The Commissioner of Police of the Metropolis v DSD and NBV & Ors

These conjoined appeals are brought in two actions for damages and declarations arising out of alleged failures by two police forces, the Metropolitan Police Service (MPS) and the Greater Manchester Police (GMP), to conduct effective investigations into allegations of crimes committed against the claimants. The claims were brought under ss.7 and 8 of the Human Rights Act 1998 (HRA). Their essence is that the failures of which the claimants accuse the police constitute violations of a duty to investigate said to be inherent in the right guaranteed by Article 3 of the European Convention on Human Rights (ECHR). As is well known Article 3 provides:

“No-one shall be subjected to torture or to inhuman or degrading treatment or punishment.”

It will make for clarity in explaining the argument if at this stage I also set out Article 1 and the first sentence of Article 2(1):

“1.

The High Contracting Parties shall secure to everyone within their jurisdiction the rights and freedoms defined in Section I of this Convention. [Section I includes Article 3.]

2(1) Everyone’s right to life shall be protected by law…”

The first of these claims to be decided was brought by two women, DSD and NBV, who were victims of the “black cab rapist”, a man called John Worboys. Between 2002 and 2008 Worboys committed over 105 rapes and sexual assaults on women who were passengers in his cab. On 28 February 2014 Green J gave judgment in favour of the claimants against the MPS. The second claim was brought by Alio Koraou, who alleged that on 23 December 2011 he was the victim of an assault at the Bar Rogue, part of the Britannia Hotel in Manchester, and part of his ear was bitten off. On 17 April 2014 HHJ Platts at the Manchester County Court dismissed the claim and gave judgment in favour of the GMP. In DSD/NBV Green J gave the MPS permission to appeal on 23 July 2014. In Koraou Lewison LJ gave the claimant permission on 30 June 2014.

In DSD/NBV the MPS assault Green J’s judgment on four grounds. (1) ECHR Article 3 does not of itself impose any obligation to investigate. To the extent that the Strasbourg court has found there to be a duty to investigate allegations of inhuman or degrading treatment, the duty springs from the positive obligation imposed by Article 1; but Article 1 forms no part of our domestic law, not being a Convention right within the meaning of the HRA. Accordingly there is no duty, cognizable in English law, to investigate alleged substantive breaches of Article 3. (2) If Ground 1 is wrong and Article 3 indeed creates a duty to investigate enforceable in our domestic law, the duty only arises where the State (or, to use the language of the HRA, a public authority) is complicit in an alleged substantive breach of the Article. (3) If Grounds 1 and 2 are both wrong and there is a duty to investigate allegations of inhuman or degrading treatment by non-State actors, then given the proper scope of the duty, there was no breach on the facts of DSD/NBV. (4) If all of Grounds 1 – 3 are wrong, Green J nevertheless erred in holding that the MPS owed a duty to NBV to investigate the perpetrator Worboys even before he attacked NBV.

In Koraou the appellant raises four grounds which in various respects attack Judge Platts’ approach to the facts. I will not enumerate them at this introductory stage. Essentially he seeks to advance a Wednesbury case ([1948] 1 KB 223): “[t]he nub of this appeal is [that] the decision to dismiss the claim in its entirety while at the same time finding a series of clear shortcomings/failings in DC Walters’ [the investigating officer] investigation is perverse” (skeleton argument, paragraph 11). The GMP of course take issue with that. They also support the MPS’ Grounds 1 and 2 in DSD/NBV.

I shall have to say more about the facts in confronting the issues, not least as regards the steps taken (and not taken) by the police in both cases. At this stage I will give a brief account of the accusations of substantive violations of Article 3 advanced by the claimants.

As Green J said at paragraph 2 of his judgment, DSD was one of Worboys’ earlier victims. She was attacked in 2003. NBV was attacked in July 2007; but there were many more victims after that. Green J proceeded to make these observations:

“6.

Between 2002 and 2008, Worboys committed in excess of 105 rapes and sexual assaults upon women whom he was carrying late at night in the back of his black cab. Over these years he developed an ever more refined methodology for administering drugs and alcohol to these women with a view to incapacitating them so that he could then assault them… The effect upon these vulnerable women was profound. In the cases of DSD and NBV… the effects of the assaults have stayed with them in a variety of ways over the ensuing years manifesting themselves in depression, feelings of guilt, anxiety, and an inability to sustain relationships…

7.

The administering of drugs of sedation and alcohol as an integral part of Worboys’ technique substantially reduced the likelihood of his apprehension and arrest. One troubling aspect of these cases is that so few of Worboys’ victims complained to police. This was partly for the reason that Worboys’ chosen modus operandi left his victims confused and disorientated and, frequently, with only a partial memory of their ordeal. The case of DSD is on point. Immediately following her attack, she was disorientated, incapacitated and vomiting. When she first came into contact with police very shortly after the assault, she appeared to be a drunk or a drug addict or both; and the police assumed as much. In an extraordinary twist of fate, she was in fact transported to the police station by Worboys himself, who had been persuaded to take DSD to the police station by a Good Samaritan third party [Kevin: he is referred to in the judgment by his first name], who also accompanied both Worboys and DSD to the station. But because she was mischaracterised as a drunk, she was not treated as a victim of crime, no-one took the name or address of Worboys or his vehicle registration. He was treated as a model citizen. And no-one took the name or address of [Kevin].”

Worboys was charged on 15 February 2008. He was tried in January 2009, convicted on 13 March 2009, and received an indeterminate sentence of imprisonment.

Koraou’s case was that he was assaulted by two men in the Bar Rogue in the early hours of 24 December 2011. He was to describe both of them as white males. One head-butted him, the other punched him in the head and neck. Both kicked him when he was on the floor. Security staff took hold of him; but as they held him, one of the men bit his ear, so that it was partially detached. Outside, he was again attacked by one of them whom the police detained. Koraou told the police that the man who attacked him in the street (subsequently identified as Wayne Maguire) was not the one who had bitten his ear. When he was taken to hospital, Koraou (on his account) told the officers who saw him there that the man who had been detained in the street had assaulted him in the bar. At length DC Walters was appointed investigating officer.

It will be convenient to address the facts of the investigations in both cases when I confront Grounds 3 and 4 in DSD/NBV, and the overall case in Koraou. I turn now to Ground 1 in DSD/NBV.

Under Ground 1 Mr Johnson QC for the MPS advances three propositions. (a) Article 3 is expressed in purely negative terms. (b) Authority shows that to the extent that there exists under the ECHR any duty to investigate substantive violations of Article 3, it arises only by force of the positive obligation to “secure… the rights and freedoms defined in Section I of this Convention” imposed by Article 1. (c) But Article 1 is not stipulated as a Convention right in the HRA. Accordingly the duty to investigate does not run in our domestic law.

Before turning to these individual propositions, there is a broader point to be made. The restrictive reading which the MPS would attribute to Article 3 allows no real weight to be given to what may be thought of as fundamentals of a civilised constitution: the rule of law, and the security and protection of the people. In the last analysis Grounds 1 and 2 in the MPS’ appeal raise issues as to the means and extent by which Article 3 gives effect to these interlocking values. It is of course not inevitable that an international treaty which distributes rights, such as the ECHR, should promote these ideals. But the preambles illuminate a large canvas (“[r]eaffirming their profound belief in those fundamental freedoms which… are best maintained on the one hand by an effective political democracy and on the other by a common understanding and observance of the Human Rights upon which they depend”); and I think the interpretation of specific measures in the ECHR should acknowledge the force of this context. In my judgment the relevant Strasbourg cases do no less.

That consideration brings me directly to the first point on Ground 1 taken by Mr Johnson. It consists as I have said in the proposition – itself incontrovertible – that the language of Article 3 is negative: “[n]o-one shall be subjected…” So, says Mr Johnson, the Article contains a bare prohibition of torture and inhuman or degrading treatment: nothing more. But this is merely to point to the literal meaning of the provision. It is blind to the impact of the jurisprudence on Article 3. The real substance of Ground 1 consists in Mr Johnson’s second proposition: that the duty to investigate substantive violations of Article 3, so far as it exists at all, arises by force of ECHR Article 1. His third proposition, that Article 1 is not stipulated as a Convention right in the HRA, is of itself as uncontentious as his first; but it is nothing to the point unless he can establish his second, to which I now turn.

The principal case relied on by Mr Johnson is Assenov v Bulgaria (1988) 28 EHRR 652. The complaint was of ill-treatment by the Bulgarian police and misconduct by other Bulgarian State officials. At paragraph 102 of Assenov the court said this:

“The Court considers that, in these circumstances, where an individual raises an arguable claim that he has been seriously ill-treated by the police or other such agents of the State unlawfully and in breach of Article 3, that provision, read in conjunction with the State’s general duty under Article 1 of the Convention to “secure to everyone within their jurisdiction the rights and freedoms defined in … [the] Convention”, requires by implication that there should be an effective official investigation. This investigation, as with that under Article 2, should be capable of leading to the identification and punishment of those responsible… If this were not the case, the general legal prohibition of torture and inhuman and degrading treatment and punishment, despite its fundamental importance…, would be ineffective in practice and it would be possible in some cases for agents of the State to abuse the rights of those within their control with virtual impunity.”

