoslavia (FRY), and that he was under 18. He was accepted as a refugee and given indefinite leave to remain (ILR). Mr Hysaj was granted naturalisation as a British citizen in 2004. Mr Bakijasi was also born in Albania. He sought asylum under a false name and date of birth, gave a false place of birth in Kosovo and falsely claimed FRY nationality. He was eventually granted ILR under the same false details, and then granted naturalisation as British citizen in 2006. When these frauds came to light the Secretary of State, on the basis of binding Court of Appeal case law, decided that in both cases the grant of citizenship was a nullity, so that the appellants were not and had never been British citizens, albeit that they remained on ILR. The appellants challenged the decisions, submitting that the earlier cases were wrongly decided. The High Court and Court of Appeal, which were also bound by this case law, upheld the Secretary of States decisions. Permission to appeal was granted by the Supreme Court on the ground that the appeals gave rise to an arguable point of law of general public importance. Unusually, the Secretary of State applied pursuant to rule 34(2) of the Supreme Court Rules for the appeals to be allowed by consent, supported with reasons. The Supreme Court unanimously allows the appeals by consent. It agrees with the reasons provided by the Secretary of State and holds that misrepresentations in an application for UK citizenship renders the applicant liable to be deprived of that citizenship pursuant to s 40 BNA. Lady Hale gives the only substantive judgment. S 40 BNA makes provision for the Secretary of State to deprive a person of citizenship if satisfied that the grant was obtained by means of fraud, false representation or concealment of a material fact. There is a right of appeal to the First tier Tribunal against most such deprivations in s 40A [7]. The original decision adopting the nullity approach, rather than a deprivation of citizenship, involved the purported grant of British citizenship to someone who was impersonating another real person (R v Secretary of State for the Home Department ex p Mahmood [1981] QB 58). Subsequent cases, including R v Secretary of State for the Home Department ex p Akhtar [1981] QB 46, Bibi v Entry Clearance Officer, Dhaka [2007] EWCA Civ 740 and the present ones, expanded this approach to persons adopting a false identity through which the characteristics needed to obtain citizenship were acquired [8 14]. The Secretary of State considers that the law took a wrong turning after Mahmood, and the nullity approach should only apply in impersonation cases [15]. The subsequent cases were based on the principle that there is a category of fraud as to identity which is so serious that a purported grant of citizenship is of no effect, but had not articulated a clear definition of such fraud. This uncertainty means the law is difficult to apply in practice and also gives rise to a number of illogical and unsatisfactory consequences. The same principle would also appear to nullify the grant of ILR, but the Secretary of State has never contended for this [16 18]. The Supreme Court agrees with this reasoning. It follows that the decisions of the Court of Appeal in Akhtar and Bibi must be overruled and the present appeals allowed by consent [19].
These appeals concern the circumstances in which a prisoner serving a life sentence or an indeterminate sentence of imprisonment for public protection (IPP), who has served the minimum period specified for the purposes of retribution and deterrence (the tariff), and whose further detention is justified only if it is necessary for the protection of the public, should be awarded damages for delay in reviewing the need for further detention following the expiry of the tariff. They are also concerned with the quantum of such damages. Since 1997, legislation has required judges to impose life sentences on a wider range of offenders than was previously the case. In addition, IPPs were introduced in April 2005. It is for the Parole Board of England and Wales (the Board) to decide whether to direct the release of a life or IPP prisoner whose tariff has expired. The prisoners case must first be referred to the Board by the Secretary of State for Justice (the Secretary of State). The increase in the number of life prisoners and the introduction of IPP sentences resulted in an increase in the Boards workload, but its resources were not increased. This resulted in delay in the consideration of post tariff prisoners cases. That delay has implications under the Human Rights Act 1998 (the 1998 Act), which gives effect to Article 5 of the European Convention on Human Rights (the Convention). Article 5(1) requires that detention must throughout its duration remain causally connected to the objectives of the sentencing court. In relation to post tariff prisoners, that objective is the protection of the public. In order to comply with Article 5(4), the Board has to review the necessity for the continued detention of post tariff prisoners speedily upon the expiry of their tariff and at reasonable intervals thereafter. The 1998 Act also provides that the remedies for a violation of a Convention right include damages. Mr Faulkner was sentenced in 2001 to life imprisonment for a second offence involving grievous bodily harm. Mr Sturnham was convicted of manslaughter in 2007 and given an IPP sentence. In each case, there was a delay in the holding of a hearing before the Board after the tariff had expired, due to administrative errors for which the Secretary of State was responsible. Both men were eventually released following Board hearings, but Mr Faulkner was twice recalled to prison in respect of allegations of which he was acquitted, and remains in custody. Each sought judicial review of the failure by the Board and the Secretary of State to conduct a review of his detention speedily, as required by Article 5(4). Mr Faulkner was unsuccessful in the High Court, but the Court of Appeal held that the Secretary of State had breached Article 5(4), that Mr Faulkner would have been released 10 months earlier than he was but for that breach, and that the Secretary of State should therefore pay him 10,000 in damages. In Mr Sturnhams case, the High Court held that there had been a breach of Article 5(4) due to a delay of 6 months, that he had been caused anxiety and distress by the delay, but that there was no prospect that he would have been released any earlier had the hearing taken place speedily. The Secretary of State was ordered to pay him 300, but that award was quashed by the Court of Appeal. In Mr Faulkners case, the Board appeals to the Supreme Court on the ground that the award of damages was excessive. Mr Faulkner cross appeals on the ground that the award was inadequate and that his imprisonment during the period of delay constituted false imprisonment at common law or a violation of Article 5(1). Mr Sturnham seeks permission to appeal against the Court of Appeals decision to quash the award of damages to him. The Supreme Court allows the Boards appeal in Mr Faulkners case, reduces the damages awarded to him to 6,500, and dismisses his cross appeal. The Court grants Mr Sturnham permission to appeal and allows his appeal. Lord Reed gives the lead judgment, with which Lord Neuberger, Lord Mance and Lord Kerr agree. Lord Carnwath delivers a concurring judgment. Mr Faulkners argument that the detention of a life prisoner constitutes false imprisonment if it continues beyond the point at which the prisoner would have been released if a hearing had been held in accordance with Article 5(4) must be rejected. That detention is still authorised by statute, and is therefore lawful until the Board directs release [16, 86]. Nor was Mr Faulkner the victim of a violation of Article 5(1). Such a violation requires exceptional circumstances warranting the conclusion that continued detention has become arbitrary, which were not present in Mr Faulkners case [17 23, 86]. On the question of the award of damages under the 1998 Act, the courts should be guided primarily by the principles applied by the ECtHR, which may be inferred from any clear and consistent practice of that court. The quantum of such awards should broadly reflect the level of awards made by the ECtHR in comparable cases brought by applicants from the UK or other countries with a similar cost of living [39]. The courts should resolve disputed issues of fact in the usual way even if the ECtHR in similar circumstances, due to the nature of its role, would not do so [39, 82]. Where it is established on the balance of probabilities that a violation of Article 5(4) has prolonged the detention of a prisoner past the point at which he would otherwise have been released, damages should ordinarily be awarded. The amount of such damages will be a matter of judgment, reflecting the facts of the case and having regard to guidance from the ECtHR and the national courts in comparable cases [75]. Pecuniary losses should be compensated in full [53, 70]. Though relevant in some circumstances, it will not ordinarily be appropriate to take into account as a mitigating factor that a claimant was recalled to prison following his eventual release [83]. Nor should damages be awarded merely for the loss of a chance of earlier release [82], or adjusted according to the degree of probability of release if the violation of Article 5(4) had not occurred [84]. Appellate courts do not ordinarily interfere with an award of damages simply because they would have awarded a different figure if they had tried the case. However, as the Court is in this case being asked to give guidance on the appropriate level of awards, and having regard to awards made by the ECtHR in other cases and to the fact that the liberty enjoyed by a person released on licence is precarious and conditional, the Court considers that an award of 6500 would adequately compensate Mr Faulkner [87]. Even where it is not established that an earlier hearing would have resulted in earlier release, there is a strong presumption that delay which violated Article 5(4) has caused the prisoner frustration and anxiety. Where such a presumption is not rebutted, an award of damages should be made, though on a modest scale [53, 67 68]. No such award should be made in cases where the frustration and anxiety were insufficiently severe to warrant an award, although that is unlikely to be the case where the delay was of around three months or more [66]. Following that approach, and having regard to ECtHR authorities, the award of 300 to Mr Sturnham was reasonable in his case [97]. Lord Carnwath concurs with the reasoning and conclusions in Lord Reeds judgment, but suggests a more selective approach to ECtHR authorities. He suggests focusing on those cases which explicitly decide points of principle, and eschewing those which are simply assessments of the facts [104 127].
The issue in this appeal is whether, where there has been a transfer of employees to which the Transfer of Undertakings (Protection of Employment) Regulations 1981 (TUPE) apply, the new employer is bound by a term of an employees contract of employment which provides that terms and conditions of employment will be in accordance with collective agreements negotiated from time to time. UK courts have previously held that Regulation 5 TUPE renders such a dynamic clause enforceable against the new employer. This appeal considers whether that approach requires to be modified in light of a decision of the European Court of Justice in Werhof v Freeway Traffic Systems GmbH & Co KG (Case C 499/04) [2006] ECR I 2397 (Werhof). Werhof interpreted Article 3(1) of the Directive which TUPE implements in domestic law (Directive 77/187/EEC on the approximation of the laws in the Member States relating to the safeguarding of employees rights in the event of transfers of undertakings, businesses or parts of businesses, as amended (the Directive). The Appellants are former employees of the London Borough of Lewisham (the council) who worked in the councils leisure department until 2002. The council subscribed to the National Joint Council for Local Government Services (the NJC). The Appellants contracts of employment provided that their terms and conditions of employment will be in accordance with the collective agreements negotiated from time to time by the NJC supplemented by agreements reached locally through the councils negotiating committees. In 2002 the part of the councils undertaking in which the Appellants worked was contracted out to CCL Ltd. In 2004, CCL Ltds undertaking was taken over by the Respondent, another private sector employer. TUPE applied to each of these transfers. The employees therefore first became employees of CCL Ltd and then of the Respondent. At the date of the transfer to CCL Ltd, there were collectively agreed terms setting out pay rates from 1 April 2002 to 31 March 2004. Subsequent NJC settlements provided for further pay increases. The Respondent refused to increase the Appellants pay in line with the NJC settlement with effect from 1 April 2006 and 1 April 2007. The Appellants brought claims for unauthorised deduction from wages. These were dismissed by the Employment Tribunal. The Appellants appeal to the Employment Appeal Tribunal was successful, but the Court of Appeal restored the Tribunals decision. It held that Werhof meant that Article 3(1) of the Directive did not bind the transferee to any collective agreement made after the transfer and that Regulation 5 TUPE did not indicate any intention to provide employees with greater protection. The Appellants appealed to the Supreme Court. The Supreme Court unanimously holds that there should be a reference to the Court of Justice of the European Union for a preliminary ruling to establish whether Article 3(1) of the Directive precludes national courts from giving a dynamic interpretation to Regulation 5 of TUPE in the circumstances of this case. Lord Hope gives the judgment, with which the other members of the court agree. Domestic legislation enacted so as to give effect to the UKs EU obligations must be construed so as to conform to those obligations, so far as possible to do so. Two questions arise: (i) whether Regulations 5(1) and 5(2) TUPE were designed to be more generous than Article 3(1) of the Directive, as interpreted in Werhof; (ii) if not, whether it is open to the national court to construe those Regulations more generously because that is not precluded by Article 3(1) of the Directive: [19] [26]. As to the first question, Parliament must be taken to have intended to do no more in enacting Regulations 5(1) and (2) TUPE than implement Article 3(1) of the Directive. Regulations 5(1) & (2) TUPE provided that contracts of employment to which it applied were to have effect after the TUPE transfer as if originally made between the employee and the transferee. Although in some respects TUPE was more generous to employees that the Directive, there was nothing indicating such an intention in Regulations 5(1) & (2). Although it is the 1981 TUPE Regulations which apply to this case, they were replaced by new regulations in 2006, which contained equivalent provisions to Regulations 5(1) & (2). It was not possible to infer from this that Parliament had intended to endorse the interpretation which had been given to Regulation 5 in the existing domestic case law: [28] [30]. As to the second question, it is open to a national court to extend the scope of national legislation implementing a Directive to areas not included within the scope of the Directive, provided that no other provisions of EU law preclude that: [32]. The scope of Article 3 has now been interpreted by the ECJ in Werhof, which was a preliminary reference from Germany. German law provides that collectively agreed rules become part of the employment contract with the content that they possess at the time when the business is transferred and are not updated after the transfer. It therefore adopts a static approach. The ECJ held such an approach was not precluded by Article 3(1) of the Directive: [37] [42] The question in this case is a different one: whether a member state is precluded from extending the protection afforded to employees on transfer, so as to provide dynamic protection, where that would be the consequence of an application of domestic contract law: [44]. It was not possible to infer from Werhof how the ECJ might have answered that question: [45]. First, the Directives aim was to promote approximation, not harmonisation, of law. It was not to affect the right of member states to introduce laws more favourable to employees: [46]. Secondly, in Werhof the ECJ had taken into account the new employers right not to join an association or union, protected by Article 11 of the European Convention on Human Rights. That had been of relevance there because of the way German employment law deals with collective agreements. It was not a concern in this case, because the matter depended entirely on the domestic law of contract: [47]. The question whether Article 3(1) of the Directive precludes national courts from giving a dynamic interpretation to regulation 5 of TUPE in the circumstances of this case should therefore be referred for a preliminary reference to the Court of Justice: [48].
