In the matter of an application by Margaret McQuillan for Judicial Review (Northern Ireland) (Nos 1, 2 and 3)

These appeals from the Court of Appeal in Northern Ireland are concerned with distressing events which occurred in the province in 1971 and 1972 at a particularly horrible and murderous period of “the Troubles”.

One appeal relates to the tragic death by shooting of Ms Jean Smyth on 8 June 1972. It arises in an application by the respondent, Margaret McQuillan, who is Ms Smyth’s sister, who seeks judicial review of among other things the failure of the Chief Constable of the Police Service of Northern Ireland (“PSNI”) and others to ensure the prompt, independent, and effective investigation of her sister’s murder. In that application Ms McQuillan has sought a declaration that the PSNI is insufficiently independent to conduct such an investigation.

The other appeal relates to the very serious ill-treatment by the Royal Ulster Constabulary (“RUC”) in August 1971 of people who were detained by the security forces for interrogation. Fourteen men, who have become known as the Hooded Men and who included Francis McGuigan and Séan McKenna, were subjected to this unacceptable treatment. Mr McGuigan and Mary McKenna, the daughter of the late Séan McKenna, seek judicial review of the decision made by the PSNI that there was no evidence to warrant an investigation, compliant with articles 2 and 3 of the European Convention on Human Rights (“the Convention”), into the allegation that the UK Government authorised and used torture in Northern Ireland in the case of the Hooded Men.

Two principal issues concerning human rights are raised on these appeals. The first is whether the domestic obligation on the UK Government to investigate a death or allegation of torture or inhuman or degrading treatment under articles 2 and 3 of the Convention, which arises as a result of the coming into force of the Human Rights Act 1998 (“HRA”) on 2 October 2000, requires there to be a “genuine connection”, including a close temporal connection, between the death or ill-treatment and the coming into force of the HRA. This question is in essence a matter of statutory interpretation, in which the Court must ascertain the meaning of the HRA. The second principal question is concerned with the tests by which the courts must assess the independence of investigations carried out or to be carried out by the PSNI into deaths in the Troubles and the ill-treatment of the Hooded Men.

This judgment is structured as follows:

The factual background to Ms McQuillan’s application for judicial review (paras 6-32).

The arrangements made to investigate deaths and ill-treatment arising during the Troubles (paras 33-51).

An overview of the Convention issues raised in these appeals (paras 52-54).

The judgments of the courts below in Ms McQuillan’s application for judicial review (paras 55-71).

The factual background to the application for judicial review by Mr McGuigan and Ms McKenna (paras 72-100).

The judgments of the courts below in the applications for judicial review by Mr McGuigan and Ms McKenna (paras 101-108).

The obligation to investigate under articles 2 and 3 of the Convention (paras 109-111).

The trigger for a further investigation: Brecknell v United Kingdom (2008) 46 EHRR 42 (paras 112-132).

The genuine connection and Convention values tests in the jurisprudence of the European Court of Human Rights (“the Strasbourg Court”) (paras 133-168).

Whether there is an obligation under article 2 or article 3 to investigate the death of Ms Smyth and the authorisation of the treatment of the Hooded Men (paras 169-192).

The Convention requirements for an independent investigation under articles 2 and 3 (paras 193-200).

Whether an investigation by the PSNI into Ms Smyth’s death would be an independent and effective investigation and whether any findings on that question can be read across into any investigation relating to the Hooded Men (paras 201-214).

Other domestic law challenges: (i) Whether there is an obligation of investigation at common law or under section 32 of the Police (Northern Ireland) Act 2000, (ii) whether the Chief Constable created a legitimate expectation of an investigation into the persons responsible for authorising the treatment of Mr McGuigan and Mr McKenna, (iii) whether the decision of the PSNI on 17 October 2014 not to investigate further was irrational and should be quashed (paras 215-252).

The question of delay (paras 253-255).

Summary and conclusions (paras 256-257).

In 1972, Jean Smyth was 24 years old. She was estranged from her husband and living with her daughter at her parents’ home in Tardree Park, West Belfast. On the evening of 8 June 1972, she went for a drink with a work colleague and friend, John Carlin. Mr Carlin collected her in his car, and they went to a pub on the Glen Road, the Glenowen Inn, where they met up with some other work colleagues.

At the end of the night, Mr Carlin offered one of those colleagues a lift home to the Lenadoon Estate, from where he drove back onto the Glen Road. Ms Smyth was sitting in the front passenger seat. At approximately 11.30pm, Mr Carlin thought he heard a noise which sounded like a tyre bursting and he got out of the car to look around. When he returned to the car, he found Ms Smyth lying in a prone position in the car, having been shot in the head.

Mr Carlin flagged down a passing black taxi, driven by James Brown. With the help of others who arrived at the scene, Ms Smyth was transferred to the taxi which left the scene, intending, it appears, to take her to the Royal Victoria Hospital (“RVH”). In the event, the taxi was driven to Andersonstown police station, which was on the route to the RVH. It is probable that by the time of the taxi’s arrival at the station, if not before, Ms Smyth had died.

Meanwhile, Mr Carlin drove his car to the home of Ms Smyth’s family. There, he spoke to her father, Mr Campbell and they left in Mr Carlin’s car to go to the RVH. They travelled directly as they were unaware that the taxi had stopped at Andersonstown police station. It appears that the taxi remained at the police station for some time.

When his taxi had been flagged down, Mr Brown’s wife had been in the taxi, and she was in shock. Once he knew that Ms Smyth was dead, Mr Brown left the police station to take his wife to the RVH for treatment, where he met Mr Carlin and Mr Campbell. Mr Brown explained to them that he believed Ms Smyth was probably already dead when she had been transferred to his taxi at the scene.

No person has ever been arrested or prosecuted in respect of Jean Smyth’s death.

Following Ms Smyth’s death, the RUC opened a criminal investigation. Witness statements were taken from Mr Carlin, Mr Brown, Mr Campbell, Mr Desmond Smyth (Ms Smyth’s husband who identified her body) and a number of police witnesses.

The police statements included a number of statements from Scenes of Crime officers (“SOCO”), who had examined Mr Carlin’s car in the grounds of the RVH and Mr Brown’s taxi, which contained Ms Smyth’s body, at Andersonstown police station. Mr Carlin’s car was again examined at Springfield Road police station. The SOCO statements recorded the opinion that Ms Smyth was killed by a bullet which entered the car through the rear off-side window and travelled at an angle towards the front passenger seat, striking Ms Smyth on the head. The SOCO statements further recorded that no other holes or bullet strike marks were found on the car. A number of photos of Mr Carlin’s car were before the lower courts. Maguire J stated that they tended to support the description of the damage in the SOCO statements. This is contested by Ms McQuillan and other family members who believe that “the car was bullet ridden”.

On 9 June 1972, a post-mortem of Ms Smyth’s body was carried out by Dr Carson, Assistant State Pathologist. The lower court judgments record that it did not appear that the report from the post-mortem was available.

The RUC investigation at the time considered paramilitary use of firearms but did not consider military involvement.

An inquest into the death of Ms Smyth was held on 9 November 1972 and an open verdict was recorded.

On 22 October 1973 an article appeared in the Belfast Telegraph under the headline “Was Jean Smyth shot by Mistake?”. The article referred to a suggestion that a youth at the scene shouted that it was the UVF who had been shooting. The journalist considered this to be unlikely. A further suggestion was that an undercover unit of the British Army, the Military Reaction Force (“MRF”) had been involved. The journalist considered this theory “to hold more water”, taking into account an incident which occurred in the same general location around two weeks after Ms Smyth’s death. Members of the MRF had shot and seriously injured four men on the Glen Road. A sergeant who was a member of the MRF was later prosecuted and acquitted. In the course of his trial, the sergeant had, according to the article, revealed that the MRF had been in operation in the area around the time of Ms Smyth’s death. The journalist further suggested that it was likely that Ms Smyth had been shot by mistake, stating (without providing any source) that “it would appear that a unit of the Provisional IRA fired on the car thinking it was army personnel”.

In 1975, it appeared that the RUC received what is described as “an intelligence report” which touched upon Ms Smyth’s death. The source of the information was unclear, but what was put forward in the report was that two named members of the Provisional IRA were responsible for Ms Smyth’s death. Those individuals had supposedly been conducting vigilante patrols in area when shots were fired from a car on the Glen Road. One of the men fired at a car on the Glen Road, killing the woman occupant. The male occupant of the car allegedly ran off. The two men, it was suggested, had stopped a taxi, ordering the driver to take the woman to hospital. Maguire J stated that there was no sign that this report led to any further action on the part of the RUC.

PSNI’s Historical Enquiries Team (“HET”) was founded in 2005 with the function of reviewing deaths arising from the conflict in Northern Ireland which had occurred in the period 1968-1998. We discuss in more detail in section 2 below the arrangements which have been made over time to investigate such deaths.

The HET carried out a review of the death of Ms Smyth during the period 2006-2008, culminating in the presentation to the family of a Review Summary Report (“RSR”) on 21 July 2008. It appears that the RSR was principally based on materials derived from the “inquest file” together with open-source material and the intelligence report.

The RSR described the original RUC investigation as “relatively basic” but noted that it “must be judged in the context of the times”. In particular, “police enquiries, which would be commonplace by today’s standards, were not always possible, and resources devoted to incidents were substantially less than would be the case today.” As regards future lines of enquiry, the RSR recorded that it had not been possible to identify the type of weapon used and that no new forensic opportunities had been identified. It also recorded that there was no intelligence prior to the incident which would have prevented the death.

The conclusion reached in the RSR was that only one shot appeared to have been fired and it described Ms Smyth’s death as a “random killing”. The review did not examine the possibility of military involvement in her death.

The RSR records that the HET had initially decided against re-interviewing Mr Carlin, but that following dialogue with family members, it decided to do so. An interview with Mr Carlin was carried out on 10 September 2008, by which time he was in poor health. Mr Carlin’s description of the incident, provided orally, was said to be virtually identical to his original statement.

On 21 November 2013, an edition of the BBC’s current affairs programme, Panorama, was broadcast, featuring the activities of the MRF in Northern Ireland in the early 1970s. According to the programme, the MRF operated over a period of some 14 months mostly in Belfast; it would have been operating at the time of Ms Smyth’s death and in the area where it occurred. Panorama claimed that some members of the unit operated outside the law and fired at and killed unarmed civilians. To a degree, interviews broadcast with former MRF members lent support to this view and the programme therefore attracted significant public attention.

The case of Ms Smyth’s death is not referred to in the broadcast. Maguire J suspected that at the date of its showing, it is likely that only a person with a very detailed knowledge of the circumstances of Ms Smyth’s death would have made a connection with the theme pursued by Panorama.

In June 2014, a researcher, in the course of conducting legacy archive research in the National Archives in Kew, found military communication records relating to the evening of 8 June 1972 and early hours of 9 June 1972. The logs took the form of operational log sheets, recording radio traffic. The logs, consisting of eight documents, were found in a file entitled “Commander’s Diary, Headquarters, Northern Ireland, 1972”.

The logs were given to Ms McQuillan’s solicitor who provided copies to the PSNI in October 2014. During the course of the judicial review proceedings, new materials including entries from further military logs were disclosed.

In his judgment, Maguire J set out the significant entries. It is not necessary to repeat these here. Maguire J identified four possible interpretations of the military logs which were reproduced in the judgment of the Court of Appeal. Neither court sought to determine which interpretation was to be preferred. The possible interpretations identified by Maguire J may be summarised as follows:

The logs provide a measure of support for the previously expressed views of family members that there was more than one shot fired at the car. For example, there is reference to the car being “machine-gunned” and there are at least two references to a “burst” of automatic fire in the context of a woman being hit.

The logs disclose an incident involving “KP19”, understood to be a code for a military sanger located nearby further along the Glen Road. This might lend support for the possibility that the sanger was fired upon and that a soldier in the sanger fired a 7.62mm round, claiming a hit.

The reference to the MRF firing on two gunmen behind a hedge in the area, may support the possibility of MRF involvement. This explanation would not be consistent with the circumstances of the death as previously understood but it is of interest that the MRF is identified as operating in the area and at the general location around the time of Ms Smyth’s death.

The reference to three men placing Ms Smyth in a taxi and the driver being told to drive to the hospital might be capable of being viewed as not discordant with the intelligence received in 1975, which had referred to men performing vigilante patrolling.

The PSNI considered the materials provided to it by Ms McQuillan. It appointed an investigating officer to review the military logs but it was not considered that they supported the view that the MRF was involved. However, the view was taken that the materials did support the possibility of military involvement in the death as a result of the reference to an incident at the KP19 sanger.

A decision was made by the PSNI in December 2015 that the death should be further reviewed within the Legacy Investigation Branch (“LIB”) which by this time had taken over the functions of the HET.

The position of the PSNI during the course of the proceedings in the lower courts was that if, in the course of its review, the LIB identified any indication that wrongdoing or a criminal offence might have been committed by a police officer, whether from the RUC or PSNI, the case would have been immediately referred to the Police Ombudsman for Northern Ireland (“PONI”) in order to carry out any investigation which it considered to be appropriate. Otherwise, the PSNI would carry out any investigation required.

The LIB did not begin the review, because, following the decision of the Court of Appeal, the Chief Constable decided to appoint independent officers to carry out the review and any necessary investigation into the death of Ms Smyth. In doing so, the Chief Constable stated that he was “deeply sorry that there were previous investigative failures into the circumstances surrounding [Ms Smyth’s] death”. The Chief Constable acted on a voluntary basis and without prejudice to the existence of a legal obligation to do so. He has appointed Jon Boutcher (the former Chief Constable of Bedfordshire Police) and his team of investigators who are from outside the PSNI. This work is currently underway under the name Operation Mizzenmast.

In 2001, the RUC was succeeded by the PSNI as the single civilian police force in Northern Ireland.

The units of the PSNI which have been involved in investigations or reviews of Troubles-related deaths in Northern Ireland have been, sequentially:

the Serious Crimes Review Team (“SCRT”) from 2003 to circa 2005/2006;

the HET from circa 2005/2006 to 2014; and

the LIB from 2015 to date.

The Stormont House Agreement of 23 December 2014 between the British and Irish Governments and the political parties in Northern Ireland envisages the establishment of an independent Historical Investigations Unit (“HIU”) which would take over responsibility for the investigation and review work of both the PSNI and PONI. The HIU has not yet been established.