The second part of this citation bears on Grounds 2 and 3. The reference in the first part to Article 1 is replicated in later cases, enumerated by Mr Johnson at footnote 2 on p. 5 of his skeleton argument. In particular it appears in paragraph 149 of MC v Bulgaria (2005) 40 EHRR 20, a case to which I must return:

“The Court reiterates that the obligation of the High Contracting Parties under Article 1 of the Convention to secure to everyone within their jurisdiction the rights and freedoms defined in the Convention, taken together with Article 3, requires States to take measures designed to ensure that individuals within their jurisdiction are not subjected to ill-treatment, including ill-treatment administered by private individuals…”

Mr Johnson, relying on these references, submitted that Article 1 amplifies the content of Article 3, which thus becomes more than a mere prohibition; by force of Article 1, it imposes a positive obligation to investigate. He draws a contrast with Article 2. In Osman v UK (1998) 29 EHRR 245 at paragraph 115 the court said this:

“The Court notes that the first sentence of Article 2 § 1 enjoins the State not only to refrain from the intentional and unlawful taking of life, but also to take appropriate steps to safeguard the lives of those within its jurisdiction… It is common ground that the State’s obligation in this respect extends beyond its primary duty to secure the right to life by putting in place effective criminal law provisions to deter the commission of offences against the person backed up by law-enforcement machinery for the prevention, suppression and sanctioning of breaches of such provisions. It is thus accepted by those appearing before the Court that Article 2 of the Convention may also imply in certain well-defined circumstances a positive obligation on the authorities to take preventive operational measures to protect an individual whose life is at risk from the criminal acts of another individual. The scope of this obligation is a matter of dispute between the parties.”

Mr Johnson’s point is that the Strasbourg court was able to derive a safeguarding or preventive obligation from the positive language of Article 2(1) alone, whereas no implicit obligation – in this case to investigate – has been (or, he would say, could be) derived from the negative language of Article 3; hence the recourse to Article 1. In Menson v UK 37 EHRR CD220 the court recalls paragraph 115 of Osman at CD229, stating that Article 2 “imposes a duty on [the] State to secure the right to life by putting in place effective criminal law provisions…, backed up by law enforcement machinery…” There was no reference to Article 1, and Mr Johnson says none was necessary. The Menson case is of greater significance for the resolution of Ground 3; but Mr Johnson submits it is grist to his mill on Ground 1.

In my judgment neither the contrasting language of Articles 2 and 3 nor the learning demonstrates that the duty to investigate ill-treatment of the gravity stipulated in Article 3 is to any extent derived from Article 1. First, Article 1 is silent as to the content of any of the substantive rights. It requires that they be secured; but they are defined, or described, elsewhere. Thus the language of Article 1 lends no support to Mr Johnson’s submission that it expands the scope of Article 3.

Secondly, on Mr Johnson’s argument there is a substantial mismatch between the scope of Article 3 guaranteed by the Convention and the scope of Article 3 enforceable, by means of the HRA, in the UK courts. The first includes an investigative duty but the second does not. In the course of argument Mr Johnson accepted that the HRA gives effect “lock, stock and barrel” to the substantive rights guaranteed by the ECHR, and that is surely right: in Quark Fishing Ltd [2006] 1 AC 529 at paragraph 34 (cited by Lord Rodger in Al-Skeini [2008] 1 AC 153, paragraph 58) Lord Nicholls stated that “[t]he [HRA] was intended to provide a domestic remedy where a remedy would have been available in Strasbourg”. This contradicts the mismatch which Mr Johnson’s argument implies. The effect of such a mismatch would anyway be bizarre. It would mean that a complaint of violation of Article 3 in the UK constituted by actual ill-treatment could be litigated here; but a complaint that the self-same Article was violated by an investigative failure would have to go to Strasbourg.

Thirdly, the omission of Article 1 from the catalogue of Convention rights in the HRA is readily explained. Article 1 is the provision by which the States Parties are obliged to secure the rights stipulated in the ECHR. S.6(1) of the HRA is in my judgment analogous (though Mr Basu QC for the GMP in the Koraou appeal submitted otherwise). It obliges public authorities in the United Kingdom to respect the Convention rights. As is well known s.6(1) provides:

“It is unlawful for a public authority to act in a way which is incompatible with a Convention right.”

There are ancillary provisions concerning proceedings and remedies (together with the process for a declaration of incompatibility – ss.4 and 10), but s.6(1) imposes the primary obligation to secure the Convention rights. The scheme of the Act is clear: those ECHR measures which state substantive rights are named as the Convention rights; other measures in the ECHR, which give the Convention effect but do not state its substance, are not. Thus Article 13 (right to an effective remedy) is omitted, as is Article 1. Mr Johnson’s argument ignores this distinction. Nothing in the cases, here or in Strasbourg, supports such an approach; the repeated references to Article 1 on which Mr Johnson relies, from paragraph 102 of Assenov onwards, do no more than identify the medium through which Article 3 has effect on the international plane.

Fourthly – here the point is a negative one – Mr Johnson can take no support from the decision of the House of Lords in Al-Skeini. In that case the House was principally concerned with the territorial scope of the HRA, and considered that that was illuminated by the territorial scope of Article 1. In his skeleton argument at paragraph 27 Mr Johnson submits that Al-Skeini supports his argument that the substantive Convention rights in the HRA should not be construed as if they were to be read in conjunction with Article 1. As a proposition that seems to me to be plainly correct (though it does not, I think, in the least depend on Al-Skeini); but in the context of the present appeal it assumes what Mr Johnson has to demonstrate, namely that the Article 3 investigative obligation has its source in Article 1. For the earlier reasons I have set out, I am clear that is not the case.

I should add that the judge below paid attention (paragraph 234) to the fact that their Lordships in Al-Skeini deployed Article 1 to cast light on the territorial scope of the HRA; however “none of the opinions expressed in that case serve to undermine the conclusion that I have arrived at in relation to the scope and effect of the HRA and Article 3”. Mr Johnson’s reference in the course of argument to the decision of Supperstone J in Morgan [2010] EWHC 2248, which with respect I need not cite, in my judgment takes the matter no further.

Like the judge, I would reject Ground 1.

Mr Johnson’s submission on Ground 2 is that a duty to investigate under Article 3 only arises where the State is complicit in an alleged substantive breach of the Article. But the Strasbourg learning places formidable obstacles in his way. I should first cite MC v Bulgaria. At paragraph 151 the court said this:

“151.

In a number of cases, Article 3 of the Convention gives rise to a positive obligation to conduct an official investigation (see Assenov and Others v. Bulgaria… §102). Such positive obligations cannot be considered in principle to be limited solely to cases of ill-treatment by State agents (see, mutatis mutandis, Calvelli and Ciglio v. Italy [GC], no. 32967/96, ECHR 2002-I).”

Like the reference to Article 1 in Assenov, this statement has been frequently repeated in later Strasbourg cases. Milanovic v Serbia is a good example, citing as it does both Assenov and MC:

“85.

The Court further recalls that where an individual raises an arguable claim that he has been seriously ill-treated in breach of Article 3, that provision, read in conjunction with Article 1 of the Convention, requires by implication that there should also be an effective official investigation capable of leading to the identification and punishment of those responsible (see Assenov… § 102). A positive obligation of this sort cannot, in principle, be considered to be limited solely to cases of ill-treatment by State agents (see MC v Bulgaria… § 151…).”

Mr Johnson’s riposte consisted in a striking submission to the effect that in the later cases the Strasbourg court had misunderstood its own judgment in MC at paragraph 151. He said that the reference in that paragraph to positive obligations (“[s]uch positive obligations cannot be considered in principle…”) did not in fact look back to the “positive obligation to conduct an official investigation” in the first sentence of the paragraph, but to a more general statement in the foregoing paragraph 150:

“Positive obligations on the State are inherent in the right to effective respect for private life under Article 8; these obligations may involve the adoption of measures even in the sphere of the relations of individuals between themselves…”

Like many a counsel of despair, this was imaginative. But the language of paragraph 151 is plainly against it: the “positive obligations” in question clearly include the investigative obligation mentioned in the first sentence. Moreover the reference in paragraph 151 of MC to the case of Calvelli v Italy tends to show that the court’s focus in MC was on the proposition that the obligation under discussion did not only arise where actual or alleged misconduct by State agents was involved. Calvelli was a case in which a new-born baby had died through a doctor’s negligence. The dismissal of a prosecution against the doctor by reason of a statutory time-bar, following delays in the criminal process, was said to constitute a violation of Article 2. The Grand Chamber held (paragraph 49) that Article 2 required “an effective independent judicial system to be set up so that the cause of death of patients in the care of the medical profession, whether in the public or the private sector, can be determined and those responsible made accountable”. This was, of course, Article 2 and not Article 3; and it was not an investigation case, but concerned other alleged deficiencies in the Italian criminal process. But I think it clear (as Ms Kaufmann QC for the respondents in DSD/NBV in effect submitted) that at paragraph 151 of MC the court was reading across, from Calvelli , a duty owed by the State under Article 3 to take steps where the primary injury has been caused or inflicted by a non-State agent.

In any case – and Mr Johnson was taxed with this in the course of argument – even if it could be said that the court in later cases had at first misunderstood its own judgment in MC, that would not avail the MPS: whether or not born of a misunderstanding, there is a clear and constant line of Strasbourg authority to the effect that “a positive obligation [to conduct an official investigation] cannot be considered in principle to be limited solely to cases of ill-treatment by State agents”. Szula v UK (2007) 44 EHRR SE19, Secic (2009) 49 EHRR 18 and C.A.S. v Romania (Application No. 26692/05) are plain examples. Repeated statements to this effect represent the considered view of the Strasbourg court.