This judgment is one of three being given simultaneously on the liabilities of the United Kingdom government for allegedly tortious acts done by HM Forces in the course of operations overseas or by foreign governments in which UK officials are alleged to have been complicit. This particular judgment deals with allegations that persons in Iraq and Afghanistan were unlawfully detained in breach of article 5 of the European Convention on Human Rights (ECHR) by HM forces engaged in peacekeeping operations in those countries under mandates from the United Nations Security Council. Serdar Mohammed (SM) was captured by British forces in Afghanistan on 7 April 2010. It is the Governments case that his capture took place in the course of a planned operation involving a ten hour firefight, from which SM was seen fleeing, discarding a rocket propelled grenade launcher and ammunition as he went. Intelligence is said to have identified him as a senior Taliban commander. SM was detained in British facilities until 25 July 2010, when he was transferred to the Afghan authorities. His detention can be divided into three periods: (i) the first 96 hours, (ii) 11 April to 4 May 2010, when he was being interrogated, and (iii) 4 May to 25 July, when he was held pending transfer to the Afghan authorities. Abd Ali Hameed Al Waheed was captured by British forces in Basrah, Iraq on 11 February 2007 at his wifes home. The Government contends that weaponry material for explosives were found on the premises. He was held at a British army detention centre for six and a half weeks, and was then released after an internal review had concluded that a successful prosecution would be unlikely. The relationship between article 5 and international law is one of three preliminary issues in Serdar Mohammed. The High Court held that British forces had no power to detain prisoners for any longer than was required to transfer them to the Afghan authorities, and then for no more than 96 hours. Accordingly, it held that the detention of SM breached article 5(1) and 5(4) of the ECHR. The Court of Appeal reached the same conclusion, albeit for different reasons. In Al Waheed, it was common ground before the High Court that, so far as the claim was based on a breach of article 5(1) of the ECHR, the judge and the Court of Appeal would be bound to dismiss it by virtue of the decision of the House of Lords in Al Jedda. The judge granted a certificate for a leapfrog appeal to the Supreme Court. Lord Toulson sat on all aspects of the appeals other than those involving the scope and procedural requirements of articles 5(1)(c) and/or (f), 5(3) and 5(4) of the Convention in relation to the detention of Serdar Mohammed, in relation to which (following Lord Toulsons retirement) Lord Hodge sat on 26 October 2016. By a majority of 7 to 2, the Supreme Court dismisses Mr Al Waheeds appeal, and allows the Governments appeal in Serdar Mohammed in part. The majority holds that British forces had power to take and detain prisoners for periods exceeding 96 hours if this was necessary for imperative reasons of security, but that its procedures for doing so did not comply with ECHR article 5(4) because they did not afford prisoners an effective right to challenge their detention. Lord Sumption (with whom Lady Hale agrees) gives the lead judgment. Lord Wilson gives a concurring judgment. Lord Mance adds a mainly concurring judgment, as does Lord Hughes (with whom Lord Neuberger agrees). On the issues with which he was involved, Lord Toulson agrees with Lord Mance, Lord Wilson and Lord Sumption. So far as he was involved, Lord Hodge agrees with Lord Sumption. Lord Reed (with whom Lord Kerr agrees) gives a dissenting judgment holding that there was authority to detain prisoners for periods exceeding 96 hours only in circumstances falling within the grounds specified in article 5(1) (which would cover the first and third periods of detention, but not the second). The first issue the Court addresses is whether British forces had legal power to detain SM in excess of 96 hours. The possible sources for such a power are customary international law and/or the authority of the UN Security Council [13]. The majority finds it unnecessary to express a concluded view on whether customary international law sanctions the detention of combatants in a non international armed conflict (NIAC) [14, 113, 148, 224]. Lord Reed concludes that no such rule currently exists as a matter of customary international law [275]. However, the majority holds that authority to capture and detain enemy combatants for imperative reasons of security was implicitly conferred by the relevant Security Council resolutions [30, 119, 164, 224]. These were UNSCR 1546 (2004) in Iraq [20]; and UNSCR 1386 (2001) in respect of Afghanistan [28, 119]. The majority (other than Lord Mance) further concludes that individual states participating in the International Security Assistance Force (ISAF) in Afghanistan were not limited by ISAFs policy of restricting detention to 96 hours, so that the United Kingdom was entitled to adopt its own detention policy [38 39]. Lord Mance considers that the resolutions conferred authority to detain on ISAF, not the contributing states [180], but arrives at the same conclusion as the majority regarding the legitimacy of the United Kingdoms detention policy on the ground that ISAF tacitly accepted of the UKs adoption of this policy [39, 188]. The next question is whether it is possible to reconcile these conclusions under public international law with article 5 ECHR. The European Court of Human Rights in Hassan v UK was able to accommodate the six permitted grounds of detention under article 5 with the power recognised under public international law to detain in the course of an international armed conflict (IAC). The majority holds that the same approach applies to a NIAC, where the source of the power to detain is a resolution of the Security Council [60, 134 6, 164, 224]. The six permitted grounds for detention in article 5(1) ECHR were formulated in relation to peacetime conditions and could not be regarded as exhaustive in conditions of armed conflict. Their object was to protect the individual from arbitrariness. This object was achieved if there was a legal basis for detention and the power to detain was not exercisable on grounds which were unduly broad, opaque or discretionary [63, 93, 164 167, 224]. The procedure governing military arrest in Afghanistan was suitably clear and precise to meet the standards of article 5(1) [93, 113, 165 167, 224]. Article 5(1) did not therefore prevent a Convention state from acting under the authority conferred by a Security Council resolution. A majority (Lord Sumption, Lady Hale, Lord Wilson, Lord Hodge, Lord Reed and Lord Kerr) considers that the detention of SM did not fall within any of the six specified grounds in article 5(1), during the second period of his detention. He was not at any time held pending extradition to the Afghan authorities, because transfer to the civil authorities within Afghanistan did not constitute an extradition within article 5(1)(f) [78, 84, 113, 235, 236, 351]. Any period when he was being detained solely for intelligence exploitation purposes could not be justified under article 5(1)(c) [81, 84, 113, 235, 236, 351], or under the relevant Security Council resolution. Lord Mance, Lord Hughes and Lord Neuberger consider that whether SMs detention during any period fell within article 5(1)(f), as modified if necessary under Hassan, should be remitted for trial [202 203, 230]. Lord Mance would also have remitted for trial the question whether article 5(1)(c) justified detention during the second period [202 203]. However, a majority (Lord Sumption, Lady Hale, Lord Wilson, Lord Mance, Lord Hughes, Lord Neuberger and Lord Toulson) holds that there should be remitted for trial issues as to: whether intelligence exploitation was in fact the sole ground for detention during the second period or whether imperative reasons of security were not also a concurrent reason, justifying detention under article 5 read with and modified as necessary under Hassan having regard to the relevant Security Council resolution [89, 113, 191 200 and 223, 224], and whether SMs detention during the third period fell within article 5(1)(c) or was justified by imperative reasons of security under article 5 read with and modified as necessary having regard to the relevant Security Council resolution [83, 94 98, 111, 113, 204, 224, 235]. The question of article 5(3) compatibility must also be left for trial [83, 94 98, 111, 113, 204, 224, 235]. A majority holds that it is unnecessary for the United Kingdom to establish a right of detention under Afghan law in order to rely on article 5 read with and modified as necessary having regard to the relevant Security Council resolution [139, 202, 233 and 343 346]. Under article 5(4), the minimum standard of protection from arbitrariness equates to that imposed by articles 43 and 78 of the Fourth Geneva Convention: an impartial body carrying out initial and regular reviews in accordance with a fair procedure [68, 134, 205 206, 224, 235]. Fairness required that SM be given an effective means of challenging his detention. A majority (Lord Sumption, Lady Hale, Lord Wilson, Lord Hodge, Lord Reed and Lord Kerr) holds that there was a breach of the requirement to provide sufficient guarantees of impartiality and fairness to protect against arbitrariness in two respects: the procedure lacked independence and it failed to provide for the participation of the detainee [104 106, 144]. Lord Mance, Lord Hughes and Lord Neuberger consider that in both respects the matter should, as the Court of Appeal concluded, be remitted for trial in the light of the views they express [212 218, 227] The majority all agree that a finding of breach of the procedural standards required by article 5(4) will not necessarily entitle SM to damages. A different review process might well have led to no more than SM remaining in UK custody or being transferred slightly to Afghan custody. [110, 113, 219 220, 223, 224, 232, 235]. In his dissenting judgment, Lord Reed concludes that the Security Council resolutions cannot be interpreted as authorising detention falling outside article 5(1)(a) to (f) of the ECHR [296]. Moreover, Hassan v UK only operates to modify article 5 where it is necessary to reconcile it with the Third and Fourth Geneva Conventions (i.e. not in the case of a NIAC, to which the Geneva Conventions have only limited application). [315 316]. SMs detention during the second period was incompatible with article 5 as it was not for one of the six specified purposes [351]. In any event, it was for a purpose outside the scope of the authority granted by UNSCR 1890; it does not remain an open question whether it was for imperative reasons of security [352 353].
The Respondent, Lodvik Guraj, pleaded guilty to offences involving the supply of heroin and money laundering and was sentenced in July 2012 to a custodial sentence and forfeiture and deprivation orders. The judge did not make a confiscation order, but postponed the determination of that issue to take place after sentence, as he was entitled to do under s.14 Proceeds of Crime Act (POCA). The judge gave directions setting a procedural timetable for a hearing to be listed in November 2015, for the Respondent to provide certain information, and for the Crown first and then the Respondent to serve statements of case. The timetable slipped badly. The Respondent provided the required information late, the Crown did not serve its statement of case until over a year later, and two hearings were aborted due to the Crowns failure to be ready. There was eventually a properly attended hearing in May 2014, at which the Respondents counsel argued that the procedural requirements under POCA had been breached with the effect that the court no longer had jurisdiction to make a confiscation order. The Respondent argued there had been two procedural breaches. The first was that the judge had made forfeiture and deprivation orders before the confiscation proceedings, in breach of s.15(2) POCA which prohibits the court from dealing with the financial or property aspects of sentence (including deprivation and forfeiture orders) until after any confiscation proceedings have been concluded. The second breach was the Crowns failure to make an application for an extension of the postponement of the confiscation hearing, before the postponement expired by November 2012 (as required by s.14(8)). The judge accepted that there had been serious procedural error, but found that no unfairness had occurred as a result. It could not be Parliaments intention that any procedural error removed the courts jurisdiction to make an order. A confiscation order was made in a sum which had by then been agreed (subject to the jurisdiction point). The Court of Appeal took the opposite view on the jurisdiction point and quashed the confiscation order. The Crown now appeals and seeks the quashed order to be restored. The Supreme Court unanimously allows the Crowns appeal. Lord Hughes gives judgment, with which the rest of the Court agrees. The judges approach was correct. No unfairness had arisen in consequence of the irregularities which occurred, and there was no obstacle to the making of the confiscation order. S.14 allows confiscation proceedings to be postponed until after sentence, for up to two years from conviction. Postponement may be applied for by the parties or may be granted by the court of its own motion. If there is a defect in procedure relating to postponement, as there was here, s.14(11) states that this alone is not sufficient to require a confiscation order made in the defective proceedings to be quashed. However, s.14(12) dis applies s.14(11) where, before the making of a confiscation order, an order has been made which s.15(2) says should not precede a confiscation order. Therefore, where forfeiture and deprivation orders have been made prior to confiscation proceedings, the rule under s.14(11) that the court is not prevented from making confiscation orders solely because there was a defect in the procedure relating to postponement does not apply [9 14]. The issue in this case was whether the dis application of the s.14(11) bar had the consequence that a confiscation order made with a defect in postponement procedure must always be quashed [18]. Preceding the insertion of ss.14(11) and (12), there was some uncertainty as to the legal consequences of procedural errors. This was clarified in R v Soneji [2005] UKHL 49 and R v Knights [2005] UKHL 50, where the House of Lords held that the dominant purpose of POCA was to make confiscation the duty of the court. It would defeat the purpose of the confiscation legislation if orders were treated as bad simply because there had been a failure to comply with procedural provisions laid down for postponement. The correct approach was to question whether the duty to make a confiscation order was removed by procedural errors which caused injustice or unfairness to the defendant. S.14(11) was subsequently introduced with the effect of clarifying on a statutory footing that a procedural error in postponement does not on its own invalidate the confiscation procedure [15 17]. The trial judges interpretation of the operation of ss.14(11) and (12) was correct. The fact that the bar to quashing a confiscation order in certain circumstances is dis applied does not give rise to a requirement to quash in those circumstances. Where s.14(11) applies, no procedural defect relating to postponement can on its own justify quashing. Where s.14(11) does not apply, and there has been a procedural defect relating to postponement, an order may be quashed, or it may not be. Applying the principles in Soneji and Knights, the position is that a procedural defect (not limited to postponement) will have the effect of making it wrong to make a confiscation order if unfairness to the defendant would ensue. If the order would give rise to no unfairness, or to none that cannot be cured, there can be no obstacle to making the order [19 30]. The Court of Appeals interpretation of s.14(12) as prescribing that an order will be invalidated for a procedural error was wrong. This was to approach s.14(11) as if it restored the position to the pre Soneji case law which regarded procedural errors as going to the courts jurisdiction to make a confiscation order. The trial judge was correct to continue to follow the Soneji and Knights approach, whose analysis still holds good [22]. That unfairness has resulted from a procedural error may be inferred in the event of a very long period of inactivity. If the statutory permitted two year postponement period is exceeded without there being exceptional circumstances, it is likely that unfairness will ensue. The present case is one where it has been accepted that the Respondent cannot point to unfairness, injury or injustice resulting from the making of the order after the prescribed timetable. The order was eventually made within the permitted period of two years. As there is no unfairness in this case, the question of curable unfairness does not arise. If it were to arise, a potential unfairness might be cured for example by adjusting a confiscation order, or by quashing a forfeiture order. Each case must depend on its own facts [31 4].
