Home » United Kingdom » Supreme Court Cases » In the matter of an application by Brigid McCaughey and another for Judicial Review (Northern Ireland)

In the matter of an application by Brigid McCaughey and another for Judicial Review (Northern Ireland)

Table of Contents

before

Lord Phillips, President
Lord Hope, Deputy President
Lord Rodger
Lady Hale
Lord Brown
Lord Kerr
Lord Dyson

JUDGMENT GIVEN ON

18 May 2011

Heard on 2 and 3 February 2011

Appellant Karen Quinlivan Jessica Simor (Instructed by Madden & Finucane)	Respondent Frank O’Donoghue QC Sean Doran BL (Instructed by Coroner’s Service for Northern Ireland)
Respondent Paul Maguire QC Dr Tony McGleenan BL (Instructed by Instructed by Crown Solicitor’s Office)	Intervener Rabinder Singh QC Fiona Doherty BL (Instructed by Northern Ireland Human Rights Commission and Equality and Human Rights Commission)
Intervener John Larkin QC David Scoffield BL (Attorney General for Northern Ireland)

LORD PHILLIPS

Introduction

These appeals require the Court to consider once again the impact of article 2 of the European Convention on Human Rights (“the Convention”) on the scope of an inquest. They arise because of a change that the Grand Chamber of the Strasbourg Court has made as to the nature of the obligations imposed by article 2. I shall start by describing briefly the nature of that change.

The Convention is a living instrument and over time the Strasbourg Court has extended the ambit of application of Convention rights in many areas. Article 2 provides a good example of this tendency. Article 2 provides that

“(1) Everyone’s right to life shall be protected by law. No one shall be deprived of his life intentionally save in the execution of a sentence of a court following his conviction of a crime for which this penalty is provided by law.”

In McCann v United Kingdom (1995) 21 EHRR 97 the Strasbourg Court held that article 2 by implication gave rise not merely to a substantive obligation on the state not to kill people but, where there was an issue as to whether the state had broken this obligation, a procedural obligation on the state to carry out an effective official investigation into the circumstances of the deaths (“the procedural obligation”).

Romania acceded to the Convention on 20 June 1994. In 1993 a pogrom had taken place in a Roma village, resulting in a number of deaths and widespread destruction of property. The State, in the form of the local police, was alleged to have been implicated. Investigations into the pogrom, and proceedings arising out of it commenced in 1993 but continued up to 2000. In Moldovan v Romania (Application Nos 41138/98 and 64320/01) (unreported) 13 March 2001 the applicants sought to invoke the procedural obligation under article 2, and a parallel obligation arising under article 3, alleging various deficiencies in the investigations. The Court held that the Convention only applied with respect to Romania after the date of its accession; it did not apply to Romania at the time of the pogrom. Because the procedural obligation to conduct an effective investigation was “derived from” the killings and the destruction of property, whose compatibility with the Convention could not be examined by the Court, it followed that the complaint of breach of the procedural obligations was also

incompatible ratione temporis with the provisions of the Convention and had to be rejected.

This reasoning was followed by the Court, when reaching similar decisions, in Voroshilov v Russia (Application No 21501/02) (unreported) 8 December 2005 and Kholodov and Kholodova v Russia (Application No 30651/05) (unreported) 14 September 2006.

The issue that the Strasbourg Court considered in these cases was echoed by an issue that arose in this jurisdiction in relation to the application of the Human Rights Act 1998 (“HRA”). In a series of decisions the House of Lords had decided that no claim lay in respect of an alleged breach of the Convention if the facts giving rise to the alleged breach predated the entry into force of the HRA. The issue then arose of whether the procedural obligation to investigate a death applied after the HRA had come into force in relation to a death that had occurred before the Act came into force. In In re McKerr [2004] UKHL 12; [2004] 1 WLR 807 the House of Lords held that it did not. Their reasoning also echoed that of the Strasbourg Court. Because the death occurred before the HRA came into force it was not within the reach of the Act. Because the obligation to hold an investigation was triggered by the death, that consequential obligation was not within the reach of the Act either. This decision was applied by the House of Lords in R (Hurst) v London Northern District Coroner [2007] UKHL 13; [2007] 2 AC 189 and Jordan v Lord Chancellor [2007] UKHL 14; [2007] 2 AC 226.

In 2009 the Grand Chamber of the Strasbourg Court took a decision which departed from the reasoning in Moldovan, Voroshilov and Kholodov and further extended the effect of article 2. In Šilih v Slovenia (2009) 49 EHRR 996 the Court ruled that article 2 imposed, in certain circumstances, a freestanding obligation in relation to the investigation of a death which applied even where the death itself had occurred before the member state ratified the Convention.

The appellants contend that the decision in Šilih has destroyed the basis of the decision of the House of Lords in McKerr. Henceforth, if an inquest is held into a death that predated the coming into force of the HRA there is none the less an obligation under the HRA to ensure that it complies with the requirements of article 2. As the HRA came into force on 2 October 2000 it might be thought that this issue is of no great moment. Such a conclusion would be wrong. These appeals relate to two of a significant number of deaths that occurred in Northern Ireland well before 2 October 2000 in respect of which inquests are still pending.

The facts

The appellants are the next of kin of Martin McCaughey and Dessie Grew who were shot and killed by members of the British Army on 9 October 1990. Allegations have been made that they were victims of a shoot-to-kill policy. I need not go into the reasons why it is that, despite the fact that they died so long ago, no inquest has yet been held into their deaths, for on these appeals no issue arises in respect of this delay. A Coroner has been assigned to the case, and on 14 September 2009 he held a preliminary hearing. At that hearing the appellants applied to the Coroner for a ruling that he would hold an inquest into the two deaths that complied with the requirements of article 2. The Coroner declined to give such a ruling. He indicated, however, that he intended to hold a vigorous, thorough and transparent inquest.

Following a further hearing, on 1 December 2009 the Coroner issued the following “preliminary definition” of the scope of the inquest that he proposed to hold:

“The Coroner will consider the four basic factual questions concerning: (a) the identity of the deceased; (b) the place of death;

Further, related to the ‘how’ question, the Coroner will examine in evidence the surveillance operation that culminated in the deaths with reference in particular to the following: (i) the purpose of the operation; (ii) the planning of the operation; (iii) the actions of those involved in the operation; (iv) the state of knowledge of those involved in the operation; (v) the nature and degree of the force used in the operation. In considering this matter, the Coroner will also examine such evidence as exists concerning the circumstances in which the deceased came to be at the locus of death at the relevant time.”

He stated that this was only preliminary and might be subject to revision at any time. He invited written representations from the parties in relation to it.

The appellants made representations to the Coroner to the effect that the scope of the inquest should cover the question of whether the operation was planned and controlled so as to minimise to the greatest extent possible recourse to lethal force. The Police Service of Northern Ireland and the Ministry of Defence made written representations which asserted that the Coroner was precluded from

investigating the planning and control of the operation. They asserted that McKerr established that there was no requirement for the inquest to comply with article 2. It followed that the scope of the inquest was restricted to establishing “by what means” rather than “in what broad circumstances” the deceased came to their deaths, ie a Jamieson inquest: see R v Coroner for North Humberside and Scunthorpe, Ex p Jamieson [1995] QB 1.

The Coroner has not yet made any ruling in relation to these representations. He is, no doubt, wisely awaiting the outcome of these proceedings. It is by no means clear that, even if this Court rules that article 2 has no application to these inquests, the Coroner will not be able lawfully to conduct an inquest which satisfies the requirements of that article or that he will not do so: see the speech of Lord Bingham of Cornhill in Jordan and my comments at para 69 in R (Smith) v Oxfordshire Assistant Deputy Coroner (Equality and Human Rights Commission intervening) [2010] UKSC 29; [2011] 1 AC 1. What is clear is that a decision of this Court is needed to prevent the delay and expense involved in interlocutory in-fighting in this and future inquests raising the same issue.

The proceedings below

In an application for permission to apply for judicial review [2009] NIQB 77 the appellants sought to persuade Weatherup J that they were entitled to a declaration that the Coroner was obliged to conduct the inquest in a way that satisfied the procedural obligation of article 2. They argued that he was not bound by McKerr because the subsequent decision of the Grand Chamber in Šilih was inconsistent with it. Weatherup J held that whether or not Šilih was inconsistent with McKerr, he was obliged by the law of precedent to follow the latter. Accordingly he refused the appellants the relief that they sought.

The Court of Appeal in Northern Ireland [2010] NICA 13 agreed with Weatherup J that McKerr was binding, even if inconsistent with Šilih. Indeed that much was conceded by Ms Quinlivan for the appellants. She did not succeed in persuading the Court that Šilih was, in fact, in conflict with McKerr. She did, however, persuade them that it was possible that this Court would “choose to extend Šilih to our domestic law”, that they should give the appellants leave to apply for judicial review and, having denied them substantive relief, grant them permission to appeal to this Court.

The Strasbourg law at the time of the decision of the House of Lords in McKerr

As a backcloth to the decision of the House of Lords in McKerr it is helpful to identify the characteristics of the procedural obligation as laid down by the Strasbourg Court at the time of that decision.

McCann, the “death on the Rock” case, was the first occasion on which the Strasbourg Court identified that article 2 gave rise to the procedural obligation. In so doing the Court was responding to a submission made by the applicants. The applicants complained about the planning, or lack of planning, that had taken place before the shooting of the IRA unit. They alleged that the shooting had either been premeditated or had resulted from negligence. And they complained that the inquest that had been held into the deaths had not constituted an adequate investigation into the circumstances of the killings. They submitted (see para 185 of the Commission’s Opinion) that

“Article 2 should be interpreted as including a procedural element, namely, the provision of an effective procedure after the event for establishing the facts… the procedures in this case were inadequate.”

The Court accepted this last submission. It held, at para 161:

“[The Court] confines itself to noting, like the Commission, that a general legal prohibition of arbitrary killing by the agents of the state would be ineffective, in practice, if there existed no procedure for reviewing the lawfulness of the use of lethal force by state authorities. The obligation to protect the right to life under this provision [article 2], 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 some form of effective official investigation when individuals have been killed as a result of the use of force by, inter alios, agents of the state.”

The Court held at para 162 that there was no need to decide on what form such investigation should take as the inquest that had been held had satisfied the procedural obligation.

The Strasbourg Court repeated the McCann formulation of the ancillary duty in Kaya v Turkey (1998) 28 EHRR 1 and Güleç v Turkey (1998) 28 EHRR 121.
In McCann no issue of jurisdiction ratione temporis arose, for both the killings and the inquest had taken place long after the ratification of the Convention by the United Kingdom. That issue arose for the first time in Moldovan.

I have already outlined the facts of Moldovan at para 3 above. The applicants complained that the police had been implicated in the pogrom and that a failure to carry out an adequate criminal investigation had deprived them of their opportunity to file a civil claim for damages against the state. The relevant passage in the Court’s judgment on admissibility is at p 13:

“In the present case, the Court notes that the killings happened in September 1993 before the entry into force of the Convention with regard to Romania, ie 20 June 1994. However, in accordance with the generally recognised rules of international law, the Convention only applies in respect of each contracting party to facts subsequent to its coming into force for that party. The possible existence of a continuing situation must be determined, if necessary ex officio, in the light of the special circumstances of each case (eg, [X and Y v Portugal (1979) 16 DR 209]). The Court must therefore verify whether it is competent ratione temporis to examine the present complaint.

It observes however that the alleged obligation under the Convention of the Romanian authorities to conduct an effective investigation capable of leading to the identification and punishment of all individuals responsible for the deaths of the applicants’ relatives is derived from the aforementioned killings whose compatibility with the Convention cannot be examined by the Court.”

This passage suggests that the Court considered that it was implicit in the applicants’ case that the procedural obligation was a continuing obligation which would persist until an investigation was held that satisfied article 2. The Court held that it was not necessary to consider whether this was correct. It was enough to deprive the Court of competence that the obligation asserted was to investigate the circumstances of deaths whose compatibility with the Convention could not be examined by the Court.