Faced with this difficulty, Mr Johnson had a fall-back position. In reply he referred to the well-known requirement of HRA s.2(1) that in “determining a question which has arisen in connection with a Convention right” our courts “must take into account” the Strasbourg jurisprudence. He submitted that we are not thereby enjoined to treat it as precedent. That is of course right; and for my part I have long thought, with respect, that needless difficulty has been caused by the treatment in this jurisdiction of Strasbourg cases almost as if they were domestic law. But where there exists so clear and constant a line of authority from Strasbourg as is to be found in this case, we must surely have very good reason to decline to apply it.

Mr Johnson submits that it should be disapplied. He says there is learning of our own courts to the effect that the Article 3 investigative duty (seen as a Convention right under the HRA) is owed only where the actual or apprehended injury is at the hands of State agents. He relies in particular on statements in three cases, P v Secretary of State [2010] QB 317, Humberstone [2011] 1 WLR 1460 and NM [2012] EWCA Civ 1182 which, he says, we are bound to follow.

Before I address these decisions I think it helpful to consider a somewhat broader canvas. Under the common law of negligence, the police owe “no general duty of care… to identify or apprehend an unknown criminal, nor… a duty of care to individual members of the public who might suffer injury through the criminal’s activities save where their failure to apprehend him had created an exceptional added risk, different in incidence from the general risk to the public at large from criminal activities…” (Hill v Chief Constable of West Yorkshire [1989] 1 AC 53, 54 (headnote); cf Brooks [2005] 1 WLR 1495). Might this rule promote a conclusion that for the purpose of the HRA the scope of any investigative duty under Article 3 does not extend to require the State (here the police), as a matter of enforceable right in the hands of a complainant, to investigate an allegation of violent crime? Mr Johnson did not so submit in terms and I would not so hold. But the question invites attention to authority, to which I will come directly, which I think illuminates significant differences between a private law claim in negligence and a suit for breach of Article 3. That is important, because it is important that the common law and the HRA should as far as possible cohere; that neither should undermine the other. It is moreover to be noted that recent statements in the Supreme Court emphasise the common law as guarantor of human rights: see for example per Lord Reed in Osborn v The Parole Board [2013] 3 WLR 1020, [2013] UKSC 61 at paragraphs 56-57.

In Van Colle v Chief Constable of Herts Police, Smith v Chief Constable of Sussex Police [2009] AC 225, two cases heard together, the complaint was that police had failed to follow up reports of threats to kill. In Van Colle the object of the threats was shot dead. In Smith he was seriously injured. The first case was brought solely under the HRA, alleging violation of Article 2. The second claimant relied only on the common law, alleging negligence by the police. The first case failed on the facts. But in the second, the claim was struck out. The contrast is striking. The relation between Strasbourg and the common law was most fully considered by Lord Brown, addressing an argument that “the common law should now be developed to reflect the Strasbourg jurisprudence about the positive obligation arising under articles 2 and 3 of the Convention” (paragraph 136). Lord Brown said this:

137 True it is that the possibility of a Human Rights Act claim now to some extent weakens the value of the Hill principle insofar as that is intended to safeguard the police from the diversion of resources involved in having to contest civil litigation. That, however, is no good reason for mirroring the Osman principle by the introduction of a common law duty of care in this very limited class of case, still less for weakening the value of the Hill principle yet further by creating a wider duty of care.

138 There is this too to be said as to why, certainly in the present context, your Lordships should not feel tempted to develop the common law ‘in harmony with’ Convention rights (as Rimer LJ put it below). As Lord Bingham pointed out in R (Greenfield) v Secretary of State for the Home Department [2005] 1 WLR 673, Convention claims have very different objectives from civil actions. Where civil actions are designed essentially to compensate claimants for their losses, Convention claims are intended rather to uphold minimum human rights standards and to vindicate those rights. That is why time limits are markedly shorter… It is also why section 8(3) of the Act provides that no damages are to be awarded unless necessary for just satisfaction. It also seems to me to explain why a looser approach to causation is adopted under the Convention than in English tort law. Whereas the latter requires the claimant to establish on the balance of probabilities that, but for the defendant's negligence, he would not have suffered his claimed loss—and so establish… that appropriate police action would probably have kept the victim safe—under the Convention it appears sufficient generally to establish merely that he lost a substantial chance of this.

139 Clearly the violation of a fundamental right is a very serious thing and, happily, since the Human Rights Act, it gives rise to a cause of action in domestic law. I see no sound reason, however, for matching this with a common law claim also. That to my mind would neither add to the vindication of the right nor be likely to deter the police from the action or inaction which risks violating it in the first place. Such deterrence must lie rather in the police’s own disciplinary sanctions (as, indeed, were applied in Van Colle) and, in a wholly exceptional case… in criminal liability. Rather I am satisfied that the wider public interest is best served by maintaining the full width of the Hill principle…”

In Michael v Chief Constable of South Wales Police [2015] 2 WLR 343 another victim of a threat to kill brought proceedings against the police, in this case alleging both negligence and breach of Article 2. It was submitted that the common law should be developed “to encompass the duties of the police under the Convention” (per Lord Toulson at paragraph 123). Lord Toulson continued:

“126 The same argument, that the common law should be developed in harmony with the obligations of public bodies including the police under the Human Rights Act 1998 and articles 2 and 3 of the Convention, was advanced in Smith as a ground for holding that the police owed a duty of care to the deceased after he reported receiving threats…

128 It is unnecessary for the purposes of this appeal to decide questions about the scope of article 3 and I would not wish to influence the Court of Appeal’s consideration of the judgment in DSD v Commissioner of Police of the Metropolis [that is, of course, this case]. It does not alter the essence of the argument which was considered and rejected by the House of Lords in Smith. I am not persuaded that it would be right for the court to depart from that decision, which itself was consistent with a line of previous authorities.”

The argument thus addressed in Van Colle/Smith and in Michael was, of course, that the common law rule should be moderated so as to accommodate the ECHR: whereas what we are considering here is the converse – that the Article 3 Convention right (within the meaning of the HRA) might properly be moderated by force of the common law. That is, perhaps, a more ambitious proposition, but in my judgment is anyway not made out. The cases show, not least through the speech of Lord Brown in Van Colle/Smith at paragraph 138, that the ECHR and the common law of negligence have different aims, and so can live together. I shall have more to say about this in addressing Ground 3, where I think it has a special importance.

I turn then to the three domestic law cases on which Mr Johnson particularly relied. In P v Secretary of State the 19-year old claimant, who was eventually diagnosed as suffering from psychopathic disorder, repeatedly harmed himself while detained in a young offender institution. If he continued to do so he might suffer life-threatening injuries. At length he brought judicial review proceedings for an order that the Secretary of State hold an inquiry into his detention, alleging an obligation to do so by force of ECHR Articles 2 and 3. No such inquiry was ordered. In this court judicial review permission was granted but the claim dismissed on the merits. Delivering the only substantive judgment, Stanley Burnton LJ (addressing Article 3) cited a lengthy passage from the judgment of Longmore LJ in AM v Secretary of State [2009] UKHRR 973, and then this from the judgment of Elias LJ in the same case at paragraph 91:

“The obligation to carry out an investigation is a procedural one which is parasitic on alleged substantive breaches of the article: see the observations of Lord Bingham of Cornhill in R(Gentle) v Prime Minister [2008] AC 1357, para. 6. The nature of that obligation is inextricably linked to the specific nature of the alleged breaches.”

Stanley Burnton LJ concluded (paragraph 58):

“Whether the Secretary of State is bound to conduct an inquiry depends on the circumstances of the case… To impose an obligation to hold a human rights inquiry has significant resource implications… Good reason for an article 3 inquiry must be shown. In the present case, all the relevant facts are known…”

Gentle, referred to by Elias LJ in AM, was a case in which the mothers of two young British soldiers killed in Iraq contended that by force of Article 2 they had an “enforceable legal right… to require Her Majesty’s Government to establish an independent public enquiry into all the circumstances surrounding the invasion of Iraq by British forces in 2003” (per Lord Bingham at paragraph 2).

Mr Johnson’s point is that the reasoning cited in P shows that any procedural rights arising out of Article 3 are parasitic upon, or adjectival to, an allegation of substantive breach; and since on any view Article 3 (indeed the ECHR as a whole) only confers rights against the State, a substantive breach may only be committed by the State. So the adjectival or parasitic duty is only owed where State agents, actually or allegedly, have perpetrated inhuman or degrading treatment contrary to the Article.

In Humberstone the claimant was arrested on suspicion of manslaughter by gross negligence following the death of her ten-year old son, who had suffered from asthma. However she was not charged. She sought public funding through the Legal Services Commission so as to be represented at the inquest into her son’s death, relying on a reference to ECHR Article 2 in the Lord Chancellor’s funding guidance. Issues concerning both her conduct and that of paramedic staff who had attended her son would or might have to be explored. The claimant succeeded at first instance and, for somewhat different reasons, in this court. Smith LJ (with whom Maurice Kay and Leveson LJJ agreed) cited at length from the judgment of Richards J, as he then was, in Goodson [2006] 1 WLR 432, and then said this at paragraph 58:

“I would summarise his conclusions by saying that article 2 imposes an obligation on the state to set up a judicial system which enables any allegation of possible involvement by a state agent to be investigated. That obligation may be satisfied in this country by criminal or civil proceedings, an inquest and even disciplinary proceedings or any combination of those procedures. This obligation envisages the provision of a facility available to citizens and not an obligation proactively to instigate an investigation. Only in limited circumstances (I depart from Richards J only so far as to decline to call them exceptional) will there be a specific obligation proactively to conduct an investigation. Those limited circumstances arise where the death occurs while the deceased is in the custody of the state or, in the context of allegations against hospital authorities, where the allegations are of a systemic nature such as the failure to provide suitable facilities or adequate staff or appropriate systems of operation. They do not include cases where the only allegations are of ‘ordinary’ medical negligence.”