In the period 2000-2003, the Strasbourg Court found breaches of the investigative obligation arising under article 2 of the Convention in a number of cases involving the United Kingdom arising out of deaths in Northern Ireland in which members of the British security forces were implicated: McKerr v United Kingdom (2002) 34 EHRR 20; Jordan v United Kingdom (2003) 37 EHRR 2; Kelly v United Kingdom (2000) 30 EHRR CD223; Shanaghan v United Kingdom (2000) 30 EHRR CD 370; McShane v United Kingdom (2002) 35 EHRR 23; and Finucane v United Kingdom (2003) 37 EHRR 29. The findings of the Strasbourg Court included both that there was a lack of independence by the RUC in some of the investigations and also investigative failings. The cases have become known as the “McKerr group”.

Pursuant to article 46(2) of the Convention, the final judgment of the Strasbourg Court in each case was transmitted to the Committee of Ministers to supervise its execution, which supervision included the issue of the independence of investigators. The Committee of Ministers conducted its scrutiny of these cases as a group.

The UK Government thereafter set in train a package of measures to remedy the identified breaches of the article 2 procedural obligation, which included, as a part of the obligation of an effective investigation, the requirement to secure the independence of the investigators. The package of measures was overseen by the Committee of Ministers.

The SCRT was established in 2003 as one of the measures designed to address the defects in previous police investigations. The UK Government reported to the Committee of Ministers that it had been discussing with the PSNI how this work might be expanded to process greater numbers of unresolved deaths and to do so in a way that commanded the confidence of the wider community. This led to the establishment of the HET.

The HET was established in September 2005, as a unit of the PSNI, to investigate 3,269 unsolved murders committed during the Troubles, specifically between 1968 and 1998. It commenced its work in 2006. Its establishment was one of the general measures proposed by the UK Government to the Committee of Ministers.

The HET was principally made up of civilian staff (many of whom were retired officers from police forces other than the PSNI) under the control of a Director who was accountable to the Chief Constable. It adopted three main objectives:

Arrangements were put in place at the time of the establishment of the HET, or evolved with experience, to secure the practical independence of the HET investigators from those implicated in the events. In some circumstances those arrangements had to take into account that the PSNI “necessarily, … inherited officers and resources” from the RUC, so that the arrangements were designed in those circumstances to secure practical independence not only from the RUC but also from the PSNI. In other circumstances, the arrangements had to take into account that the investigation in question was into the security forces. In those circumstances, the arrangements were designed to secure practical independence from the security forces. The Court of Appeal set out those arrangements in paras 35-67 of their judgment. We summarise them briefly: (i) the HET had a large degree of operational independence as it reported directly to the Chief Constable; (ii) until September 2013 a PSNI legacy investigation was commanded by two senior staff from outside the PSNI; (iii) the existence between 2006 and 2014 of a team, known as “the White team”, which was based in London and staffed by police officers from outside Northern Ireland to examine cases in which there were allegations of collusion, which worked alongside the HET’s Complex Enquiry Team, which also was staffed by investigators who had no prior link with the RUC or the PSNI; and (iv) declarations and undertakings by former RUC and PSNI officers at the stage when a case was being allocated to reveal any prior involvement and proposals for vetting such declarations.

The Committee of Ministers examined the progress of the HET from its inception until 19 March 2009. On that date, having been satisfied by the information provided to it, the Committee of Ministers closed its examination into the issue of the independence of investigations into historical cases “as the HET has the structure and capacities to finalise its work”.

The HET was disbanded in 2014, and its workload was transferred to the LIB.

The LIB is the present unit within the PSNI responsible for legacy cases, which includes reviewing the circumstances of all Troubles-related deaths in Northern Ireland and conducting investigations into such deaths, if there is sufficient evidence to do so.

On 18 September 2014, the Chief Constable met with the Performance Committee of the Policing Board and outlined that, due to severe budgetary pressures, the PSNI was considering drawing together its legacy operations under a single command, provisionally referred to as the “Legacy Branch”. This Branch was to be integrated into the Crime Operations Department, accountable to the Chief Constable through the Assistant Chief Constable.

On 5 January 2015, the LIB was established as a unit within the PSNI operating under the direction of the Assistant Chief Constable, Crime Operations. The Chief Constable stated that the PSNI supported the establishment of the HIU, but recognised that until that unit was established, he had the statutory responsibility for the legacy caseload. The LIB was therefore to be a temporary measure pending the establishment of the HIU. Terms of reference for the LIB were adopted by the PSNI on 20 May 2015. In February 2016, the LIB was transferred from the PSNI’s Crime Operations Department to the newly established Legacy and Justice Department within PSNI.

The number of staff in HET varied between 80 and 180 during the period between 2006 and 2009. At its establishment, it was anticipated that the LIB would have a total of 70 staff. The LIB has the role of reviewing and if appropriate investigating deaths related to the security forces arising out of the Troubles between 1969 and 2004. However, its case load has included not only some 900 cases not completed by the HET, but also the Bloody Sunday enquiry, the Military Reaction Force enquiry, the Boston College tapes enquiry, the “on the runs” enquiry, and approximately 70 other murder enquiries, previously within the remit of PSNI C2 (Crime Operations Department), for which the LIB became responsible.

Detective Superintendent Jason Murphy deposed on 11 April 2016 that the number of staff involved in reviews and investigations in LIB was then 72. Detective Superintendent Harrison deposed on 5 December 2017 that the number had fallen to 55. Of the 55 LIB members of staff on 5 December 2017, 27 had former service in the RUC. Two of those individuals had previously worked within RUC Special Branch or its successor within the PSNI, one of whom was a police officer and the other worked in a secretarial position. Within these 27 members of LIB staff with RUC service, six individuals worked as LIB support staff and not as primary investigators or review officers.

As we have said (para 35 above), the Stormont House Agreement of 23 December 2014 provided for the establishment of the HIU. The HIU would be a new independent body to take forward investigations into outstanding Troubles-related deaths. It would take on outstanding cases from the HET process and the legacy work of PONI.

The agreement did not specify a timeframe within which the new HIU was to be established. At the time, the Chief Constable indicated that he expected it to be two years before the new unit was to be ready to start work. The HIU has not yet been established.

In order to set in context our summaries of the judgments of the Northern Ireland courts, it may be helpful to give a very brief overview of the Convention issues which they addressed. As we explain more fully below, there is implied into article 2 of the Convention an obligation on the state promptly to conduct some form of official investigation when an individual is killed by the use of force. A similar investigative obligation may arise under article 3 of the Convention where there is a reasonable suspicion that a person has been subjected to torture or inhuman or degrading treatment. The jurisprudence of the Strasbourg Court has established a requirement that such an investigation should be effective. One aspect of this requirement is that the investigation must be independent.

There is also jurisprudence of the Strasbourg Court, which lays down pre-conditions of the investigative obligation, where a death or ill-treatment had occurred some years before a state accepted the right of individual petition to the Strasbourg Court. That jurisprudence provides that there must be a “genuine connection” between the death or ill-treatment and the date when the right of individual petition was accepted. Alternatively, there must be an extraordinary situation in which the need to ensure effective protection of the guarantees and underlying values of the Convention constitutes a sufficient basis for such a connection. This is recognised by all parties to these appeals. There is also jurisprudence from the courts of the United Kingdom, which is challenged in these appeals, that, where a challenge is made under the HRA, similar requirements exist in relation to events which occurred some years before the coming into force of that Act on 2 October 2000.

A further issue which arises on these appeals relates to the circumstances in which the obligation to investigate under article 2 or 3 of the Convention may revive years after an event on the discovery of new and relevant evidence of sufficient cogency.

Ms McQuillan applied for judicial review in an application dated 16 June 2015 in which she sought among other things an order requiring an effective investigation of Ms Smyth’s death in accordance with article 2 of the Convention. Maguire J issued two judgments dated 3 March 2017 and 13 April 2018. In his first judgment Maguire J accepted the gravamen of Mr Southey QC’s arguments on behalf of Ms McQuillan which were, in summary, that the previous investigations by the RUC and the HET appeared to lack impartiality, as there had been no enquiry into the possibility of military involvement in Ms Smyth’s death. Mr Southey also drew attention to the criticisms of the HET by HM Inspectorate of Constabulary (“HMIC”) in a report in 2013 both for failing to check the self-declarations of its officers as to any involvement or interest in a case which they would be investigating, and for its dependence on intelligence provided by the PSNI Intelligence Branch, some of whose staff were former Special Branch Officers. He also criticised the movement of investigations from the HET to the LIB for budgetary reasons, which created further difficulties in relation to the public perception of the independence of such investigations.

Maguire J concluded that the LIB at that time lacked the necessary element of independence to enable it to make further investigations into Ms Smyth’s death (para 105). He emphasised the paramount importance of the need to have regard to public perception (para 106) and saw the issue as being how the matter is reasonably perceived (para 107). The failure of the HET to look at the possibility that soldiers were involved in Ms Smyth’s death might have a benign explanation, but it might also reflect a reluctance to investigate a possibility that would traduce the reputation of the security forces (para 115). He concluded that, in the light of the inadequacy of the earlier investigations, the transfer of investigations from the HET to the LIB, and the criticisms of the HET by HMIC, a fair-minded and informed observer would not perceive the PSNI as passing the test for independence for this purpose (para 123). He made a declaration that the proposed investigation by the LIB into Ms Smyth’s death was in conflict with the requirements of article 2 of the Convention, as the LIB lacked the requisite independence to perform an article 2 compliant investigation in respect of this death.

The first judgment proceeded on the basis of a concession by the PSNI that, although Ms Smyth’s death had occurred 28 years before the HRA came into force, the military logs constituted new evidence coming to light within the principles set out by the Strasbourg Court in Brecknell v United Kingdom (2008) 46 EHRR 42 (“Brecknell”), which we discuss in paras 116-118 below, and that this required the PSNI to take further investigative measures concerning her death. It was and is common ground that the military logs met the Brecknell evidential threshold. The PSNI also conceded that the investigative measures were subject to the article 2 investigative obligation and that those responsible for carrying out the investigation were required to be independent of those implicated in the events.

On appeal, the PSNI accepted that the military logs would otherwise satisfy the evidential threshold set out in Brecknell and undertook to conduct a review of the death which would comply with article 2 standards. It has renewed that undertaking before this court. But it sought successfully in the Court of Appeal to withdraw the concessions to which we have referred in the previous paragraph. The Court of Appeal therefore remitted the case to Maguire J for reconsideration.

Ms McQuillan was permitted to expand her Order 53 statement to add further common law grounds of challenge to the claim under article 2 of the Convention.

In his second or supplementary judgment, Maguire J dismissed Ms McQuillan’s application, holding:

that an investigative obligation under article 2 of the Convention did not arise as a matter of domestic law under the HRA because there was not a genuine connection between Ms Smyth’s death and the coming into force of the HRA and the “Convention values” test, which we discuss in paras 145-146 below, had not been met; there had therefore been no breach of article 2;

that the PSNI was not bound by any form of legitimate expectation in domestic law; and

that there was no parallel to article 2 existing at common law requiring an independent investigation so there was no breach of the common law.

Ms McQuillan appealed Maguire J’s second judgment. The Court of Appeal (Morgan LCJ, Stephens LJ and Sir Paul Girvan) [2019] NICA 13, [2020] NI 583 allowed her appeal holding:

that under the HRA as a matter of domestic law an article 2 investigative obligation did arise in relation to Ms Smyth’s death;

that the decision of the Strasbourg Court in Brecknell was authoritative in relation to the institutional independence of the PSNI;

that the critical date for the purposes of the “genuine connection” test in the context of section 6(1) of the HRA is 2 October 2000, on which date that Act came into force;

that the PSNI had not demonstrated practical independence on the part of the LIB so that it had the capacity to carry out an investigation into Ms Smyth’s death; and

that the challenges at common law failed: Maguire J had been correct to conclude that the PSNI was not bound by any form of legitimate expectation, that there was no parallel obligation to article 2 at common law, and that the Chief Constable had not acted irrationally in the exercise of discretion concerning the future conduct of any investigation into Ms Smyth’s death.

On the question of the passage of time, the Court of Appeal, recognising that Ms Smyth’s death had occurred 28 years before the HRA entered into force on 2 October 2000, stated that there had to be a genuine connection between the death and the entry into force of the HRA (para 131). It held that the lapse of time between the relevant death and the entry into force of the HRA must remain reasonably short if it is to comply with the genuine connection standard but that ten years or less was not an immutable requirement. It held that little if any weight should be attached to the passage of time in any case where, as in Brecknell, plausible and credible evidence comes to light after 2 October 2000: absent exceptional circumstances, the death would meet the genuine connection test (paras 132, 135 and 137-138). It also held that the second condition inherent in the “genuine connection” test, regarding the carrying out of investigative activity after the “critical date”, was satisfied. In this case there had been no investigation of military involvement in Ms Smyth’s death and the lack of that investigation might be the result of incompetence or something which was not benign, such as positive obstruction by the RUC (paras 135-136). On that basis it held that the “genuine connection” test was satisfied.

On the question of the independence of the PSNI, the Court of Appeal referred to the obligation under section 2 of the HRA to take account of the clear and constant line of decisions of the Strasbourg Court and the Committee of Ministers. Those decisions established the need for an effective investigation where individuals had been killed by the use of force. The central purpose of the investigation was to secure the effectiveness of domestic laws which protect the right to life and to ensure that state agents or bodies are accountable for deaths occurring under their responsibility. To be effective, the investigation had to be independent from those implicated in the events. Such independence had two elements: (i) there must be a lack of hierarchical or institutional connection between the investigators and those implicated in the events and (ii) the investigation must have practical independence. The investigation had to be structured in such a way that the state authorities were able to demonstrate that it was an independent investigation.

The Court of Appeal, in addressing the need for a lack of hierarchical or institutional connection, recognised that there was no requirement of a complete hierarchical or institutional disconnection where the state could take steps to introduce independence into the relevant structures. It analysed the police reforms which had been effected since 1998, following the Belfast Agreement. It narrated the principal recommendations of the Patten Report, the acceptance of those recommendations by the UK Government and the enactment of the Police (Northern Ireland) Act 2000, which resulted in the creation of the PSNI. Section 31A of that Act established the core principle that police officers were to exercise their functions with the aims of securing the support of the local community and of acting in cooperation with the local community. The Court of Appeal referred to the personnel changes to the PSNI through early retirement packages, the policy of 50/50 recruitment from the Protestant and Catholic communities, the dissolution of the former Special Branch, the establishment of a new Northern Ireland Policing Board in 2001 and the creation of the Northern Ireland Department of Justice in 2010.