Moldovan left unanswered the question of whether, if there is an initial obligation under article 2 to conduct an adequate investigation, that obligation continues until such an investigation has been held. That question had, however, been answered by the Commission in a decision on admissibility in McDaid v United Kingdom (1996) 85-A DR 134. 14 residents of Derry made an application

alleging inter alia that there had been a breach of the procedural obligation under article 2 to hold a full investigation into the “Bloody Sunday” killings in 1972. They alleged inadequacies in the Widgery Report, an investigation conducted by the RUC and the inquest that had been held into the deaths. In an attempt to get round the six month time limit for bringing a complaint they submitted that this was a continuing obligation. In rejecting that submission the Commission said this, at p 140:

“In so far as the applicants complain that they are victims of a continuing violation to which the six month is inapplicable, the Commission recalls that the concept of a ‘continuing situation’ refers to a state of affairs which operates by continuous activities by or on the part of the state to render the applicants victims (see, eg, [Montion v France (1987) 52 DR 227; Hilton v United Kingdom (1988) 57 DR 108; A P v United Kingdom (Application No 24841/94) (unreported) 30 November 1994]). Since the applicants’ complaints have as their source specific events which occurred on identifiable dates, they cannot be construed as a ‘continuing situation’ for the purposes of the six month rule. While the Commission does not doubt that the events of ‘Bloody Sunday’ continue to have serious repercussions on the applicants’ lives, this however can be said of any individual who has undergone a traumatic incident in the past. The fact that an event has significant consequences over time does not itself constitute a ‘continuing situation’ ”.

A precursor to the judicial review proceedings brought by Mr McKerr which ended in the House of Lords was an application that he made to the Strasbourg Court (2001) 34 EHRR 553. It was heard together with three other applications which raised similar issues. His complaint related to the killing of his father on 11 November 1982 by a special unit of the RUC. He alleged, inter alia, that there had been a failure to satisfy the article 2 procedural obligation. Specific complaint was made of a police investigation, a criminal trial, an independent police inquiry and an inquest. These complaints were, in large measure, upheld by the Court. At para 111 the Court expanded on the purpose of the procedural obligation beyond the explanation given in McCann:

“The essential purpose of such an investigation is to secure the effective implementation of the domestic laws which protect the right to life and, in those cases involving state agents or bodies, to ensure their accountability for deaths occurring under their responsibility. What form of investigation will achieve those purposes may vary in different circumstances. However, whatever

mode is employed, the authorities must act of their own motion, once the matter has come to their attention.”

This then became a standard part of the summary of the Strasbourg law given by the Court in cases where the procedural obligation was in issue.

By way of summary of this jurisprudence, the procedural obligation was initially closely related to the substantive obligation that article 2 imposed on the state. The object of the procedural obligation was to check whether there had been a breach of the substantive obligation. If the substantive obligation did not exist the procedural obligation could not exist either. The expanded reason for the procedural obligation given by the Court in McKerr arguably extended the obligation to circumstances in which there was no allegation that the death had resulted from breach by the state of its primary duty under article 2. None the less the instances where there had been a complaint of breach of the procedural obligation had always involved allegations of state implication in the death. Furthermore the complaints had focussed on alleged deficiencies in the historic investigations that had been held. In no case had it been held that there was a continuing obligation to hold an adequate investigation, and in McDaid the Commission had held that there was not.

There is one discordant note. In Balasoiu v Romania (Application No 37424/97) (unreported) 20 April 2004 the applicant complained of being beaten up by police in 1993, before Romania had acceded to the Convention, but also of delay in the procedural investigation, which was still ongoing. The Court held the latter complaint admissible, without explaining how this could be reconciled with the reasoning of the Court in Moldovan.

The reasoning of the House of Lords in McKerr

I must now consider the decision in McKerr in greater detail. This is necessary because the question that arises on the current appeals is whether the reasoning in McKerr applies in the light of the change in nature of the procedural obligation that the decision in Šilih has produced.

Having secured a finding by the Strasbourg Court of historic failures to comply with the procedural obligation imposed by article 2, Mr McKerr returned to the domestic courts. In judicial review proceedings he contended that the procedural obligation was a continuing obligation and that failure to provide an article 2 compliant investigation was unlawful and a breach of section 6 of the HRA. The Court of Appeal in Northern Ireland upheld this contention and made a

declaration that the Government had failed to comply with article 2. The majority in the House of Lords addressed an appeal advanced on the premise that article 2 gave rise to a continuing obligation to hold an adequate investigation that had persisted for over 20 years from the time of the death of Mr McKerr’s father: see paras 25, 48, 61, 76. This premise was, however, questioned by Lord Hoffmann in para 66, Lord Rodger at para 80 and Lord Brown at paras 92 to 95, rightly in my view: see para 22 above.

Section 6 of the HRA provides that it is unlawful for a public authority to act in a way which is incompatible with a Convention right. The issue before the House was whether, on true interpretation of the HRA, this provision applied to the assumed continuing procedural obligation, notwithstanding that it had its origin in a death that occurred before the HRA had come into force. By the time of McKerr it was well established that the death itself could not give rise to any breach of the HRA because the Act did not apply to conduct that took place before it came into force: see R v Kansal (No 2) [2001] UKHL 62; [2002] 2 AC 69; R v Lambert [2001] UKHL 37; [2002] 2 AC 545; Wilson v First County Trust Ltd (No 2) [2003] UKHL 40; [2004] 1 AC 816.

The Lords were unanimous in holding that the obligations created by the HRA did not, on the true interpretation of that Act, extend to the assumed continuing procedural obligation. The reasoning of Lord Nicholls appears in the following passages:

“19. … The application of section 6(1) of the Human Rights Act to a case of an unlawful killing is straightforward. Section 6(1) applies if the act, namely, the killing, occurred after the Act came into force. Section 6(1) does not apply if the unlawful killing took place before 2 October 2000. So much is clear.

The position is not so clear where the violation comprises a failure to carry out a proper investigation into a violent death. Obviously there is no difficulty if the death in question occurred post-Act. The position is more difficult if the death occurred, say, shortly before the Act came into force and the necessary investigation would fall to be held in the ordinary course after the Act came into force. On which side of the retrospectivity line is a post-Act failure to investigate a pre-Act death?

In my view the answer lies in appreciating that the obligation to hold an investigation is an obligation triggered by the occurrence of a violent death. The obligation to hold an investigation does not exist

in the absence of such a death. The obligation is consequential upon the death. If the death itself is not within the reach of section 6, because it occurred before the Act came into force, it would be surprising if section 6 applied to an obligation consequential upon the death. Rather, one would expect to find that, for section 6 to apply, the death which is the subject of investigation must itself be a death to which section 6 applies. The event giving rise to the article 2 obligation to investigate must have occurred post-Act.

I think this is the preferable interpretation of section 6 in the context of article 2. This interpretation has the effect, for the transitional purpose now under consideration, of treating all the obligations arising under article 2 as parts of a single whole. Parliament cannot be taken to have intended that the Act should apply differently to the primary obligation (to protect life) and a consequential obligation (to investigate a death). For this reason I consider these judicial review proceedings are misconceived so far as they are sought to be founded on the enabling power in section 7 of the 1998 Act.”

At para 25 Lord Nicholls explained why it was that some of the lower courts had erred in reaching a contrary conclusion in earlier cases. This was

“by failing to keep clearly in mind the distinction between (1) rights arising under the Convention and (2) rights created by the 1998 Act by reference to the Convention. These two sets of rights now exist side by side. But there are significant differences between them. The former existed before the enactment of the 1998 Act and they continue to exist. They are not as such part of this country’s law because the Convention does not form part of this country’s law. That is still the position. These rights, arising under the Convention, are to be contrasted with rights created by the 1998 Act. The latter came into existence for the first time on 2 October 2000. They are part of this country’s law. The extent of these rights, created as they were by the 1998 Act, depends upon the proper interpretation of that Act. It by no means follows that the continuing existence of a right arising under the Convention in respect of an act occurring before the 1998 Act came into force will be mirrored by a corresponding right created by the 1998 Act. Whether it finds reflection in this way in the 1998 Act depends upon the proper interpretation of the 1998 Act.”

Lord Steyn at para 48 appears to have decided the issue on the simple basis that there was no continuing article 2 obligation to hold an investigation. Any breach of the procedural obligation occurred before the HRA came into force and could give rise to no remedy under domestic law.

Lord Hoffmann at para 66 reasoned as follows:

“the fallacy of the reasoning lies in the notion of a ‘continuing breach’ of articles 2 and 3. The judge was concerned with the rights of the claimants in domestic law. Before 2 October 2000, there could not have been any breach of a human rights provision in domestic law because the Act had not come into force. So there could be no continuing breach. There may have been a breach of article 2 as a matter of international law and this may have ‘continued’ after 1 October 2000, although, for the reasons given by my noble and learned friend, Lord Brown of Eaton-under-Heywood, I think it unlikely. But that is irrelevant to whether the claimants had rights in domestic law, for which there can be no source other than the 1998 Act. The Act did not transmute international law obligations into domestic ones. It created new domestic human rights. The simple question is whether as a matter of construction, those rights applied to deaths which occurred before the Act came into force.

67. Your Lordships’ House has decided on a number of occasions that the Act was not retrospective. So the primary right to life conferred by article 2 can have had no application to a person who died before the Act came into force. His killing may have been a crime, a tort, a breach of international law but it could not have been a breach of section 6 of the Act. Why then should the ancillary right to an investigation of the death apply to a person who died before the Act came into force? In my opinion it does not. Otherwise there can in principle be no limit to the time one could have to go back into history and carry out investigations. In R (Wright) v Secretary of State for the Home Department Jackson J was prepared to accept the possibility of investigations into breaches of article 2 ‘during the 50- year period between the UK’s accession to the Convention and the coming into force of the [1998 Act]’. But that was because he regarded an international law right under the Convention as a necessary (and sufficient) springboard for a domestic claim on the basis of a ‘continuing breach’. In my opinion, however, the international law obligation is irrelevant. Either the Act applies to deaths before 2 October 2000 or it does not. If it does, there is no reason why the date of accession to the Convention should matter. It

would in principle be necessary to investigate the deaths by state action of the Princes in the Tower.”

Lord Hoffmann added at para 69:

“In my opinion Parliament intended section 6 of the 1998 Act to be enforced, but enforced only in respect of breaches occurring after it came into force.”

I have not, with respect, found all of this reasoning easy to follow. Some of it does not appear to focus on the basis of the claim, which was that a continuing obligation under the Convention had given rise to an obligation under domestic law once the HRA came into force. He appears to have concluded, however, that any breach of the Convention that had taken place occurred at the time of the death, which was before the HRA had come into force, so that it gave rise to no breach of domestic law.
Lord Rodger’s reasoning appears in the following paragraphs of his speech: “79. What the applicant claims, however, is an article 2 Convention

right under the Act to have his father’s death investigated even

though, as he accepts, the killing did not violate, and is not to be regarded as having violated, any article 2 Convention right under the Act. Such a claim is fatally flawed and must be rejected.

Like Lord Brown I am doubtful whether, even in international law terms, there was by October 2000 any continuing breach of the relatives’ right to an effective investigation of Gervaise McKerr’s death under article 2 of the Convention. But, even supposing that there was, that continuing breach of an international obligation was not turned into a continuing breach of an article 2 Convention right in domestic law when the Act came into force. Any breach that there was remained a breach in international law and nothing more. The applicant relies on the Act as part of the domestic law of Northern Ireland. Under the Act the right to an investigation, deriving from an article 2 Convention right, presupposes that the killing could have been in violation of that selfsame Convention right. So, when the applicant’s father was killed in 1982, his relatives had no right to an investigation under the Act. Moreover, since the Act is not retroactive, they are not now to be regarded as having had such a right in 1982 or at any time after that. Conversely, the Secretary of

State is not to be regarded as having been in breach, or continuing breach, of such a right either in 1982 or at any time after that.