The Respondents were each detained as suspects for questioning at a police station under sections 14 and 15 of the Criminal Procedure (Scotland) Act 1995. Their detentions took place prior to the decision of this Court in Cadder v HM Advocate [2010] UKSC 43, and they did not have access to legal advice either before or during their police interviews. In the course of their interviews, they each made statements which were later relied on by the Crown at their trials. They were convicted and sentenced to various periods of imprisonment. They appealed, and their appeals were still current when the judgment in Cadder was delivered on 26 October 2010. The Respondents argued, on the basis of Cadder, that the leading of evidence of the statements they made during their police interviews was a breach of their rights under Articles 6(3)(c) and 6(1) of the European Convention on Human Rights, and that, in terms of section 57(2) of the Scotland Act 1998, the Lord Advocate had no power to lead that evidence. For Birnie, it was also submitted that the reliance by the Crown upon his admissions in these circumstances deprived him of a fair trial, to which he was entitled under Article 6(1) and at common law. The Crowns objections to the devolution issue were repelled by the Appeal Court, and the Crown appealed to the Supreme Court. On the question whether the Respondents had waived their right to legal assistance, the Lord Advocates position before the Court was that the important point in these appeals was that raised in the case of Birnie. Unlike the other two Respondents, Birnie made an unsolicited statement following his police interview, having declined the opportunity to have access to a lawyer prior to and while making it. The outstanding matters before the Court were therefore: (i) Whether the time bar referred to in section 100(3B) of the Scotland Act 1998, as amended, applies to Judes appeal; (ii) Whether Birnie waived his right of access to a lawyer when he made his unsolicited statement following his police interview; and (iii) Whether the reliance by the Crown upon the appellants admissions in these circumstances deprived him of his right to a fair trial under Article 6(1) of the Convention. The Supreme Court unanimously dismisses the Crowns appeal on the question whether section 100(3B) of the Scotland Act 1998 applies in this case. It unanimously dismisses its appeals on the issue as to waiver in regard to the police interviews of Jude and Hodgson. By a majority of 4 1, it allows the appeal on the question whether it was incompatible with Birnies right to a fair trial for the Crown to lead and rely on the evidence of the statement which he made following his police interview and remits that matter for determination by the High Court of Justiciary. Lord Hope gives the leading judgment. Lord Kerr gives a partly dissenting judgment. (1) Time Bar: Section 100(3B) of the Scotland Act 1998, as amended, provides that any proceedings brought on the ground that an act of a member of the Scottish Executive is incompatible with the Convention rights must be brought before the end of the period of one year beginning with the date on which the act complained of took place. The question is whether that section applies to proceedings brought by way of an appeal under the 1995 Act [6]. A criminal appeal under section 57(2) of the Scotland Act falls plainly into the category of a proceeding that is by virtue of the Scotland Act. The fact that the procedure under which the complaint is made is provided by the 1995 Act is irrelevant as far as this point is concerned, and does not render section 100(3B) inapplicable[13]. The opening subsection of section 100 makes the same distinction as that found in section 7(1) of the Human Rights Act 1998 between bringing proceedings on the basis of Convention rights, and relying on Convention rights in any such proceedings. The wording is not exactly the same in the two Acts, but the assumption is that they have the same effect. The time bar under section 7(5) of the Human Rights Act refers only to proceedings under section (1)(a) and not those under (1)(b) [15]. The time bar in section 100(3B) has the same effect, so it does not apply to proceedings of the kind referred to in section 100(1)(b). The point is that proceedings under that section are proceedings that have been brought by someone other than the person who maintains that the act in question is incompatible with the Convention rights [16]. An appeal against conviction or sentence is still part of the prosecution process that has been brought by the Lord Advocate [17]. Further, the 1995 Act contains its own system of time limits for the bringing of appeals. It would be very odd if an appeal were subject to two different time limits under two different Acts [18]. (2) Waiver. Birnie was offered rights of access to a solicitor before he made his statement and was also asked whether he wished to have a solicitor present while he was making it. He expressly declined both offers [26]. There is no absolute rule that the accused must have been given legal advice on the question whether or not he should exercise his right of access to a lawyer before he can be held to have waived it: see McGowan (Procurator Fiscal, Edinburgh) v B [2011] UKSC 54 [28]. It was not suggested in the course of argument that an absolute rule requiring reasons for the accuseds decision to waive his right to legal assistance is to be found in the jurisprudence of the Strasbourg court. The only question for this Court is whether the absence of such an inquiry amounted in itself to a breach of a Convention right. It is not for the Supreme Court to say how the law and practice respecting crimes should be developed by the common law in Scotland. The fact that the waiver was made without legal advice and without reasons being requested may be taken into account in the assessment as to whether Birnie understood the right that was being waived. But Strasbourg does not require the Court to hold that it would necessarily be incompatible with Article 6 to rely on statements made to police just because it was not ascertained why the suspect did not want to speak to a lawyer. The Strasbourg court has been careful, in general, to leave the national authorities to devise a more Convention compliant system without itself imposing specific requirements on the State. The Supreme Court should be no less careful in the way that it deals with Scottish criminal law and procedure [29]. There is room for argument as to whether Birnies statement was truly voluntary and in any event whether, taking all the circumstances into account, it was fair to admit this evidence. The question of overall fairness for the purposes of Article 6(1) must be examined in the light of all the facts and circumstances, and is therefore a matter for determination by the High Court of Justiciary [33]. For Lord Kerr, it is an indispensable prerequisite that there must be some means of ascertaining the reason that the right to legal assistance has been waived [53]. On the available evidence, it has not been established that there was an effective waiver by Birnie of his right to legal assistance [57].
On 6 February 2007, police officers carried out observations on the Appellant from about 0835 hours to about 1200 hours. He was seen leaving his car and entering the block of flats in which he lived, leaving the block carrying a bag and entering a car which then drove off. He was observed leaving various other locations and cars in Glasgow and then, carrying a bag which appeared to be heavy, entering a taxi which was later seen parked outside his brothers home. The police approached the taxi, and the Appellant and his brother were detained. Various searches were carried out and large sums of money were recovered by the police. On 16 December 2010 the Appellant was found guilty on indictment in Glasgow Sheriff Court of money laundering offences [3 4]. At a preliminary stage, the Appellant had lodged a devolution minute. He referred to article 8 of the European Convention on Human Rights which provides that everyone has a right to respect for his private life. He argued that the police had acted unlawfully because they had failed to obtain authorisation to conduct covert surveillance on him and his associates under the Regulation of Investigatory Powers (Scotland) Act 2000 (the 2000 Act). He asked the Court to hold that the surveillance was unlawful and that the fruits of that surveillance were inadmissible as evidence. The Sheriff refused the devolution minute and refused leave to appeal his decision [3 and 5 6]. Following his conviction, the Appellant appealed to the High Court of Justiciary. His first ground of appeal was that the Sheriff should have granted leave to appeal his decision to refuse the devolution minute. The Appellant conceded that the Sheriff was bound by the decision of the Appeal Court in Gilchrist v HM Advocate (which he said the prosecution had relied on when opposing the devolution minute) but argued that it was wrongly decided. The Appellants second ground of appeal was that the trial Sheriff was wrong to reject his no case to answer submission [8]. In the High Court of Justiciary, leave to appeal on both grounds was refused at the first and second sifts. On 2 November 2011 the Appeal Court, having heard counsel for the Appellant and without giving reasons, granted leave to appeal to the Supreme Court [9]. The parties agreed that the issue whether the observations of the police breached the Appellants rights under article 8 arose in the appeal to the Supreme Court. The Appellant maintained that the issue whether the act of leading that evidence was incompatible with the Appellants rights under article 8 and article 6 (to a fair trial) and therefore unlawful under the Scotland Act 1998 also arose, but the Respondent did not accept this [10]. The Supreme Court unanimously dismisses the appeal. There has been no interference with the Appellants rights under articles 8 and 6 of the Convention. The judgment is given by Lord Hope with whom all the other Justices agree. Taking it on its own terms, the devolution minute did not appear to raise a devolution issue at all. The question of whether the police acted in a way that is incompatible with the Appellants Convention rights is not a devolution issue. The only relevant devolution issue would have been whether the act of the Lord Advocate in leading the surveillance evidence would have been incompatible with the Appellants Convention rights. But no mention of the issue whether the Lord Advocate leading such evidence would have breached the Appellants article 6 right was made or appears to have been considered at any stage of the proceedings. There was no determination of the issue in the High Court of Justiciary because the question it raises was not before it. In terms of the Scotland Act 1998, the Supreme Court does not have an original jurisdiction in these matters. Except in regard to devolution issues as defined in the Scotland Act 1998, every order of the High Court of Justiciary is final and conclusive and not subject to review by any court whatsoever [11 13]. The proper course might well have been to dismiss this appeal as incompetent. But, with considerable hesitation, the Court decided that it should hear argument on the issue for three reasons in particular. First, the prosecution did not oppose the Appellants motion for leave to appeal to the Supreme Court. Second, the Appeal Court gave leave to appeal to the Supreme Court. Third, the Appellant was really seeking to re examine the correctness of the decision in the Gilchrist case (which was that surveillance evidence obtained without a valid 2000 Act authorisation led by the Lord Advocate was admissible). However, the Supreme Courts decision to hear the appeal should not be taken as an indication that it is not aware of the limits to its jurisdiction, or of its responsibility to ensure that those limits are respected. Devolution minutes should say what they mean [14]. Any breach of article 8 in obtaining the surveillance evidence in this case was due to acts of the police, not the Lord Advocate. The fact that evidence is irregularly obtained because there is no authorisation under the 2000 Act does not of itself make that evidence inadmissible at common law. Nor does the fact that the evidence is obtained in breach of article 8 necessarily mean that it would be incompatible with article 6 for that evidence to be led at the trial. Nevertheless, the key to the whole argument lies in what one makes of the underlying article 8 issue [15 17]. The Strasbourg Court has not yet considered the situation where a persons movements in a public place are noted down by the police as part of their investigations when they suspect the person of criminal activity. But it could not reasonably be suggested that a police officer who came upon a person who has committed a crime in a public place and simply noted down his observations in his notebook was interfering with the persons article 8 right. In this case, notes of the Appellants movements in public were kept by the police over a period of hours in a covert manner as part of a planned operation. However, there is nothing to suggest that the Appellant could reasonably have had any expectation of privacy. He engaged in his activities in places where he was open to public view by neighbours, by persons in the street or by anyone else who happened to be watching what was going on. He took the risk of being seen and of his movements being noted down. The criminal nature of what he was doing, if that was what it was found to be, was not an aspect of his private life that he was entitled to keep private [20 21]. On the first issue in the appeal, there are no grounds for holding that the actions of the police amounted to an infringement of the Appellants rights under article 8. It is plain that the absence of a reasonable expectation of privacy was the basis for the decision in Gilchrist, which was rightly decided in this respect. On the second issue in the appeal, it follows that there has been no breach of article 6, since the only ground for arguing this was that there had been a breach of article 8 [21 22].