Those reforms were the factual background to the decision of the Strasbourg Court in Brecknell that there was a lack of institutional connection between the PSNI and the RUC, a decision which the Court of Appeal treated as authoritative (paras 146 and 192). Similarly, the Court of Appeal (paras 158-167) recounted the decisions of the Committee of Ministers taken under article 46 of the Convention between 2007 and 2009 which resulted in the decision of the Committee of Ministers on 19 March 2009 to close its examination of the independence of the PSNI investigators because the HET had the structure and capacities to allow it to finalise its work.

Where the Court of Appeal found the PSNI’s investigation of the death of Ms Smyth to be wanting was in relation to its practical independence. The structures and capacities of the PSNI to conduct independent investigations had changed since the decision of the Committee of Ministers on 19 March 2009; in particular, the HET had ceased to exist and the HIU had not been established. The Court of Appeal accepted an approach that the court should await the outcome of an investigation before deciding whether it was independent, “provided it has the capacity to fulfil the procedural requirement of independence”, taking account of both presently envisaged arrangements and any arrangements which might be put in place as the investigation proceeds (para 173). There was therefore a strong presumption against a judicial review application before the conclusion of an investigation unless there was some exceptional circumstance, such as “a compelling case that the arrangements that are presently envisaged and any sensible alteration to them as the investigation proceeds, will obviously lead to a requirement for a fresh investigation” (para 174). But this did not mean that the Chief Constable should not give the victim or the victim’s family details of the practical independence of an envisaged investigation: the Belfast Agreement envisaged such accountability. The prompt provision of such information was a requirement to enable a victim or a victim’s family to determine whether the practical arrangements for an investigation were fundamentally and obviously flawed (para 175).

The Court of Appeal summarised its conclusions as to the legal principles in relation to independence in para 176. So far as relevant to practical independence it stated:

The Court of Appeal interpreted Maguire J’s judgment as being a finding that the LIB at that time lacked practical independence. That judgment was informed by (i) the past investigatory failures, including that conducted by the HET in 2006-2008, (ii) the lack of proper arrangements for the provision of intelligence information to the investigator and (iii) the inadequate vetting of the relevant staff. Because, as Maguire J had found, there had been a real possibility of bias in the initial RUC investigation and the HET investigation, it was necessary that clear and practical arrangements be put in place to secure that the future investigation was capable of being independent. The Chief Constable had failed to set out the practical arrangements which he proposed to put in place to conduct a further review or investigation. As a result, it had not been demonstrated that the further review or investigation by the LIB had the capacity to satisfy the article 2 requirement of practical independence.

The Court of Appeal granted declarations that the Chief Constable is (i) obliged to conduct the further investigations into Ms Smyth’s death in a way which satisfies the State’s procedural obligation under article 2 of the Convention, and (ii) bound promptly to take steps to secure the practical independence of the investigators so that they have the capacity to carry out an article 2 compliant, effective investigation into the death of Ms Smyth.

The Chief Constable of the PSNI, the Secretary of State for Northern Ireland and the Northern Ireland Department of Justice appeal to this court with the leave of the Court of Appeal. In addition to defending the Court of Appeal’s reasons, Ms McQuillan advances two further arguments to uphold the Court of Appeal’s decision, namely that there is an obligation at common law to make sure that an investigation of Ms Smyth’s death is independent and that a similar obligation arises under section 32 of the Police (Northern Ireland) Act 2000.

As discussed more fully in para 202 below, the PSNI seeks to introduce in this court further affidavit evidence to update the court on the current composition of the LIB, and the measures which the PSNI has taken to address the criticisms of the HET which were made by HMIC, including the development of new policy documents. The Committee of Ministers in March 2021 re-opened its supervision of the McKerr group of cases in view of their concerns about, among other things, the failure to implement the Stormont House Agreement by setting up the HIU, the findings of this court in In re Finucane [2019] UKSC 7; [2019] NI 292; [2019] 3 All ER 191, the findings of the Court of Appeal in the McQuillan case, HMIC’s criticisms of the HET for treating deaths involving the state differently from other deaths in a manner inconsistent with article 2 of the Convention, and the lack of resources available to the PSNI and the PONI.

In these proceedings Francis McGuigan and Mary McKenna seek judicial review of the decision made by the PSNI that there was no evidence to warrant an investigation, compliant with articles 2 and 3 of the Convention, into the allegation that the UK Government authorised and used torture in Northern Ireland in the case of the Hooded Men. The factual background is set out in considerable detail in the judgment of Maguire J (at paras 1-137) and in the judgment of Morgan LCJ and Stephens LJ in the Court of Appeal (at paras 3-37). For present purposes we draw particular attention to the matters set out in the following paragraphs.

Between 1920 and March 1972 Northern Ireland was governed by a devolved Parliament and administration pursuant to the Government of Ireland Act 1920. In March 1971, against a background of increasing fatalities, injuries and civil unrest, the devolved Northern Ireland Government entered into discussions with the UK Government on the introduction of internment without trial and the establishment of an interrogation centre in Northern Ireland. Training was provided by the Army to the RUC. This included training in five interrogation techniques which had been used in British colonies (“the five techniques”): (i) wall-standing; (ii) hooding; (iii) subjection to noise; (iv) deprivation of sleep; and (v) deprivation of food and drink. These techniques were taught by officers of the British Military Intelligence Centre to members of the RUC at a seminar in April 1971. Military Standing Orders were drawn up to govern the operation of the interrogation centre and the conduct of interrogations. Interrogations were supposed to be conducted in accordance with the Joint Intelligence Committee Directive JIC(65)15, which set out guidelines on the approach to interrogation in internal security situations outside the United Kingdom. They had been formulated in 1965 and amended in 1967 following an investigation into allegations of abuse of detainees by the Army in Aden. The JIC Directive did not itself contain any reference to the five techniques.

On 9 August 1971 Francis McGuigan and Séan McKenna, the father of Mary McKenna, were among some 350 persons detained and interned by the security forces. Between 11 and 17 August 1971 Mr McGuigan and Mr McKenna were two of the 14 men, subsequently known as the Hooded Men, who were subjected by the RUC to what was described as “interrogation in depth” which involved the five techniques. In its judgment in Ireland v United Kingdom (1979-80) 2 EHRR 25 the Strasbourg Court described these techniques as follows (at para 96):

Within a few days of the commencement of internment, allegations were made by and on behalf of those detained of brutality and ill-treatment by the security forces. These were published in newspapers, first in Ireland and then in England and Northern Ireland. In the period following the commencement of internment, the UK Government stated publicly that the decision to conduct deep interrogation and use the five techniques in Northern Ireland had been authorised by the Northern Ireland Government in concurrence with the UK Government. Statements in Parliament towards the end of 1971 indicated that Ministers knew the interrogation would be conducted within the guidelines in JIC(65)(15) and that the methods would be the same as had been used on numerous occasions in the past. In its 1978 judgment the Strasbourg Court stated (at para 97):

On 31 August 1971, the Home Secretary, Mr Reginald Maudling, responded to allegations of mistreatment of detainees by establishing a non-statutory Committee of Inquiry chaired by Sir Edmund Compton. Its terms of reference were:

On 17 November 1971 the House of Commons debated both the content of the Compton Report and the propriety of using the interrogation techniques which it identified. In response to questions as to ministerial knowledge or authorisation, Lord Balniel, Minister of State in the Ministry of Defence, stated that “the methods of interrogation have been used for many years” and that “the same Ministerial concurrence … was given to the same methods of interrogation used in Aden, Malaysia and Borneo”. He also stated:

In opening the debate on the Compton report on 16 November 1971 the Home Secretary stated that the Prime Minister had decided to establish a further committee of inquiry to examine the interrogation procedures “currently authorised” and whether any policy change was required (Hansard (HC Debates) 16 November 1971, col 217). The Parker Committee was established by the Home Secretary with terms of reference to consider:

The majority interpreted its terms of reference as requiring it to enquire generally into the interrogation in custody of a person suspected of terrorism in the future and not specifically in connection with Northern Ireland. They considered whether the five techniques complied with the JIC(65)15 Directive and whether they should be authorised. They found the boundaries between hardship, humiliating treatment and torture were matters of fact and degree and would ultimately attract different opinions. They concluded that the use of the techniques in certain circumstances should not be ruled out, with Ministerial authorisation, within the scope of the JIC(65)15 Directive and subject to safeguards against excessive use. They made no findings as to whether the use of these techniques had been authorised by a Minister, commenting (at para 12):

The minority report of Lord Gardiner considered that the procedures to be examined were those described in the report of the Compton Committee and that it was, as a result, necessary to establish of what those procedures consisted, by whom they were authorised, what their effects were and whether they required amendment. He expressly rejected the definition of “brutality” adopted by the Compton Committee. He considered whether the five techniques had been authorised by “any document or Minister”, commenting that the “only evidence before us on this point was that it could not be said that UK Ministers had ever approved them specifically, as opposed to agreeing the general principles set out in the Directive on Military Interrogation”: para 8. He considered that if any document or Minister had purported to authorise the use of the techniques, the authorisation would have been invalid, as the procedures were illegal under domestic law. He identified the possibility that use of the techniques may have contravened international law but refrained from expressing a view on this issue. He considered that the use of the techniques was not morally justifiable, even in emergency situations. He opposed any change in legislation that would allow such ill-treatment in the future.

The Parker Report was published and debated in Parliament on 2 March 1972. The Prime Minister stated that the techniques would not be used in the future as an aid to interrogation and that parliamentary approval would probably be required if they were to be so used.

All 14 men on whom the five techniques had been used made civil claims for damages, including claims alleging unlawful conspiracy, against Ministers. Dr Denis Leigh, Consultant Psychiatrist to the Army, to whom reference will be made below, acted as a defence medico-legal expert in the civil claims. All of the claims were eventually settled for sums ranging from £10,000 to £25,000.

On 16 December 1971 the Government of Ireland submitted an inter-State application to the European Commission of Human Rights (“the Commission”) against the United Kingdom contending that persons detained in Northern Ireland, including the Hooded Men, had been subjected to treatment in breach of article 3 of the Convention, carried out by the security forces of the United Kingdom, and that the treatment had been an administrative practice of the State and a continued series of executive acts which had exposed a section or sections of the population to torture or inhuman and degrading treatment. The application was not limited to the treatment of the Hooded Men in August 1971. The Irish Government maintained that the alleged acts were not isolated in time and place and had not been duly punished. It also drew attention to the failure of the UK Government to inform the Commission of the authority which ordered the application of the techniques. The Commission held that the application was admissible.

The position of the UK Government was that, as it had been acknowledged that the techniques were contrary to law, that they had been abandoned and that they would not be used again, it was not necessary to reach a finding on whether they were in breach of article 3. It advised its witnesses not to answer questions on the use of the five techniques, asserting that this was out of concern for the safety of the witnesses involved.

The Commission heard evidence in 16 “illustrative cases” which included the cases of two of the Hooded Men, Mr Patrick McClean and Mr Pat Shivers, both of whom gave evidence. The Commission heard evidence from 119 witnesses over a period of more than a year. The Commission was aware of the fact and the outcome of the civil proceedings brought by the Hooded Men. The Commission heard conflicting expert evidence in relation to the psychiatric effects of the five techniques. Professor Robert Daly and Professor Jan Bastiaans on behalf of the Irish Government maintained that both Mr McClean and Mr Shivers “would continue for a long time to have considerable disability shown by bouts of depression, insomnia and a generally neurotic condition resembling that found in victims of Nazi persecution”. Dr Denis Leigh on behalf of the UK Government maintained that “the acute psychiatric symptoms developed by witnesses during the interrogation had been minor and that their persistence was the result of everyday life in Northern Ireland for an ex-detainee carrying out his work travelling to different localities. In no sense could the witnesses’ experiences be compared with those of the victims of Nazi persecution”.

On 25 January 1976 the Commission published its report which concluded unanimously that the combined use of the five techniques constituted a practice of inhuman and degrading treatment and torture contrary to article 3 of the Convention. The Commission found that no physical injuries had resulted from the infliction of the five techniques as such but weight loss and the development of acute psychiatric symptoms during interrogation were recorded in the medical and other evidence. With regard to the conflict of expert evidence in the cases of Mr Shivers and Mr McClean, the Commission was unable to establish the exact degree of the psychiatric after-effects which the use of the five techniques might have had on these witnesses or generally on persons subjected to them. It was satisfied, however, that depending on the personality of the person concerned, the circumstances in which he found himself and the conditions of everyday life in Northern Ireland at the relevant time, some after-effects resulting from the application of the techniques could not be excluded.

The Irish Government requested that the case should be referred to the European Court of Human Rights in order to affirm the decision of the Commission. A friendly settlement could not be achieved as the two governments were unable to reach agreement on the possibility of initiating prosecutions or disciplinary proceedings against the officers involved in conducting the interrogations. The Irish Government requested that the Court make two consequential orders against the United Kingdom: first, an order that it refrain from reintroducing the five techniques as a method of interrogation or otherwise; and secondly, that criminal or disciplinary proceedings be undertaken against members of the security forces who had perpetrated, condoned or tolerated the breaches of article 3 found by the Commission. The first request was withdrawn following a solemn undertaking given by the UK Government on 8 February 1977 that the five techniques would not in any circumstances be reintroduced as an aid to interrogation. The second request was maintained.

The UK Government did not contest the findings of the Commission that the breaches of article 3 which it had found amounted to both inhuman and degrading treatment and torture. Nevertheless, the Strasbourg Court decided to adjudicate for itself upon the allegations of whether the conduct of the UK authorities amounted to a State practice of both inhuman and degrading treatment and of torture, in breach of article 3.