What the applicant is really saying, therefore, is that, when the Act came into force, it conferred on him a right under article 2 to have his father’s death investigated even though his killing was not, and is not to be regarded as having been, in breach of any article 2 Convention right under the Act. Therefore, the applicant is not asking the courts to apply the Act according to its terms, but to amend them so as to fit this case. That cannot be done. If Parliament had intended the rights under article 2 to be split up, with the Act applying differently to the different aspects, then it would have provided for this expressly. The potential objections are obvious. It would be curious to give a right, under the Act, to an investigation of a killing to which the Act did not apply. If there were to be such a right to an investigation, how far back would it go? Speculation is fruitless: what matters is that Parliament could have made, but did not make, any such transitional provision. The obvious conclusion is that the right to an investigation under the Act is confined to deaths which, having occurred after the commencement of the Act, may be found to be unlawful under the Act. The applicant seeks to contradict the policy of Parliament.”

There were a number of strands in this reasoning. The one that most closely echoed the reasoning of the Strasbourg Court in Moldovan was that if the Act did not apply at the time of the death, the Act could not sensibly impose an obligation to have an investigation into the circumstances of the death. But Lord Rodger also based his decision on an objective assessment of what Parliament must have intended.

Lord Brown, at paras 88 and 89, referred to the object of the procedural investigation as set out by the Strasbourg Court in McKerr at para 111. He observed that the article 2 obligation asserted was

“a procedural obligation properly to be regarded as secondary or ancillary or adjectival to the substantive obligation to protect life, an obligation arising directly out of the loss of a life.”

Lord Brown then held that because the obligation to investigate was linked to the death it could not arise under domestic law:

“The duty to investigate is, in short, necessarily linked to the death itself and cannot arise under domestic law save in respect of a death occurring at a time when article 2 rights were enforceable under domestic law, ie on and after 2 October 2000.”

The issue in McKerr was whether, on true construction of the HRA, the article 2 procedural obligation could give rise to obligations under the Act after the Act had come into force notwithstanding that the death that triggered that obligation had occurred before the Act came into force. The House of Lords gave a negative answer to that question. In doing so their reasoning echoed that of the Strasbourg Court when giving a negative answer to the different question of whether the procedural obligation could arise under the Convention after it came into force in a particular state when the death that triggered the obligation occurred before the Convention had come into force in respect of that state. The echo was, however, unconscious. Moldovan was not referred to in McKerr and no analogy was drawn with the principle established by that decision.

Further developments at Strasbourg before Šilih

In Voroshilov the Court applied, and extended, the reasoning in Moldovan in relation to an application that alleged breach of the procedural obligation in respect of a breach of the substantive obligation under article 3. The applicant alleged that he had been tortured by the police in 1997, the year before the Convention came into force in respect of the Russian Federation. Criminal proceedings were commenced in 1997 but had not concluded. The applicant claimed that the state had failed to conduct an adequate investigation into his treatment. The Court held the applicant’s complaint inadmissible for the following reasons:

“The Court observes that the procedural obligation under article 3 arises where an individual makes ‘a credible assertion’ of having suffered treatment contrary to article 3 (see Labita, cited above). However, since the Court is prevented from examining the applicant’s assertions relating to the events outside its jurisdiction ratione temporis, it is unable to reach a conclusion as to whether the applicant has made a ‘credible assertion’ as required by the above provision. Accordingly, it cannot examine whether or not the Russian authorities had an obligation under the Convention to conduct an effective investigation in the present case (see Moldovan v Romania (Application No 41138/98, 13 March 2001) Likewise the alleged failure to conduct the investigation cannot be held to constitute a continuous situation raising an issue under article 3 in

the present case, since the Court is unable to conclude that such an obligation existed.”

The Court applied similar reasoning to the same effect in Kholodov.

The Strasbourg Court gave further consideration to the object of the procedural obligation in Angelova and Iliev v Bulgaria (Application No 55523/00) (unreported) 26 July 2007. The applicants were the mother and brother of a Roma man who had been stabbed to death by a gang of teenagers. There was no suggestion that the killing itself was accompanied by any direct involvement of the state. The applicants complained, however, of inadequacies in the investigation of the crime carried out by the police. The Court held at para 93 that the absence of any direct state responsibility for the death did not exclude the applicability of section 2. The Convention imposed a duty to secure the right to life by putting in place effective criminal law provisions to deter the commission of offences backed by law-enforcement machinery. The Court continued at para 94:

LORD HOPE

The decision of the Grand Chamber of the Strasbourg Court in Šilih v Slovenia (2009) 49 EHRR 996 requires us to take a fresh look at the present state of our domestic law about the holding of inquests in cases where the death occurred before the Human Rights Act 1998 was commenced on 2 October 2000. There are two issues. The first is whether article 2 of the European Convention on Human Rights as interpreted in McCann v United Kingdom (1995) 21 EHRR 97, para 161 gives rise to a procedural obligation on the state to carry out an effective public investigation into the circumstances of a death where agents of the state are, or may be, in some way implicated, even though because the death occurred before 2 October 2000 the substantive obligation does not apply to it in domestic law. The second is whether, if there is no such obligation in domestic law but the state nevertheless decides to carry out an investigation into a pre-commencement death of that kind, the investigation which it carries out must meet the procedural requirements of article 2 as explained in R (Middleton) v West Somerset Coroner [2004] UKHL 10, [2004] 2 AC 182.

These questions require us to attempt to understand what the Grand Chamber’s decision in Šilih means, and then to see how the message that it conveys can be fitted into domestic law. As we are concerned with pre- commencement deaths, the focus must be on the transitional provisions in section 22(4) of the 1998 Act. The struggle which some members of the House of Lords, including myself, had with those provisions are now a distant memory: see R v Kansal (No 2) [2002] 2 AC 69; R v Lambert [2002] 2 AC 545; Wilson v First County Trust Ltd (No 2) [2004] 1 AC 816. But they must now be addressed again, as the decision in Šilih appears to have detached the procedural obligation under article 2 from the substantive obligation. The question is whether giving effect to that decision in either of the two ways referred to above would contravene the policy choice by Parliament to which section 22(4) of the Act gave effect.

The policy choice was that, save to the limited extent referred to in section 22(4), the Act was not to apply retrospectively. As Lord Nicholls of Birkenhead said in Wilson v First County Trust Ltd, para 12, sections 6 to 9 are forward looking in their reach: one would not expect a statute promoting human rights values to render unlawful acts which were lawful when done. I would add that, as section 6(6) provides that “an act” includes a failure to act, one would not expect it to apply to failures to act which were not unlawful when the alleged failure occurred.

The deaths with which these appeals are concerned took place in October 1990. The papers were passed to the Coroner in 1994, but they were incomplete as they omitted statements from the soldiers who committed the killings. Those

statements were not provided to him until 2002. It was not until 14 September 2009 that the Coroner held the preliminary hearing in which he was asked to hold an inquest which complied with the procedural requirements of article 2. It is common ground that, as the deaths occurred before article 2 was made part of domestic law, the substantive aspects of that article cannot be applied to them under the Human Rights Act. Section 22(4) of the Act precludes this. Sections 6(1) and 7(1)(a) of the Act do not apply because the killings occurred before the Act came into force. Any attempt that might have been made in domestic law prior to 2 October 2000 to require the Coroner to carry out an investigation into them that met the requirements of article 2 would have been bound to fail. Human rights had not yet been brought home. The simple fact is that from the date when the deaths occurred to the date immediately before the Act came into force there was no obligation to investigate these deaths in the manner that meets the procedural requirements of article 2 under domestic law. The House of Lords held in In re McKerr [2004] 1 WLR 807 that, where there had been no breach of the procedural obligation before 2 October 2000, there could be no continuing breach thereafter.

But the proceedings that the Coroner was asked to undertake when the papers were passed to him were still pending when the Human Rights Act came into force, and they are still pending. In Šilih, para 159 the Grand Chamber held that the procedural obligation to carry out an effective investigation under article 2 has evolved into a separate and autonomous duty. It can give rise to a finding of a separate and independent interference with the article 2 right which is capable of binding a State even when the death took place before it ratified the Convention. In so doing, as Judges Bratza and Türmen pointed out in their joint dissenting opinion, it departed from the general rule that the provisions of the Convention do not bind a contracting party in relation to any act or fact which took place before the date of their entry into force with regard to that party.

The question is whether section 22(4) of the Human Rights Act permits that novel approach to be adopted in domestic law. In para 21 of McKerr Lord Nicholls said that the obligation to hold an investigation is an obligation triggered by the occurrence of a violent death:

“The obligation to hold an investigation does not exist in the absence of such a death. The obligation is consequential upon the death. If the death itself is not within the reach of section 6, because it occurred before the Act came into force, it would be surprising if section 6 applied to an obligation consequential upon the death.”

So, for the new cause of action in section 6 to apply to it, the death which is the subject of investigation must itself be a death to which section 6 applies. The event giving rise to the article 2 obligation to investigate must have occurred after the

Act came into effect. He went on in para 22 to say that in this way, for the transitional purpose, all obligations arising under article 2 must be treated as parts of a single whole:

“Parliament cannot be taken to have intended that the Act should apply differently to the primary obligation (to protect life) and a consequential obligation (to investigate a death).”

The reasons given by Lord Hoffmann at para 67, by Lord Rodger of Earlsferry at para 81 and by Lord Brown of Eaton-under-Heywood at para 89, as I read them, are to the same effect. Lord Rodger said that, if Parliament had intended the rights under article 2 to be split up, with the Act applying differently to the different aspects, it would have provided for this expressly.

When it made those observations the House was, of course, approaching the matter as one of domestic law. But it was doing so at a time when there was no indication in the jurisprudence of the Strasbourg Court that the procedural obligation under article 2 was detachable from the substantive obligation. The matter was being approached on the assumption, as Lord Nicholls made clear in para 21 of his speech, that it was not. The question whether or not those obligations can be detached from one another is pre-eminently a matter for Strasbourg. Section 2(1) of the Act tells us that in determining any question that has arisen in connection with a Convention right we must take into account any judgment of the European Court of Human Rights so far as, in our opinion, it is relevant to the proceedings in which the question has arisen.

The judgment in Šilih is such a judgment. If we can find a clear ruling on the point in that judgment we should follow it, even if this contradicts the assumption on which the House based its decision in McKerr. We are not absolutely bound to do so, as section 2(1) does not go that far. But our practice is to follow the guidance of the international court unless there are strong reasons for not doing so. As Lord Bingham of Cornhill said in R (Ullah) v Special Adjudicator [2004] UKHL 26, [2004] 2 AC 323, para 20, its case law is not strictly binding but courts should, in the absence of some special circumstances, follow any clear and constant jurisprudence of the Strasbourg court as the Convention is an international instrument, the correct interpretation of which can be authoritatively expounded only by that court. He added that it followed that a national court subject to a duty such as that imposed by section 2 should not without strong reason dilute or weaken the effect of the Strasbourg case law.

Only the most starry-eyed admirer of the Strasbourg court could describe the guidance that the Grand Chamber offered in para 163 of its judgment in Šilih

as clear. Judge Lorenzen’s criticisms in his concurring opinion of the reasoning of the majority and the powerful dissenting opinion of Judges Bratza and Türmen show that it has many shortcomings. But I do not think that it is difficult to identify the point that matters for our purposes. In para 159 the Court concluded from the fact that it had consistently examined the question of procedural obligations under article 2 separately from the question of compliance with the substantive obligation, and where appropriate found a separate violation of article 2 on that account, as revealed by the decisions listed in para 158, that the procedural obligation to carry out an effective investigation under article 2 had evolved into a separate and autonomous duty. Although it was “triggered” by the acts concerning the substantive aspects of article 2, it could be considered to be a “detachable” obligation arising out of that article which was capable of binding a state even when the death took place before it ratified the Convention.