In NM the claimant was a 19-year old prisoner who was sexually assaulted by a fellow prisoner during association in his cell. His claim against the Secretary of State for Justice was wide-ranging, but the only issue remaining in this court was an allegation that in breach of Article 3 the incident had not been adequately investigated by the prison authorities. The claimant had made it clear that “he did not want the police involved” (per Rix LJ at paragraph 8). The claim failed at first instance and in this court. At paragraph 29 Rix LJ, with whom Lewison LJ and I agreed, said this:

“In article 3 cases, therefore, the alternatives of civil and criminal proceedings, and ombudsman enquiries, are important available sources of sufficient investigation, where such investigation may be needed: see also R (P) v. Secretary of State for Justice, approving Longmore LJ's analysis in AM… It is only or primarily where there is credible evidence of treatment, sufficiently grave to come within article 3, inflicted ‘by or with the connivance of the state’ that the investigative obligation arises (see Sedley LJ in AM at [4]). In the absence of state complicity, the essential obligation of the state is only to provide a system under which civil wrongs may be remedied in litigation or criminal wrongs investigated and prosecuted: see MC v Bulgaria…, Secic v Croatia… The investigative obligation, particularly under article 3, is highly fact sensitive and subject to resource implications (… AM at [107], and P at [58]). ‘Where the line is to be drawn is a matter of fact and degree’ (per Richards LJ in R (Mousa) v. Secretary of State for Defence [2010] EWHC 3304 (Admin).”

Mr Johnson’s submission is that taken together these cases show that it has been accepted in this jurisdiction, for the purpose of giving effect to Article 3 as a Convention right under the HRA, that a specific investigative obligation only arises where the State has been or is alleged to have been complicit in a substantive violation of the Article.

Green J below made these observations at paragraph 237:

“… I do not interpret the existing case law of the Court of Appeal as inconsistent with Strasbourg case law. Mr Johnson QC took me to a series of cases. In each of these cases however the facts did not concern the responsibility of the State to investigate a crime committed by a private person of such severity that it could be categorised as torture or degrading or inhuman treatment where there was no element at all of State complicity. They covered cases where the State was directly or indirectly complicit in the violence. In the domestic context he referred to: R (NM) v Secretary of State for Justice… and in particular the dictum of Rix LJ at [29]; and to R (Humberstone) v Legal Services Commission… These were not cases where the facts involved violence by private parties with no State complicity and, moreover, as I explain below R(NM) actually recognises the existence of the free standing duty that I have concluded exists in cases with facts such as the present.”

At paragraph 239 the judge, referring to paragraph 29 of the judgment of Rix LJ in NM, noted in terms that the reasoning there set out demonstrated

“a recognised duty on the State ‘in the absence of State complicity’ to investigate and prosecute criminal wrongs. The judge cited MC v Bulgaria, and Secic v Croatia, both of which – amongst many other cases – confirm the existence of a free-standing obligation upon the police to investigate quite irrespective of complicity or connivance upon their part in the underlying violent crime. The ‘system’ referred to is clearly the overall legal and operational system deployed by police to investigate.” (original emphasis)

In my judgment Green J’s reasoning in these paragraphs, addressing Mr Johnson’s argument on the domestic authorities, was entirely correct. But before I elaborate my own conclusions on Ground 2 I should give some account of the argument for the respondent.

Ms Kaufmann submitted that the Strasbourg learning disclosed three distinct categories of investigative obligation in the Article 3 context. The first is what she called a systems duty – the State’s duty to introduce and maintain a judicial system that includes process for the investigation of actual or alleged events giving rise to issues touching Article 2, 3 or 4. The second is an adjectival duty – triggered only where there is an arguable case that the State itself has violated Article 2, 3 or 4. The third is a criminal investigative duty – requiring the effective investigation of conduct sufficiently grave to meet the threshold of Article 2, 3 or 4 whether or not perpetrated by State agents. Ms Kaufmann submits that this is the duty that was owed by the MPS to her clients.

The utility of this classification from Ms Kaufmann’s point of view is that it enables her to isolate this third duty – the only duty relevant to her case – and to submit that observations in the authorities, notably those in this jurisdiction, which are said to tell against her are concerned only with the first or second duty class. Thus she points to Smith LJ’s reference to two duties at paragraph 52 of Humberstone:

“[T]he [Strasbourg] cases describe two different obligations arising under article 2. First, there is a duty imposed on the state to set up an effective judicial system by which any death, which might possibly entail any allegation of negligence or misconduct against an agent of the state may be adequately investigated and liability established. That will apply in a wide range of circumstances. Second, there is a duty proactively to conduct an effective investigation into the circumstances of a death in a much narrower range of circumstances where the evidence suggests a possible breach of the state’s substantive duty to protect the life of those in its direct care.”

Ms Kaufmann submits that the distinction there drawn is between the first two duties in the triad which she described; neither this passage at paragraph 52, nor indeed any part of the Humberstone case, has anything to do with the third duty – the duty owed to her clients. Nor did NM: in that context Ms Kaufmann attached significance to the fact that the claimant “did not want the police involved”.

Ms Kaufmann says that the third duty class is well supported by the Strasbourg cases, indeed by the clear and constant line of authority which I have described. She placed some emphasis on Menson, to which I have briefly referred in dealing with Ground 1, and also on the Grand Chamber case of O’Keeffe v Ireland (2014) 59 EHRR 15, which was concerned with sexual abuse at a Catholic school. At paragraph 172 of that decision “[t]he Court recalls the principles outlined in CAS v Romania to the effect that art.3 requires the authorities to conduct an effective official investigation into alleged ill-treatment inflicted by private individuals which should, in principle, be capable of leading to the establishment of the facts of the case and to the identification and punishment of those responsible…”

The Strasbourg learning plainly establishes that the duty thus summarised in O’Keeffe is inherent in Article 3, and in my judgment Ms Kaufmann is right to submit that the English cases do not require a different approach to the article when it functions as a Convention right under the HRA. That is enough to dispose of this ground of appeal, concerning State complicity, in the respondents’ favour. But I should make it plain that I do not accept Ms Kaufmann’s tripartite division of the investigative duty, and explain the reasons. The point is of some importance in seeing how the common law and the ECHR fit – an issue I have already visited in discussing Van Colle/Smith, and to which I must return in considering Ground 3 – and also in order to articulate an accurate overall view of the scope and nature of the Article 3 right.

Ms Kaufmann’s three classes are too permeable; they flow into each other too far for each to be treated as a self-standing duty category. In particular, Ms Kaufmann did not identify, at least to my satisfaction, any principled difference between the investigative element in the systems duty (duty no.1) and the criminal investigative duty (no.3). In my judgment an appreciation of the reach and nature of the investigative duty that is part of Article 3 demands a broader consideration of the aims of this part of the ECHR.

The rights which the Convention guarantees are enjoyed against the State, and only the State. It is important to recognise that ill-treatment by a non-State agent, however grave, does not of itself constitute a breach of Article 3. This is sometimes glossed over in the language of the cases, as for instance at paragraph 85 of Milanovic, cited above at paragraph 22. Likewise a killing does not of itself violate Article 2, nor an act of enslavement Article 4, if it is not perpetrated by an agent of the State. But it is surely inherent in the Convention’s purpose that the State is to protect persons within its jurisdiction against such brutalities, whoever inflicts them. It is therefore no surprise that the Strasbourg court has interpreted Article 3 so as to provide safeguards that are broader than the bare prohibition of acts of torture or gross ill-treatment by servants of the State.

Reading the cases, one might be forgiven for supposing that Article 3 comprises a series of loosely connected rights given effect by loosely connected duties owed by the State. But it is important to keep in mind the Article’s overall, strategic, safeguarding purpose. One consequence is that it is misleading to regard investigative processes as always “ancillary” or “adjectival” to the “substantive” right guaranteed by Article 3. Language of that kind more or less fits the case where there is a credible allegation of ill-treatment by State agents: then, there is a “substantive” breach by the State, whose investigation may reasonably be regarded as “adjectival”. But that model is inapt where there is ill-treatment by non-State agents. In such a case there is no antithesis between what is substantive and what is adjectival: the “substantive” act does not of itself violate the Convention. In such a case Article 3 generally requires a proper investigation, and criminal process if that is where the investigation leads. The idea at the core of the Article is that of safeguarding or protection in all the myriad situations where individuals may be exposed to ill-treatment of the gravity which the Article contemplates.

There is perhaps a sliding scale: from deliberate torture by State officials to the consequences of negligence by non-State agents. The energy required of the State to combat or redress these ills is no doubt variable, but the same protective principle is always at the root of it. The margin of appreciation enjoyed by the State as to the means of compliance with Article 3 widens at the bottom of the scale but narrows at the top. At what may, without belittling the victim, be called the lower end of the scale where injury happens through the negligence of non-State agents, the State’s provision of a judicial system of civil remedies will often suffice: the individual State’s legal traditions will govern the means of compliance in the particular case. Serious violent crime by non-State agents is of a different order: higher up the scale. In these cases, which certainly include DSD/NBV, a proper criminal investigation by the State is required. I will explain what I mean by “proper” when I come to Ground 3.