The Strasbourg Court handed down its judgment on 18 January 1978: Ireland v United Kingdom (1979-80) 2 EHRR 25. It expressly recognised that “[f]rom the start, it has been conceded by the respondent Government that the use of the five techniques was authorised at ‘high level’” (at para 97). The Court concluded that there was a State practice to use the five techniques during the interrogation of detainees, finding that it was “inconceivable that the higher authorities of a State should be, or at least should be entitled to be, unaware of the existence of such a practice” (at para 159). The Court agreed with the Commission that the combined use of the five techniques amounted to “inhuman and degrading treatment” contrary to article 3 but the majority found that they did not amount to torture. The majority concluded (at para 167):

There is no evidence that anyone involved in the authorisation or operation of the five techniques has ever been the subject of criminal charges. On 1 December 1971 the Attorney General of England and Wales indicated that no prosecutions would result from these events. It appears that the same position was adopted in Northern Ireland.

Following the disposal of the proceedings before the European Court of Human Rights in 1978 the case of the Hooded Men lay dormant for a considerable period of time until 2003 when government documents relating to the matter began to be deposited at the UK National Archives at Kew, pursuant to the 30-year rule. Prior to their release a review of papers was carried out by the Northern Ireland Office and other Government departments in 2000-2002 to determine whether there were any reasons why the papers should not be released or should be redacted. A note prepared by officials in 2000 stated that “there are various papers that show that Ministers and senior officials were indeed aware of the interrogation methods being used”. A further NIO document dated 15 February 2000 stated that some of the papers “contained implicit acknowledgement of authorisation of the ‘five techniques’ at Government level”. The records were released, beginning in April 2003.

The release of those materials seems to have gone unnoticed for some while. In time, however, they triggered the interest of various parties including the Pat Finucane Centre and the National University of Ireland Galway, which carried out research. Research was also carried out by the Irish national broadcaster RTÉ which on 4 June 2014 broadcast a documentary entitled “The Torture Files” about the Hooded Men and the use of the five techniques. It referred to a variety of documents which appeared to have been located in UK National Archives, which it said were newly discovered and had not been before the Commission or the Strasbourg Court. In particular, the documentary alleged that the United Kingdom had withheld from the Commission and the Court: (1) materials, including medical reports of Dr Leigh, which tended to undermine the evidence given on behalf of the United Kingdom that the psychological effects of the five techniques on the Hooded Men were not likely to be long-lasting or severe; and (2) information about who was responsible for the use of the five techniques, including the role of Ministers.

By applications dated 20 January 2015 and 19 May 2015 respectively, Mr McGuigan and Ms McKenna applied to the High Court in Northern Ireland for judicial review of decisions taken by the authorities in the light of the RTÉ documentary.

On 4 December 2014, the Government of Ireland made an application to the Strasbourg Court pursuant to rule 80 of its Rules of Court, requesting a revision of the Court’s 1978 judgment in Ireland v United Kingdom. Rule 80(1) provides:

The revision application was based upon material which the Irish Government had obtained from RTÉ and which had been used in the preparation and production of “The Torture Files”. Ireland maintained that these materials had been withheld from the Commission and the Strasbourg Court during the course of the main proceedings, that they were not known to the Court at the time of its judgment, and that they would or might have had a decisive influence on the specific question of whether or not the use of the five techniques amounted to torture. Ireland relied upon two grounds for the revision application:

The UK Government had information within its possession, including medical reports from Dr Leigh, demonstrating that the effects of the five techniques could be substantial, severe and long-lasting while that government, through the evidence of the same Dr Leigh, before the Commission had alleged in the Convention proceedings that the said effects were minor and short term.

The archive material revealed the extent to which, at the relevant time, the UK Government had adopted and implemented a policy of withholding information from the Commission and the Court about key facts concerning the five techniques, including that their use had been authorised at ministerial level and their purpose in doing so.

In a judgment issued on 20 March 2018 the Strasbourg Court, by a majority of six to one, dismissed the request to revise the 1978 judgment to substitute a finding of torture for one of inhuman and degrading treatment.

In rejecting the first ground, the majority noted that Dr Leigh’s report on Mr McKenna in June 1975 was the only document relied upon which contained direct proof of Dr Leigh’s medical views on one of the men who was subjected to the five techniques:

In relation to the second ground the majority observed that some of the documents relied upon demonstrated that the use of the five techniques constituted an administrative practice which had been authorised at ministerial level, not only at a “high level” as admitted by the United Kingdom in the original proceedings. The majority accepted that a number of the documents demonstrated that the then Government of the United Kingdom was prepared to admit that the use of the five techniques had been authorised at “high level” to avoid any detailed inquiry into the issue and that it was opposed to the hearing of witnesses in respect of the five techniques in order to avoid exposing ministers involved. However, “while the documents shed more light on the attitude of the then respondent”, the majority did not find that the relevant facts as such, which Ireland characterised as the withholding of information about the five techniques by the United Kingdom, were “unknown” to the Court at the time of the original proceedings (at para 115). It continued:

After the RTÉ documentary was broadcast, on 3 July 2014 a Board member of the Northern Ireland Policing Board asked whether there was to be an investigation by the PSNI. Assistant Chief Constable Drew Harris stated that the PSNI was aware from the documentary of the existence of a memorandum dated 31 March 1977 by Mr Merlyn Rees, who was then Home Secretary, to the Prime Minister, Mr James Callaghan (“the Rees Memo”), which referred to the use of the five techniques as “torture” and to their approval by Ministers, and in particular by Lord Carrington, who was then Secretary of State for Defence. ACC Harris explained that the PSNI wished to source the original document and other materials by visiting the Public Records Office in order to confirm or further clarify its contents. Once that had been done, the PSNI would contact the Public Prosecution Service to seek advice as to next steps. He stated that the matter had been passed to the HET. Further, in response to a written question at the meeting of the Northern Ireland Policing Board on 3 July 2014, which asked what action the Chief Constable had taken in relation to the assertion that Lord Carrington had authorised torture in Northern Ireland, the Chief Constable stated in a written answer:

Research was carried out by an officer with the HET in the National Archives at Kew to locate relevant documents. The investigating officer concluded that the use of torture was never authorised at any level and that there would be no useful purpose in taking the matter further. His report was considered at senior officer level in the PSNI. This resulted in a decision on 17 October 2014 by the Assistant Chief Constable not to take the matter any further on the basis that no evidence had been found which supported the allegation that the UK Government had authorised the use of torture in Northern Ireland. We discuss these matters more fully when we discuss the challenges to the decision of 17 October 2014 based on allegations of breach of legitimate expectations and irrationality in paras 218-252 below.

By their applications for judicial review dated 20 January 2015 and 19 May 2015 respectively, Mr McGuigan and Ms McKenna sought, among other things, to challenge (1) the decision of the PSNI, represented by the Chief Constable of Northern Ireland, that there was no evidence to warrant an investigation, compliant with article 3 of the Convention, into the allegation that the UK Government authorised the torture of Mr McGuigan and Mr McKenna; and (2) the continuing failure of the Secretary of State for Northern Ireland, the Minister of Justice for Northern Ireland and the Chief Constable to order or ensure a full, independent and effective investigation into their torture, in compliance with obligations under article 3, under international law and at common law.

In his judgment dated 27 October 2017 Maguire J held in relation to the article 3 temporal issues as follows:

Two questions arose. The first was whether it was likely that the Strasbourg Court would find that the investigative obligation under article 3 was breached. The second was whether it was open to the domestic court to hold that there was such a breach (paras 237-238).

In relation to the first question, for present purposes the “critical date” was the coming into force of the HRA on 2 October 2000 (para 240).

In considering whether the genuine connection test was met two aspects required to be met if the temporal problem was to be overcome, namely the time factor and the balance of the process of investigation as between the period prior to the critical date and period after it. With regard to the time factor, he held that “the distance in time in the present cases is simply too long to establish the existence of a genuine connection”. The gap was upwards of 40 years which exceeded by a wide margin the norm of ten years, even if this period was made the subject of a generous extension. So far as the balance of the process of investigation was concerned, the essence involved the question whether much of the investigation into the relevant event took place or ought to have taken place in the period following the critical date. In these cases, this requirement was not satisfied because “the great bulk of the activity in respect of the events here at issue occurred in the period 1971-78. Thereafter there was a long period when the issues were dormant and the court struggles to conclude that post 2014 there have been extensive investigative measures taking place.” In this regard he noted that what article 3 required at the time was the subject of exhaustive analysis in the course of the Strasbourg proceedings. Accordingly, he concluded that the two aspects of the genuine connection test, both of which would ordinarily have to be passed, had not been passed (paras 241-246).

Maguire J observed that, in considering whether the Convention values test was met, the court “is not fixed with the outlook of yester-year. As is well known, the Convention is a living instrument and falls to be interpreted in the light of present day conditions”. In this regard, “it seems likely to the court that if the events here at issue were to be replicated today the outcome would probably be that the [Strasbourg Court] would accept the description of torture in respect of these events as accurate”. Accordingly, the Convention values test had been passed (paras 251-258).

So far as the Brecknell test was concerned, Maguire J was satisfied that the new material came within the description of “plausible or credible allegation, piece of evidence or item of information relevant to the identification, and eventual prosecution or punishment of the perpetrator”. Accordingly, given that the Convention values test had been surpassed, this was sufficient to cause a revival of the obligation under article 3 to carry out an effective official investigation (paras 259-263).

He concluded, therefore, that “it is likely the [Strasbourg Court] would regard these cases as ones in which the article … 3 obligations remain to be fulfilled” (para 263).

In relation to the second question, however, Maguire J held that as a matter of domestic law the decision of the House of Lords in In re McKerr ([2004] 1 WLR 807) precluded the possibility of an investigative obligation under articles 2 or 3 as a matter of domestic law, where the events pre-dated the commencement of the HRA on 2 October 2000 (paras 264-273).

The issue of the temporal application of article 3 was, therefore decided against Mr McGuigan and Ms McKenna.

On the challenge to the independence of the PSNI, Maguire J favoured taking a broad approach which viewed the circumstances in the round (para 278). The RUC had been significantly involved in the use of the five techniques and the issue was whether the PSNI was sufficiently detached from the RUC (para 280). He referred to his judgment in the McQuillan case on this issue and acknowledged that it was then under appeal. He therefore confined himself to expressing provisional and obiter views. In concluding that the PSNI lacked the requisite independence of the RUC, he emphasised the importance of public perception and the need to maintain public confidence: there had been considerable public concern regarding the quality of the investigations into Troubles-related deaths, including those carried out by the HET, which was associated with the PSNI, and the level of scrutiny and the identity of the investigator required heightened care against this background (para 283).

Maguire J dealt with the common law challenges succinctly. Following the judgment of the House of Lords in In re McKerr, he rejected the submission that there was a parallel common law duty to investigate (para 292). But he concluded that the PSNI’s decision of 17 October 2014 not to investigate further whether the British Government had authorised the use of torture in Northern Ireland was unreasonable or the product of a legal misdirection (paras 306-311). That decision therefore fell to be quashed. He declined to make a final determination of the question whether the Chief Constable’s written answer to the Northern Ireland Policing Board gave rise to a legitimate expectation, which had been defeated by the decision of 17 October 2014 not to take forward the investigation. He observed that on one view all that the Chief Constable had said in effect was that he would do his duty, so that the real issue might be better viewed as whether the decision was one which lawfully fell within the ambit of his discretion; but that in any event, as the irrationality challenge had been upheld, this ground of challenge added nothing of substance (para 314).

Following the decision of Maguire J and before the hearing of the appeal against his decision to the Court of Appeal, two important legal developments occurred. First, on 20 March 2018 the Strasbourg Court delivered its revision judgment in Ireland v United Kingdom. Secondly, on 27 February 2019 the Supreme Court handed down its decision in In re Finucane’s Application for Judicial Review (Northern Ireland) [2019] UKSC 7; [2019] NI 292 (“Re Finucane”).

The Court of Appeal (Morgan LCJ, Stephens LJ and Sir Donnell Deeny) delivered its judgment on the appeal by Mr McGuigan and Ms McKenna on 20 September 2019, Sir Donnell Deeny dissenting in part: [2019] NICA 46; [2021] NI 15. The reasoning of the majority in relation to the article 3 temporal issues may be summarised as follows:

While the decision of the House of Lords in In re McKerr had not been overruled, it was now necessary to revise its application in the light of the decision of the Supreme Court in Re Finucane. It was now established that the Brecknell test could provide a basis for the revival of the procedural obligation as a matter of domestic law, provided that the genuine connection test or the Convention values test was also satisfied (para 96).

The “critical date” for the establishment of Convention rights in domestic law is 2 October 2000 when the HRA was commenced (para 96).

It was necessary to review the judge’s findings on Brecknell in the light of the revision judgment in Ireland v United Kingdom. Although the overall issue before the Strasbourg Court was whether there was material which satisfied the very high test as to whether there were exceptional circumstances justifying a revision of its previous judgment, the basis upon which it was contended that the judgment should be revised required determination of what the new material was and how it was relevant to a revision of the original findings (paras 99-100).

There must be a trigger before the obligation to conduct a procedural investigation arises. Any other approach would offend the principle of legal certainty. In Janowiec v Russia (2014) 58 EHRR 30 (“Janowiec”) the Grand Chamber of the Strasbourg Court had described the Brecknell test as depending upon new material emerging which should be sufficiently weighty and compelling to warrant a new round of proceedings. It was therefore necessary to examine what material was available by the time of the delivery of the judgment in Ireland v United Kingdom in 1978 and what difference to the obligation to investigate had been established by the material newly released into the National Archive (paras 102-103). Taking into account the analysis in the revision judgment, the majority considered that the Brecknell test had not been met (para 106).

Furthermore, the genuine connection test had not been met for the reasons given by Maguire J (para 107).

The majority found the application of the Convention values test more difficult. The majority accepted that the application of the five techniques amounted to torture in current circumstances. It observed that that conclusion reflected the development of the Convention as a living instrument and was a feature that should be taken into account in determining whether any proposed investigation was required by Convention values. However, the majority questioned whether that was the only feature. There was, at the very least, a role to be played in the application of the Convention values test by considering the nature of those values at the time when the relevant events took place. That would necessarily mean taking into account the conclusions of the Strasbourg Court in 1978 and would require one to recognise the investigations which did take place through the Compton Enquiry, the Parker Committee, the debates in Parliament, the investigation by the Commission and the consideration by the Court. However, the resolution of this issue was not necessary given its conclusion that the Brecknell test was not satisfied (paras 108-109).