Support for the view that this was the essential basis for the decision is to be found in para 9 of the dissenting opinion, where Judges Bratza and Türmen said that they could not agree with the reasoning of the majority which was founded on the alleged “detachability” of the procedural obligation from the substantive obligation. In para O-IV14 they said that, while they had no quarrel with the idea that the procedural obligation had evolved into a “separate and autonomous duty”, they differed from them as to their view that it was “detachable” from the death which gives rise to it. While the working out of the concept as indicated in para 163 of the Court’s judgment is far from clear, the ruling that the procedural obligation is detachable from the death is not.

I return then to the two questions which I identified in para 65, above. As to the first question, I see no reason to disagree with the way it was answered in McKerr. One must distinguish between rights arising under the Convention and rights created by the Act by reference to the Convention. The effect of section 22(4) of the Act is that the rights created by the Act came into existence for the first time on 2 October 2000: Lord Nicholls in McKerr, para 25. The right to an investigation under the Act in domestic law is confined to deaths which, having occurred after the commencement of the Act, may be found to be unlawful under the Act: Lord Rodger in McKerr, para 81. The “trigger” that gives rise to the procedural obligation under the Convention is prevented from operating in domestic law in the case of a pre-commencement death by the bar that has been applied by section 22(4). I agree with Lord Rodger (see para 161, below) that, if it were otherwise, a time limit to identify which deaths trigger the duty in domestic law, and which do not, would have had to have been written into the Act. There is none, and I do not think that we should be deflected from holding fast to that position by the way the Strasbourg court has attempted to work out the limits of its own temporal jurisdiction for its own purposes in para 163 of Šilih. The concept of a temporal connection with a pre-commencement event, so as to bring the event

itself within the scope of domestic law, is entirely alien to the system that the Act lays down.

As I see it, however, the second question can and does admit of a different answer. We are told by Strasbourg that the procedural obligation, as now understood, has a life of its own as it is detachable from the substantive obligation. Furthermore, there is no need for a trigger to bring the obligation into operation in this case, as it has been decided that an inquest is going to be held into these deaths. The objection that this would be giving retrospective operation to section 6 of the 1998 Act does not arise. The question whether the inquests must satisfy the procedural requirements of article 2 otherwise they will be unlawful in terms of that section is being directed to something that has yet to take place. The answer to it is not to be found in McKerr, as the House treated the procedural and the substantive obligations in that case as inseparable.

Lord Rodger says (see para 155, below) that to approach the issue in this way does not reflect the decision in Šilih. I, for my part, think that it does. It is true that it does not say this in terms. What the decision seeks to do in para 163 is to identify those pre-ratification deaths that will bring the procedural obligation into effect after the date of ratification. Its concern is with the circumstances that the Strasbourg court will accept jurisdiction in such cases. The question whether there is an article 2 obligation to investigate these deaths in domestic law is a different question. But the holding of inquests into the deaths in this case will be a procedural act which the state itself has decided should take place and, as the deaths were the result of acts by agents of the state, the circumstances meet the test for an article 2 inquiry that was identified in R (Middleton) v West Somerset Coroner [2004] 2 AC 182, para 3.

These pre-commencement deaths could not have given rise to any violation of the obligations of the state under article 2 in domestic law. But I do not think that we can ignore the possibility that they may have violated the deceased’s article 2 rights under the Convention. That certainly is how the matter would be viewed in Strasbourg. There is no doubt that these deaths fall within the jurisdiction of the Strasbourg court, as the events that have happened since the appellants lodged their application with that court have shown. The effect of Šilih is to breathe life into the procedural obligation post-commencement in a way that domestic law can recognise and give effect to.

It may be said that to extend the procedural obligation to these cases would be to give a more generous interpretation to the judgment in Šilih than it deserves. I think however that it would be unduly cautious for us not to do this. The whole idea of bringing rights home was to enable effect to be given to the Convention rights in domestic law. I do not think that we need any further guidance on this

matter from Strasbourg. As there is nothing in the wording of the 1998 Act to prevent us from directing that when he conducts these inquiries the Coroner must comply with the procedural obligation under article 2, I would hold that we should do so.

I would allow the appeals and direct the Coroner to hold an inquiry in each case that complies with the state’s procedural obligation under article 2 of the Convention.

LADY HALE

In Šilih v Slovenia (2009) 49 EHRR 996, the Grand Chamber of the European Court of Human Rights decided that the procedural obligation under article 2 of the Convention, to investigate deaths where the state might have been in breach of its obligation to protect the right to life, was separate or “detachable” from the substantive obligation to protect life. In certain ill-defined circumstances, therefore, that procedural obligation might still be in existence after the date on which a new Member State had acceded to the Convention, and the Court would have jurisdiction to investigate whether or not it had been broken, even though the death to which it related had taken place before that date.

This controversial decision undoubtedly broke new ground. But at first sight it has nothing to do with the United Kingdom. The United Kingdom accepted the right of individual petition to the Strasbourg Court as from 14 January 1966. The deaths in question in these appeals took place on 9 October 1990. So there is no doubt that the Strasbourg Court has jurisdiction to decide whether the United Kingdom is in breach of any of its obligations under article 2. The Šilih case makes no difference to that. The obligations of the United Kingdom in international law have not changed. The question, however, is whether the obligations of United Kingdom public authorities in United Kingdom domestic law have changed.

This would not matter if all inquests were automatically sufficient to comply with the procedural obligation in article 2. Unfortunately, there remain some differences between those which do, and those which do not, have to comply. Those differences centre on the scope of the available verdict as to “how” the deceased met his death: in a conventional inquest, “how” means only “by what means” whereas in an article 2 compliant inquest it must also encompass “in what broad circumstances”: see R (Middleton) v West Somerset Coroner [2004] UKHL 10, [2004] 2 AC 182. As Lord Bingham made clear in Jordan v Lord Chancellor [2007] UKHL 14, [2007] 2 AC 226, para 37, the scope of the inquiry is a different matter:

“. . . the purpose of an inquest is to investigate fully and explore publicly the facts pertaining to a death occurring in suspicious, unnatural or violent circumstances, or where the deceased was in the custody of the state, with the help of a jury in some of the most serious classes of case. The coroner must decide how widely the inquiry should range to elicit the facts pertinent to the circumstances of the death and responsibility for it.”

Furthermore, at para 39, there is nothing to prevent “a jury finding facts directly relevant to the cause of death which may point very strongly towards a conclusion that criminal liability exists or does not exist”. And, at para 40, if those “factual findings point towards the commission of a criminal offence, or it appears to the coroner that an offence may have been committed, the coroner’s duty . . . is to report promptly to the Director of Public Prosecutions . . .”

In this case the coroner gave a preliminary ruling on the scope of the inquest which clearly encompassed the purpose and planning of the surveillance operation which led to these deaths. The appellants were content with this but the Police Service of Northern Ireland argued that it should be made clear that this would be a conventional inquest, concerned in producing a verdict on “how” the deceased met their deaths, with “by what means rather than in what broad circumstances”. Hence, in addition to a declaration on delay, which is no longer an issue, the appellants sought a declaration that the coroner is obliged to conduct the inquest in an article 2 compliant manner. It is the refusal of that declaration which has led to these appeals.

The Human Rights Act 1998 came into force on 2 October 2000. It became unlawful for a public authority to act in a way which is incompatible with a convention right: section 6(1). A person claiming that a public authority has acted unlawfully may (a) bring proceedings against the authority, or (b) rely on the Convention right or rights concerned in any legal proceedings: section 7(1). However, with one exception, this does not apply to an act taking place before section 7(1) came into force: section 22(4). If the act in question is the death itself, then there is no remedy for deaths taking place before 2 October 2000 (In re McKerr [2004] UKHL 12, [2004] 1 WLR 807) nor does the interpretative obligation in section 3(1) apply (R (Hurst) v London Northern District Coroner [2007] UKHL 13; [2007] 2 AC 189).

The exception applies to “proceedings brought by or at the instigation of a public authority”; in such proceedings, a victim may rely on a Convention right whenever the act in question took place: section 22(4) read with section 7(1)(b). It was argued in Hurst that an inquest was “proceedings brought by or at the instigation of a public authority” and that the mother of the deceased was therefore

entitled to rely on her Convention right to an article 2 compliant inquiry whenever the death took place. That argument was roundly rejected (at paras 60 to 64) on the basis that the exception was meant to allow people against whom a prosecution or civil proceedings had been brought to rely upon his Convention rights to defend himself against the state: “Convention rights may be used as a shield to defeat proceedings brought against victims by public authorities, but not as a sword” (Lord Brown of Eaton-under-Heywood, at para 62).

In Hurst, the mother was seeking to use the Human Rights Act to compel the coroner to re-open an inquest when he had thought that it would serve no useful purpose. Unless she succeeded, there would be no proceedings. Similarly, in McKerr, the son of the deceased was seeking to use the Act to compel the holding of an inquiry which would not otherwise take place. In this case, by contrast, there is indeed to be an inquest. If the application of the exception in section 22(4) had first arisen in this context, I do wonder whether the result would have been the same. Section 7(1)(b) does not require that the proceedings are brought against the victim: it simply says that the victim may rely upon the Convention right “in any legal proceedings”. And it can plausibly be argued that, for the purpose of section 22(4), an inquest is a legal proceeding brought by a public authority, the coroner. Thus, even if the act in question is the death, were it not for Hurst, I would have been inclined to hold that where there is an inquest on foot it must be conducted in an article 2 compliant way, whenever the death took place. Unsurprisingly, in the light of Hurst, that is not how Ms Quinlivan puts the appellants’ case. I mention the point only because it leads to the same result as that at which I have arrived by another route.

This is where the decision in Šilih comes in. The obligation to hold an article 2 compliant inquiry into certain deaths has now to be understood to be separate from the obligation to protect life. The Attorney General complains that this is wrong: the object of the duty to hold an inquiry is to find out what happened, who was responsible, and to hold those responsible to account for the breach of their substantive obligations under article 2. If what they did cannot be an actionable breach of those obligations, because the Human Rights Act was not then in force, it makes no sense to oblige the state to hold an inquiry into whether or not there was a breach. That seems to me to take an unduly narrow view of the purpose of an article 2 investigation. Where deaths take place for which the state may bear some responsibility, a fortiori where, as here, deaths take place at the hands of state agents, there is always a useful purpose in finding out, so far as is possible, what took place, so that everyone, but in particular the relatives, may learn the truth about what took place, and lessons may be learned for the future. As the Court observed in Šilih, at para 156, “the procedural obligation has not been considered dependent on whether the State is ultimately found to be responsible for the death”.
The serious criticism of the Šilih decision is that it leaves so much uncertainty about when the investigative duty continues. As the Court says in paragraph 161 of its judgment, the obligation cannot be open-ended. But, with respect, I agree with the concurring opinion of Judge Lorenzen, when he complains that the criteria laid down in paragraph 163 of the Court’s judgment (quoted by Lord Phillips at para 44 above) are not easy to understand. However, I also agree with him that in one respect, they are quite clear. That is, in his words, “where the event occurred and an investigation was initiated before the entry into force of the Convention, but a significant part of that investigation was only carried out after that date” (p 1045, para O-I4). That is this case. The Coroner began his inquiries at the very latest once the Director of Public Prosecutions had announced on 2 April 1993 that there was to be no prosecution. But for a variety of reasons things have proceeded very slowly since then and a significant part of the investigation, in particular the inquest, has still to take place.