The case concerns whether Zipvit, a trader selling vitamins and minerals by mail order, is entitled when accounting for VAT on its sales to make deductions of input VAT (the tax paid by the trader on goods and services purchased in connection with its business, as opposed to output VAT, which is the tax charged to the consumer by the trader on its goods or services) in respect of the price of postal services supplied to it by Royal Mail. Under Royal Mails terms and conditions, Zipvit was required to pay the commercial price for the supply plus such amount of VAT (if any) as was chargeable. At the time of supply, both Royal Mail and Zipvit understood that the supply was exempt from VAT, so Zipvit was only charged and only paid a sum equal to the commercial price for the supply. Royal Mails invoices treated the supplies as exempt. However, the Court of Justice of the European Union (the CJEU) subsequently held that such a supply of individually negotiated mail services should in fact have been treated as standard rated for VAT. If that had been appreciated at the time of the supplies, Royal Mail would have charged Zipvit VAT and would have accounted for this to HM Revenue and Customs (HMRC). The present proceedings are a test case in respect of supplies of services by Royal Mail where the same mistake was made. Zipvit now claims that under article 168(a) of the Principal VAT Directive (2006/112/EC) (the Directive) it is entitled to deduct as input VAT the VAT due in respect of these supplies or a VAT element deemed by law to be included in the price paid to Royal Mail for each supply. HMRC contend that on the proper interpretation of the Directive: (a) there was no VAT due or paid for the purposes of the Directive; and/or (b) since Zipvit at no point held invoices which showed that VAT was due and its amount, in compliance with article 226(9) and (10) of the Directive, Zipvit is not entitled to recover input tax. Zipvit made claims to HMRC for the deduction of input VAT, which were rejected by HMRC. Zipvit appealed against HMRCs decision to the First tier Tribunal (Tax Chamber), which dismissed the appeal. Zipvit appealed to the Upper Tribunal (Tax Chamber), which dismissed the appeal. The Court of Appeal dismissed Zipvits appeal from the Upper Tribunal. Zipvit now appeals to the Supreme Court. The Supreme Court unanimously decides that the legal position under the Directive is not clear. It is common ground that at this stage in the process of the UKs withdrawal from the EU, in a case involving an issue of EU law which is unclear, the Supreme Court is obliged to refer that issue to the CJEU to obtain its advice on the point. Therefore, the Supreme Court makes an order for a reference and sets out the questions for the CJEU. Lord Briggs and Lord Sales give the judgment, with which all other members of the Court agree. Zipvit appealed on two issues: first, the due or paid issue, and second, the invoice issue. The Court has decided that neither issue can be regarded as acte clair (so obvious as to leave no scope for any reasonable doubt) and that a reference should be made to the CJEU. The due or paid issue arises out of article 168(a) of the Directive, which provides that a trader who is a taxable person has an entitlement to deduct from VAT which he is liable to pay the VAT due or paidin respect of supplies to him of goods or services, carried out or to be carried out by another taxable person. Zipvit contends that the commercial price it paid Royal Mail for the supplies of postal services must be treated as having contained an element of VAT, even though the invoice purported to say that the services were exempt from VAT. Alternatively, even if this embedded element of VAT is not to be regarded as having been paid, it should be regarded as being due [26] [27]. HMRC contend that there is nothing in the Directive which requires or justifies retrospective re writing of the commercial arrangements between Royal Mail and Zipvit. Royal Mail did not issue further invoices to demand payment of VAT, cannot be compelled to issue such further invoices, and has not accounted to HMRC for any VAT in respect of the services. HMRC could not take action to compel Royal Mail to account for any VAT in respect of the supply of services. As the courts below found, if Zipvit were to succeed it would gain an unmerited financial windfall at the expense of the taxpayer [31] [32]. On the invoice issue, Zipvit submits that CJEU case law indicates that there is an important difference between the substantive requirements to be satisfied for a claim for input tax and the formal requirements that apply in relation to such a claim. The approach is strict for the substantive requirements, but departure from the formal requirements is permissible if alternative satisfactory evidence of the VAT which was paid or is due can be produced. Zipvit contends that it has produced alternative satisfactory evidence of the VAT paid, in the form of payment of the embedded VAT [36] [38]. Against this, HMRC submit that the regime in the Directive requires particular importance to be attached to the requirement of the production of an invoice showing that VAT is due and in what amount. A valid claim for the deduction of input tax cannot be made in the absence of a compliant VAT invoice [39] [40]. The Supreme Court refers four questions to the CJEU. The first asks whether, in circumstances like those of Zipvit, the effect of the Directive is that the price actually paid by the trader is to be regarded as the combination of a net chargeable amount plus VAT thereon, thus allowing the trader to claim to deduct input tax under article 168(a) of the Directive in the amount of VAT which was in fact so paid by it in respect of that supply [42(1)]. The second asks whether, in those same circumstances, the trader can claim to deduct input tax under article 168(a) as VAT which was due in respect of that supply [42(2)]. The third asks whether, where a tax authority, the supplier, and the trader misinterpret European VAT legislation and treat a taxable supply as exempt, resulting in a non compliant VAT invoice which stated that no VAT was due, the trader is entitled to claim to deduct input tax under article 168(a) [42(3)]. Finally, in answering the prior three questions, the Court asks whether it is relevant to investigate whether the supplier (Royal Mail) would have a defence to any attempt by the tax authority to issue an assessment requiring it to account for a sum representing VAT in respect of the supply, and whether it is relevant that the trader (Zipvit) knew at the same time as the tax authority and the supplier that the supply was not in fact exempt, and could have offered to pay the VAT due, but omitted to do so [42(4)].
GAFTA Form 49 is the standard form of FOB sale contract of the Grain and Feed Trade Association for goods delivered from central or eastern Europe in bulk or bags. (Under an FOB sale contract, the seller agrees to pay to deliver the goods free on board the purchasers chosen transporter.) This appeal is about the Default Clause in GAFTA 49, which provides in part: DEFAULT In default of fulfilment of contract by either party, the following provisions shall apply: (a) The party other than the defaulter shall, at their discretion have the right, after serving a notice on the defaulter, to sell or purchase, as the case may be, against the defaulter, and such sale or purchase shall establish the default price. (b) If either party be dissatisfied with such default price or if the right at (a) above is not exercised and damages cannot be mutually agreed, then the assessment of damages shall be settled by arbitration. (c) The damages payable shall be based on, but not limited to, the difference between the contract price and either the default price established under (a) above or upon the actual or estimated value of the goods, on the date of default, established under (b) above. Nidera BV, the buyers, entered into a contract with Bunge SA, the sellers, under which they agreed to buy 25,000 tonnes (+/ 10% in buyers option) of Russian milling wheat crop 2010, FOB Novorossiysk (a port in Russia). The shipment period was 23 30 August 2010. The contract incorporated GAFTA 49. On 5 August 2010, Russia introduced a legislative embargo on exports of wheat from its territory, which was to run from 15 August to 31 December 2010. The sellers notified the buyers of the embargo on 9 August 2010 and purported to declare the contract cancelled under GAFTA 49s Prohibition Clause. The buyers maintained that the sellers were not yet entitled to cancel the contract and treated the purported cancellation as a repudiation of the contract, which they accepted on 11 August 2010. The sellers offered to reinstate the contract on the same terms on 12 August 2010, but the buyers refused and began arbitration proceedings under the GAFTA rules. At the arbitration, the parties agreed that the Default Clause applied to anticipatory repudiation, that the buyers had not bought against the sellers pursuant to sub clause (a), that the date of default for the purpose of sub clause (c) was 11 August 2010, and that the difference between the contract and the market price at that date was US$3,062,500. The sellers no longer dispute that they were in breach of contract and the only issue between the parties concerns damages. The first tier panel held that the buyers were not entitled to a substantial damages award, because the embargo was still in place when the time for delivery arrived and so the contract would have been cancelled in any event. The GAFTA Appeal Board accepted that the contract would have been cancelled in any event, but held that the buyers were entitled under sub clause (c) of the Default Clause to a damages award of US$3,062,500, reflecting the difference between the contract price and the market price on the agreed date of default. The sellers appeal was dismissed by both Hamblen J and the Court of Appeal. The Supreme Court unanimously allows the appeal. Lord Sumption gives the leading judgment. Lord Toulson agrees with Lord Sumption but sets out his reasoning in his own words. Lord Neuberger, Lord Mance and Lord Clarke all agree with both Lord Sumption and Lord Toulson. Lord Sumption first addresses the position at common law. The fundamental principle of the common law of damages is the compensatory principle. The cases have addressed two questions relating to the calculation of damages at common law following an anticipatory breach. First, where there is an available market, the prima facie measure of damages is the difference between the contract price and the market price of the goods at the time when they ought to have been delivered, unless the buyer should have mitigated by going into the market and entering into a substituted contract at some earlier stage, in which case damages will be assessed with reference to the market price at that earlier date. Once the relevant market price has been determined, any subsequent change in the market price is irrelevant. Second, as the House of Lords explained in The Golden Victory [2007] 2 AC 353, it will be relevant to take account of contingencies other than a change in the market price if subsequent events known at the time of the judges or arbitrators assessment would have reduced the value of performance, perhaps to nothing, even without the defaulters renunciation. This principle applies to contracts both for the supply of goods or services over a period of time and for one off sales. [12 23] Lord Sumption then considers the effect of the GAFTA 49 Default Clause. Damages clauses are not necessarily intended to be complete codes for the assessment of damages. In this case, the clause applies in default of contractual performance, and to anticipatory breach as well as actual breach. The combined effect of sub clauses (a), (b) and (c) of the Default Clause differs from the common law, first, in that they give the injured party a discretion about whether to go into the market to buy or sell against the defaulter, so that damages are required to be assessed as at the date when the injured party accepted an anticipatory repudiation only if he actually went into the market to fix a price at that date; and, secondly, in that they provide that the relevant comparator is the actual or estimated value of the goods rather than their market price. [24 28] The Default Clause provides a complete code for determining the market price or value of the goods that falls to be compared with the contract price. It does not deal at all with the effect of subsequent events which would have resulted in the original contract not being performed in any event, to which the common law still applies. [29 31] Applying The Golden Victory to this case, the buyers in fact lost nothing and should receive only nominal damages in the sum of US$5. [35 36] Lord Toulson agrees with Lord Sumption and holds that the language of the Default Clause is not sufficiently clear to preclude the application of The Golden Victory. The most reasonable interpretation is that the clause is concerned only with placing a mathematical value on the goods, assuming the contract to be capable of performance, and that it is not intended to oust the application of ordinary common law principles where that assumption is inappropriate. He also rejects the argument that the clause precludes the operation of the common law mitigation of loss principle, but the application of this principle was not the subject of argument before this court. [58 62] Where a contract is discharged by reason of one partys breach, and there exists an available market in which the innocent party could obtain a substitute contract, the innocent partys loss will ordinarily be measured by the extent to which his financial position would be worse off under the substitute contract than under the original contract. Whether the innocent party in fact enters into a substitute contract is a separate matter. The assessment should be made on the facts as known at the date of the assessment, as the House of Lords held in The Golden Victory and consistently with the fundamental compensatory principle. The Golden Victory applies to one off sales. [63 89]
Under section 188 of the Housing Act 1996 (the 1996 Act) local authorities have a duty to secure that accommodation is made available for applicants who are homeless and have priority need. Priority need is defined in section 189(1) of the 1996 Act and includes at paragraph (c) persons who are vulnerable as a result of old age, mental illness or handicap or physical disability or other special reason, or with whom such a person resides or might reasonably be expected to reside. The Appellants applied for accommodation on the basis that they had priority need. The First Appellant has very significant learning difficulties and symptoms of depression and PTSD. He is cared for by his brother. Southwark Borough Council (Southwark) refused his application on the grounds that, if homeless, he would be provided with the necessary support by his brother. The Second Appellant has multiple physical problems as well as psychotic symptoms and suicidal ideation. He was deemed by Southwark not to be in priority need because he would not be at a greater risk of injury or detriment than an ordinary street homeless person due to the ability of his wife and son to fend for the whole household. The Third Appellant claimed to be vulnerable because he had become addicted to heroin while in prison and was in poor physical and mental health. Solihull Metropolitan Borough Council (Solihull) found that he was not in priority need on the basis that he would not be less able to fend for himself than an ordinary homeless person. The First and Third Appellants were unsuccessful in the courts below. The Second Appellant succeeded in the County Court but lost in the Court of Appeal. Three issues arise in the present appeal: (1) Does the assessment of whether an applicant is vulnerable for the purposes of section 189(1)(c) of the 1996 Act involve an exercise in comparability, and, if so, by reference to which group of people is vulnerability to be determined? (2) When assessing vulnerability, is it permissible to take into account the support which would be provided by a family member to an applicant if he were homeless? (3) What effect, if any, does the public sector equality duty under section 149 of the Equality Act 2010 (the 2010 Act) have on the determination of priority need under the 1996 Act in the case of an applicant with a disability or any other protected characteristic? Lord Neuberger (with whom Lord Clarke, Lord Wilson and Lord Hughes agree) dismisses the First Appellants appeal, but Lady Hale would have allowed his appeal. All five Justices allow the Second Appellants appeal and dismiss the Third Appellants appeal. On the first issue in the appeal, vulnerable in section 189(1)(c) connotes that the applicant must be significantly more vulnerable than an ordinary person who happened to be in need of accommodation [55, 59]. The decisions of the Court of Appeal on this issue have all accepted that vulnerability has to be assessed comparatively [48] [50]. This is correct; vulnerable carries a necessary implication of relativity. It can fairly be said that anyone who is homeless is vulnerable. So, it follows that section 189(1)(c) must contemplate homeless people who would be more vulnerable than many others