Sir Donnell Deeny delivered a separate judgment, dissenting in part:

He agreed with the majority that neither the Brecknell test nor the genuine connection test was satisfied (paras 13-15).

However, he did not consider it appropriate to make a finding that the treatment at issue, which occurred in 1971, should now be characterised as torture, for the following reasons. First, it altered a finding of Maguire J for no good reason. Secondly, it ran counter to the substance of the finding of the Strasbourg Court in its revision judgment which chose not to make that finding, applying the appropriate test. Thirdly, it contradicted the principle of legal certainty. Fourthly, it was an unnecessary and otiose finding. He preferred the approach of the trial judge, namely to find that the conduct of the State in 1971 had a larger dimension than an ordinary criminal offence and would amount to the negation of the very foundations of the Convention (para 12).

Nevertheless, Sir Donnell Deeny joined with the majority in questioning whether the Convention values test was satisfied bearing in mind the conclusion of the Strasbourg Court in Ireland v United Kingdom and the extent of the investigation that had already taken place (para 16). The Convention values test required more than that the triggering event was of a larger dimension than an ordinary criminal offence and amounted to the negation of the very foundations of the Convention. Referring to Janowiec at para 150, he considered that, however deplorable, the treatment in this case was not to be equated with war crimes, genocide or crimes against humanity (para 17). The Convention values test contemplated extraordinary situations where there was a need to ensure the real effective protection of the guarantees and the underlying values of the Convention. This matter had been twice before the Strasbourg Court, thus meeting any such need (para 18). The passage of time and the fact that those who authorised the techniques were either dead or very elderly also pointed against the application of the test which should be kept for something more exceptional (para 19).

The Court of Appeal dealt briefly with the other grounds of challenge. It agreed with Maguire J that there was no duty to investigate at common law as the House of Lords had held in In re McKerr and as the Supreme Court confirmed in R (Keyu) v Secretary of State for Foreign and Commonwealth Affairs [2015] UKSC 69; [2016] AC 1355 (“Keyu”) (para 110). Unlike Maguire J, the majority of the Court of Appeal considered that the Chief Constable’s statement at the Northern Ireland Policing Board meeting had given rise to a legitimate expectation. Like Maguire J, the majority considered that the investigation which the researcher carried out was unduly narrow and that the PSNI’s approach to the investigation was irrational. The majority held that the expectation, from which the Chief Constable had not resiled, remained unfulfilled; the way in which the decision of 17 October 2014 not to investigate further was taken could raise an issue about whether there would be public confidence in an investigation “without practical independence from the PSNI”; and Maguire J had been correct in quashing that decision, “so that a proper investigation in which the public could have confidence could proceed” (paras 111-115). The majority of the Court of Appeal stated that, in light of the Court of Appeal’s decision in McQuillan regarding lack of independence on the part of the PSNI, it did not need to comment on the submissions relating to the independence of the PSNI (para 71). It held that in light of the manner in which the investigation had been pursued, it appeared unlikely that an investigation by the LIB or its successor would engender confidence (para 116(ix)). Sir Donnell Deeny dissented on this point also. In his view, the statement by the Chief Constable was simply a paraphrase of the duty of investigation imposed on the PSNI by section 32 of the Police (Northern Ireland) Act 2000 which created no distinct enforceable legitimate expectation (para 25). He also expressed the view that the decision in McQuillan regarding the lack of independence of the PSNI was distinguishable on the facts (para 29).

The jurisprudence of the Strasbourg Court which underpins the obligation on the State to investigate a death, or allegation of torture or inhuman and degrading treatment under articles 2 and 3 of the Convention is well established. (In this judgment, when convenient to do so, we will refer to this investigative obligation as “the article 2/3 investigative obligation”):

Articles 2 and 3 of the Convention enshrine two of the basic values of democratic societies making up the Council of Europe. Article 2, which safeguards the right to life and sets out the circumstances in which deprivation of life may be justified, ranks as one of the most fundamental provisions of the Convention: Anguelova v Bulgaria (2004) 38 EHRR 31, para 109; Jordan v United Kingdom (2003) 37 EHRR 2, para 102. Article 3, which provides that “[n]o one shall be subjected to torture or to inhuman or degrading treatment or punishment”, is also one of the core provisions of the Convention from which no derogation is permitted even in time of war or other public emergency.

As the State has a general duty under article 1 of the Convention to secure to everyone the rights and freedoms defined in the Convention, the combination of articles 1 and 2 requires by implication that there be some form of official investigation when individuals have been killed by the use of force: McCann v United Kingdom (1996) 21 EHRR 97, para 161; Nachova v Bulgaria (2006) 42 EHRR 43, para 110 (Grand Chamber); Tunç v Turkey (Application No 24014/05) [2016] Inquest LR 1, para 169 (Grand Chamber). The essential purpose of such an investigation is two-fold. It is to secure the effective implementation of the domestic laws that protect the right to life; and, in cases involving State agents or bodies, it is to ensure their accountability for deaths occurring under their responsibility: Nachova (above) para 110; Jordan (above), para 105.

A similar duty of investigation arises under article 3 of the Convention where there is a reasonable suspicion that a person has been subjected to torture or inhuman or degrading treatment: El-Masri v Former Yugoslav Republic of Macedonia (2013) 57 EHRR 25, para 182; Al Nashiri v Romania (2019) 68 EHHR 3, para 638; R (Mousa) v Secretary of State for Defence (No 2) [2013] EWHC 1412 (Admin); [2013] HRLR 32.

An adequate and prompt investigation is essential to maintain public confidence in the adherence of the State authorities to the rule of law and in preventing any appearance of complicity or collusion in or tolerance of unlawful acts: McKerr v United Kingdom (2002) 34 EHRR 20, para 114; Brecknell, para 65; Al Nashiri v Romania (above), para 641. Victims, their families and the general public have a right to the truth, which necessitates public scrutiny and accountability in practice: El-Masri v Former Yugoslav Republic of Macedonia (above), para 191; Al-Nashiri v Romania (above), para 641. The authorities must act of their own motion, once the matter is brought to their attention: McKerr v United Kingdom (above), para 111.

There must be a sufficient element of public scrutiny of the investigation or its results in order to secure accountability in practice. The degree of public scrutiny that is required will vary from case to case but the next of kin or victim must be involved in the procedure to the extent necessary to safeguard his or her legitimate interests: McKerr v United Kingdom (above), para 115; Anguelova v Bulgaria (above), para 140; Jordan (above), para 109.

There is an obligation to ensure that the investigation is effective; this is an obligation of means rather than result. The investigation must be effective in the sense that it is capable of leading to a determination of whether the force used by an agent of the State was or was not justified in the circumstances and to the identification and punishment of those responsible: Jordan (above), para 107; Nachova (above), para 113; Ramsahai v Netherlands (2008) 46 EHRR 43, para 324. For the investigation to meet this criterion, the authorities must take whatever reasonable steps they can to secure the evidence and reach their conclusions on thorough, objective and impartial analysis of all relevant elements: Giuliani and Gaggio v Italy (2012) 54 EHRR 10, paras 301-302.

Another aspect of an effective investigation, which is the focus of one of the central issues in these appeals, is that the persons responsible for carrying out the investigation must be independent of those implicated in the events. The Strasbourg Court has emphasised, as we discuss more fully below, that this requires not only a lack of hierarchical or institutional connection but also practical independence. See McKerr v United Kingdom (above), para 112; Jordan (above), para 106; Ramsahai (above), para 325. In Nachova (above), para 112, the Grand Chamber stated: “For an investigation into alleged unlawful killing by state agents to be effective, the persons responsible for and carrying out the investigation must be independent and impartial, in law and in practice.” In support of that proposition the Grand Chamber cited Güleç v Turkey (1999) 28 EHRR 121, paras 81-82; Őgur v Turkey (2001) 31 EHRR 40, paras 91-92; and Ergi v Turkey (2001) 32 EHRR 18, paras 83-84.

In our view, the importance of such independence and impartiality where agents of the State are implicated in a death or alleged ill-treatment cannot be understated. The Strasbourg Court stated in Ramsahai (above), para 325, that “what is at stake here is nothing less than public confidence in the state’s monopoly on the use of force”. In Northern Ireland in the terrible years of the early 1970s the State did not have such a monopoly. On the information currently available to the court it is not established that a shot fired by a member of the security forces was the probable cause of Ms Smyth’s death; it is one of several possibilities. But that does not diminish the need for an investigation to be independent.

Nonetheless, the degree of independence that is required depends on the circumstances of the specific case. In Tunç v Turkey (above) the Strasbourg Court stated (para 223):

In both the McQuillan case and the Hooded Men case the first set of issues in the appeal concerns the temporal effect of Convention rights in the Convention and under the HRA.

Under section 6(1) of the HRA public authorities have a duty to act compatibly with Convention rights. The Convention rights are those set out in articles of the Convention which are reproduced in Schedule 1 to the HRA. However, as explained below, the Convention rights as embodied in the HRA are domestic rights which are distinct from the rights in the Convention but which follow those rights so far as their substantive content is concerned. Accordingly, when section 6(1) applies and an investigative obligation arises under article 2 or article 3, relevant public authorities will have a duty to undertake investigations in accordance with that obligation.

The early cases in Strasbourg regarding the investigative obligations under article 2 and article 3 of the Convention concerned deaths or ill-treatment which occurred at a time after the relevant contracting state had become a party to the Convention and had accepted a right of individual petition to the Strasbourg Court. It was possible in this situation to analyse the investigative obligation as parasitic or consequential upon the substantive rights which the deceased person or the person subject to ill-treatment enjoyed under the Convention at the time of death or the ill-treatment. That was the analysis adopted by the House of Lords in In re McKerr: see in particular para 21 per Lord Nicholls of Birkenhead. The House of Lords also held that the Convention rights under the HRA are domestic rights created by that Act, and as such are distinct from the equivalent rights under the provisions of the Convention itself. It followed that the investigative obligation did not arise under section 6(1) of the HRA in relation to a death which occurred before 2 October 2000, the commencement date for the relevant provisions of the HRA.

However, two later developments in the Strasbourg case law meant that this analysis could not be sustained. First, in Brecknell the Strasbourg Court explained that in certain circumstances the article 2 investigative obligation could revive after an investigation had closed if new evidence came to light. Secondly, and related to this, the Strasbourg Court addressed the situation where new evidence came to light regarding a death which occurred before the relevant contracting state entered into the Convention: see Silih v Slovenia (2009) 49 EHRR 37 (“Silih”) and Janowiec. In view of the similar basis for the investigative obligation under article 3, the same principles apply. The significance of these developments for domestic law under the HRA was examined by this court first in In re McCaughey (Northern Ireland Human Rights Commission intervening) [2011] UKSC 20; [2012] 1 AC 725 and then in Re Finucane. However, in the present appeal Mr Hugh Southey QC makes a submission on behalf of the claimants which was not the subject of detailed analysis in those cases.

In Brecknell the Strasbourg Court examined the extent to which, when an investigation into a suspicious death appeared to be closed, the coming to light of new evidence about the death might revive the investigative obligation under article 2. The case concerned a terrorist murder in Northern Ireland which took place in 1975. (Since it is relevant to the discussion below, it should be noted that this was after the United Kingdom had entered into the Convention in 1953 and after it had accepted the right of individual petition to the Strasbourg Court on 14 January 1966.) Investigations at the time and through to 1981 failed to result in evidence sufficient to support a prosecution and the investigation was taken no further. However, in 1999 allegations emerged that the murder had been carried out with the collusion of the police. The wife of the deceased brought a complaint before the Strasbourg Court that there had not been an investigation of the death by independent law enforcement officials in light of the new material, contrary to the requirements of the investigative obligation under article 2. The Strasbourg Court held that the investigative obligation could be revived if new information was uncovered, even after the conclusion of criminal proceedings in relation to the death. It found that the investigative obligation had revived in this case in 1999; the allegations of police collusion with paramilitary groups were plausible, so it was incumbent on the state to verify the information and consider whether to launch a full investigation with a view to issuing criminal proceedings (para 75); and a breach of the investigative obligation had occurred because the initial inquiries were carried out by the RUC, which was the police force accused of collusion and hence was not independent as required by article 2 (para 76).

The Strasbourg Court set out the relevant principles at paras 65-72. It pointed out (para 66) that the obligation to carry out an effective investigation “comes into play, primarily, in the aftermath of a violent or suspicious death”. The Strasbourg Court explained (para 68) that “events or circumstances may arise which cast doubt on the effectiveness of the original investigation and trial or which raise new or wider issues and an obligation may therefore arise for further investigations to be pursued”, but the nature and extent of the investigative obligation at that stage “would inevitably depend on the circumstances of each particular case and might well differ from that to be expected immediately after a suspicious or violent death has occurred”.

While expressing doubts (at para 70) as to “whether it is possible to formulate any detailed test which could usefully apply to the myriad of widely differing situations that might arise”, the Strasbourg Court said this at para 71 (omitting footnotes):

As we have said, in the McQuillan case it is common ground that the information in the military logs was sufficient to trigger a fresh investigative obligation in accordance with the criteria set out in Brecknell. The question whether the Brecknell test is met is raised, however, on the appeals by Mr McGuigan and Ms McKenna in the case of the Hooded Men.

Mr Southey on behalf of Mr McGuigan and Ms Karen Quinlivan QC on behalf of Ms McKenna submit that an article 3 duty of investigation arises in the case of the Hooded Men because significant new material has been identified since the entry into force of the HRA so as to trigger that obligation. In their submissions before this court, they maintain that the Brecknell test is satisfied by new material relating to the authorisation of the five techniques. The task of presenting a common set of submissions on this issue fell to Mr Southey.

First, Mr Southey points to the determination of Maguire J that the new material was sufficient to cause a revival of the obligation under article 3 to carry out an effective official investigation. The judge was satisfied (at paras 260-261) that the new material fell within the broad description referred to in Brecknell of a “plausible or credible allegation, piece of evidence or item of information relevant to the identification, and eventual prosecution or punishment of the perpetrator”. Mr Southey submits, in reliance on the decision of the Supreme Court in DB v Chief Constable of Police Service of Northern Ireland [2017] UKSC 7; [2017] NI 301 at paras 78-80, that that conclusion was not clearly erroneous and should therefore have been left undisturbed by the Court of Appeal, which acted in improper reliance on the 2018 judgment of the Strasbourg Court in Ireland v United Kingdom.