I do not see this as involving the retrospective operation of the 1998 Act. As public authorities, the coroner and the court have now to act compatibly with the Convention rights. The question is what the Convention rights now entail. It has always been clear that the content of the Convention rights can evolve over time. When the 1998 Act was passed, Parliament must be taken to have known of the jurisprudence which described the Convention as a “living instrument” in Tyrer v United Kingdom (1978) 2 EHRR 1; which implied further rights into those expressed in Golder v United Kingdom (1975) 1 EHRR 524; which developed autonomous concepts in Engel v The Netherlands (1976) 1 EHRR 647; which recognised positive obligations in Marckx v Belgium (1979) 2 EHRR 330; and which insisted that rights be made “practical and effective” rather than “theoretical and illusory” in Airey v Ireland (1979) 2 EHRR 305.

In the light of that well-known jurisprudence, it cannot have been Parliament’s intention that the Convention rights enshrined in the 1998 Act were to remain set in stone as they were when the Act was passed or when it came into force. It must have been intended, as Lord Bingham famously put it in R (Ullah) v Special Adjudicator [2004] UKHL 26, [2004] 2 AC 323, para 20, that the national courts would, at the very least, “keep pace with the Strasbourg jurisprudence as it evolves over time”. If the evolutive interpretation of the Convention rights means that they now mean something different from what they meant when the 1998 Act was passed, then it is our duty to give effect to their current meaning, rather than to the one they had before.

In Bellinger v Bellinger [2003] UKHL 21, [2003] 2 AC 467, to take an obvious example, the House of Lords made a declaration that section 11(c) of the Matrimonial Causes Act 1973 was incompatible with the Convention rights, because in Goodwin v United Kingdom (2002) 35 EHRR 447, decided in July 2002, the Strasbourg court had finally held that the non-recognition of a change of

sex was in breach of articles 8 and 12 of the Convention. This was a change, albeit well telegraphed in advance, from its previous jurisprudence in Rees v United Kingdom (1986) 9 EHRR 56, Cossey v United Kingdom (1990) 13 EHRR 622, and Sheffield and Horsham v United Kingdom (1998) 27 EHRR 163. No-one suggested that the Act did not apply.

At the same time, of course, we are not obliged to follow that jurisprudence if there are good reasons to depart from it. We have not so far failed to follow a decision of the Grand Chamber; as Lord Rodger equally famously put it in Secretary of State for the Home Department v AF (No 3) [2010] 2 AC 269, para 98, “Argentoratum locutum, iudicium finitum”. But the day might come when we would find good reasons to do so. Despite the urgings of the Attorney General, however, I do not think there are such good reasons here. This case fits into the limited class of case identified by Judge Lorenzen in Šilih. Accepting that this inquest must comply with the procedural requirements of article 2 does not require that old inquests be re-opened (unless there is important new material) or that inquiries be held into historic deaths. The one case which does not quite fit into Judge Lorenzen’s formula is where there is a death before the relevant date and the decision to hold an inquest or other inquiry is taken after that date. To my mind that would still fit into the criterion of “a significant proportion of the procedural steps required by this provision . . . will have been . . . carried out after the critical date”. In other words, if there is now to be an inquiry into a death for which the state may bear some responsibility under article 2, it should be conducted in an article 2 compliant way.

The coroner himself obviously wishes to conduct these inquests in way which complies with article 2. It will save a great many procedural arguments if he does. His concern before us was with the wider potential ramifications of Šilih. I hope that these concerns are allayed by our judgments, which seek to provide a principled basis for drawing the line.

For these reasons, I would allow this appeal and make the declaration sought by the appellants.

LORD BROWN

In 1995 the European Court of Human Rights decided that article 2 of the European Convention on Human Rights gives rise to a procedural obligation to hold an inquiry to specified standards into those deaths which occur in circumstances where the state’s responsibility is or may be engaged. Hitherto it has been understood that this obligation is secondary and ancillary to the death in

question. True, it was recognised by Strasbourg as a “free-standing” obligation. But that meant no more than that it was not dependent on the existence of a substantive violation of article 2.

It was on that understanding that the House of Lords decided in In re McKerr [2004] 1 WLR 807 that no such procedural obligation can arise in respect of a death occurring before 2 October 2000 when the Human Rights Act 1998 took effect.

The recent, post-McKerr, decision of the Grand Chamber in Šilih v Slovenia (2009) 49 EHRR 996 requires, however, submit the appellants, a fundamental change in that approach. Now it appears that the article 2 procedural obligation is not correctly to be understood as merely ancillary to a particular death but is rather to be seen as “a separate and autonomous duty”, “a detachable obligation arising out of article 2 . . . even when the death took place before the critical date” (para 159 of the court’s judgment). In short, the court held that in point of time (and it is time which is all important in the present domestic context just as it was in the international context in which the court in Šilih was determining its own temporal jurisdiction over subscribing states) the obligation may arise subsequent to the death requiring investigation and is not to be regarded as outwith the court’s jurisdiction merely because the death itself preceded the court’s assumption of jurisdiction.

I acknowledge, as Lord Rodger points out in his powerful dissenting judgment, that we are here concerned with what Parliament chose to enact in the 1998 Act, not with what it might have chosen to enact. Equally, of course, we are concerned to decide the domestic court’s jurisdiction under that Act, not the Strasbourg Court’s jurisdiction under the Convention. All that said, by the same token that had McCann v United Kingdom (1995) 21 EHRR 97 been decided after, rather than before, the 1998 Act, we should still have accepted the need under domestic law to give effect to the newly discovered ancillary article 2 procedural obligation, so too it now seems to me right to construe the Act as recognising a domestic law obligation to give effect to what we now learn is a detachable article 2 procedural duty.

As for the precise circumstances in which this detachable duty should henceforth be recognised to arise, this, I acknowledge, in the light of the deeply unsatisfactory purported delineation of the duty to be found at paras 161-163 of the Grand Chamber’s judgment in Šilih, is more difficult. Our essential task must be to construe the Act in the context of Parliament’s underlying intention that Convention obligations arising after 2 October 2000 should be enforceable here rather than have to be litigated in Strasbourg, but provided always that no particular difficulty results from events having occurred before 2 October 2000 –

hence Parliament’s decision not to apply the Act retroactively save to the limited extent provided for by section 22(4). These considerations are, I believe, essentially those described by Lord Phillips (at paras 58 and 59 of his judgment) respectively as “the mirror principle” and “the non-retroactive principle”.

On this approach I too, in common with Lord Phillips (para 61) and Lord Hope (para 77), would hold that any inquests still outstanding, even, as in these cases, in respect of deaths occurring before 2 October 2000, must so far as remains possible comply with the relatives’ article 2 Convention rights.

Such a conclusion will not to my mind present any great difficulties. There are, we were told, 16 existing “legacy inquests” (involving 26 deaths) currently outstanding on the coroner’s books, a further six incidents (involving eight pre- 2000 deaths) referred by the Attorney General to the Coroner for inquests pursuant to section 14 of the Coroners Act (Northern Ireland) 1959 and a further 7 deaths (between August 1994 and January 2000) not yet the subject of inquests. Moreover, not merely will there therefore be only comparatively few inquests affected by this ruling but it may be doubted whether in reality there is all that much difference between an article 2 compliant inquest (a Middleton inquest: see R (Middleton) v West Somerset Coroner [2004] 2 AC 182) and one supposedly not (a Jamieson inquest: see R v Coroner for North Humberside and Scunthorpe, Ex p Jamieson [1995] QB 1) – a topic exhaustively discussed by this court in R (Smith) v Oxfordshire Assistant Deputy Coroner [2011] 1 AC 1, see particularly Lord Phillips’ judgment at paras 73-78 and my own at paras 152-154.

I too, therefore, would allow these appeals.

LORD KERR

On 9 October 1990 Martin McCaughey and Desmond Grew were shot and killed by members of the British Army. Some twenty one years later an inquest into their deaths has not been held. These bare facts are testament to the difficulties that beset the investigation of controversial killings in Northern Ireland. Those difficulties are, of course, by no means unique to that province. Lord Phillips in his review of Strasbourg jurisprudence has amply illustrated the huge problems encountered in many member countries of the Council of Europe in the conduct of inquiries into how people come to be killed. The stream of cases that have flowed throughout domestic courts and in Strasbourg paint a disheartening picture. But perhaps the decision of this court in these appeals will mark the end of at least one area of controversy. Lord Phillips has said that a decision of this court is needed to prevent the delay and expense involved in interlocutory in-fighting in future

inquests raising the same issue as arises here. I share his hope that we can achieve that goal.

Before the hearing of this appeal, the appellants (who are the next of kin of Mr McCaughey and Mr Grew) lodged an application with the European Court of Human Rights (ECtHR) complaining of substantive and procedural breaches of article 2 of the European Convention on Human Rights and Fundamental Freedoms (ECHR). That court asked the government of the United Kingdom to provide its written observations on the admissibility and merits of the appellants’ application. The government applied to the court for suspension of the time limit within which those observations must be submitted while the decision of this court was awaited. The Strasbourg court has agreed to this request.

As Lord Rodger has pointed out, ECtHR unquestionably has jurisdiction to deal with a complaint that there has been a failure by the United Kingdom to fulfil its duty to investigate the violent deaths of Mr McCaughey and Mr Grew. If the appellants do not succeed in enforcing that duty in the courts of this jurisdiction, therefore, it is to be presumed that, if there is a failure on the part of the government to institute an article 2 compliant inquiry, the appellants will proceed with the application that they have already lodged. This unsatisfactory and anomalous situation can be avoided if the decision in Šilih v Slovenia (2009) 49 EHRR 996 that the procedural obligation to investigate deaths is detachable has the effect in domestic law for which the appellants contend.

Whatever may be said about the reasoning in Šilih (and, as Lord Dyson has said, it has been the subject of trenchant criticism, not least by the Attorney General for Northern Ireland, intervening in these appeals) the essential ratio of the case is clear. It is to the effect that article 2 involves two distinct and separable obligations, one procedural, the other substantive. Where a death occurs before the accession of a member state to the ECHR, the procedural obligation to investigate the circumstances of the death will, in certain circumstances, require the acceding state to comply with article 2.

The real difficulty that the decision of the Grand Chamber creates is in defining the circumstances in which the procedural obligation arises. Clearly not every death that occurred before the ratification of ECHR by a particular state can be subject to the duty to investigate under article 2. Must there be a temporal connection between the death and the procedural obligation to investigate? If so, how close to the time of accession must the death have occurred? Or is another form of connection sufficient to generate the link? Is it sufficient that the ratifying state decides, after it has acceded to the Convention, to investigate a death that occurred before ratification? In that event, can the ratifying state avoid inspiring the procedural obligation by deciding not to investigate the death? Many of these

questions remain unanswered by Šilih. It seems to me probable that they will ultimately require to be addressed by ECtHR and, for that reason alone, one cannot hope to provide a comprehensive statement on the precise, indispensable elements of the necessary connection but some aspects of this question require to be considered in order to determine whether, if Šilih is to be applied to the HRA, these appeals can succeed.

First, however, it is necessary to grapple with the issue whether, in light of Šilih, the article 2 Convention right under the HRA applies to the investigation, after 2 October 2000, of violent deaths which occurred before that date. Lord Rodger makes a powerful case for the proposition that the temporal application of the Convention is irrelevant for purposes of deciding the temporal application of the HRA but that, as it seems to me, does not provide an answer to the essential question. The majority of the Grand Chamber in Šilih elided the criticism of the dissenting minority that the effect of the majority judgment was to make the Convention retroactive. It did so by finding that the procedural right was detachable. The invocation of that right therefore could and did occur after Slovenia had ratified the Convention. In this way, no impact on the temporal jurisdiction of the Convention was involved. The duty arose – or at least continued to exist – after the Convention had been ratified. The right could be asserted when Slovenia was squarely within the temporal jurisdiction of the Convention and no question of retroactive effect was involved. This process of reasoning brings to centre stage the decisive finding that the separable right either comes to life or continues to exist after the critical date.