This application of a single principle with varying degrees of rigour represents, I think, the true sense of Article 3. The nuance which this necessarily involves explains the different voices in which the cases speak. So much is reflected in this court’s decision in Allen [2013] EWCA Civ 967, where however the court is said to have misplaced the degree of rigour required at a point on the scale relevant to the present case. But whether or not it did so is a question for Ground 3.

I would reject Mr Johnson’s submissions on Ground 2 for all the reasons I have given.

I have already anticipated a large part of my answer to this part of the case in stating, in relation to Ground 2, that serious violent crime by non-State agents generally requires a proper criminal investigation by the State. But there is more to be said, not only out of deference to counsel’s submissions but also because Ground 3 provides the proper context in which to try and resolve the question of the common law’s coherence with the Convention rights.

Mr Johnson’s case is (to use, if I may, my language rather than his) that in the circumstances the judge below placed the degree of rigour required of the police investigation by Article 3 too high on the scale. He should have been guided by six principles which taken together tend to show that the MPS did not fall short of the standard of investigation which was required in the circumstances. The principles, said to be derived from the cases, are enumerated by Mr Johnson at paragraph 42 of his skeleton argument:

A number of these factors (a better term, I think, than principles) run into each other, notably the fourth and fifth: these two, moreover, point towards features that are relevant to the coherence between the common law and the ECHR. The first and third factors assert a greater rigour for some classes of case (Article 2, State agents) over others (Article 3, non-State agents), but the list of factors taken together does not, I think, quite convey the broad effect of the single principle of protection with varying degrees of rigour according to the gravity of the case which, as I have said, represents the true sense of Article 3. I must return to that, but first there are specific points to be made about the first factor (Article 2 imposes a greater duty than Article 3), the third (the obligation is less extensive in a non-State agent case) and the sixth (successful conviction demonstrates the efficacy of the investigation).

As regards the first factor, Mr Johnson cites Banks v United Kingdom (2007) 45 EHRR SE2, which concerned assaults and ill-treatment suffered by prisoners at HMP Wormwood Scrubs:

“In the context of Article 2 of the Convention, the obligation to conduct an effective investigation into allegations of the unlawful use of force attracts particular stringency in situations where the victim is deceased and the only persons with knowledge of the circumstances are officers of the State. It is important, with a view to ensuring respect for the rule of law and confidence of the public, that the facts, and any unlawfulness, are properly and swiftly established. In the context of Article 3, where the victim of any alleged ill-treatment is, generally, able to act on his own behalf and give evidence as to what occurred, there is a different emphasis and… it will not always be necessary, or appropriate, to examine the procedural complaints under the latter provision. The procedural limb of Article 3 principally comes into play where the Court is unable to reach any conclusions as to whether there has been treatment prohibited by Article 3 of the Convention…” (p.22)

In fact the court noted in Banks that there was no lack of any investigation capable of establishing the facts and attributing responsibility; and it was held that there were “no issues” arising under the “procedural” head of Article 3. Ms Kaufmann cited Jordan v United Kingdom (2003) 37 EHRR 2, a case in which a young man had been shot and killed by an officer of the Royal Ulster Constabulary. At paragraph 107 the court said:

“The investigation must also be effective in the sense that it is capable of leading to a determination of whether the force used in such cases was or was not justified in the circumstances… and to the identification and punishment of those responsible… This is not an obligation of result, but of means. The authorities must have taken the reasonable steps available to them to secure the evidence concerning the incident, including inter alia eye witness testimony, forensic evidence and, where appropriate, an autopsy… Any deficiency in the investigation which undermines its ability to establish the cause of death or the person or persons responsible will risk falling foul of this standard.”

Ms Kaufmann’s point is best made by a comparison of the language of that citation with what was said in Vasilyev (Application No 32704/04), an Article 3 case, at paragraph 100:

“Even though the scope of the State’s positive obligations might differ between cases where treatment contrary to Article 3 has been inflicted through the involvement of State agents and cases where violence is inflicted by private individuals…, the requirements as to an official investigation are similar. For the investigation to be regarded as ‘effective’, it should in principle be capable of leading to the establishment of the facts of the case and to the identification and punishment of those responsible. This is not an obligation of result, but one of means. The authorities must have taken the reasonable steps available to them to secure the evidence concerning the incident, including, inter alia, eyewitness testimony, forensic evidence, and so on. Any deficiency in the investigation which undermines its ability to establish the cause of injuries or the identity of the persons responsible will risk falling foul of this standard, and a requirement of promptness and reasonable expedition is implicit in this context…”

The two cases effectively use the same language, and it is replicated elsewhere. In my judgment this is a strong indication that the nature, scope and rigour of the investigative exercise do not in principle shift as between Articles 2 and 3, and I would so hold. Of course there may be practical differences. The fact that in an Article 3 case the victim has survived may be very important if he is able to give his own account. The citation from Banks points the contrast with an Article 2 case where “the only persons with knowledge of the circumstances are officers of the State”. The reality is that all these cases are (to use an over-used phrase) fact-sensitive. The weakness of Mr Johnson’s argument is that it seeks to elevate potential practical differences into rigid differences of principle. The sliding scale of cases, to which I will return, has a nuance which needs to be accommodated: the investigation of ill-treatment does not necessarily require more effort or commitment where the victim is dead.

This rigidity in Mr Johnson’s argument applies also to factor (3), asserting that the investigative obligation is less extensive in a non-State agent case. Of course the investigative requirements of transparency and independence will be especially pressing where there is a serious case that State agents have killed or injured in violation of Article 2 or 3. But in the end these factors are driven by the exigencies of the particular case.

That brings me to factor (6): the supposedly conclusive effect of a prosecution and conviction. In Menson, where the State was accused of racism in relation to its pursuit of allegations concerning the death of a black man, the court regarded the ultimate conviction of the perpetrators as “decisive” (CD230), because it showed the State’s capacity to bring the criminals to book “irrespective of the victim’s racial origin”. In O’Keeffe, the investigation opened after a complaint of sexual abuse was made to the police, and at length the abuser pleaded guilty to a number of charges. No breach of the investigative obligation was found (paragraphs 173-174).

The judge below said this:

“220… The assessment of the efficiency and reasonableness of an investigation also takes into account whether the offender was adequately prosecuted. In this respect, a successful prosecution within a reasonable period of time will render prior operational failures irrelevant (non-justiciable). However a prosecution that is brought after an unreasonable point of time does not in and of itself expunge the legal effect of prior operational failures (Menson ibid).”

There is a subtlety in Green J’s choice of words. I do not understand him to mean that a successful prosecution within a reasonable period of time must always have the result that prior operational failures cannot constitute a violation of the investigative duty under Article 3. If that were what is meant, I would with respect disagree with it: there might, for example, have been an abject failure of investigation but then an unexpected and complete confession by the perpetrator. However such a successful prosecution will generally bring closure to the case, so that an examination of investigative failure may be of little utility to the victim (though it may still be desirable in the public interest).

In these circumstances I do not accept Mr Johnson’s sixth proposition. Neither Menson nor O’Keeffe (nor, so far as I can see, any other authority) supports a rule that a timeous and successful prosecution necessarily demonstrates that there has been an effective investigation.

These qualifications to Mr Johnson’s six “principles” brings the argument back to where I left it at the end of Ground 2: the existence of a single protective principle with varying degrees of rigour – a sliding scale. In this context I should address the decision in Allen, on which Mr Johnson relies; Ms Kaufmann says we are not bound to follow it and should not do so. Gross LJ (with whom Ryder LJ and my Lord the Master of the Rolls agreed) said this at paragraph 43:

“In principle, [the investigative obligation under Article 3] is not limited to cases of ill-treatment by state agents: MC v Bulgaria… at [151]. Importantly, however, the nature of the investigation required, is fact sensitive and will depend on the context: see, R (Takoushis) v Inner North London Coroner [2005] EWCA Civ 1440; [2006] 1 WLR 461, at [104] – [105]. Thus the scope of the State’s obligation may well differ depending on whether the violation of Art. 3 rights is inflicted by agents of the State or private individuals: Beganovic v Croatia (Application no. 46423/06) 25 June 2009, at [69]. By way of an obvious example, the investigation required where there has been systematic torture by State agencies (one end of the spectrum) will differ from that required in respect of misconduct by private individuals narrowly surmounting the minimum threshold for the engagement of Art. 3 (the other end of the spectrum). Thus, in some cases, the State will discharge its investigative obligation through the totality of available procedures, including a criminal investigation and the possibility of criminal, civil and disciplinary proceedings: Takoushis (supra), at [105]; R (NM) v Secretary of State for Justice [2012] EWCA Civ 1182 , per Rix LJ, at [29]. Manifestly, not every arguable breach of Art. 3 calls for a full independent inquiry; there must be a sense of proportion: see, R(P) v Justice Secretary [2009] EWCA Civ 701; [2010] QB 317, per Stanley Burnton LJ, at [51] et seq, including the extensive and valuable citation from the judgment of Longmore LJ in R(AM) v Secretary of State for the Home Department [2009] EWCA Civ 219; [2009] UKHRR 973, at [74] et seq. Furthermore, there is, generally, a difference of emphasis between Art. 2 and Art. 3; in Art. 2 death is involved, whereas, again generally, the victim of a breach of Art. 3 is alive and knows of the acts or omissions said to contravene his ECHR rights: R(P), at [51].”