In our view, the Court of Appeal did not fall into error in the manner suggested. It expressly referred (at para 99) to the guidance given by the Supreme Court in DB that it is only on the rarest occasions when the appeal court is convinced by the plainest of circumstances that it should interfere with the finding of the first instance judge on a matter of this kind. Nevertheless, it was correct in considering it significant that the judge was addressing how the Strasbourg Court might interpret the circumstances and in concluding that its decision on Ireland’s revision application, which was not available to the judge, gave substantial guidance on that issue. As the Court of Appeal pointed out (para 100), the question the court was required to address in relation to the issue of ministerial authorisation was whether that was known to the original court at the time of the hearing in 1978. Although the overall issue was whether there was material which satisfied the very high test as to whether there were exceptional circumstances justifying a revision of the judgment, the basis on which it was contended that the judgment should be revised required determination of what the new material was and how it was relevant to a revision of the original findings. The judgment in the Ireland v United Kingdom case was, therefore, an important development of which the Court of Appeal was required to take account. In particular, the conclusion of the Strasbourg Court that the new material did not demonstrate facts relating to the level of authorisation which were unknown to the Court when it delivered its original judgment was clearly relevant to whether the Brecknell test was satisfied.

Mr Southey’s principal submission is that the Brecknell evidential threshold is met because important new information relevant to the identification and prosecution of those responsible for the authorisation of the use of the five techniques came to light in 2014. He submits that the material discovered in the National Archives was fresh evidence which for the first time identified senior Ministers - the Secretary of State for Defence (Lord Carrington), the Home Secretary (Mr Reginald Maudling), the Minister for Home Affairs in Northern Ireland (Mr Brian Faulkner) and the Minister of State for Defence (Lord Balniel) - as having given informed authorisation to carry out criminal acts and torture. In particular, he submits that the new information confirms that Lord Carrington was told not only about the safeguards found in JIC(65)15 but also about the supporting methods designed to heighten the subject’s desire to communicate with his fellow human beings including isolation, fatigue, white sound and deprivation of sense of place and time. Furthermore, the documents are said to demonstrate that the techniques were deployed only after RUC officers had received assurances from the highest places that those officers would not be sanctioned. By contrast, he submits, the information reasonably available to Mr McGuigan and Ms McKenna prior to 2014 did not identify those responsible for authorising the five techniques, did not indicate the circumstances surrounding the authorisations and did not explain the extent to which they were informed about what they would involve.

In the same way, Mr Southey submits that, although the Strasbourg Court in its judgment in Ireland v United Kingdom in 1978 recorded (at para 97) that “it has been conceded by the respondent Government that the use of the five techniques was authorised at ‘high level’”, this did not identify individuals, did not suggest they were Ministers, and did not explain the extent of their knowledge or the circumstances surrounding the authorisation.

In considering whether the new material relied upon is “sufficiently weighty and compelling to warrant a new round of proceedings” (Janowiec, para 144) it is necessary to assess the new material against the material that was available at the date of the previous investigation. We are assisted in this regard by the following summary provided by the Court of Appeal which has not been challenged by any of the parties before us.

With regard to the issue of ministerial authorisation, we would draw particular attention to statements made by Lord Balniel in the House of Commons which are referred to in para 77 above. On 17 November 1971 he stated that “the methods of interrogation have been used for many years” and that “the same Ministerial concurrence … was given to the same methods of interrogation used in Aden, Malaysia and Borneo”. He also stated that Ministers knew that the interrogation would be conducted within the guidelines laid down in 1965 and 1967 and that the methods would be the same as had been used on numerous occasions in the past. On 9 December 1971 Lord Balniel stated in the House of Commons that the interrogation was authorised by the Northern Ireland Government with the knowledge and concurrence of Her Majesty’s Government.

We would also draw attention, as did the Court of Appeal, to the observations on the subject of authorisation by Lord Gardiner in the minority report of the Parker Committee. As we have seen, he explained that he found this a difficult point because the terms of reference of the Committee appeared to assume that the procedures were authorised. “The only evidence before us on this point was that it could not be said that UK Ministers had ever approved them specifically, as opposed to agreeing the general principles set out in the Directive on Military Interrogation” (at para 8). He then went on to explain that if any Minister had purported to authorise them it would have been invalid because the procedures were illegal by domestic law. It is clear, therefore, that the question of authorisation at Ministerial level was a live issue in the investigations which took place up to 1978. It was one which was not pursued at that time as a matter of policy.

The new material undoubtedly provides a considerable amount of detail in relation to the authorisation of the five techniques which was not previously publicly available. In particular, it identifies the part played by individual Ministers. It also casts light on the policy decision by the UK Government not to pursue criminal or disciplinary proceedings against individuals. However, that is not sufficient to give rise to an investigative obligation. What is critical here is not the inconclusive nature of earlier investigations but whether there now exists such weighty and compelling new evidence as to require a fresh investigation. In our view, the new material does not add significantly to the state of knowledge in relation to this matter as it stood in 1978 nor does it alter its substance. As the Court of Appeal put it:

Our conclusion that the Brecknell test is not satisfied in the present circumstances is supported by the judgment of the Strasbourg Court on the application to revise its earlier judgment in Ireland v United Kingdom. Although not concerned directly with the application of the Brecknell case, that judgment is highly relevant to the present issue for the reasons explained above. The Strasbourg Court noted (at para 114) that some of the documents demonstrated that the use of the five techniques constituted an administrative practice which had been authorised at ministerial level, not only at a high level as admitted by the United Kingdom in the original proceedings. However, it went on to observe (at para 117) that the United Kingdom had conceded from the start that the use of the five techniques had been authorised at a high level and taught to members of the RUC. Furthermore, there had been a practice adopted of using the techniques. In these circumstances the Strasbourg Court concluded (at para 118) that the documents did not demonstrate facts that were unknown to the Court when the original judgment was delivered.

We do not understand Mr McGuigan or Ms McKenna to rely on further material concerning the medical evidence of Dr Leigh in support of their case that the Brecknell test is satisfied. In our view this material cannot be relevant to the identification and eventual prosecution or punishment of a perpetrator of conduct in breach of article 3. In any event, we note that in its judgment on the revision application in Ireland v United Kingdom the Strasbourg Court rejected the submission that Dr Leigh had misled the Commission.

Finally on this issue, we accept the submission of Sir James Eadie QC on behalf of the Secretary of State that there is a close analogy between the present case and Chong v United Kingdom (2019) 68 EHRR SE2. In December 1948, a group of 24 unarmed civilians were shot and killed by a British Army patrol in Selangor which at that time was a British Protected State in the Federation of Malaya. In 2011 relatives of the deceased sought judicial review in England and Wales of the authorities’ refusal to hold an inquiry into the deaths. The claim based on article 2 of the Convention failed before the domestic courts in the Keyu case. In subsequent proceedings the Strasbourg Court held the application before it inadmissible on the ground that it had been lodged significantly out of time. It observed (at para 93):

For these reasons we conclude that the Brecknell test is not satisfied in the case of the Hooded Men.

In Silih the question arose how the article 2 investigative obligation applied in relation to a death which occurred before the relevant contracting state (Slovenia) became a party to the Convention, where an investigation commenced prior to the date of accession continued after that date. The Strasbourg Court had to consider whether it had temporal jurisdiction as a matter of international law to determine a complaint alleging violation of the article 2 investigative obligation in relation to such a death. The Strasbourg Court held (para 159) that the investigative obligation was “a separate and autonomous duty” which was “detachable” and was not parasitic upon the substantive obligation in article 2. On that basis, it held that it did have temporal jurisdiction to consider the complaint and found a violation of the investigative obligation in respect of steps taken after the date of accession. We examine the reasoning of the Strasbourg Court in more detail below.

In Re McCaughey this court, by a majority, held that the decision of the House of Lords in In re McKerr fell to be reconsidered in the light of Silih. Since the article 2 investigative obligation was autonomous and was detachable from the substantive obligation under article 2 to protect life, it could not be treated as parasitic on that substantive obligation as the House of Lords had done.

The interaction of the investigative obligation revival principle in Brecknell and the principle of detachability of the investigative obligation established in Silih was examined by the Grand Chamber of the Strasbourg Court in Janowiec. That case was concerned with, among other things, claims that Russia was subject to an investigative obligation under article 2 in relation to the mass killing of Polish soldiers by the Soviet Union in the Katyn Forest in 1940, ie long before Russia became a party to the Convention. By a majority of 13 votes to 4, the Grand Chamber held that the Strasbourg Court had no temporal jurisdiction and was not competent to examine the complaint. The Grand Chamber explained that, in accordance with Silih, the Strasbourg Court’s temporal jurisdiction in relation to such a claim requires either (1) a “genuine connection” with the death which constitutes the triggering event for the obligation consisting of (a) a reasonably short period of time between the death and the entry into force of the Convention for the state in question, not in excess of ten years, and (b) a requirement that the major part of the investigation must have been or ought to have been carried out after the entry into force of the Convention for that state, or (2) in extraordinary situations which do not meet the “genuine connection” test, where there is a need to ensure that the guarantees and the underlying values of the Convention are protected (the “Convention values” test). Although the case involved allegations which potentially fell within the class of case covered by the Convention values test, the Grand Chamber held that this test could not be applied to events which occurred prior to the adoption of the Convention in 1950.

The Grand Chamber reiterated (para 128) that “the provisions of the Convention do not bind a Contracting Party in relation to any act or fact which took place or any situation which ceased to exist before the date of the entry into force of the Convention with respect to that Party (the critical date)”. It noted (para 132) that, as held in Silih, the investigative obligation under article 2 is autonomous and detachable and so is capable of binding the state even when the relevant death took place before the critical date; however, it observed (para 133) 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”. The Grand Chamber acknowledged (para 140) that the application of the “genuine connection” test and the “Convention values” test to be derived from the Silih judgment had given rise to some uncertainty and therefore proceeded at paras 141-151 to clarify the position.

At paras 142-144 the Grand Chamber gave guidance in relation to procedural acts and omissions in the period after the Convention comes into force for a contracting state as follows (omitting certain footnotes):

As regards the “genuine connection” test, the Grand Chamber explained (paras 146-148, omitting footnotes):

In relation to the “Convention values” test, the Grand Chamber said this (paras 149-151):

In Keyu, as outlined above, this court considered claims that the respondent Secretaries of State were obliged to hold a public inquiry into the deaths of 24 unarmed civilians killed by British soldiers in Malaya in 1948, pursuant to an investigative obligation arising under article 2, customary international law or under the common law. These claims all failed.

As regards the investigative obligation under article 2, the primary argument for the Secretaries of State was based on the temporal limits of the operation of the Convention itself. They argued (para 77) that for the purpose of determining those limits the critical date referred to in Janowiec is the date on which the relevant state first recognised the right of individual petition (which in the case of the United Kingdom, including in respect of events in Malaya at the relevant time, was 1966) rather than the date on which the Convention came into force in relation to the relevant territory (which was 1953). This submission was accepted by a majority of the court: paras 78-87; and this was later confirmed as correct by the Strasbourg Court when the Keyu case went to Strasbourg: Chong v United Kingdom (above), para 84. On that basis, the deaths occurred more than ten years before the critical date, so the “genuine connection” test was not satisfied. The “Convention values” test was not applicable, because the deaths occurred before the Convention came into being (Keyu, para 88, citing Janowiec, para 151). Lord Neuberger of Abbotsbury ruled that the claim based on article 2 therefore failed because the Strasbourg Court would rule it inadmissible (para 89). After referring to Re McCaughey, he left open the question whether, if the Strasbourg Court would have held that the claimants were entitled to complain about a breach of the investigative obligation under article 2, yet the death was more than ten years before the commencement of the HRA in October 2000, there would have been an equivalent enforceable obligation under section 6(1) of the HRA: paras 92-98. That is the issue which arises in these appeals and is addressed in this section of the judgment.

Lord Neuberger did, however, comment (para 93) that the argument of the Secretaries of State that there would be no jurisdiction for a domestic court to entertain a claim under the HRA in such circumstances was “on the face of it, … a very powerful contention”, since it is clear from section 22(4) of the HRA that it was not intended to have retrospective effect and the contention had the support of all five opinions in the House of Lords in In re McKerr. We return to these points below.

In Re Finucane this court confronted the question whether a claimant could maintain an action based on the article 2 investigative obligation and section 6(1) of the HRA in circumstances where the triggering death had occurred 11 years and eight months prior to the coming into force of the HRA in October 2000. The leading judgment was given by Lord Kerr of Tonaghmore, with whom the other members of the court agreed on this issue. The argument before the court and its own analysis proceeded on the footing that temporal jurisdiction in respect of a complaint of violation of the Convention rights in the HRA is determined by reference to the coming into effect of the HRA on 2 October 2000, that being the critical date for the purposes of application of the test in Silih and Janowiec in the context of a claim made under section 6(1) of the HRA: see paras 101 and 107-108. On that basis Lord Kerr held (paras 110-111) that it had been established in Re McCaughey that the approach in Silih should be adopted in domestic law for determining whether a domestic court had jurisdiction in relation to a claim under section 6(1) of the HRA complaining of a breach of the investigative obligation under article 2. Following the approach in Silih, which in this respect Lord Kerr considered had not been modified by Janowiec and had been adopted into domestic law in Re McCaughey, he held (paras 108-109) that the decision as to whether there is a genuine connection between a triggering death and the coming into force of the relevant obligation (in domestic law, the obligation arising under the HRA) involved a multi-factorial exercise in which the weight to be attached to each factor will vary according to the circumstances of the case, and that it is not an immutable requirement that the period between them be less than ten years. On that basis, Lord Kerr considered (paras 114-137) that new material had come to light which satisfied the test in Brecknell for revival of the article 2 investigative obligation in that case; it had not been subject to proper investigation; most of the investigations into the death had occurred after the critical date; and in the particular circumstances the court should hold that there was a continuing obligation to investigate the death under section 6(1) of the HRA taken with article 2. That obligation had not been complied with (paras 138-147) and a declaration was made to that effect (para 153).