That central finding, that the procedural right is detachable and not linked inexorably to the death, cannot now be ignored in deciding whether, after the coming into force of HRA, there is a procedural obligation to investigate a death that occurred before that date. The decision in McKerr depended crucially on the indissolubility of the procedural and substantive rights. The passages from Lord Nicholls’ speech which have been quoted by Lord Phillips in para 27 of his judgment make that unmistakably clear. In particular, Lord Nicholls’ statement that the interpretation that he proposed had the effect of “treating all the obligations arising under article 2 as parts of a single whole” makes the point decisively. That stance is no longer possible. Strasbourg has positively said that the obligations arising under article 2 are not parts of a single whole. That inescapable conclusion must inform this court’s approach to the application of the HRA. The procedural right has the capacity to be animated or to continue to exist after the coming into force of HRA. If it does come alive or if it continues to have life, then there is no question of HRA being applied retrospectively. It is applied to an extant right.

Of course, it is undoubtedly true, as Lord Rodger has said, that Parliament had it in mind that Convention rights would only be accessible domestically for

acts that occurred after 2 October 2000 (apart from the closely defined exceptions in section 22(4)). But the phenomenon of an Act having an unintended consequence is by no means unusual. Moreover, the divisibility of the article 2 rights which has now been recognised eliminates any conflict that might otherwise have occurred with Parliament’s intention that only rights that accrue after the coming into force of HRA should be enforceable in this jurisdiction. The procedural right to an article 2 compliant investigation, if it accrues or continues to exist after the coming into force of HRA, does not clash with that legislative intention. It is entirely consonant with Parliament’s manifest intention that HRA should apply to acts or omissions that take place after the date of the coming into force of HRA. The essence of the decision in Šilih is that the procedural obligation is either animated or that it endures after the critical date. I therefore agree with all the judgments of the majority in this case that HRA requires that the procedural obligation under article 2 of the Convention, if it arises or endures after the coming into force of HRA, must be complied with notwithstanding that the death in respect of which the obligation arises occurred before October 2000.

How does one determine whether the procedural obligation exists after the critical date – in the case of Šilih the date of ratification of the Convention by Slovenia, in this case the coming into force of HRA? It is, I think, important to note that the Grand Chamber in Šilih reiterated the well established principle that death in suspicious circumstances normally triggers the procedural obligation under article 2 – see para 157. But the court also found that this obligation binds the State throughout the period in which the authorities can reasonably be expected to take measures with an aim to elucidating the circumstances of death and establishing responsibility for it – also para 157. So, although the procedural obligation is detachable, it is triggered by the death of the deceased. In this case the starting point for the existence of the obligation was plainly before the coming into force of HRA. One must focus, therefore, on the question whether, by reason of its detachability from the substantive right, it continued to have an existence following the coming into force of HRA such as to keep alive after that date the procedural duty to conduct an article 2 compliant inquiry.

The fact that the right has been triggered before the critical date is in no sense determinative of whether it continues to have life after that date. In para 159 the court said that “Although [the right] is triggered by the acts concerning the substantive aspects of article 2 it can give rise to a finding of a separate and independent ‘interference’ within the meaning of the Blečić judgment”. (The decision in Blečić v Croatia (2006) 43 EHRR 1038 is discussed by Lord Phillips in paras 40 and 41 of his judgment). Although the court found that separate interference can arise after the critical date, it also made clear that only procedural acts and omissions occurring after that date can fall within the Court’s temporal jurisdiction – para 162. It would appear, therefore, that procedural obligations spanning the period before and after the coming into force of HRA must be

segregated in order to determine which are amenable to article 2 requirements and which are not. This might well have implications in relation to a complaint of delay in holding an article 2 inquiry although, as Lord Phillips has pointed out, this is no longer an issue in the present appeals.

The critical passage in the majority’s judgment in Šilih is contained in para 163:

“… there must exist a genuine connection between the death and the entry into force of the Convention in respect of the respondent State for the procedural obligations imposed by article 2 to come into effect.

Thus a significant proportion of the procedural steps required by this provision – which include not only an effective investigation into the death of the person concerned but also the institution of appropriate proceedings for the purpose of determining the cause of the death and holding those responsible to account (Vo, cited above, § 89) – will have been or ought to have been carried out after the critical date.

A number of points should be noticed about the statements contained in this passage. First, the majority stipulates that there must be a “genuine” connection between the death and the critical date. It does not, however, specify what “genuine” connotes in this context. True it is that it is stated that a significant proportion of the procedural steps required by article 2 will have been or ought to have been carried out after the critical date. But it is not at all clear that the court is there postulating that this is an indispensable requirement for the connection to be established or merely observing that this is the consequence of the existence of the connection. In any event, the catch-all final section appears to contemplate that the connection could be established by circumstances quite unrelated to any temporal proximity between the death and the critical date. Indeed, this final part of the paragraph is drawn in such wide and general terms that it is difficult to forecast the range of cases that might fall within its embrace.
The best, I think, that one can make of this paragraph is that a connection must exist between the death and the critical date; that where much of the investigation into the death occurs or should occur after the critical date, this will be evidence of the existence of the necessary connection; and that there are other unspecified circumstances in which, although the death is not proximate temporally to the critical date, the need to protect the basic guarantees and principles of the Convention dictates that such a connection should be recognised.

That this provides a less than clear prescription for all the circumstances in which a sufficient connection is to be recognised has already been made clear by the concurring opinion of Judge Lorenzen in the Šilih case. At pp 1044-1045, paras O-I3 to O-I4 of the report he said:

“… I fail to see that the criteria established by the majority in paragraph 163 are in conformity with this requirement [of legal certainty]. Thus, it is not easy to understand what is meant by the requirement for ‘a genuine connection’ between the death and the entry into force of the Convention in respect of the respondent State for the procedural obligations imposed by article 2 to come into effect. Furthermore the fact that the majority seem ready to accept such a connection ‘based on the need to ensure that the guarantees and the underlying values of the Convention are protected in a real and effective manner’ appears to confirm that the jurisdictional limits will be difficult to identify, if they exist at all. I find it incompatible with the declared intention to respect the principle of legal certainty to define the Court’s temporal jurisdiction in such a vague and far-reaching way.

In my opinion, there must be a clear temporal connection between on the one hand the substantive event – death, ill-treatment, etc – and the procedural obligation to carry out an investigation and, on the other, the entry into force of the Convention in respect of the respondent State. This will be the case where the event occurred and an investigation was initiated before the entry into force of the Convention, but a significant part of that investigation was only carried out after that date. Likewise where the event occurred or was only discovered so close to the critical date that it was not possible to commence an investigation before that date. Where on the other hand no investigation was carried out despite knowledge of the event or where the investigation was terminated before the critical date, I would say that the Court would have jurisdiction only where an obligation to carry out investigative measures was triggered by relevant new evidence or information (see, mutatis mutandis,

Brecknell v United Kingdom, no. 2457/04, §§ 70-71, 27 November 2007).”

The statement at the beginning of the second paragraph of this passage that “there must be a clear temporal connection” between the substantive event and the procedural obligation on the one hand and the critical date on the other hand suggests, at first sight, that Judge Lorenzen considered that unless the substantive event preceded the critical date by a short period, the necessary connection would not be present. On this analysis, the appellants in this case could not succeed for the deaths preceded the coming into force of HRA by some ten years. But Judge Lorenzen appears in the succeeding sentence of the second paragraph to have made an important qualification to his statement about the need for a temporal connection. He suggests that the two elements will be sufficiently connected temporally where the event occurred and an investigation was initiated before the critical date, but a significant part of that investigation was carried out subsequent to that date. On this formulation the appellants’ cases would qualify because an investigation into the deaths of Mr McCaughey and Mr Grew was begun before the coming into force of HRA but a significant part – indeed, the most significant part, the inquest – will take place subsequently.

It is perhaps unwise to parse the judgments of the majority and of Judge Lorenzen too closely in order to produce a set of precise principles from views which have been expressed at a level of some generality. But it seems to me that the following may reasonably be deduced from both judgments:

There must be a connection between the substantive event (the death) and the critical date (in this case, the coming into force of HRA) – the majority and Judge Lorenzen;

A close temporal link (in other words where the death has preceded the critical date by a short period) will provide the necessary connection – the majority and Judge Lorenzen;

Where much of the investigation into the death occurs after the critical date, the connection is present – the majority;

Where a significant part of the investigation ought to take place after the critical date, this will be sufficient to make the connection – the majority;
Where an investigation begins before the critical date but a significant part of this takes place after the critical date, the connection is present – Judge Lorenzen. Although expressed slightly differently from the manner in which the majority put it in para 163, this formulation approximates to that set out in (iii) above;

In certain unspecified circumstances the connection might be based on the need to ensure that the guarantees and the underlying values of the Convention are protected in a real and effective manner

– the majority. Judge Lorenzen expressly disagreed with this statement;

If the investigation has been completed before the critical date or if, despite knowledge of the event, no investigation was conducted, there will be no sufficient connection. In that event the procedural duty will only arise after the critical date if triggered by relevant new evidence or information – Judge Lorenzen.

The present appeals come within (iii) and (v) of this list of the principles to be gleaned from the judgments of the majority and Judge Lorenzen. On that account I too would allow the appeals.

LORD DYSON

The decision of the majority of the ECtHR in the Grand Chamber decision of Šilih v Slovenia (2009) 49 EHRR 996 has, with justification, been the subject of trenchant criticism. But I agree with Lord Hope (paras 73 and 74 of his judgment) that the core of the decision is clear enough. It is that the procedural obligation to investigate a death under article 2 of the European Convention on Human Rights (“the Convention”) is not only distinct from the substantive aspect of the article but is autonomous and detachable from it. It is not profitable to condemn the decision on the grounds that the language used by the majority, in particular at paras 161 to 163 of their judgment, is obscure. Their decision is not a one-off aberration. It is an important decision of the Grand Chamber which is now well entrenched in the Strasbourg jurisprudence: see, for example, Varnava v Turkey (2008) 50 EHRR 467, paras 136 to 138 (another decision of the Grand Chamber). There is no good reason not to follow Šilih.

Both Lord Rodger and Lord Hope say that Šilih is a decision about the temporal jurisdiction of the Convention and that it is therefore irrelevant to the interpretation and application of the Human Rights Act 1998 (“HRA”) and, in

particular, to the temporal application of the HRA. To hold otherwise, they say, in effect involves adding to the HRA a transitional provision that is not to be found in section 22 of the act. It involves an impermissible application of the HRA. I shall call this the “section 22 argument”. Lord Hope would allow the appeals on the narrow basis that, if the state decides to hold an investigation post-HRA into a pre- HRA death, then it must do so in a way which is compatible with article 2 of the Convention.

The position under the Convention

In order to see whether the section 22 argument is right, it is necessary first to examine the essential reasoning of the majority decision in Šilih. The death in that case occurred before Slovenia ratified the Convention. A question arose as to whether the ECtHR had temporal jurisdiction to adjudicate on the question whether Slovenia had discharged its procedural obligation under article 2 to investigate the death. The fundamental principle that the ECtHR has no temporal jurisdiction to examine a complaint of breach of a Convention obligation by a state which is alleged to have occurred on a date before it ratified the Convention (referred to as “the critical date”) was not in issue. There is no doubt that the Convention does not have retroactive effect. Indeed, this important principle was asserted and acknowledged in Šilih as being the source of the problem that arose in that case: how was the principle to be applied where the death occurs before the critical date and the investigation is conducted wholly or in part after the date of ratification? Thus at para 146, the majority referred to Blečić v Croatia (2006) 43 EHRR 1038 (para 82) as confirming that “in order to establish the Court’s temporal jurisdiction it is therefore essential to identify, in each specific case, the exact time of the alleged interference”.