I have referred to a sliding scale; Gross LJ employed the metaphor of a spectrum. The idea, plainly, is the same. But Ms Kaufmann submits that the reasoning in this paragraph from Allen is wrong to suggest that the criminal investigative duty (her duty no.3) can in some circumstances be discharged by other means, such as civil process. She says that no such issue was argued, and indeed it was not: passages from the submissions made in Allen, which were provided to us, demonstrate as much. We were also shown authority on the doctrine of precedent – Morelle v Wakeling [1955] 2 QB 379 and Iqbal v Whipps Cross University NHS Trust [2007] EWCA Civ 1190 – to equip us to decide whether we might properly depart from Allen.

For my part I have concluded, however, that the reasoning in Allen is not as stark as Ms Kaufmann’s submission suggests. Allen supports the application of a single principle with varying degrees of rigour, which as I have suggested represents the true sense of Article 3’s investigative duty. It is consistent with the proposition that serious violent crime by non-State agents generally requires a proper criminal investigation by the State. It is true that Gross LJ states that “in some cases, the State will discharge its investigative obligation through the totality of available procedures, including a criminal investigation and the possibility of criminal, civil and disciplinary proceedings… [m]anifestly, not every arguable breach of Art. 3 calls for a full independent inquiry”. He cites NM, P v Secretary of State and AM. But those cases do not show, and with respect Gross LJ did not mean (nor did he state), that for the purposes of Article 3 the State has a general and open-ended choice of the means by which it will confront a credible accusation of serious violent crime by non-State agents. Moreover it is plainly right that not every allegation of ill-treatment which meets the Article 3 threshold calls for a full criminal investigation. There will be cases where all the facts are known (see for example P v Secretary of State, paragraph 58); where the actual or putative victim does not want the police involved (as in NM: paragraph 8); or (and I have already referred to this) where the harm is caused by negligence, and there is no criminal act.

I do not mean to suggest that in no such case will a proper criminal investigation ever be mandatory. That would amount to an over-classification at least as rigid as Mr Johnson’s six principles or Ms Kaufmann’s three duties. I mean only that these instances (and there are no doubt others) present features which may, depending no doubt on the details, yield the conclusion that a full criminal investigation is unnecessary or inappropriate or disproportionate. As I have said the notion of a single protective principle, applied with varying degrees of rigour, possesses a nuance: a nuance which explains the different voices in which the cases speak. But this does not undermine the mandatory requirement of a proper criminal investigation in a typical or paradigm case of serious violence.

However the existence of such a mandatory requirement, and (as I would hold) its undoubted application to DSD/NBV, does not exhaust the debate on Ground 3. The need for a proper criminal investigation locates the Article 3 duty within a certain bracket on the scale or spectrum; but it does not tell us what standard the duty’s performance must attain. What constitutes a breach of the duty? What is a “proper” investigation? This brings me back to the contrast between common law claims in negligence and the ECHR.

It will be recalled that in Van Colle/Smith at paragraph 138 Lord Brown observed that “[a]s Lord Bingham pointed out in [Greenfield]…, Convention claims have very different objectives from civil actions…” Greenfield was a case in which a prisoner alleged breaches of ECHR Article 6 arising out of the way in which charges brought against him under the Prison Rules had been dealt with. The Secretary of State accepted the breaches; the question before the House of Lords was whether the appellant should recover damages. Lord Bingham said (paragraphs 3-4):

“The primary aim of the European Convention was to promote uniform protection of certain fundamental human rights among the member states of the Council of Europe… the focus of the Convention is still on securing observance by member states of minimum standards in the protection of the human rights specified in the Convention.”

Then at paragraph 19 he stated that “the [HRA] is not a tort statute. Its objects are different and broader.”

There are important differences between the ECHR’s strategic purpose to secure minimum standards of human rights protection, and the English private law purpose (as Lord Brown described it in Van Colle/Smith) of compensation for loss. It is elementary that in a negligence claim at common law, the court asks whether the defendant owes a duty of care to the claimant: that is, a duty to take reasonable care; and “reasonable” care is generally what a “reasonable” man – traditionally the passenger on the Clapham omnibus – would take it to be (though where the duty is owed by an expert, such as a doctor, the court considers the standard set by his profession). If the duty is established, the question will be whether any act or omission relied on by the claimant (a) constitutes a breach of the duty and (b) has caused the claimant loss; loss is a defining element of the tort.

The process by which a human rights claim is adjudicated is quite different. The starting-point is not the relationship between the claimant and the (State) defendant. It is to ascertain whether the case is within the scope of any of the rights or freedoms which the ECHR requires the State to secure; and then, if it is, to decide whether the State has or has not violated the Article or Articles in question. The possibility of compensation for the individual complainant is secondary: the provision for “just satisfaction” (ECHR Article 41, discussed by Lord Bingham in Greenfield; cf HRA s.8) is essentially discretionary. The focus is on the State’s compliance, not the claimant’s loss.

These points of departure between the ECHR and the common law are not merely theoretical. They mark important differences in practice. The contrast between damages as of right and compensation at the court’s discretion is one. But another, in my judgment, goes to the standard applicable to the ascertainment of breach of the Article 3 investigative duty, as compared with what might constitute breach of a common law duty of care. Because the focus of the human rights claim is not on loss to the individual, but on the maintenance of a proper standard of protection, the court is in principle concerned with the State’s overall approach to the relevant ECHR obligation. This emphasis is in my judgment behind much of the language used in the cases cited to us (the emphasis in what follows is mine): the investigation “should in principle be capable of leading to the establishment of the facts of the case and to the identification and punishment of those responsible. This is not an obligation of result , but one of means … Any deficiency in the investigation which undermines its ability to establish [those matters] will risk falling foul of this standard…” ( Vasilyev paragraph 100, cited above: see also Sigarev [2013] ECHR 17116/04 paragraph 121 and other cases). Lord Bingham’s reference in Greenfield to “minimum standards in the protection of the human rights” is of a piece with these formulations.

I should say that the judge below considered (paragraph 225(iii)) that the use of the term “risk” was “simply loose language”. I do not think so. In my view all of these expressions are intended to emphasise that the enquiry into compliance with the Article 3 duty is first and foremost concerned, not with the effect on the claimant, but with the overall nature of the investigative steps to be taken by the State. This circumstance, moreover, is consonant with the fact that Strasbourg accords a margin of appreciation to the State as to the means of compliance with Article 3. As I have said, the margin widens at the bottom of the scale but narrows at the top. While the doctrine of the margin of appreciation has its origin in the international character of the court, which inevitably stands at some distance from the differing exigencies of the individual States Parties, I have no doubt that we should accord a like margin (more often described on the domestic front as a margin of discretion) in the adjudication of claims under the HRA. Such a margin of discretion is, however, quite foreign to the adjudication of common law claims: once the court has ascertained what the relevant duty of care requires, its remaining task is to decide whether there has been a breach of the duty causing damage. No margin of discretion enters into the exercise.

The practical result of this approach is, I think, reflected in these observations of Green J at paragraph 226 of his judgment:

“A failure to perform an individual act that really could have been performed will not trigger liability [for violation of Article 3] if: (a) notwithstanding that omission the investigation viewed in the round did in fact lead to the arrest of the suspect within a reasonable time; or (b) the investigation (even absent a prosecution) may still be said to encompass a series of reasonable and efficient steps. This is an important point since the Strasbourg case law repeatedly emphasises that the police must be accorded a broad margin of appreciation in the choice of means of investigation.”

I have not so far described what the police did and did not do in the course of investigating Worboys’ crimes. Manifestly this part of the case is critical to the result of the appeal. I may deal with it largely by reference to what was said by Green J, since in my judgment it is inescapable that he was right to find a violation of Article 3.

Green J set out the stated policy of the MPS for the investigation of rape and serious sexual assaults at paragraphs 88-112. Special Notice 11/02 (9 August 2002) was the relevant guidance at the time DSD was attacked. Green J considered, rightly, that “the mere failure to adhere to internal standards and operating procedures [did not engage] liability”, but was nevertheless relevant. I need not set out the policy, but may turn to the judge’s findings of breach. At paragraphs 244-284 he describes what he calls “systemic failures”, and at paragraphs 285-298 “operational failures in the case of DSD’. I can do no better than draw on his account, and I turn first to the former.

At paragraph 245-246 Green J said this:

“245 In my view these ‘systemic’ failings can be accounted for in five different areas: (i) failure properly to provide training; (ii) failure properly to supervise and manage; (iii) failure properly to use available intelligence sources; (iv) failure to have in place proper systems to ensure victim confidence; and (v) failure to allocate adequate resources. In my view these systemic failings are sufficient in themselves to trigger liability… I have also (at (vi) below) benchmarked these five systemic failings against a further IPCC Report into systemic failings identified following an investigation into another serial rapist, Kirk Reid.

246 Each of these systemic failings is recognised by the MPS and the IPCC in their subsequent dissections of what went wrong. Each is at least to some degree interrelated…”

I set out here only the core passages from the judge’s treatment of these systemic failings. This makes for something of a disjointed narrative, and the whole text of Green J’s treatment of the subject repays attention.

“(i)

Failure to provide training to relevant officers

247 First, the officers who investigated the cases of DSD and NBV were not properly trained and, in consequence, failed to take steps which could have resulted in the early apprehension of Worboys. This… explains in large measure why officers in the case of DSD failed adequately to investigate her case and equally why thereafter her case was not reopened until 2008. It also largely explains why NBV was assaulted at all and why her investigation was mishandled.