With respect to Lord Kerr, he did not identify any clear principle by which one could tell when and to what extent it might be appropriate to water down a strict ten-year requirement as the Grand Chamber of the Strasbourg Court had appeared to lay down in Janowiec, para 146. We have reservations as to whether Lord Kerr was right to interpret Janowiec as he did. This court has not been invited to depart from its decision in Re Finucane but we note that the extension beyond ten years allowed in Re Finucane involved less than two more years. It would significantly undermine the legal certainty which the Grand Chamber sought to achieve in Janowiec if longer extensions than this were to be contemplated or permitted. Moreover, in Janowiec, para 146, the Grand Chamber emphasised that the time factor is the “most crucial indicator” in relation to the “genuine connection” test and that the test requires that “the lapse of time between the triggering event and the critical date must remain reasonably short”. In our judgment, an extension beyond the normal ten year limit of up to two years is permissible where there are compelling reasons to allow such an adjustment constituted by circumstances that (a) any original investigation into the triggering death can be seen to have been seriously deficient and (b) the bulk of such investigative effort which has taken place post-dates the relevant critical date. If in these circumstances there is an extension of no more than two years beyond the ten-year limit mentioned in Janowiec, it remains possible to describe the lapse of time as “reasonably short” in accordance with the guidance in that judgment at paras 146 and 148.

As we have stated in our discussion of Janowiec in para 139 above, an alternative basis on which the temporal application of the HRA may be established is the Convention values test. In Janowiec (at para 149) the Strasbourg Court accepted that “there may be extraordinary situations which do not satisfy the ‘genuine connection’ standard …, but where the need to ensure the real and effective protection of the guarantees and the underlying values of the Convention would constitute a sufficient basis for recognising the existence of a connection.” The Grand Chamber explained (at para 150) that:

As we have said, in the present proceedings Maguire J held that the Convention values test had been satisfied. The Court of Appeal (at paras 108-109) found the question difficult and gave no concluded answer. We return to this issue below.

Against this background we now turn to the question whether 14 January 1966 or 2 October 2000 should be taken to be the “critical date” for the purposes of the genuine connection test derived from Silih and Janowiec as that test falls to be applied in the context of a claim under section 6(1) of the HRA, read with the article 2/3 investigative obligation. Mr Southey submits that the “critical date” for these purposes should be taken to be the same “critical date” identified in the Strasbourg jurisprudence in relation to the Convention, which is 14 January 1966, that being the date on which the United Kingdom accepted the right of individual petition under the Convention. Against this, Sir James Eadie submits that for the purposes of a claim under section 6(1) of the HRA the “critical date” is 2 October 2000, the date of commencement of that Act.

We have already noted that in Keyu, para 93, Lord Neuberger indicated that he saw considerable force in the latter submission and that in Re Finucane Lord Kerr and the other members of the court all assumed that this is the correct position and based their analysis upon it. We would add that Lord Kerr appears to have read Re McCaughey as supporting this approach, though Lord Neuberger was more circumspect about that. Certainly, two members of the court in Re McCaughey thought that this was the proper analysis: see paras 112 and 119 (Lord Kerr) and paras 135 and 139 (Lord Dyson). However, the other judgments of those in the majority (Lord Phillips of Worth Matravers, Lord Hope of Craighead, Baroness Hale of Richmond and Lord Brown of Eaton-under-Heywood; Lord Rodger of Earlsferry dissented) are not so clear on this point. The issue was not the subject of full debate or analysis in Re McCaughey and was not determined. As Lord Neuberger noted in Keyu, the question of what is the “critical date” for the purposes of a claim under section 6(1) of the HRA and the interaction of the position which would apply in Strasbourg and that which applies in relation to a claim in the domestic courts remains open. The question arises for decision in these appeals.

Three features of the HRA are significant in this context. First, although the Convention rights created in domestic law by the HRA are defined by reference to the Convention, they are distinct from the rights in the Convention itself. This was clearly and authoritatively explained in In re McKerr. Lord Nicholls emphasised (para 25) that, in analysing the effect of Convention rights in domestic law under the HRA, it was necessary “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”:

The other members of the Appellate Committee drew the same distinction between the rights under the Convention, which are binding on the state under international law, and the rights under the HRA, which are binding on public authorities in domestic law: see paras 48 (Lord Steyn), 62-66 (Lord Hoffmann), 80 (Lord Rodger) and 90 (Lord Brown). As Lord Hoffmann said (para 62), the obligations under the Convention and the obligations under section 6(1) of the HRA “belong to different legal systems; they have different sources, are owed by different parties, have different contents and different mechanisms for enforcement”; and (para 63) the Convention, as an international treaty, is not part of domestic law, and that has not been changed by the HRA:

Secondly, it was established well before In re McKerr that the HRA, in particular section 6, does not have retrospective effect: see In re McKerr, paras 16 (Lord Nicholls), 48 (Lord Steyn), 67 (Lord Hoffmann), 77 (Lord Rodger) and 88 (Lord Brown). The general presumption is that a statute which creates rights and obligations does not have retrospective effect. This reflects values of fairness, legal certainty and the rule of law. It is desirable that people, including public officials and public authorities, should be able to determine their legal rights and obligations at the time of acting or omitting to act. It is generally unfair to treat people as subject to obligations of which they were not on notice at that time. The applicability of the general presumption to the HRA is emphasised by section 22(4) of the Act, which provides that, subject to one exception, section 7(1)(b) of the HRA (which is the remedial provision in relation to section 6(1) of the Act) “does not apply to an act taking place before the coming into force of that section”. However, as Lord Nicholls observed in In re McKerr (para 16), the general proposition that the HRA does not have retrospective effect can give rise to as many questions as it answers.

In the context of a statute with such wide-ranging application and effects as the HRA, the presumption that it does not have retrospective effect operates more as a general principle capable of nuanced application rather than as a simple bright-line rule. The position was examined in some detail in Wilson v First County Trust Ltd (No 2) [2003] UKHL 40; [2004] 1 AC 816 (“Wilson”), to which reference was made in In re McKerr. Wilson concerned the application of the HRA in relation to proceedings taken after October 2000 concerning enforcement of a consumer credit agreement entered into in 1999, shortly before the HRA came into effect. Under regulations, the agreement was unenforceable because the lender had failed to comply with certain formal requirements in relation to consumer credit agreements. The Court of Appeal, however, held that the loan agreement should be enforced because to fail to do so would involve violations of the lender’s Convention rights. On appeal the House of Lords overturned that decision, holding that the HRA had to be interpreted in light of the presumption against the retrospective effect of legislation and that the correct interpretation of the HRA (in that case, section 3(1) of the HRA) is that it did not affect the rights and obligations of the parties to an agreement made before the HRA came into force.

Lord Nicholls pointed out (para 19) that the presumption against retrospective effect is vague and imprecise and, drawing on the judgment of Staughton LJ in Secretary of State for Social Security v Tunnicliffe [1991] 2 All ER 712, 724, observed that it is based on the idea that Parliament is presumed not to have intended to alter the law applicable to past events and transactions in a manner which is unfair to those concerned in them: “[i]t is not simply a question of classifying an enactment as retrospective or not retrospective. Rather it may well be a matter of degree - the greater the unfairness, the more it is to be expected that Parliament will make it clear if that is intended.” See also para 208 per Lord Rodger, which is to similar effect. Lord Hope said (para 98) that the presumption “is based on concepts of fairness and legal certainty”, and the weight attaching to those matters will vary according to context. Lord Hobhouse of Woodborough emphasised (para 126) the distinction between the domestic rights under the HRA and rights under international law as set out in the Convention and explained (para 130) that the domestic rights create legal liabilities for public authorities in domestic law and therefore raise a potential issue of retrospectivity, in relation to which section 22(4) carries the clear implication that the HRA in general does not have retrospective effect as against public authorities (see also para 160 per Lord Scott of Foscote). Lord Scott likewise explained (para 153) that the presumption is based on fairness and justice, and operates where unfairness or injustice may be produced “if persons are subjected on account of … past events to liabilities that they were not previously subject to”; but the presumption can be rebutted if it sufficiently appears that it was indeed the intention of Parliament to produce the result in question. In his view (para 161), “[t]he arguments against allowing the Act to have a general retrospective effect [are] very powerful”; transactions might have continuing effects, such as a lease granted for 99 years, and might in some respects be affected by later legislation, “[b]ut it would be unusual for [later] legislation to alter the rights and obligations of the parties resulting from events that had already taken place, such as a forfeiture notice already served, a damages claim already instituted, rent review machinery already in train, and so on.”

This court in Re McCaughey and Re Finucane did not depart from the general position that the HRA does not have retrospective effect, but treated it as qualified to the extent that it could be inferred that Parliament intended that there could be a limited application of the article 2/3 investigative obligation in respect of a triggering event which occurred before 2 October 2000 by way of an analogy to be drawn with the principles identified in Silih and Janowiec. We discuss below how this bears on the question we have to decide.

Thirdly, in general terms it can be said that Parliament intended that the domestic rights created by the HRA in relation to public authorities should mirror the rights in the Convention, applicable in international law to the United Kingdom as a contracting state. This mirror principle has been noted many times and was highlighted in the judgment of Lord Phillips in Re McCaughey, para 59: “[t]he object of the Act was to bring human rights home. This will only be achieved if claimants are able to bring in this jurisdiction claims that they would otherwise be permitted to bring before the Strasbourg court.” Lord Phillips observed that this principle was taken as a guide in relation to the territorial ambit of the HRA in R (Quark Fishing Ltd) v Secretary of State for Foreign and Commonwealth Affairs [2005] UKHL 57; [2006] 1 AC 529 and R (Al-Skeini) v Secretary of State for Defence (The Redress Trust intervening) [2007] UKHL 26; [2008] AC 153 (“Al-Skeini”).

However, there clearly are limits to the mirror principle. The extent to which it applies depends upon the proper interpretation of the HRA, as Lord Nicholls was careful to point out in In re McKerr, para 25 (quoted above). The proper interpretation of the HRA must take into account the fact that it created a set of rights and obligations distinct from those applicable in international law pursuant to the Convention to the United Kingdom as a state and the fact that Parliament intended the general presumption against retrospective effect to govern the application of the HRA. It is inherent in the scheme of the HRA that it would not mirror the position in Strasbourg in all cases: there would be many cases in which prior to 2 October 2000 a public authority might act in a way which engaged the responsibility of the United Kingdom under international law and constituted a violation of a right under the Convention for which a remedy could be sought in Strasbourg, yet there would be no remedy obtainable under the HRA. Since the rights in the Convention and the HRA are distinct and the persons subject to obligations under them and the obligations themselves are different, the mirror principle can only provide general guidance; it is inherent in the nature of the legal regime produced by the HRA that Parliament could not have intended the mirroring effect of the Act to be exact and uniform. That is particularly so in relation to the time at which Convention rights are to apply.

Moreover, the mirror principle does not provide a sure guide in every case and eminent judges have treated it as less than decisive in various contexts. In Al-Skeini, Lord Bingham of Cornhill, dissenting on the issue of the territorial ambit of the HRA, considered that there were insufficient indications in the HRA to outweigh the usual presumption that a statute has territorial effect only within the United Kingdom. In Al-Skeini Lord Rodger considered that the mirror principle prevailed so as to make the HRA applicable to the same territorial extent as the Convention, saying (para 57) that “in case of doubt, the Act should be read so as to promote, not so as to defeat or impair, its central purpose” of enabling Convention rights to be vindicated in the domestic courts without the need to resort to the Strasbourg Court. But in Re McCaughey he dissented, disagreeing with the view of the majority that section 6(1) of the HRA should be read by analogy with the approach of the Strasbourg Court in Silih and Janowiec so as to give effect to an investigative obligation under article 2 applicable to a death which occurred before the HRA came into effect. He forcefully drew attention to the fact that Parliament had chosen not to provide a remedy under the HRA for matters occurring before 2 October 2000 even though it knew that claimants would be able to pursue complaints about them before the Strasbourg Court: paras 157-158.

The interpretative exercise in the present case involves weighing the specific intention of Parliament (as underlined by section 22(4)) that section 6(1) of the HRA should not have retrospective effect against the general intention that the rights protected under that provision should mirror those applicable under the Convention against the United Kingdom as a state. Lord Brown framed the issue helpfully in Re McCaughey, para 100: “[o]ur 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).”

Mr Southey contends that no significant issue of retrospective application of section 6(1) of the HRA arises on the submission which he makes. At the time of the relevant events in the McQuillan case and the Hooded Men case, which post-dated the acceptance of the right of individual petition in 1966, the United Kingdom was subject to obligations under article 2 and article 3 of the Convention, respectively, including obligations of investigation under those provisions. On Mr Southey’s argument, those investigative obligations have since been revived pursuant to the principle in Brecknell. Therefore, he says, the claimants in the present cases would be entitled to take their complaints about breaches of those obligations to the Strasbourg Court if they are not afforded a remedy in the domestic legal system; and there is nothing retrospective or which pre-dates the HRA about those current obligations. The HRA should be interpreted in line with the mirror principle, and since the claimants would be able now to seek to vindicate their rights under the article 2/3 investigative obligation before the Strasbourg Court they should be able to do the same before the domestic courts under the HRA.

We do not accept this argument. The consequence of it, as Mr Southey submits, is that for the purposes of the application of the genuine connection test, as transposed into domestic law in relation to the HRA, the relevant “critical date” is the date on which the United Kingdom accepted the right of individual petition under the Convention in 1966. In our view, this involves a mistaken approach to the application of the genuine connection test as applied in the context of the distinct legal instrument which is the HRA.

The genuine connection test was articulated by the Strasbourg Court in Silih to specify how the principle of non-retroactivity of the Convention (identified at para 140) should be applied in the context of the article 2 investigative obligation where the triggering death occurred before the entry into force of the Convention with respect to the state party (see also Janowiec, para 128). The Strasbourg Court observed (para 141) that in applying the principle of non-retroactivity it was prepared “to have some regard to facts which occurred prior to the critical date because of their causal connection with subsequent facts” which form the basis of the complaint. Although the article 2 investigative obligation is a detachable obligation, it remains the case that “it is triggered by the acts concerning the substantive aspects of article 2” (para 159), which is to say the death itself (see also Janowiec, para 132). The Strasbourg Court stated (para 161) that, although the investigative obligation is detachable, “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” (see also Janowiec, para 133); and it was in that context that the Strasbourg Court held (para 163) that “there must exist a genuine connection between the death and the entry into force of the Convention … for the procedural obligations imposed by article 2 to come into effect.” As appears from this analysis, the investigative obligation is detachable not in the sense of being completely free-standing, but in the more limited sense that it is capable of arising notwithstanding that the death which triggers the obligation preceded the Convention coming into effect, provided there is a sufficient connection between that event and the death.