At paras 149 to 151, the majority referred to the varying approaches taken by different Chambers of the Court to the issue of temporal jurisdiction in the particular context of a death occurring before the critical date and an investigation into the death occurring wholly or partly after the critical date. These included the decision in Moldovan v Romania (Application Nos 41138/98 and 64320/01) (unreported) where the court held that it had no temporal jurisdiction to deal with the procedural obligation under article 2 as that obligation derived from killings which had taken place before Romania ratified the Convention. At para 152 of their judgment in Šilih, the majority said that having regard to the varying approaches taken by different Chambers of the Court to the problem:

“the Grand Chamber must now determine whether the procedural obligations arising under article 2 can be seen as being detachable from the substantive act and capable of coming into play in respect of deaths which occurred prior to the critical date or alternatively

whether they are so inextricably linked to the substantive obligation that an issue may only arise in respect of deaths which occur after that date.”

This was the question that the court set itself to answer. It is clear, therefore, that it was not the aim of the court to undermine or vary in any way the temporal jurisdiction principle. The question was how that principle was to be applied where the death occurs before the date of ratification and the investigation is conducted wholly or in part after that date.

The majority answered that question by reinterpreting the investigative obligation created by article 2. It was now declared to be a distinct free-standing obligation. As the majority put it at para 159 of the judgment, the procedural obligation to carry out an effective investigation under article 2 had “evolved into a separate and autonomous duty” which could give rise to a “separate and independent ‘interference’ within the meaning of the Blečić judgment” which can be considered to be a “detachable” obligation arising out of article 2 capable of binding the State even when the death took place before the critical date.

This new interpretation of the procedural obligation in article 2 had an important consequence for the temporal jurisdiction of the court. It meant that the court would now have temporal jurisdiction to deal with complaints that investigations into deaths before the critical date were not being conducted in accordance with article 2. Previously, this was not possible because the investigative obligation was seen as deriving from the death and an integral part of the substantive article 2 obligation. If the court’s temporal jurisdiction prevented it from dealing with a complaint of breach of the substantive obligation, it inevitably followed that for the same reason it was prevented from dealing with a complaint of breach of the procedural obligation. But the principle that the Convention does not have retroactive effect was left untouched by Šilih.

It is worth noting that in Šilih an objection of the dissenting judges was that the majority interpretation of article 2 would be “tantamount to giving retroactive effect to the Convention”: see para 9 of the Joint Opinion of Judges Bratza and Türmen. But the majority did not see it that way. Their approach was that they were reinterpreting article 2 by detaching the investigative obligation from the death. The consequence of this was that in certain circumstances there could now be an obligation to investigate deaths that occurred before the critical date. But this was achieved by a substantive interpretation of the article and not by departing from the fundamental principle that the Convention does not have retroactive effect.
I have already said that the core reasoning of the decision of the majority in Šilih is that the investigative obligation is distinct and detachable from the substantive obligation. But the court recognised that there had to be limits to this distinct obligation in the particular context of the obligation to investigate deaths occurring before the critical date. The court was alive to the need to exclude from the procedural obligation (i) investigations into deaths that occurred many years before the critical date and (ii) investigations into deaths that occurred (even shortly) before the critical date but which are not started until many years after the critical date. Thus the majority said in terms at para 161 that, having regard to the principle of legal certainty, the court’s temporal jurisdiction as regards compliance with the procedural obligation of article 2 in respect of deaths that occur before the critical date is not open-ended.

They defined the limits at paras 162 and 163 in language which, in parts, is extremely obscure. Lord Phillips has quoted the passages at para 44 above. The use of the word “genuine” in the first sentence of para 163 to describe the nature of the connection that must exist between the death and the critical date is puzzling, but it is not necessary to decide what this means for the purposes of determining the present appeal. There must, however, be a temporal connection between the death and the investigative obligation on the one hand and the critical date on the other hand. Thus if the date of death was not long before the critical date and a significant proportion of the procedural steps required by article 2 falls to be carried out after the critical date, then the investigation should be carried out in accordance with article 2 and the ECtHR will entertain a complaint of non- compliance. This is made clear at para 165 of the judgment of the majority in Šilih, where they noted that the death occurred little more than a year before the ratification of the Convention by Slovenia and that, with the exception of the preliminary investigation, all the relevant proceedings were initiated and conducted after that date. It was in the light of these factors that they concluded that the alleged interference with the procedural article 2 rights fell within the court’s temporal jurisdiction. It was the closeness of the temporal connection between the death and the critical date that led Judge Lorenzen to agree at p 1045, para O-I3 of the EHRR report, that the court had jurisdiction to examine the procedural complaint under article 2.

It does not, therefore, necessarily follow from the mere fact that an investigation is to take place after the critical date into a death that occurred before the critical date that the investigation must comply with the article 2 procedural obligation. There must be some temporal connection between the investigation and the critical date. That connection will exist where, for example, an investigation was initiated before the critical date, but a significant part was conducted after the critical date.

The position under the HRA

So much for the position under the Convention. But how does this affect the position in our domestic law under the HRA? I return to the section 22 argument. Unquestionably, the effect of section 22 of the HRA is that there is no remedy under the HRA for breach of the substantive aspect of article 2 of the Convention in respect of deaths occurring before 2 October 2000. In In re McKerr [2004] 1 WLR 807, the House of Lords decided that, in the context of the article 2 investigative duty, the obligation in section 6(1) of the HRA did not apply to a death which occurred before the Act came into force. It is sufficient to refer to the reasoning of Lord Nicholls which Lord Phillips has set out at para 28 above. In particular, at para 22 Lord Nicholls said that this interpretation of section 6:

“has the effect, for the transitional purpose now under consideration, of treating all the obligations arising under article 2 as parts of a single whole. Parliament cannot be taken to have intended that the Act should apply differently to the primary obligation (to protect life) and a consequential obligation (to investigate a death).”

But this reasoning cannot stand in the light of Šilih. It is no longer right to treat all the obligations arising under article 2 as parts of a single whole. It has now been explained that the procedural obligation is distinct and detachable from the substantive obligation.

I agree with Lord Phillips that this is where the “mirror principle” becomes relevant. Subject to the limits which I have mentioned, the ECtHR has now defined the procedural obligation in article 2 as being detachable from the substantive obligation. That definition must be reflected in our domestic law. This does not make the HRA retroactive in circumstances not permitted by section 22 any more than Šilih made the Convention retroactive contrary to the long-standing principle of temporal jurisdiction.

As Lord Hope said in Aston Cantlow and Wilmcote with Billesley Parochial Church Council v Wallbank [2004] 1 AC 546, para 44 “the purpose of [sections 6 and 7 of the HRA] is to provide a remedial structure in domestic law for rights guaranteed by the Convention”. One of those rights is to have an article 2 compliant investigation after the critical date into a death that occurred before the critical date. Translated into terms of the HRA, that must mean that there is a right to have such an investigation after 2 October 2000 into a death that occurred before that date if such a right would be recognised by the ECtHR.
In my view, it is nothing to the point to speculate what Parliament might have chosen to do if Šilih had been decided before the enactment of the HRA. The same question might be asked about any development of the Convention by the ECtHR post-HRA. The fact is that Parliament chose to incorporate the Convention and must be taken to have known that, in doing so, it would be likely to be re- interpreted by the ECtHR from time to time. The Convention is a “living instrument” which evolves over time as a result of interpretative decisions of the ECtHR. The procedural obligation implicit in article 2 is itself a good example of such evolution. It was recognised by the ECtHR in McCann v United Kingdom (1995) 21 EHRR 97 when, for the first time, the court held that the obligation to protect life under article 2 “requires by implication that there should be some form of effective official investigation when individuals have been killed as a result of the use of force by, inter alios, agents of the State” (para 161).

Suppose that the decision in McCann itself had postdated the coming into force of the HRA. It would not have been arguable on the basis of the section 22 argument that, since the re-interpretation of article 2 had occurred after the HRA came into force, section 6(1) did not require an article 2-compliant investigation.

The present case

The deaths occurred on 9 October 1990. The papers were passed to the Coroner in 1994, but they were incomplete as they omitted statements from the soldiers who committed the killings. These were not provided until 2002 and it was not until 14 September 2009 that the Coroner held the preliminary hearing in which he was asked to hold an inquest which complied with the procedural requirements of article 2.

The deaths were 10 years before the HRA came into force. That is a relevant factor to be taken into account when considering whether there is a sufficient connection between the deaths and the coming into force of the Act. But Šilih shows that it is not the only factor. In particular, of considerable importance is the fact that at that date the investigation had been initiated, but a significant proportion of the procedural steps required to be taken had not yet been taken. In that respect, the facts of the case are similar to the facts in Šilih. This is the feature of Šilih which is emphasised by the majority at para 165 and by Judge Lorenzen at p 1045, para O-I4 of the EHRR report.

I would hold that the inquests into the deaths should be conducted in accordance with the requirements of article 2. I also agree with Lord Phillips that this conclusion is reinforced by the mirror principle. It would be unsatisfactory for

the Coroner to conduct an inquest which did not satisfy article 2 leaving open the possibility of a claim against the United Kingdom in the ECtHR.

Conclusion

I would allow these appeals. I would add that I hope that before long the ECtHR will have an opportunity to clarify the meaning of para 163 of the judgment of the majority. If nothing else, the present appeal has served to highlight some of its obscurity and the difficulties of its application.

DISSENTING JUDGMENT

LORD RODGER

The appellants are the next-of-kin of Martin McCaughey and Dessie Grew, who were shot and killed by the Army on 9 October 1990 – long after the United Kingdom had ratified the European Convention and recognised the right of individual petition under it. So at all relevant times those who can be regarded as victims of the killings could have applied either to the Commission or, after 1999, to the Court in Strasbourg if they had considered that the United Kingdom had violated article 2 of the Convention and the matter had not been remedied by the domestic courts.

The Director of Public Prosecutions decided that no-one should be prosecuted for the deaths and he passed papers to the coroner in 1994. While preparatory steps have been taken, so far no inquest has been held.

In 1995 the European Court first identified a positive duty inherent in article 2 of the Convention that requires states to investigate relevant deaths: McCann v United Kingdom (1995) 21 EHRR 97. Assuming, therefore, for the sake of the argument, that there had been a failure by the United Kingdom to fulfil its duty to investigate the violent deaths of McCaughey and Grew, the European Court would unquestionably have had jurisdiction ratione temporis to deal with it and to grant any appropriate remedy..

By contrast, the widow of Gervaise McKerr, who was shot and killed by police officers in November 1982, did make an application to Strasbourg, alleging inter alia a violation of article 2. In McKerr v United Kingdom (2002) 14 EHRR

553 the European Court held that there had been a violation of article 2 in respect of failings in the investigative procedures relating to that death. The court also awarded McKerr’s son, who had taken over the case on his mother’s death,

£10,000 as just satisfaction in respect of the frustration, distress and anxiety which he must have suffered. The Government paid that sum and presented a package of proposals to the Committee of Ministers with responsibility for supervising execution of judgments.

The Human Rights Act 1998 (“HRA”) came into force on 2 October 2000, some ten years after the deaths of McCaughey and Grew. The issue in the present appeals is whether any failure by the authorities to investigate these deaths can give rise to a breach of the appellants’ article 2 Convention rights under the HRA. In other words, even if the United Kingdom would be in violation of its international law obligation to investigate the deaths under article 2 of the Convention, would the same facts also constitute a breach of a domestic law duty of the relevant public authorities under the HRA to investigate those deaths?

In the case of McCaughey the House of Lords has already answered that question in the negative in Jordan v Lord Chancellor; McCaughey v Chief Constable of the Police Service of Northern Ireland [2007] 2 AC 226. Lord Bingham simply observed, at p 240, para 4 and p 256, para 35, that, since the deaths occurred well before the HRA came into force, the decision of the House of Lords in In re McKerr [2004] 1 WLR 807 meant that the HRA did not apply to any investigation of them. So, although no plea has been taken by the Chief Constable, in substance the point raised in the present appeals is res judicata, so far as the first appellant and the Chief Constable are concerned.