in the same position [51]. Parliament probably did not intend vulnerability to be judged by reference to what a housing officer thought to be the situation of an actual homeless person. Such an approach would be more likely to lead to arbitrary and unpredictable outcomes. The comparator could not be an ordinary homeless person in the area of the relevant authority as this could lead to unacceptable outcomes with vulnerable people being put out on the streets [56]. The 1996 Act does not refer to street homeless as a category or distinguish between the situations which may constitute homelessness; this calls into question the authority making use of the term in assessing their duty to an applicant [42] As to the relevance of support from family members, an applicants vulnerability under section 189(1)(c) has to be assessed by reference to his situation if and when homeless, which involves a contextual and practical assessment of the applicants physical and mental ability when homeless. As such, any services and support that would be available to the applicant if he were homeless must be taken into account [62]. This conclusion is supported by the purpose of the legislation in question. Those who are more vulnerable in practice if they are homeless can be expected to receive priority treatment. It would be contrary to common sense to ignore any aspect of the actual or anticipated factual situation when assessing vulnerability [63]. It does not matter whether the support is provided pursuant to a legal obligation, but housing authorities can only take third party support into account where they are satisfied that it will be provided on a consistent and predictable basis [65]. The primary focus of section 189(1)(c) is on the applicant, not the benefit of the third party and it would place an excessive burden on housing authorities if family support were disregarded. However, the mere fact that support is available does not of itself prevent the applicant from being vulnerable; there must be a case specific analysis of whether the support can obviate the vulnerability [69] [70]. On the third issue in the appeal, the weight and extent of the public sector equality duty are highly fact sensitive and dependent on individual judgment [74]. The authoritys equality duty was complementary to its duty under the 1996 Act. Each stage of the decision must be made with the equality duty well in mind and the officer must focus very sharply on: (i) whether the applicant has a relevant protected characteristic, (ii) its extent, (iii) its likely effect, when taken together with any other features, on the applicant if and when homeless, and (iv) whether the applicant is vulnerable as a result [78]. Lady Hale would have allowed the First Appellants appeal. She concludes that, while any statutory services which will be available to an applicant should be taken into account when assessing his vulnerability, family support should not [93]. It is not consistent with the intention of the statute to take into account help which may be available from other members of the household. Both the vulnerable person and their non vulnerable family member qualify as being in priority need. The 1996 Act permits the non vulnerable family to apply on behalf of both themselves and the vulnerable person. Parliament did not intend applications to be made by a family member who was not looking after the vulnerable person [95]. There is House of Lords authority for this proposition and none for the suggestion that the existence of a carer within the same household can mean that a person who is otherwise obviously vulnerable is not to be treated as such [99].
E challenged JFSs (formerly the Jews Free School) refusal to admit his son, M, to the school. JFS is designated as a Jewish faith school. It is over subscribed and has adopted as its oversubscription policy an approach of giving precedence in admission to those children recognised as Jewish by the Office of the Chief Rabbi of the United Hebrew Congregation of the Commonwealth (the OCR). The OCR only recognises a person as Jewish if: (i) that person is descended in the matrilineal line from a woman whom the OCR would recognise as Jewish; or (ii) he or she has undertaken a qualifying course of Orthodox conversion. E and M are both practising Masorti Jews. E is recognised as Jewish by the OCR but Ms mother is of Italian and Catholic origin and converted to Judaism under the auspices of a non Orthodox synagogue. Her conversion is not recognised by the OCR. Ms application for admission to JFS was therefore rejected as he did not satisfy the OCR requirement of matrilineal descent. E challenged the admissions policy of JFS as directly discriminating against M on grounds of his ethnic origins contrary to section 1(1)(a) of the Race Relations Act 1976 (the 1976 Act). Alternatively, E claimed that the policy was indirectly discriminatory. The High Court rejected both principal claims. The Court of Appeal unanimously reversed the High Court, holding that JFS directly discriminated against M on the ground of his ethnic origins. JFS appealed to the Supreme Court. The United Synagogue also appealed a costs order made against it by the Court of Appeal. The Supreme Court has dismissed the appeal by The Governing Body of JFS. On the direct discrimination issue, the decision was by a majority of five (Lord Phillips, Lady Hale, Lord Mance, Lord Kerr and Lord Clarke) to four (Lord Hope, Lord Rodger, Lord Walker and Lord Brown). The Majority held that JFS had directly discriminated against M on grounds of his ethnic origins. Lords Hope and Walker in the minority would have dismissed the appeal on the ground that JFS had indirectly discriminated against M as it had failed to demonstrate that its policy was proportionate. Lords Rodger and Brown would have allowed JFSs appeal in its entirety. The Supreme Court unanimously allowed in part the United Synagogues appeal on costs. The Majority Judgments The judgments of the Court should not be read as criticising the admissions policy of JFS on moral grounds or suggesting that any party to the case could be considered racist in the commonly understood, pejorative, sense. The simple legal question to be determined by the Court was whether in being denied admission to JFS, M was disadvantaged on grounds of his ethnic origins (or his lack thereof) (paras [9], [54], [124] and [156]). Direct Discrimination General Principles In determining whether there is direct discrimination on grounds of ethnic origins for the purposes of the 1976 Act, the court must determine, as a question of fact, whether the victims ethnic origins are the factual criterion that determined the decision made by the discriminator (paras [13], [16], [20] and [62]). If so, the motive for the discrimination and/or the reason why the discriminator considered the victims ethnic origins significant is irrelevant (paras [20], [22], [62] and [142]). Where the factual criteria upon which discriminatory treatment is based are unclear, unconscious or subject to dispute the court will consider the mental processes of the discriminator in order to infer as a question of fact from the available evidence whether there is discrimination on a prohibited ground (paras [21], [64], [115] and [133]). It is only necessary to consider the mental processes of the discriminator where the factual criteria underpinning the discrimination are unclear (para [114]). To treat an individual less favourably on the ground that he lacks certain prescribed ethnic origins constitutes direct discrimination. There is no logical distinction between such a case and less favourable treatment predicated upon the fact that an individual does possess certain ethnic origins (paras [9] and [68]). Direct discrimination does not require that the discriminator intends to behave in a discriminatory manner or that he realises that he is doing so (para [57]). There is no need for any consideration of mental processes in this case as the factual criterion that determined the refusal to admit M to JFS is clear: the fact that he is not descended in the matrilineal line from a woman recognised by the OCR as Jewish. The subjective state of mind of JFS, the OCR and/or the Chief Rabbi is therefore irrelevant (paras [23], [26], [65], [78], [127], [132], [136], [141] and [147] [148]). The crucial question to be determined is whether this requirement is properly characterised as referring to Ms ethnic origins (paras [27], [55] and [65]). Application in This Case The test applied by JFS focuses upon the ethnicity of the women from whom M is descended. Whether such women were themselves born as Jews or converted in a manner recognised by the OCR, the only basis upon which M would be deemed to satisfy the test for admission to JFS would be that he was descended in the matrilineal line from a woman recognised by the OCR as Jewish (para [41] per Lord Phillips). It must also be noted that while it is possible for women to convert to Judaism in a manner recognised by the OCR and thus confer Orthodox Jewish status upon their offspring, the requirement of undergoing such conversion itself constitutes a significant and onerous burden that is not applicable to those born with the requisite ethnic origins this further illustrates the essentially ethnic nature of the OCRs test (para [42] per Lord Phillips). The test of matrilineal descent adopted by JFS and the OCR is one of ethnic origins. To discriminate against a person on this basis is contrary to the 1976 Act (para [46] per Lord Phillips). The reason that M was denied admission to JFS was because of his mothers ethnic origins, which were not halachically Jewish. She was not descended in the matrilineal line from the original Jewish people. There can be no doubt that the Jewish people are an ethnic group within the meaning of the 1976 Act. While JFS and the OCR would have overlooked this fact if Ms mother had herself undergone an approved course of Orthodox conversion, this could not alter the fundamental nature of the test being applied. If Ms mother herself was of the requisite ethnic origins in her matrilineal line no conversion requirement would be imposed. It could not be said that M was adversely treated because of his religious beliefs. JFS and the OCR were indifferent to these and focussed solely upon whether M satisfied the test of matrilineal descent (paras [66] and [67] per Lady Hale). Direct discrimination on grounds of ethnic origins under the 1976 Act does not only encompass adverse treatment based upon membership of an ethnic group defined in the terms elucidated by the House of Lords in Mandla v Dowell Lee [1983] 2 AC 548. The 1976 Act also prohibits discrimination by reference to ethnic origins in a narrower sense, where reference is made to a persons lineage or descent (paras [80] [84] per Lord Mance). The test applied by JFS and the OCR focuses on genealogical descent from a particular people, enlarged from time to time by the assimilation of converts. Such a test is one that is based upon ethnic origins (para [86] per Lord Mance). This conclusion is buttressed by the underlying policy of the 1976 Act, which is that people must be treated as individuals and not be assumed to be like other members of a group: treating an individual less favourably because of his ancestry ignores his unique characteristics and attributes and fails to respect his autonomy and individuality. The UN Convention on the Rights of the Child requires that in cases involving children the best interests of the child are the primary consideration (para [90] per Lord Mance). The reason for the refusal to admit M to JFS was his lack of the requisite ethnic origins: the absence of a matrilineal connection to Orthodox Judaism (para [112] per Lord Kerr). Ms ethnic origins encompass, amongst other things, his paternal Jewish lineage and his descent from an Italian Roman Catholic mother. In denying M admission on the basis that he lacks a matrilineal Orthodox Jewish antecedent, JFS discriminated against him on grounds of his ethnic origins (paras [121] [122] per Lord Kerr). It might be said that the policy adopted by JFS and the OCR was based on both ethnic grounds and grounds of religion, in that the reason for the application of a test based upon ethnic origins was the conviction that such a criterion was dictated by Jewish religious law. The fact that the rule adopted was of a religious character cannot obscure or alter the fact that the content of the rule itself applies a test of ethnicity (paras [129] [131] per Lord Clarke). The fact that a decision to discriminate on racial grounds is based upon a devout, venerable and sincerely held religious belief or conviction cannot inoculate or excuse such conduct from liability under the 1976 Act (paras [35], [92], [113] and [119] [120]). Further Comments It is not clear that the practice based test adopted by JFS following the Court of Appeals judgment will result in JFS being required to admit children who are not regarded by Jewish by one or more of the established Jewish movements (para [50] per Lord Phillips). It may be arguable that an explicit exemption should be provided from the provisions of the 1976 Act in order to allow Jewish faith schools to grant priority in admissions on the basis of matrilineal descent; if so, formulating such an exemption is unquestionably a matter for Parliament (paras [69] [70] per Lady Hale). Indirect Discrimination As the case is one of impermissible direct discrimination it is unnecessary to address the claim of indirect discrimination (para [51] per Lord Phillips). Direct and indirect discrimination are mutually exclusive; both concepts cannot apply to a single case concurrently. As this case is one of direct discrimination it could not be one of indirect discrimination (para [57] per Lady Hale). Ex hypothesi, if the case was not direct discrimination, then the policy was indirectly discriminatory (para [103]). The policy pursued the legitimate aim of effectuating the obligation imposed by Jewish religious law to educate those regarded by the OCR as Jewish (paras [95] [96]). However, JFS had not, and on the basis of the evidence before the court could not, demonstrate that the measures it adopted, given the gravity of their adverse effect upon individuals such as M, were a proportionate means of pursuing this aim (paras [100] [103], [123] and [154]). The Minority Judgments Direct Discrimination In identifying the ground on which JFS refused to admit M to the school the Court should adopt a subjective approach which takes account of the motive and intention of JFS, the OCR and the Chief Rabbi (para [195] [197] per Lord Hope). In the instant case JFS, the OCR and the Chief Rabbi were subjectively concerned solely with Ms religious status, as determined by Jewish religious law. There is no cause to doubt the Chief Rabbis frankness or good faith on this matter (para [201] per Lord Hope). The availability of conversion demonstrates that the test applied is inherently of a religious rather than racial character (para [203] per Lord Hope). It is inapt to describe the religious dimension of the test being applied by JFS as a mere motive (paras [201] per Lord Hope; [227] per Lord Rodger). The appropriate comparator for M in this case is a child whose mother had converted under Orthodox Jewish auspices. The ground of difference in treatment between M and such a child would be that the latters mother had completed an approved course of Orthodox conversion (paras [229] [230] per Lord Rodger). Indirect Discrimination Lords Hope and Walker Clearly, children who were not of Jewish ethnic origin in the matrilineal line were placed at a disadvantage by JFSs admission policy relative to those who did possess the requisite ethnic origins (para [205]). JFSs policy pursued the legitimate aim of educating those regarded as Jewish by the OCR within an educational environment espousing and practising the tenets of Orthodox Judaism (para [209]). The 1976 Act placed the onus on JFS to demonstrate that in formulating its policy it had carefully considered the adverse effect of its policy on M and other children in his position and balanced this against what was required to give effect to the legitimate aim which it sought to further (para [210]). There is no evidence that JFS considered whether less discriminatory means might be adopted which would not undermine its religious ethos: the failure to consider alternate, potentially less discriminatory, admission policies means that JFS is not entitled to a finding that the means which it has employed are proportionate (paras [212] and [214]). Lords Rodger and Brown The objective pursued by JFSs admission policy educating those children recognised by the OCR as Jewish was irreconcilable with any approach that would give precedence to children not recognised as Jewish by the OCR in preference to children who were so recognised. JFSs policy was therefore a rational way of giving effect to the legitimate aim pursued and could not be said to be disproportionate. (para [233] per Lord Rodger; para [256] per Lord Brown). The United Synagogue Costs Appeal The United Synagogue must pay 20 per cent. of Es costs from the Court of Appeal but not those incurred in the High Court. The 20 per cent. of Es costs in the High Court previously allocated to the United Synagogue must be borne by JFS in addition to the 50 per cent. that it has already been ordered to pay (para [217]).