…

249 … [H]ad they been properly trained… it is entirely possible, and indeed probable, that at least a significant number of the serious failings would not have occurred at all and Worboys might very well have been apprehended and prosecuted very much earlier…

250 … [C]irca 25% of all rapes are first reported at the police counter… It is therefore obvious that such staff need to be especially carefully trained… If in the case of DSD the counter staff at Holloway police station had – because of proper training - been aware that a person presenting in the incapacitated and incoherent state of DSD might… be a victim of DFSA [drug facilitated sexual assault] as opposed to being a drunk or an addict then the officer might have focused upon evidence collection as well as DSD’s medical welfare…

251 … [I]f the frontline officer had taken Worboys’ name and address and/or his vehicle registration number then (i) he might have been arrested earlier or, if not, (ii) the mere fact that his details had been recorded… might have deterred him from continuing with his assaults; but (iii), in any event his details should then have been recorded on a database… If on 7^th May 2003 the front desk at Holloway had taken 30 seconds to record the cab driver’s details and his cab registration then many or even all of the rapes and assaults that ensued might never have occurred.

252 A second example concerns forensic evidence. Forensic evidence might not prove terribly useful in DFSA cases. Semen will rarely be present if the perpetrator uses a condom; some drugs do not stay in the system very long so might not show up or might not be recognized… the effect of many drugs will be magnified if combined with alcohol so that the forensic evidence might well indicate that the complainant had consumed excess alcohol and was ‘drunk’. In the case of both DSD and NBV the forensic reports and the toxicology reports were inconclusive.

…

254 A trained officer thoroughly sensitised through proper training to the particularities of DFSA would not have placed over reliance upon forensic and toxicology reports and would have focused upon super-expedited, and elementary, evidence collection…

255 A third example is the fact that, once again due in my view to an absence of specific training in DFSA, the officers either mischaracterised DSD or failed ultimately to take her complaints seriously. Even after she had awoken in the Whittington and reported to police, there remained the clear view in the minds of officers that she was simply a drunk…

256 In fact, of course, the suggestion [made in police records in the CRIS log] that ‘all steps’ had been taken to identify Worboys was incorrect: the police had from the outset failed to record Worboys’ name or vehicle registration number, or take Kevin’s details and he was never subsequently interviewed; nor had they checked the CCTV of vehicles coming to and from the police station. In my view the long and short of the investigation was a premature conclusion that DSD was a drunk with a coke habit…

...

259 … NBV… also presented to police exhibiting classic features of a victim of DFSA. Yet rather than being cognisant that this was what they were facing the complaint was not even recorded as a serious sexual assault. It was, on the contrary, recorded as a ‘critical incident’ and accordingly no closing report had to be prepared…

260 The failure to provide training was of course not just in relation to the front-line officers. Supervising officers had also not received specialist training. Had they been given this training then it is much more probable that they would have passed it on or ensured that junior officers adhered to the procedures. Once again had this occurred then it is quite possible that Worboys would have been apprehended and prosecuted earlier.

(ii)

Failures in supervision and management: Inappropriate ‘clear up’ pressures/failures to consult the CPS

261 The MPS and IPCC both found in their reports systemic failures to supervise and manage in an effective manner. In my view there are two main reasons for this: First, inadequate training of more senior officers; Secondly, inappropriate pressure from the very highest level of Borough management not to focus upon sexual assaults…

262 In relation to the failure to provide training this was not just in relation to the front-line officers. Supervising officers had also not received specialist training. Had they been given this training then it is much more probable that they would - in accordance with the guidelines - have ensured that junior officers adhered to the procedures. Once again had this occurred then it is quite possible that Worboys would have been apprehended and prosecuted very much earlier than in fact he was…

263 The second issue relating to failures of supervision and management concerns the inappropriate pressure which appears to have emanated from the very highest levels of Borough management not to focus upon sexual assaults, as opposed to other, less complex, offences. This is in the context of the pressure on the MPS to meet performance targets. The first hint of this arose out of the summary of interview of a DI with the IPCC on 26^th June 2009. This concerned alleged failures by officers in relation to the case of NBV. [Green J proceeds to give details.]

…

267 … [F]rom the highest levels of management pressure was imposed which had the effect of incentivising more junior officers not to pursue allegations of sexual assault with the seriousness and intensity that they so manifestly demanded and in encouraging supervising officers to be more willing than they should have been to close files. This will, in my judgment, have contributed materially to the systemic and other operational failings which I have identified. They created an environment in which such failings could thrive… I would add finally that this appears to have been a factor in the case of NBV but there was no evidence before the Court to indicate that it was a relevant consideration in the case of DSD in 2002/2003…

(iii)

Failure to use intelligence resources

268 The third systemic defect concerns the failure to use (or use to any effective ends) available intelligence. This is a much more serious criticism in the case of NBV than in the case of DSD. This is because by the time Worboys assaulted NBV in 2007 about 100 women had already been subjected to his predatory designs and the computer databases should have been brimming with details of vulnerable women being subjected to drug rapes and assaults by a taxi driver…

…

271 It is a quite remarkable fact that the search carried out on 7^th February 2008 was recorded as being ‘routine’… but this uncovered almost immediately 4 allegations of assault with a strikingly similar MO [modus operandi] and this led to the apprehension of Worboys within days.

272 The obvious question to ask is why these links were not identified earlier? The obvious answer is that the systems were not in place which would lead officers to record investigative steps properly and the same officers were not trained to conduct adequate computer cross-checking to seek out links…

273 The particular failings were exacerbated by the fact that individual steps in individual investigations were not taken which, had they been taken and recorded properly, would then have been entered onto computer systems. All of the various failings are interconnected…

(iv)

Failure to maintain confidence with victims

274 Only a tiny fraction of Worboys’ victims reported their assaults to the police prior to February 2008. Of the more than 80 victims who contacted police following the arrest of Worboys, over 60 never reported the incident to police. Originally 12 offences were identified as part of the enquiry into Worboys but following a media appeal in February 2008 about 81 offences were identified of which 72 had occurred in the Metropolitan area.

275 The MPS and IPCC recognise that efficient policing of sexual assault cases depends upon victims feeling able to report their ordeals to the police. A deterrent to this is a perception that their complaints will not be treated seriously or sympathetically…

276 The MPS and IPCC both recognised that the question of victim confidence was at the heart of the problems they faced…

277 The IPCC identified the following which had not hitherto been done but which needed to be done in the future: provision of standard information for victims in terms of what to expect from the investigation and process (time frames, court proceedings, etc). Provision of regular updates and support whilst the case is ongoing; increased provision of public information to encourage other victims to come forward; the provision of more information to local agencies to ‘promote public safety, prevent and detect crimes’; increased liaison and cooperation with the voluntary sector; increased quality checking of front-line training with input from the voluntary sector and from ‘specialist advocates’.

278 In the present case there is tangible evidence of both DSD and NBV not feeling supported or believed… There is evidence that victims (for instance NBV) were fed information that was simply inaccurate about whether her case file had been submitted to the CPS. In the case of DSD she was far too quickly categorised as a drunk whose case could not be prosecuted.

…

280 In terms of determining whether, had a proper system been in place which instilled greater confidence in victims, matters would have been different and the hypothetical improved system would have been capable of identifying, arresting and prosecuting Worboys, then it is possible to identify ways in which this could have occurred…

(v)

Failures to allocate appropriate resources

281 … Had the MPS known the nature and extent of the problem I am quite certain that they would have allocated substantial resources to the capture of Worboys. The failure to deploy adequate resources is hence one component of the systemic failures which characterise this case. The obstacles placed in the way of the allocation of adequate resources are multiple…

(vi)

Benchmarking the systemic failures: The case of Kirk Reid

282 Support for the conclusion that failures in the Worboys case were systemic is found in the Commissioners' report (June 2010) into the MPS investigation into allegations against Kirk Reid. Reid was found guilty on 26^th March 2009 at Kingston Crown Court of 27 sexual offences and two cases of possession of indecent images of children… He would mainly attack lone women during the hours of darkness. The number of offences committed by Reid was estimated to be between 80-100. It appears that most of the offences were committed between August 2001-2008. He and Worboys were prowling the streets at the same time…

283 … From the limited information provided in the IPCC report it appears that a host of operational failings, akin to those occurring in the Worboys case, were perpetrated by officers in relation to Reid. For present purposes it suffices to record that the Commissioner identified the same systemic failings in the Reid case as occurred in the Worboys case…

284 It is, in my view, a significant corroborating factor to my conclusions in the Worboys case that similar systemic and operational failings were identified in the case of Reid and that these were treated by the IPCC as systemic across the entirety of the MPS.”

The judge then turned to what he termed operational failures, first in the case of DSD and then NBV. As regards DSD, he held that the relevant timespan was the 6-year period from May 2003 when DSD first presented to the police until 2009 when Worboys was convicted. Green J divided this period into three. The failures are set out in detail between paragraphs 289 and 297. They are all, as it were, examples of the systemic failures in action. Those attributable to 2003-2004 are as follows (paragraphs 289-293): failure of front desk reception staff to record relevant facts; failure to interview Kevin; failure to collect relevant CCTV evidence; failure to believe DSD or take her complaint seriously; failure properly to supervise. Then from 2004-2008 (paragraph 295): failure to use intelligence sources. From 2008-2009 (paragraph 297): consequential, and “wholly unacceptable” delay in the prosecution and conviction of Worboys.