In McKerr the issue in the appeal by McKerr’s son was the same as in the present appeal. The House of Lords held, unanimously, that the obligation under section 6(1) of the HRA and the article 2 Convention right in Schedule 1 to the Act, to carry out a proper investigation into a violent death, did not apply to a death which occurred before the HRA came into force on 2 October 2000. A significant element in the reasoning of the members of the Appellate Committee was their view that the investigative obligation on a state under article 2 of the Convention was designed to ensure that the state had procedures which would discover whether, in the case of a particular death, there had been a violation of the victim’s right to life under that article. It followed that there was no obligation under article 2 of the Convention to investigate a death which could not involve a substantive violation of article 2. Their Lordships concluded that, similarly, there was no obligation under section 6 of, and Schedule 1 to, the HRA to investigate a death which could not itself be unlawful under the same provisions.
The appellants do not suggest that in McKerr the House of Lords misunderstood the international law as it stood at the time. Rather, they point out that, for the first time, in Šilih v Slovenia (2009) 49 EHRR 996 the Grand Chamber of the European Court declared that the investigative obligation under article 2 of the Convention is not only independent of the substantive obligation (that had long been recognised) but also detachable from it. Therefore the investigative obligation was capable of binding a state, even when the death had taken place before the date of the state’s ratification of the Convention: 49 EHRR 996, 1030, para 159. Similarly, Ms Quinlivan argued, this court should now hold that, for the purposes of the HRA, the investigative obligation under section 6 and under the article 2 Convention right was capable of binding the relevant public authorities even when the death had taken place before the HRA came into force.

Having held that an obligation of investigation under article 2 could arise in respect of deaths which had occurred before the state ratified the Convention, the European Court added, at paras 161 – 163 of its judgment in Šilih, that this obligation was limited in some way:

“161. However, having regard to the principle of legal certainty, the court’s temporal jurisdiction as regards compliance with the procedural obligation of article 2 in respect of deaths that occur before the critical date is not open-ended.

First, it is clear that, where the death occurred before the critical date, only procedural acts and/or omissions occurring after that date can fall within the court’s temporal jurisdiction.

Second, there must exist a genuine connection between the death and the entry into force of the Convention in respect of the respondent state for the procedural obligations imposed by article 2 to come into effect.

Thus a significant proportion of the procedural steps required by this provision – which include not only an effective investigation into the death of the person concerned but also the institution of appropriate proceedings for the purpose of determining the cause of the death and holding those responsible to account – will have been or ought to have been carried out after the critical date.

However, the court would not exclude that in certain circumstances the connection could also be based on the need to ensure that the

guarantees and the underlying values of the Convention are protected in a real and effective manner” (internal citation omitted).

It would be a work of supererogation for me to criticise the court’s legal analysis in para 163 or to emphasise the blow to legal certainty which it has struck. All that need be said on these matters has been said – concisely, authoritatively and trenchantly – by the best-qualified of critics, Sir Nicolas Bratza, in his joint dissenting opinion with Judge Türmen, O-IV: 49 EHRR 996, 1050 – 1055.

Since the majority of the Grand Chamber adopted the analysis in para 163 of its judgment in Šilih despite this powerful dissent and despite the problems pointed out by other judges, they were clearly determined to bring at least some deaths occurring before a state’s ratification of the Convention within the court’s temporal jurisdiction. Predictably, the Strasbourg court has gone on to apply its new approach in other cases, including Varnava v Turkey (2008) 50 EHRR 467 where it held that it had jurisdiction over a disappearance in 1974, long before Turkey ratified the Convention. Doubtless, similar decisions will follow in future.

This court has nothing to do with the temporal jurisdiction of the European Court. Even on the most extensive interpretation of what it said in Šilih, the European Court might hesitate to assert jurisdiction over the investigation of deaths before January 1966 when the United Kingdom recognised the right of individual petition. But, even if it did assert jurisdiction, there would be nothing which this court could do about it.

In a novel twist to the thinking behind the HRA, however, Ms Quinlivan submits that this court should bring the many problems created by Šilih home from Strasbourg. In other words, this court should abandon its clear decision in McKerr and should hold that, in the light of Šilih, the article 2 Convention right under the HRA applies to the investigation, after 2 October 2000, of violent deaths which occurred before that date. Understandably perhaps, she declined to formulate a version of the Šilih “principles” which would define the scope of this new obligation under the HRA. It was enough, she said, that the obligation would apply to the inquests into the deaths of McCaughey and Grew since they are to be held at a time when the HRA is in force. According to counsel, Šilih means that, even if the public authorities are not under an obligation to hold an investigation, if they actually choose to do so after 2 October 2000, the investigation must comply with the relatives’ article 2 Convention rights. She drew a comparison with criminal proceedings: while the DPP may not be under an obligation to prosecute someone, if he chooses to do so, after 2 October 2000 the prosecution and trial must comply with the defendant’s article 6 Convention rights.
Whatever else may be said about this carefully crafted and deliberately low- key submission, it does not reflect the decision in Šilih. Article 2 is not concerned simply with the procedures to be adopted in any investigation which a state may choose to hold after ratification: on the contrary, as para 163 of the decision makes clear by the use of “ought”, according to the European Court, in an appropriate case the Convention requires the State to hold an article 2-compliant investigation into pre-ratification deaths. So, if this court were to apply the reasoning of the European Court by analogy – and there is no other pretended justification for overruling McKerr – it would have to hold that the HRA requires the relevant public authorities to hold an investigation which complies with the article 2 Convention right into violent deaths that occurred before 2 October 2000. As the court was told, quite a few violent deaths in Northern Ireland have not been investigated in this way.

Plainly, if Šilih had been decided before McKerr, some of the reasoning of the Appellate Committee would have been different. But it may be more instructive to suppose that Šilih had been decided before royal assent was given to the HRA in November 1998. In other words, suppose it had been known then that article 2 of the Convention could apply to the investigation of certain deaths that had occurred before a state ratified the Convention or recognised the right of individual petition. In that situation Parliament would have known that individuals could apply to the Strasbourg court and allege that an investigation should be held into certain deaths occurring within the jurisdiction of the United Kingdom before 14 January 1966. Perhaps – let us suppose – some such applications had already succeeded. What difference, if any, would this have made to Parliament’s decision as to the appropriate temporal application of the HRA? It is hard to see why it should have made any difference at all.

The decision as to the temporal application of the HRA cannot have depended on the technicalities of the analysis of the various rights. Rather, recognising that it was about to make a major change to our domestic legal systems, Parliament had to decide how that major change was to be carried through. So, for example, Parliament decided that the HRA should not come into operation for almost two years after royal assent – in order to allow time for appropriate preparations to be made. And then, except as provided by section 22(4), the Act was not to be retroactive. Another question was whether, and, if so, how, the HRA was to apply to situations – such as the situations in these appeals are said to be – which were ongoing when it came into operation on 2 October 2000. These were broad policy questions. I find it impossible to believe that, if the Government and Parliament had been aware of the decision in Šilih in 1998, this would have had any effect at all on their choice of policy.

After all, the temporal application of the Convention was irrelevant for purposes of deciding the temporal application of the HRA. Moreover, the

Government and Parliament knew very well, for example, that, between royal assent and 2 October 2000, there were liable to be many alleged violations of, say, articles 6 and 8 of the Convention. It would have been easy to bring them within the scope of the Act by a suitable transitional provision saying that the Act applied to acts occurring on or after 8 November 1998. But Parliament introduced no such transitional provision: subject to section 22(4), these acts would continue to be justiciable only in Strasbourg.

Making the HRA apply to the investigation of violent deaths occurring as far back as 1980 or 1990 would have raised particularly sensitive questions. Is it really to be supposed that Parliament would have decided to introduce a transitional provision that would have imposed a duty to investigate all such pre- commencement deaths merely because the European Court thought that it had jurisdiction in respect of violent deaths that had occurred before a state ratified the Convention? Surely not. The reality is that including a transitional provision to cover the investigation of deaths in the 1980s and 1990s would have had significant practical effects. These effects would have been felt, in particular, in Northern Ireland where – as was well known from the McKerr case in Strasbourg among others – there was pressure for such investigations to be held. The Good Friday Agreement was still in its uncertain infancy. Whether to bring past deaths within the scope of the Act was therefore a policy question, with potentially far- reaching ramifications, for the executive and the legislature. And, as the House of Lords held in McKerr, the plain text of the HRA shows that Parliament decided not to bring them within its scope. If, by contrast, Parliament had intended to include them, the relevant provisions would have had to be drafted differently in order to reflect the difference in the way that section 6(1) was intended to apply both in relation to different article 2 Convention rights and in relation to article 2 Convention rights as opposed to, say, article 8 Convention rights. See McKerr [2004] 1 WLR 807, 814, para 22, per Lord Nicholls, and p 831A—B, para 81 of my speech.

This court is concerned with what Parliament chose to enact, rather than with what it might have chosen to enact. Since the only transitional provision in the HRA is section 22(4), the inevitable inference is that, with this exception, all the provisions, including section 6(1) and the Convention rights in Schedule 1, were intended to apply only to events occurring on or after 2 October 2000. So there was to be no article 2 Convention right to an inquiry into a death that occurred before the Act commenced. If – as the House of Lords held in McKerr – that is indeed the correct interpretation of the temporal application of the HRA when it was passed, it is both incoherent and impossible to suggest that its temporal application can have been altered by the poorly reasoned and unstable decision of the Strasbourg court on the Convention in Šilih more than ten years later.
To hold otherwise involves adding to the HRA a transitional provision that never was. That is no small matter since drafters know very well that they must painstakingly consider how all the provisions of the legislation are to apply to circumstances as they will exist on commencement: G C Thornton, Legislative Drafting (4th edition, 1996), p 383. Moreover, if the compelling force for introducing this imaginary transitional provision is para 163 of Šilih, then it can safely be said that no Parliamentary counsel would ever have inserted a transitional provision that even remotely resembled the supposed “principles” in that paragraph. But, if the imaginary transitional provision does not reflect – and fully reflect – that paragraph, including its open-ended tailpiece, then the simple fact is that this court would be overruling McKerr by inventing its own transitional provision which is designed to insert into the Act a backwards time-limit that Parliament did not enact. In McKerr [2004] 1 WLR 807, 827, para 67, Lord Hoffmann commented that, if the HRA applies to pre-commencement deaths, it would in principle be necessary to investigate the deaths by state action of the Princes in the Tower. That was a vivid illustration of the point that – provided the claimant was a “victim” – there would be nothing in the Act to limit the deaths which would have to be investigated. The campaigns by relatives of soldiers who were court-martialled and shot during the First World War and by relatives of certain people who were executed in the 1950s suggest that the victim requirement could be satisfied long after the event. A time-limit for the deaths to be investigated would therefore have been essential if the HRA had been intended to apply to the investigation of pre-commencement deaths. The Act contains no time- limit and nothing to provide a proper basis for inferring the existence of one. This was one of the crucial reasons for the decision in McKerr and it remains as powerful today as it was seven years ago.

For all of these reasons I would dismiss the appeals. If, having deciphered Šilih, Parliament feels moved to amend the HRA so as to impose an obligation on public authorities to investigate deaths which occurred before the HRA came into force, it has every opportunity to do so. It has not done so over the last two years. Somehow, I would be surprised if it did so in future.

In the matter of an application by Brigid McCaughey and another for Judicial Review (Northern Ireland)

In the matter of an application by Brigid McCaughey and another for Judicial Review (Northern Ireland)

LORD HOPE

LADY HALE

LORD BROWN

LORD KERR

LORD DYSON

DISSENTING JUDGMENT

Related Posts:

More Posts

Section 47 EFCC Act 2004: Short Title

Section 46 EFCC Act 2004: Interpretation

Section 45 EFCC Act 2004: Savings

Section 44 EFCC Act 2004: Repeal of Act No. 5 2002

Leave a Reply Cancel reply