This appeal concerns the inter relationship between the sentencing provisions of the Criminal Justice Act 1991 (the 1991 Act) and the Criminal Justice Act 2003 (the 2003 Act). Under the 1991 Act it became mandatory for the Secretary of State to release prisoners part way through the period of their sentence. Home Detention Curfew (HDC) was introduced by the Crime and Disorder Act 1998, by which prisoners could be released on licence after they had served a requisite period. The 2003 Act replaced the sentencing regime for sentences over 12 months on 4 April 2005. However, its provisions governing early release for sentences of less than 12 months have never been brought into force. Thus when a prisoner was sentenced to consecutive sentences, including terms both over and under 12 months, it was necessary to have regard to the Criminal Justice Act 2003 (Commencement No 8 and Transitional and Savings Provisions) Order 2005 (the 2005 Order) to determine which scheme should apply. The appellant, Rebecca Noone, was sentenced on 23 May 2007 for five offences to (i) a term of 22 months imprisonment, (ii) three concurrent terms of four months imprisonment to be served consecutively, and (iii) a one month, also to be served consecutively. The policy of the Secretary of State was that her release on HDC should be calculated by treating the longest sentence as subject to the 2003 Act and as commencing first (because it was pronounced first by the sentencing judge), and the shorter sentences as subject to the 1991 Act. This produced an HDC date over three months later than the approach urged by the appellant, which combined all the sentences and took the half way point of the combined term as the conditional release date from which HDC was calculated. The High Court held that the policy of the Secretary of State was unlawful, but on appeal the Court of Appeal held that, although the matter could not be determined by policy, the policy had in fact correctly reflected the position in law. The Supreme Court unanimously allowed the appeal. The substantive judgments were given by Lord Phillips (President) and Lord Mance, with whose approaches the other Justices agreed. Lord Phillips stated that the interpretation of Paragraph 14 of the 2005 Order lay at the heart of the appeal. It raised the question, where sentences of under and over 12 months were ordered to be served consecutively, of how they were to be linked together and how the provisions as to early release on HDC and licence were to operate in relation to each sentence [para 22]. The approach of the Court of Appeal opened the door to the possibility of capricious results, placed a near intolerable burden on the sentencer and did not readily cater for the position where a series of sentences was imposed of which some were over and some were under 12 months (a mixed sentences case) [29]. The words in brackets in Paragraph 14 were drafted too economically to reflect the clear intention that a mixed sentences case was to receive different treatment from sentences of less than 12 months, namely that the 2003 Act should apply to it [para 33]. Consecutive sentences were subject to s 244(3)(d) of the 2003 Act, and it was necessary to refer to s 262(2) to identify the requisite custodial period. Although this did not define custodial period for sentences under 12 months, it was obvious that this was half the sentence that the prisoner would have had to serve before release, had his sentence not been imposed consecutively with an over 12 month sentence. The relevant custodial period was the amalgam of all the individual custodial periods [para 35] and from that eligibility for HDC could be determined [para 36]. Lord Mance observed that the Secretary of States approach meant that the transitional provisions, bringing the 2003 Act into force in many respects, but keeping the 1991 Act in force in some others, achieved a result which Parliament did not intend by either Act [para 60]. He did not agree that the wording of Paragraph 14 compelled it. In his view Paragraph 14 was clearly drafted to be limited to cases where the only sentences in the arena had a term of less than 12 months [para 70]. The draftsman must have overlooked the fact that it left a gap in the definition of the custodial period in a mixed sentences case, but it was clear what was intended to happen (by reference to the continued application of s 33(1)(a) of the 1991 Act to all cases with sentences under 12 months) namely that the provisions of s 263(3) and s 264(2) and (3) should apply [para 71]. Lord Judge deplored the fact that so much intellectual effort, as well as public time and resources, have had to be expended in order to discover a route through the legislative morass to what should be, both for the prisoner herself, and for those responsible for her custody, the prison authorities, the simplest and most certain of questions the prisoners release date [para 87].
The Appellant Mark Roberts and his brother John Roberts were beneficiaries of a will made by their grandmother, Mrs. Alice Roberts. A clause in the will provided that if John Roberts paid all the inheritance tax due on Mrs. Roberts death then a property known as the Lower Hellingtown Farm would pass to him, and another property known as the Coppice would pass to the Appellant. The considerable value of the farm meant that it would be to the advantage of John Roberts if he complied with the clause. Upon the death of Mrs Roberts on 27 July 1995, John Roberts was granted the right to administer the estate in the place of Mrs Roberts executors, who had decided not to take up office. In order to obtain his position as administrator, John Roberts paid some of the inheritance tax due on the estate, but not all of it. In July 1996, John Roberts, as administrator, transferred ownership of Lower Hellingtown Farm to himself as beneficiary and in 1997 the property was sold. The majority of the proceeds of sale were paid to John Roberts, the remainder being used to discharge some of the estates liabilities. Two firms of solicitors advised John Roberts. The First Respondent, Gill & Co, advised John Roberts on the transfer of the property and the Second Respondent, Whitehead Vizard, advised him on the sale of the farm. On 30 October 2000, John Roberts was replaced as administrator by the Appellants solicitor. In a claim brought on 27 November 2002, the Appellant brought proceedings against the First and Second Respondents for negligence, alleging broadly that they had assisted, in breach of the provisions in the will, in the transfer and sale of the property without John Roberts having paid all the inheritance tax due. The claim was, however, framed in such a way as to allege that the duty of care owed by the firms of solicitors was owed to the Appellant personally. The correct legal position (which was not disputed by any of the parties on appeal), was that a firm of solicitors advising a person administering an estate does not owe a duty of care to the beneficiaries of that estate personally; rather the duty of care is owed to the estate of the dead person. Normally the proper person to bring any claim for negligence, therefore, would be the person administering the estate. A beneficiary of a will may bring a claim on behalf of the estate, but only where special circumstances exist. On 25 August 2006, the Appellant applied to amend his claim so as to continue it both in his own personal capacity and on behalf of the estate. The First and Second Respondents resisted the application on the grounds (a) that the amendment was barred as being out of time under section 35 of the Limitation Act 1980 (the Act) and rule 19.5 of the Civil Procedure Rules (the CPR), and (b) that there were no special circumstances which entitled the Appellant, as a beneficiary, to continue the claim on behalf of the estate. The High Court refused the application, holding that there were no special circumstances. The Court of Appeal held by a majority that there were special circumstances but that the amendment was time barred. The Appellant appealed. The Supreme Court unanimously dismissed the appeal. Lord Collins gave the leading judgment, dismissing the appeal on the basis that the amendment was time barred. Lords Rodger and Walker agreed with the entirety of Lord Collins judgment. Lords Hope and Clarke declined to decide the case on the grounds that the amendment was time barred but nonetheless ruled in favour of the First and Second Respondents on the ground that there were no special circumstances which entitled the Appellant to carry on the claim on behalf of the estate. The main question in relation to ground (a) was whether, in order to be able to carry on his claim, the Appellant would need not only to alter the claim so that he was suing on behalf of the estate, but also to add the administrator as a defendant. If he did have to add the administrator, a further question arose: did he have to add him at the time at which he altered his claim, or could he do so later? [para 44]. Rule 19.5 of the CPR stated that a new party could be added after the limitation period only where to do so was necessary for the determination of the original litigation. The addition of the administrator was clearly not necessary for determining the Appellants personal claim: there was no possible basis for any suggestion that the administrator would be a proper or necessary party [para 43]. If the Appellant was able to make the application to change the capacity in which he sued first, that would then enable him to subsequently add the administrator as a party, as it would then be necessary for the determination of the proceedings brought on behalf of the estate for the administrator to be joined [para 44]. The Appellant therefore needed to be able to demonstrate either that the administrator did not need to be added at all, or that he could be added after the Appellant had successfully altered the claim. Neither was possible. The administrator needed to be added at the outset of the proceedings [paras 63, 71] and it would be contrary to principle for the court to grant permission to alter the claim first before considering the addition of the administrator [para 71]. The appeal would accordingly be dismissed on ground (a) [paras 77, 86, 95]. Although ground (b) did not directly arise for decision given the conclusion on ground (a), there were no special circumstances that would entitle the Appellant to carry on a claim on behalf of the estate. The judge had a wide latitude in evaluating whether there were special circumstances, had taken all the relevant circumstances into account, and had conducted the enquiry in a way with which an appellate court should not have interfered [para 76]. Lords Hope and Clarke, in the minority on ground (a), disagreed that the rule that the administrator must be joined was quite as absolute as Lord Collins suggested [paras 79, 115]. The rule could be departed from if it was necessary to avoid injustice [paras 84, 116]. While on the facts of the case it was difficult to justify a departure from the rule [para 84], Lords Hope and Clarke both preferred to decide the case on the basis that there were no special circumstances [paras 78, 114].