As regards the operational failures in NBV’s case, the judge found (paragraph 300) that the relevant timeframe was that of the whole investigation into Worboys, again divided into three periods. The first of these was 2003-2007, before NBV was attacked in the early hours of 26 July 2007. The judge held this period to be relevant for three reasons. First, but for the failures during this period it was probable that NBV would not have been raped at all (paragraph 302). Secondly (paragraph 303) the MPS’ own guidance recognizes the need to record the progress of an investigation, because rapes are prone to repetition and the identification of linked MO may assist the prevention of serial rapes. Thirdly (paragraph 304) the very policy which underscores the Article 3 investigative obligation – protection, and therefore prevention – supports this approach.

The operational failures in NBV’s case are described at paragraphs 305-311. Failing adequately to collect intelligence marked the first period (paragraph 305). In the second period there were four “particularly serious” operational errors: failure to conduct proper searches, to conduct proper interviews of Worboys, to follow up CCTV evidence, and to record the NBV incident as a serious sexual offence (paragraphs 306-310). The last period (four months only, between October 2007 and February 2008 when NBV’s case was re-opened) is included because Worboys should, and but for the serious failings in relation to NBV would, have been arrested and prosecuted earlier (paragraph 311).

And so the judge found violations of the Article 3 investigative duty in the case of both claimants. Mr Johnson has no quarrel with his findings or, subject to Ground 4, his evaluation of the facts. If the applicable legal principles are as I have stated them, then (again subject to Ground 4) Green J’s conclusion on liability was in my judgment inevitable.

Aside from his points on the merits Mr Johnson submits (skeleton paragraph 57) that it was not part of NBV’s case that Worboys should have been detected before she was assaulted, and that the judge’s finding went well beyond her pleaded case. I am not impressed with this. The overall case was put in opening without objection on behalf of the MPS.

I have already summarized the judge’s reasons, given at paragraphs 302-304 of the judgment, for holding that the period (2003-2007) before the assault on NBV was part of the relevant timeframe in her case. The first reason was that but for the failures during this period it was probable that NBV would not have been raped at all. Mr Johnson takes issue with this. But it seems to me that the judge’s conclusion was properly open to him. As I have shown the judge found multiple systemic failures, and serious operational failures in DSD’s case, occurring before the attack on NBV.

Green J’s emphasis on the serial nature of Worboys’ crimes is I think important. O’Keeffe v Ireland, to which I have already referred, was also such a case. The court’s observations at paragraph 173 are worth noting in the present context:

“[T]he procedural obligations arise once a matter has been brought to the attention of the authorities… In the present case, once a complaint about the sexual abuse by LH of a child from Dunderrow National School was made to the police in 1995, the investigation opened. The applicant was contacted for a statement which she made in early 1997… LH was charged on 386 counts of sexual abuse involving 21 pupils from Dunderrow National School. LH pleaded guilty to 21 sample charges. He was convicted and imprisoned. It is not clear from the submissions whether the applicant’s case was included in the sample charges: however, she did not take any issue with the fact that LH was allowed to plead guilty to representative charges or with his sentence…”

It is right, as I have said, that no breach of the investigative obligation was found in that case. But there is no suggestion that had there been such failures before the individual applicant made her statement, she would have been unable to rely on them. As Green J said at paragraph 302:

“Nothing in the Strasbourg case law indicates that the timeframe must always start with the assault on the applicant or complainant and common sense indicates that in the case of serial rapists the timeframe for a duty to investigate should be longer and should attach to the conduct of the criminal not the ordeal of the victim.”

Moreover this approach is clearly of a piece with the fact, as I have put it in addressing Ground 3, that the enquiry into compliance with the Article 3 duty is first and foremost concerned, not with the effect on the claimant, but with the overall nature of the investigative steps to be taken by the State.

I would therefore reject Mr Johnson’s argument on Ground 4.

For all the reasons I have given I would dismiss the MPS’ appeal.

Mr Barton for the appellant in Koraou adopted Ms Kaufmann’s submissions on the law, as to which I have set out my conclusions in addressing the DSD/NBV appeal. He made it clear that he had no quarrel with Judge Platts’ primary findings of fact. He also made it clear, as I have already stated (paragraph 4 above) that “[t]he nub of this appeal is [that] the decision to dismiss the claim in its entirety while at the same time finding a series of clear shortcomings/failings in DC Walters’ [the investigating officer] investigation is perverse” (skeleton argument, paragraph 11). Although this is the language of Wednesbury, what Mr Barton has to show is not that the judge’s conclusions were irrational or unsupported by evidence but that on my view of the law, if my Lords agree with it, his findings should have led him to a different result.

As I have said four grounds of appeal are articulated. (1) The judge incorrectly applied the “capability” test. Mr Barton refers in particular to Green J’s statement at paragraph 226 of his judgment, cited by Judge Platts, that “the case law uses the concept of capability in a more proximate and immediate sense as indicating an act which in a material and reasonable way is capable of leading to a positive outcome”. (2) The judge wrongly surmised (my word) that the outcome would have been the same had the proper lines of investigation been carried out. (3) The judge wrongly concluded that certain deliberate or knowing failures by DC Walters were not wholly unreasonable (and therefore, presumably, not to be forgiven by reference to the margin of discretion available to the GMS). (4) The judge should have held that a failure by the police to “clarify matters” with the appellant (concerning the identity of his assailant) amounted, alone or in combination with other failures, to a breach of the investigatory obligation.

The reality is, as Mr Barton accepted in the course of his submissions at the hearing, that the case is “all about” the judge’s exercise of judgment. As to that, after citing Green J in DSD/NBV at some length, Judge Platts addressed the evidence in great detail at paragraphs 7-51 of his judgment. He made findings along the way, including the following. The appellant did not tell the police that Maguire had been involved in the attack when his ear was bitten (paragraphs 23-25), though later he said that he had told officers that Maguire had been involved (paragraphs 46-47). He told Constable Swindells that the assault had been committed by one of two white males in the club (paragraph 28), while the police had information from door staff at the club that the persons involved with the Appellant in the bar were black males (ibid.). DC Walters gave inaccurate and misleading information to her supervising officer that a press release had been issued (paragraph 40). The judge found the appellant’s evidence to be “unreliable in a number of respects” (paragraph 25). He said this of DC Walters:

“35 Generally, I found Constable Walters to be extremely defensive when giving her evidence, no doubt in the light of the number of criticisms which were being made of her investigation, some of which she had to accept; and, in my judgment, that defensiveness on occasions caused her to elaborate her evidence so that in some instances it was not an accurate recollection of events…”

At length the judge encapsulated his criticisms of the investigation:

“72 So the shortcomings in the investigation, which I accept to a greater or lesser extent are, therefore, (1) failure to get statements from the door staff, (2) failure to get statements from or clarify matters with the officers at the scene when it became apparent that there was a conflict or some confusion, (3) failure to ensure that a press release had been done, (4) failure to seek, if it existed, CCTV footage covering the ejection of the male from the rear of the bar, both external and internal, and (5) failure to note the punch thrown by Maguire as the Claimant came out of the bar. I add to that her [DC Walters’] failure to take a full statement from Maguire in August in line with the recommendation of the informal resolution. These are all possible lines and avenues of enquiry which she should or the Defendant in general could have carried out but did not.”

The judge considered that the case was “on the margin of what might properly be described and amount to inhuman or degrading treatment… [T]here is a need to avoid an unacceptable burden being imposed on the police and that is a reason for adopting a cautious approach to the law and not to setting the bar for liability at too low a level…” (paragraph 78). The inconsistency between the appellant’s account that a white male had bitten his ear and the independent evidence from the door staff that the offender was black was “a large stumbling block in the overall investigation right at the outset [which]… the Defendant’s officers were entitled to take… into account when deciding what steps to take in the investigation” (paragraph 79). Then this:

“81 Against that background I then ask the question: was in all the circumstances the investigation carried out by the police reasonable? … [T]his was not the most serious of cases and each of the allegations made by the Claimant has a large question mark hanging over its reliability. It is, therefore, not a case where, in my judgment, it would have been reasonable to leave no stone unturned in the investigation of the crime. Account has to be taken of the fact that police resources are limited… Further, this is not a case where the police did nothing. There were a number of positive steps taken by the Defendant’s officers, particularly at the scene… The CCTV footage was sought and investigated… I acknowledge that there were a number of shortcomings in Detective Constable Walters’s involvement… and I accept that those shortcomings were probably as a consequence of her taking an early view at an early stage in the investigation that the identification and prosecution of the offender was going to be difficult. In my judgment, that was not a wholly unreasonable view for her to take… Constable Walters was supervised by senior officers. They did ask her to carry out certain further investigations and she did not ignore those requests, albeit she did not always carry them out accurately or in the way that they might have expected but, at the end of the day senior officers agreed that this file should be closed and, given the evidential difficulties that there were always going to be in this case, I cannot conclude that that decision of itself was unreasonable.”

Judge Platts asked the compendious question, whether in all the circumstances the police investigation was “reasonable”. This is perhaps a loose approach, but in my judgment his overall treatment of the case is in line with the scope and nature of the Article 3 duty as I have sought to describe it. The duty is first and foremost concerned, not with the effect on the claimant, but with the overall nature of the investigative steps to be taken by the State; and the State enjoys a margin of discretion as to the means of compliance with the duty – a margin which widens at the bottom of the scale (negligence by non-State agents) but narrows at the top (deliberate torture by State officials). Judge Platts has weighed the proved deficiencies of the investigation, its difficulties as he found them to be, and the gravity of the case. In my judgment his conclusion cannot sensibly be faulted.

I would dismiss the appeal in Koraou.

I agree.

I also agree.