Mr McDonald (the respondent) worked as a miner for British Coal. He joined the British Coal Staff Superannuation Scheme on 11 December 1978 and began contributing to it. He married Mrs McDonald (the appellant) on 22 March 1985. Shortly afterwards, he retired early on grounds of ill health and exercised his right to receive a pension income before his normal retiring age. As a result, between 11 December 1978 and 10 August 1985 Mr McDonald was a member of and contributor to the scheme; since then he has been a member in receipt of income benefits under the scheme. Mrs McDonald seeks a pensions sharing order under section 8(1)(baa) of the Family Law (Scotland) Act 1985 (the 1985 Act) on her divorce from Mr McDonald on the basis that his pension forms part of the matrimonial property which is taken into account in fixing financial provision. This appeal raises questions of statutory interpretation both in relation to the 1985 Act and The Divorce etc. (Pensions) (Scotland) Regulations 2000 (the 2000 Regulations), which were made under section 10(8) of the 1985 Act as amended. Section 10(5) of the 1985 Act treats as matrimonial property the proportion of any rights or interests of either personin any benefits under a pension arrangement which is referable to the period [during the marriage but before the relevant date]. The relevant date is the final date of separation, 25 September 2010 being the relevant date in the present case when the parties ceased to cohabit. The 2000 Regulations, which apply to occupational pension schemes and personal pension schemes of all kinds, provide for the valuation of a persons rights or interests in a pension arrangement for the purposes of section 10(5) by reference to what is known as the cash equivalent transfer value. Regulation 4 of the 2000 Regulations contains the relevant formula: A x B/C where A is the value of these rights or interests in any benefits under the pension arrangement which is calculated, as at the relevant date, in accordance with paragraph (2) of regulation 3 above; and B is the period of C which falls within the period of the marriage of the parties before the relevant date and, if there is no such period, the amount shall be zero; and C is the period of the membership of that party in the pension arrangement before the relevant date The dispute between the parties relates to that formula. The words which fall to be interpreted are the words in the definition of factor C, namely the period of membership of that party in the pension arrangement. Mr McDonald argues that the court should apportion the value of his pension rights by reference only to the period in which he was an active member of the scheme, that is the period during which he was making contributions to the scheme. On that basis, the value of his interest in the pension benefits which is matrimonial property would be 10,002. Mrs McDonald argues that the cash equivalent transfer value should be apportioned by reference to the period of Mr Macdonalds membership of the scheme, both when in pensionable employment and also when drawing a pension, that value being 138,534. An Extra Division of the Inner House dismissed Mrs McDonalds appeal. The majority based their reasoning on the general rule found in section 10(4) of the 1985 Act, which states matrimonial property is confined to assets acquired during the marriage but before the relevant date. They also relied on the formula in the 2000 Regulations. The Supreme Court unanimously allows Mrs McDonalds appeal. Lord Hodge gives the judgment, with which the other Justices agree. The period of membership in regulation 4 of the 2000 Regulations refers to the period of the persons membership of the pension arrangement, whether or not contributions are being made in that period [31]. There are four reasons why membership should not be confined to active membership of pension scheme while the member was contributing to it [25]. First, interpreting regulation 4 as confined in such a way involves adding words which are not there. The person who drafted the 2000 Regulations was clearly aware of the different categories of membership, as can be observed from the differentiation between categories of membership in regulation 3, and chose not to differentiate in regulation 4 between classes of membership [26]. Secondly, the 2000 Regulations apply to both occupational pension schemes and personal pension schemes. The definition of active membership in section 124(1) of the Pensions Act 1995 makes no sense in relation to personal pension schemes. It must be assumed that Parliament intended the Regulations to operate sensibly in respect of differing pension schemes. Further, it would prove difficult to ascertain the point at which a party who has made occasional contributions to a personal pension scheme had chosen to cease to make contributions [27]. Thirdly, the reading of the word active or contributing into regulation 4 cannot be supported by referring to the focus in section 10(4) of the 1985 Act to the acquisition by the parties of assets during the marriage but before the relevant date. Section 10(5) of the 1985 Act deals specifically with pensions, and the opening words of section 10(4), which defines matrimonial property, state that the definition provided in section 10(4) is subject to subsection (5) below. Parliament chose to deal with pensions differently by making a separate provision for them in section 10(5). It follows from the creation of that separate provision that the definition in section 10(4) should not be considered to apply to pensions and, therefore, the majority of the Extra Division of the Inner House erred in its reliance on it. Fourthly, it is not persuasive that membership in regulation 4 must mean active membership in order to give meaning to the statement that factor B can be zero. If the person drafting the wording of factors B and C in regulation 4 intended to confine membership in such a way that would be remarkably indirect. There is no hint of such an intention in the words of the Regulations. Further, confining the period of membership to the period when contributions were made and apportioning the value of the rights or interests in the benefits by reference to time, as section 10(5) requires, may often create an apportionment of the rights of interests in benefits in personal pension schemes which bears no relationship to the relative value of the rights acquired before and during the marriage [30]. This interpretation does not mean that the value of an interest in a pension must be shared equally. Section 9(1) of the 1985 Act contains other principles which inform the courts decision making and introduces flexibility into the award of financial provision. Further flexibility is introduced by the recognition in section 10(1) that there may be special circumstances for departing from the equal sharing of matrimonial property [13, 32].
In 1999 the Inland Revenue [now known as Her Majestys Revenue and Customs, HMRC] published a booklet known as IR20 and entitled Residents and Non Residents Liability to tax in the United Kingdom, which offered general guidance on the word residence and the phrase ordinary residence for the purposes of an individuals liability for UK income and capital gains tax. IR20 remained operative until 2009. The Appellants contend that, on its proper construction, IR20 contained a more benevolent interpretation of the circumstances in which an individual becomes non resident and not ordinarily resident in the UK than did the ordinary law; alternatively that prior to 2005 it was the settled practice of HMRC to adopt such a benevolent interpretation of IR20. Either the construction or the practice gave rise (so they say) to a legitimate expectation that the benevolent interpretation would be applied to determinations of their status for tax purposes and consequently HMRC should not have determined that, during the years relevant to them, they were resident or ordinarily resident in the UK. The First Appellants, Mr Davies and Mr James, contend that prior to 6 April 2001 they left the UK for the settled purpose of establishing and working full time for a Belgian company. Although their wives and Mr Davies daughters remained resident in the UK and although they returned frequently to the UK, albeit for short periods, they contend that they are entitled to be treated as non resident and not ordinarily resident in 2001 2002 by reference to paragraph 2.9 of IR20 since they had gone abroad for a settled purpose and had remained abroad for at least a whole tax year. The situation of the Second Appellant, Mr Gaines Cooper, is different from that of the First Appellants in that it has already been conclusively determined, by reference to the ordinary law, that he was resident and ordinary resident in the UK in the years relevant to him. He contends, however, that his status should instead be determined by reference to paragraphs 2.8 and 2.9 of IR20 or to the alleged settled practice and that, on either basis, he was not resident in the UK from 1993 to 2004 nor ordinarily resident here from 1992 to 2004. The High Court refused the Appellants permission to apply for judicial review of the determinations by HMRC that they were resident and ordinarily resident in the UK in the relevant years. The Court of Appeal granted them permission but dismissed their substantive applications. The Appellants appeal to the Supreme Court. The Supreme Court, by a 4 1 majority, dismisses the two appeals on the grounds that the proper construction of IR20 does not support the Appellants contentions and that there is insufficient evidence of any settled practice on the part of the HMRC by way of departure from the IR20 guidance. Lord Wilson gives the leading judgment; Lords Hope, Walker and Clarke give short concurring judgments. Lord Mance gives a dissenting judgment. An individuals status as being resident and ordinarily resident in the UK largely determines his liability for UK income tax and capital gains tax. In law an individual who has been resident in the UK ceases to be so resident only if he ceases to have a settled or usual abode in the UK per Levene v Inland Revenue Comrs [1928] AC 217 [13 Section 334 of the Income and Corporation Taxes Act 1988 (now replaced) also provided that an individual would nevertheless be deemed to have remained resident in the UK if he had left the UK for the purpose only of occasional residence abroad [15 17]. At law, an individual needs to effect a distinct break in the pattern of his life in the UK in order to become non resident per Reed v Clark [1986] Ch 1 [18 19]; this mandates a multifactorial evaluation of his circumstances [20]. But an individuals pursuit of full time employment abroad is likely to be sufficient to cause him to cease to be a UK resident and not to be deemed under the statute still to be a UK resident [21]. HMRC issued guidance on residence and ordinary residence in IR20. HMRC accepts that it is bound by whatever might be the proper construction of the guidance and that the guidance gave rise to a legitimate expectation that it would appraise any individuals case by reference to such guidance even if it failed to reflect the ordinary law [27]. The First Appellants contend that HMRC represented in IR20 that non residence was achieved if an individual left the UK to take up full time employment abroad, or left the UK permanently or for at least three years, or went abroad for a settled purpose and remained abroad for at least a whole tax year, provided in each case that any visits to the UK totalled less than six months in any one year and averaged less than 91 days each year [the day count proviso] [30]. The Second Appellant contends that HMRC thereby represented that it was sufficient for an individual to live abroad for at least three years and to satisfy the day count proviso, thus eliminating any need for consideration of whether he had effected a distinct break in the pattern of his life in the UK [31]. The majority holds that the proper construction of IR20, when read as a whole, does not support the Appellants contentions [45, 64]. Paragraph 2.1 indicated that an individuals claim to non residence would generate consideration of various aspects of his life with a view to the identification of its usual location [35]. The heading to paragraphs 2.7 to 2.9 namely Leaving the UK permanently or indefinitely required consideration of the quality of his absence from the UK [37]. Paragraph 2.9, which stated that if an individual had gone abroad for a settled purpose, he would be treated as not resident and not ordinarily resident if his absence from the UK had covered at least a whole tax year and he had met the day count proviso, could not be construed as a freestanding route to non residence since there was an express link to paragraph 2.8, which required an individual to leave indefinitely [41]. Although its exposition of how to achieve non residence should have been much clearer, IR20, taken as a whole, informed the ordinarily sophisticated taxpayer that he had to leave the UK permanently, indefinitely or for full time employment; had to do more than to take up residence abroad; and had to relinquish his usual residence in the UK. It also informed him that any subsequent returns to the UK had to be no more than visits and that any property retained in the UK by him for his use had to be used for the purpose only of such visits rather than as a place of residence [45]. He will have concluded that such requirements in principle demanded, and might well in practice generate, a multifactorial evaluation of his circumstances [45, 64] and, in summary, that he had to make a distinct break [45]. Alternatively, IR20 was so unclear as to communicate nothing to which legal effect might be given [47]. The majority holds that there was insufficient evidence that HMRC had departed from IR20 as a matter of settled practice [58]. Such a contention requires evidence that the practice was so unambiguous, so widespread, so well established and so well recognised as to amount to a specific commitment of treatment in accordance with it [49] but the Appellants evidence to this effect was far too thin and equivocal [58]. Lord Mance, dissenting, holds that the references to going abroad permanently or living outside the UK for three years or more in paragraphs 2.7 2.8 referred to the taxpayers intention regarding the duration of his absence rather than the quality of any absence or the nature of any return visits or continuing UK connections [89]. Paragraph 2.9 was designed to assist taxpayers who never intended to leave permanently or indefinitely, but went abroad for a settled purpose to engage in an overseas activity for an extended period of time of lesser duration [89]; or where the taxpayer could subsequently show he had acquired an intention to leave the UK permanently or that his actual absence covered three years from departure [90]. It would be remarkable if there were a requirement for a distinct break from life in the UK when no such requirement was clearly expressed [93] and other factors, including the day count proviso, militated against such a requirement [95; 96].
The appellant, an Algerian national, entered the United Kingdom in 1996 and was refused asylum. He married a French national in 1997. He was granted a residence permit, and had acquired a right of permanent residence by February 2003. He had two children with his wife but they were estranged by July 2004 and she returned to France in late 2005. By the end of January 2012, the Appellant had acquired 28 criminal convictions for 48 offences, including one 23 month sentence. The Home Secretary unsuccessfully attempted to deport him for that reason in January 2007. But he continued to offend, and was sentenced to 20 weeks imprisonment for an offence of theft on 25 January 2012. On 3 April 2012, just as the appellant was due to be released from custody for that offence, the Secretary of State served him with notice of her intention to make a deportation order against him under the Immigration (European Economic Area) Regulations 2006, on grounds that he would pose a genuine, present and sufficiently serious threat to the interests of public policy if he remained. He was detained from 3 April 2012 to 6 September 2012 (on bail from 6 June) under regulation 24(1) and Schedule 3 of the Immigration Act 1971 pending a decision being made on whether to deport him. He was served with a fresh Notice to that effect on 7 September 2012, and was again detained from 7 September 2012 until 2 January 2013. The appellant contended that his detention pending removal was unlawful, and sought judicial review. He argued that his detention contravened article 27(1) of the Citizens Directive (2004/38/EC) and that regulation 24(1) was incompatible with European law and unlawful because it discriminated against him on the basis of nationality without lawful justification contrary to Article 18 TFEU (there being no equivalent provision for pre decision detention in relation to family members of British nationals or non EEA nationals). The Upper Tribunal and the Court of Appeal dismissed his claim and appeal respectively. Before the Supreme Court, the appellant raised four essential points of challenge, namely that (i) the power to detain under regulation 24(1) was discriminatory without lawful justification, (ii) the power was unnecessary and disproportionate, (iii) the absence of a time limit for detention infringed the Hardial Singh principle and (iv) regulations 21 and 24 failed to accurately transpose the safeguards in articles 27 and/or 28 of the Directive. The Supreme Court unanimously dismisses Mr Nouazlis appeal, thereby holding that the appellants pre decision detention was not unlawful. It further declines to make a preliminary reference to the CJEU. Lord Clarke gives the judgment, and Lord Carnwath writes a concurring judgment. The legal framework for detention pending a decision to deport comprises Articles 27 and 28 of the Citizens Directive, as transposed by the EEA Regulations 2006. EEA Nationals or their family members exercising EU rights benefit from powerful protections against their expulsion from the UK, and can only be removed if certain limited circumstances apply, including where there are grounds of public policy, public security or public health [30, 36]. The power to detain under regulation 24(1) does not discriminate without lawful justification against EEA nationals and their family members. The general principle is that Article 18 TFEU is only concerned with the way in which EU citizens are treated in member states other than their states of nationality, and not the way in which member states treat nationals of other countries residing within their territories see the decision of the European Court of Justice in Vatsouras and Koupstantze v Arbeitsgemainschaft (AGRE) Nurnnerg (Joined Cases C 22/08, C 23/08) [2009] ECR I 4585. Third country nationals are not appropriate comparators for testing discrimination: such discrimination is simply a function of the limited scope of the EU legal order, into which third country nationals do not fall [39 49]. Nor is there discrimination between EU nationals and third country nationals contrary to article 21(1) of the EU Charter of Fundamental Rights [50 51, 61, 104]. The appellants new way of putting this argument was that discrimination occurs between British nationals and EEA nationals (exercising treaty rights) who each have third country spouses, since the spouse of the EEA national who is liable to be detained might be hypothetically deterred from exercising their own free movement rights the principle in Surinder Singh (R v Immigration Appeal Tribunal Ex p Secretary of State for the Home Department (Case C370/90) [1992] 3 All ER 798). But this argument also fails, since there is no basis for holding that the actual or hypothetical rights of this appellants spouse, who was long since estranged, would be so affected in this case [52 60, 104] As to proportionality, it is not in dispute that regulation 24(1) must be applied proportionately, but it was not argued that it was applied disproportionately in this case [62]. In this case, the absence of a specified time limit for detention does not infringe the principles in R v Governor of Durham Prison, Ex P Hardial Singh [1984] 1 WLR 704. That approach is fact sensitive, and the clear statutory framework here provides sufficient judicial scrutiny. The Hardial Singh approach, moreover, is entirely consistent with European law [63 78, 105]. Finally, regulations 21 and 24 do not fail accurately to transpose the safeguards in articles 27 and 28 of the Directive and are compliant with it [80 84, 106]. The recent CJEU decision in JN v Staatssecretaris van Veiligheid en Justitie (Case C 601/15), brought to the Courts attention in written submissions after the conclusion of the hearing, concerns a different Directive that is not binding on the United Kingdom. It also materially differs from the Citizens Directive because it contains an express freestanding power of detention for applicants for international protection, and not detention pending a deportation decision. It does not affect the disposal of this appeal [88 96]. Lord Carnwath writes a concurring judgment, setting out the appellants four overall challenges and agreeing with Lord Clarke that they should be dismissed [97 107].
On 18 January 2010, Dundee Council granted planning permission for the construction of an Asda superstore on a site at Myrekirk Road, Dundee. The Appellants operate a supermarket at a site on South Road, Dundee, some 800m from the proposed Asda site. Scottish Executive policy guidance states that, subject to material considerations indicating otherwise, proposed sites should be considered in the following descending order of preference: (a) town centre, (b) edge of town, (c) other commercial centres identified in the development plan and (d) out of centre locations that can be accessed by various transport modes (the sequential test). Effect is given to the sequential test by the policies set out in the statutory development plan, comprising the structure plan and the local plan. In considering Asdas application, the Council noted that the proposed site was out of town and fell therefore in the least desirable category. There was an available site at Methven Road in the Lochee district of Dundee, but the Council discounted it as being too small for the proposed store. There was no other available site within categories (a), (b) or (c) that would have been suitable. The Council accepted that the proposal failed to comply with the sequential test, but given that (i) it did not undermine the core land use strategies of the development plan; and (ii) it had a number of other planning, economic and social benefits, permission was granted. Tesco applied for judicial review of the Councils decision arguing that it had improperly interpreted and applied the development plan and that it had failed to consider its own policy in respect of the Lochee district. The petition was dismissed by the Lord Ordinary and a reclaiming motion against his interlocutor was refused by the Inner House. Tesco now appeals to this court. The Supreme Court dismisses the appeal. The lead judgment is given by Lord Reed, with whom the other justices agree. Lord Hope adds a brief concurring judgment. The Appellants contended that the Respondents had misinterpreted a criterion in one of the policies set out in the structure plan, and its equivalent in the local plan. The requirement in the policy reads as follows: In keeping with the sequential approach to site selection for new retail developments, proposals for new or expanded out of centre retail developments in excess of 1000 sq m gross will only be acceptable where it can be established that: no suitable site is available, in the first instance, within and thereafter on the edge of city, town or district centres [] [13]. The Appellants submitted that if there was a dispute about the meaning of a policy in the development plan it was for the court to determine what the words were capable of meaning. If the planning authority attached a meaning to the words which they were not properly capable of bearing, then it made an error of law, and had failed properly to understand the policy [13]. In the present case, the Respondents Director had interpreted suitable as meaning suitable for the development proposed by the applicant. But suitable meant suitable for meeting identified deficiencies in retail provision in the area. As no such deficiency had been identified, it was inappropriate to undertake the sequential approach and the Respondents had proceeded on an erroneous basis. They had failed to identify correctly the extent of the conflict between the proposal and the development plan and their assessment of whether other material considerations justified a departure from the plan was inherently flawed [13]. They had compounded their error by treating the proposed development as definitive when assessing whether a suitable site was available [14]. In response, the Respondents submitted that it was for the planning authority to interpret the relevant policy, exercising its planning judgment. The planning authority would only make an error of law if it attached a meaning to the words of the policy document which they were not capable of bearing. In the present case, the relevant policies required all the specified criteria to be satisfied. The Respondents had considered that the proposal failed to accord with the second and third criteria. In those circumstances, they had correctly concluded that the proposal was contrary to the policies in question. [15]. So far as concerned the assessment of suitable sites, Asdas retail statement had reflected a degree of flexibility: it considered smaller sites and sites which could accommodate only food retailing, whereas its application was also for non food retailing [16]. The Supreme Court considers that, in this area of public administration as in others, policy statements should be interpreted objectively in accordance with the language used, read as always in its proper context. That is not to say that such statements should be construed as if they were statutory or contractual provisions. Many of the provisions of development plans are framed in language whose application to a given set of facts requires the exercise of judgment. Such matters fall within the jurisdiction of planning authorities, and their exercise of their judgment can only be challenged on the ground that it is irrational or perverse. Nevertheless, planning authorities do not live in the world of Humpty Dumpty: they cannot make the development plan mean whatever they would like it to mean [18 19, 35]. In the present case, the question what the word suitable means cannot be answered by the exercise of planning judgment: it is a logically prior question as to the issue to which planning judgment requires to be directed [21]. Moreover, where, as here, it is concluded that the proposal is not in accordance with the development plan, it is necessary to understand the nature and extent of the departure from the plan which the grant of consent would involve in order to consider on a proper basis whether such a departure is justified by other material considerations [22]. The Supreme Court considers that the Respondents were correct to proceed on the basis that the word suitable meant suitable for the development proposed by the applicant, subject to the qualification that flexibility and realism must be shown by developers [24, 28, 37 38]. The Supreme Court makes this finding for the following reasons: (1) this is the natural reading of the policies [25]; (2) the interpretation favoured by the Appellants conflates the first and third criteria of the policies in question [26]; and (3) the policies were intended to implement the guidance given in National Planning Policy Guidance 8, which focuses upon the availability of sites which might accommodate the proposed development, rather than upon addressing an identified deficiency in shopping provision [27]. In the present case, it is apparent that a flexible approach was adopted [30]. An error in interpreting the policies would be material only if there was a real possibility that the determination might have been different. The Court is not persuaded that in the present case there was any such possibility [31].
This case concerns the proper test to be applied by the Parole Board when determining whether to direct the release of a person subject to a sentence of imprisonment for public protection (IPP). On 19 May 2006 the appellant punched a man during a fight outside a pub. The man fell backwards, struck his head on the ground and died the next day. The appellant was duly convicted of manslaughter. The judge concluded that the appellant was dangerous; he was forceful, physically strong and considered that he had the right to respond with violence to any tendered or threatened towards him. Accordingly the judge decided to impose a sentence of imprisonment for public protection on the basis that there was a significant risk to members of the public of serious harm occasioned by the commission by [the offender] of further specified offences (Criminal Justice Act 2003, section 225(1)(b)). The judge fixed a minimum term (tariff period) of 2 years 108 days. The tariff period expired on 19 May 2009. Following the expiry of this tariff period it was the responsibility of the Parole Board to decide whether the appellant should be released on licence. In conducting this assessment the Parole Board are required not to order release unless satisfied that it is no longer necessary for the protection of the public that the prisoner should be confined (Crime (Sentences) Act 1997, section 28(6)(b)). The Parole Board review took place on 10 May 2010. The Parole Board concluded that the appellant had made significant progress but continued to present a low risk of re offending and a medium risk of serious harm. As a result the Parole Board declined to order release. The appellant issued proceedings for judicial review, claiming (a) that the Parole Board had applied the wrong test; and (b) damages for the delay of almost one year in holding the review. The claim for damages was disposed of by the Supreme Court by judgment dated 1 May 2013, [2013] UKSC 23. Consequently, only the claim that the wrong test was applied remains for determination. Subsequent to the commencement of the judicial review proceedings, the Parole Board directed that the appellant be released on licence. As a result the present appeal has no direct significance for the appellants detention. Nonetheless, the matter was of significance for the Parole Board when it came before the Court of Appeal and may have a continuing significance in future cases. Consequently, the appellant seeks permissions to appeal to the Supreme Court. The Supreme Court grants permission to appeal but unanimously dismisses the appeal. The judgment of the Court is given by Lord Mance. Section 225(3) of the Criminal Justice Act 2003 (the 2003 Act) introduced a new form of indeterminate sentence, imprisonment for public protection, based on offending which was either of a kind for which a life sentence was not available or not of such seriousness as to justify a life sentence [21]. Sentences of IPP were fitted into the pre existing framework established for mandatory or discretionary life sentences [22], under which the criteria for imposition of a discretionary life sentence were, broadly, the commission of a very serious offence and a conclusion that the offender was a serious danger to the public and likely to remain so for an indeterminate period. In imposing a discretionary life sentence the court makes, at least in part, a predictive judgment as to the risk the offender will pose in the post tariff period [32]. Nothing in section 225(1)(b) of the 2003 Act suggests a distinction between the approach required when imposing a discretionary life sentence and when imposing a sentence of IPP [33]. It is difficult to square the reasoning in R v Smith (Nicholas) [2011] UKSC 37 that no such predictive judgment is involved when imposing a sentence of IPP with other case law or to see why Smith needed to address that point. The reasoning in Smith is thus questionable, but since it was not challenged on this appeal and is not ultimately decisive, no more need be said about it. [34 38]. The suggestion in R v Parole Board, Ex parte Bradley [1991] 1 WLR 134, 143F, 144H, 145F G and 146A C that the reference to future offending being likely involves a test of mathematical probability is unsound. It is not helpful to define significant risk in terms of numerical probability, whether as more probable than not or by any other percentage of likelihood [17]. A test of good grounds is more appropriate [28]. The test to be applied by the Parole Board when considering whether to direct release on licence from IPP is not the same as the test applied by the sentencing judge when imposing the sentence of IPP in the first place [39]. The two tests are, both in their terms and in their default position, substantially different [41]. By introducing a sentence of IPP into the framework applicable to discretionary life sentences, Parliament must on the face of it have intended that release from a sentence of IPP should be subject to the like test as release from a discretionary life sentence [42]. There is no reason why the statutory scheme should not involve a high threshold for imposition of a sentence of IPP than for continuing detention post tariff [44, 48]. The European Convention on Human Rights does not require a different result. Strasbourg case law accepts a sufficient causal connection between the imposition of a sentence of IPP and the deprivation of the offenders liberty in the post tariff period when release is contingent on him demonstrating to the Parole Board that he no longer poses a risk [48]. There was not basis for interfering with the decisions below that the appellant had not established that the Parole Board wrongly took into account directions by the Secretary of State on the test to apply, which the Secretary of State had no power to give [50 53].