Thus the point of the genuine connection test articulated in Silih and in Janowiec is to identify when a death before the critical date is capable of triggering an investigative obligation under the article 2 right set out in the Convention which the Strasbourg Court will recognise as falling within its jurisdiction ratione temporis. Contrary to the suggestion by Mr Southey, the genuine connection test and the assessment of what should count as the “critical date” cannot be separated from the issue of non-retroactivity (so far as concerns the Convention) and the issue of non-retrospectivity (so far as concerns the HRA). Although the article 2 investigative obligation is detachable, it remains the case that it has to be anchored in the circumstances of a death which has occurred in suspicious circumstances and which the court is prepared to treat as relevant. This means that there is inevitably an issue of retroactivity where the death happened before the relevant article 2 obligation was brought into operation.

Turning to the distinct regime of rights in domestic law as set out in the HRA, the question is how the limited qualification of the principle of non-retroactivity of the Convention as explained in Silih and Janowiec falls to be applied by analogy in relation to the HRA which, like the Convention, does not have retrospective effect. As explained in In re McKerr, the article 2 right in the HRA is not the same as the article 2 right in the Convention. It is a distinct domestic law right which applies to domestic public authorities in the context of a domestic statute. The genuine connection test falls to be applied by analogy, but the analogy has to be developed with due regard to the context of the domestic Convention rights regime in the HRA.

Under the HRA, the substantive right under article 2 has no application before 2 October 2000, by contrast with the substantive right under article 2 of the Convention. Under the Convention, the substantive right under article 2 has no relevant application in relation to the United Kingdom before 1966. In both cases the same issue arises, as to when a death which occurred before the inception of the relevant right under article 2 can trigger an investigative obligation under article 2 which is capable of recognition and enforcement by a domestic court or the Strasbourg Court, respectively. For the purposes of application of the genuine connection test, the “critical date” is the date of inception of the relevant right under article 2. In the case of the Convention, so far as concerns the United Kingdom, it is 14 January 1966; and different dates are relevant for other contracting states, depending on the date of inception of the right under article 2 for them. In the case of the HRA, applying the same principle, the critical date is 2 October 2000.

This conclusion is supported by the clear choice made by Parliament as to the non-retrospective effect of the HRA, as emphasised in specific terms by section 22(4) and the absence of any transitional provision to carry forward from 2000 any obligation to investigate to the demanding Convention standards any suspicious deaths going back to 1966. As Lord Rodger observed in Re McCaughey, para 159, “[m]aking the HRA apply to the investigation of violent deaths occurring as far back as 1980 or 1990 would have raised particularly sensitive questions … [which] would have had significant practical effects”, especially in relation to Northern Ireland, which Parliament can hardly be thought to have overlooked when it decided that the HRA should not have retrospective effect. We would add that these points apply with even more force when it is recognised that the submission of Mr Southey takes the relevant starting date back to ten years or so before 1966 and that pursuant to it the range of investigative obligations is expanded to cover cases of alleged conduct contrary to the standards in article 3 as well as deaths. Referring to Lord Brown’s formulation in Re McCaughey, para 100, quoted above: to construe the HRA by reference to a critical date of 14 January 1966 rather than 2 October 2000 would create major practical difficulties in its application which Parliament cannot have intended should arise.

Further, to interpret the HRA in this way would require the diversion of significant resources now to investigate matters to Convention standards which were not applicable in domestic law at the time of the events in question. This would involve a significant element of unfairness for the relevant public authorities, which would have been able to deal with matters of investigation far more efficiently, more effectively and at less cost had they been on notice of such obligations when the relevant events occurred. This again points to the applicability and force in this context of the general presumption against retrospective effect of a statute (see paras 151-153 above), particularly as underlined by section 22(4). There is inevitably a retrospective dimension in a case of this sort for the simple reason that the article 2/3 investigative obligation only arises because of, and with respect to, a particular event (or alleged event) which occurred in the past, before the HRA came into effect.

In our view, against the background of the general position of non-retrospectivity of the HRA and in recognition of the distinct nature of the rights established by the HRA, the only plausible inference is that Parliament intended only a modest qualification of the non-retrospectivity of the domestic law regime in current circumstances by analogy with the principle in Silih and Janowiec, treating 2 October 2000 as the “critical date” for the purposes of application of the genuine connection test articulated in those cases. Interpreting the HRA in this way gives effect to the general mirror principle of the Act, by following the principle laid down in the jurisprudence of the Strasbourg Court, while at the same time taking account of the non-retrospective effect of the Convention rights in the Act, which is a specific and cardinal feature of the Act. In that regard, the difference in the view of Lord Rodger between Al-Skeini and Re McCaughey is instructive: there were no specific indications in the HRA to show that the mirror principle underlying the Act should not outweigh the usual presumption of territorial effect of a statute; by contrast, there were strong indications in the Act that the presumption against non-retrospective effect was intended to have real force. As general guidance, the mirror principle cannot be taken to outweigh the specific choice that Parliament has made that the HRA should not have retrospective effect, other than by treating that effect as qualified to the limited extent we have set out. The investigative obligation under article 2 only arises where there is a death which is a relevant triggering event, and as Lord Scott observed in Wilson (para 161) it is unusual for legislation to alter rights and obligations resulting from events that have already taken place. In the context of interpretation of the HRA, the mirror principle can only justify a departure from that general position to the limited extent we have indicated.

Therefore, after full examination of the issue, we confirm our agreement with the provisional view of Lord Neuberger in Keyu, the view of Lord Kerr and Lord Dyson in Re McCaughey and the view of the entire court in Re Finucane (see paras 133-154 above) that the “critical date” for the purposes of the application of the genuine connection test as drawn from Silih and Janowiec in the context of section 6(1) of the HRA is 2 October 2000.

On the footing that, as we have concluded, for the application of the genuine connection test in the context of section 6(1) of the HRA the “critical date” is 2 October 2000, it is clear that in neither the McQuillan case nor the Hooded Men case is the genuine connection test satisfied.

In the McQuillan case, the putative triggering event for the article 2 investigative obligation was a murder in 1972, some 28 years before the “critical date”. That is a period very considerably in excess of that contemplated in Silih, Janowiec and Re Finucane. Therefore the first condition inherent in the genuine connection test is not satisfied.

As regards the second condition inherent in the genuine connection test, Maguire J held that this was not satisfied either. There was a significant investigation at the time of the death, albeit it did not cover the matters raised by the military logs. In the judge’s view, it could not be said “that the majority of the investigation” had, or ought to have been, conducted after the “critical date” of 2 October 2000, and this meant that the second condition had not been met (see his second judgment in the McQuillan case, paras 26-28). The Court of Appeal disagreed, holding that the further investigation of military involvement in the death was a significant matter which fell squarely within the guidance in Silih, para 163, 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”.

The proper approach to the second condition is that set out in Janowiec, para 147 (quoted in para 139 above), which in broad terms affirms what the Strasbourg Court said in Silih, para 163, but also explains the second condition in more detail. It is unfortunate that in this part of their judgments neither the judge nor the Court of Appeal directed themselves specifically by reference to Janowiec, para 147.

In our view, the judge erred in the approach he adopted, by looking to see when, in simple terms, the majority of the investigation took place. According to Janowiec, para 147, under the second condition it will be sufficient if “much of the investigation into the death took place or ought to have taken place” after the critical date, which may include “a significant proportion of the procedural steps that were decisive for the course of the investigation”. The investigation of military involvement in the death of Jean Smyth is a major aspect of the case which is potentially decisive for the course of the investigation and which will only be carried out after the relevant critical date. In our view, this means that the second condition is satisfied in this case. Para 147 of Janowiec also posits a long-stop test, which involves asking whether by reason of “a major part of the proceedings or the most important procedural steps” taking place before the critical date, the court’s ability to make an overall assessment of the effectiveness of the investigation is “irretrievably undermine[d]”. We do not consider that this can be said to be the case here.

In the case of the Hooded Men, so far as the first limb of the genuine connection test is concerned (temporal connection), the triggering event occurred in 1971, some 29 years before the critical date. Even if a generous extension were to be made to the norm of ten years, it is clear that this element of the test is not satisfied in the present case.

So far as the second limb of the genuine connection test is concerned (procedural acts and omissions), very extensive enquiries into the matter took place during the 1970s. These included the Compton Committee, the Parker Committee, Parliamentary debates, inter-State proceedings before the Commission and proceedings before the Strasbourg Court. All of these extensive investigations and enquiries were completed by 1978. It is clear that the centre of gravity of the investigations which took place or should have taken place is firmly fixed in the decade of the 1970s, more than 20 years before the critical date. Furthermore, we agree with the observation of Maguire J (at para 245) that while it is correct that these investigations and inquiries did not concentrate on the issue of identifying those responsible for acts with a view to bringing them to justice, this reflected the prevailing legal situation at that time. What article 3 required at that time was the subject of exhaustive analysis in the course of the Strasbourg proceedings in the 1970s. The limited additional material which has come to light in 2014 adds little to the picture, does not constitute a significant new dimension of investigation of the treatment of the Hooded Men and cannot be regarded as “decisive for the course of the investigation”. We also consider that in this case, by reason of the extent of the investigative steps taken in the 1970s, the court’s ability to make an overall assessment of the effectiveness of the investigation into the treatment of the Hooded Men is now irretrievably undermined, so that the long-stop test referred to in Janowiec, para 147, referred to above, is satisfied.

Accordingly, we agree with Maguire J and the Court of Appeal that neither limb of the genuine connection test has been passed in the cases of Mr McGuigan and Ms McKenna.

In view of our conclusion that the “critical date” for the purposes of application of the genuine connection test in the context of section 6(1) of the HRA is 2 October 2000, it must follow in the McQuillan case that the relevant public authorities are not subject to any article 2 investigative obligation under section 6(1). In Janowiec, para 144 (quoted above), the Grand Chamber made it clear that if the triggering event for an article 2 investigative obligation “lies outside the Court’s jurisdiction ratione temporis, the discovery of new material after the critical date may give rise to a fresh obligation to investigate [pursuant to the revival principle in Brecknell] only if either the ‘genuine connection’ test or the ‘Convention values’ test … has been met.” In other words, the revival of an article 2/3 investigative obligation pursuant to the principle in Brecknell is subject to either the genuine connection test or the Convention values test being satisfied in respect of the relevant trigger event (death or alleged ill-treatment, as the case may be) in the particular case. By analogy, the same approach applies in relation to section 6(1) of the HRA, with the “critical date” for that purpose being 2 October 2000. As we have explained, neither of those tests is satisfied in the McQuillan case, so there is no scope for application of the Brecknell principle in that case.

With respect to the Court of Appeal in the McQuillan case, they overlooked this part of the judgment of the Grand Chamber in Janowiec. They fell into error by ruling, in effect, that the “genuine connection” test is subject to the Brecknell test, which is the opposite of the ruling in Janowiec, para 144. That paragraph reflects the logic of the Grand Chamber’s reasoning in Silih, as explained in Janowiec. In the light of that reasoning, Sir James Eadie rightly pointed out that the Court of Appeal’s approach is wrong in principle. If correct, it would have the effect that in any Brecknell-type case where new evidence emerges in relation to a suspicious death or alleged ill-treatment going back decades, perhaps to the promulgation of the Convention in 1950, the article 2/3 investigative obligation would apply; but that would be destructive of the legal certainty which the Grand Chamber was concerned to achieve by its judgments in Silih and Janowiec and would be contrary to the statements in those judgments that the investigative obligation in relation to deaths that occur before the relevant “critical date” for any contracting state is not open-ended.

As we have explained (paras 119-132 above), we agree with the Court of Appeal that the test laid down in Brecknell for revival of an article 2/3 investigative obligation is not satisfied in the Hooded Men case.

Even if that test were satisfied, the reasoning we have set out in relation to the McQuillan case would mean that the genuine connection test could not be satisfied in the Hooded Men case because of the passage of time.

It was not suggested that the Convention values test (para 139 above) is relevant in the McQuillan case. In support of her submission on behalf of Ms McKenna (and as adopted by Mr Southey on behalf of Mr McGuigan) that the Convention values test is satisfied in the case of the Hooded Men, Ms Quinlivan submits that aspects of this case clearly relate to the core standards that should govern State activity and that the Convention values test provides an important means for identifying and safeguarding such standards. She makes the point that the prohibition of torture and inhuman or degrading treatment is a value of civilisation closely bound up with respect for human dignity (Bouyid v Belgium (2016) 62 EHRR 32, para 81) and that respect for human dignity forms part of the very essence of the Convention (Vinter v United Kingdom (2016) 63 EHRR 1, para 113).

Ms Quinlivan submits that the Convention values question is one that also engages common law principles and peremptory norms of customary international law. This, she submits, is a point about mutually reinforcing legal standards and the irreducible nature of the prohibition of torture in contemporary law. Here, she is able to rely on the speech of Lord Bingham in A v Secretary of State for the Home Department (No 2) [2005] UKHL 71; [2006] 2 AC 221, para 33:

Ms Quinlivan is also able to invoke the following pronouncement by the Strasbourg Court in Soering v United Kingdom (1989) 11 EHRR 439, para 88, cited by Lord Bingham in A (No 2) (above), para 29, which is of particular relevance in the present context.

The boundary between torture and inhuman and degrading treatment is one of degree and is often difficult to draw. The conclusion of the Strasbourg Court in Ireland v United Kingdom in 1978 (at para 167) was that the treatment accorded to the Hooded Men, although undoubtedly amounting to inhuman and degrading treatment, “did not occasion suffering of the particular intensity and cruelty implied by the word torture as so understood”. However, the law in this regard has developed since 1978. In Selmouni v France (2000) 29 EHRR 403, it was submitted on behalf of France (at para 94) that, in the light of the Court’s case law and the circumstances of the case, the ill-treatment allegedly inflicted by the police officers in that case did not amount to torture within article 3. The Strasbourg Court observed (at para 101, omitting footnotes):

In its judgment on the revision application in Ireland v United Kingdom the Strasbourg Court drew attention to this passage in Selmouni and made clear that, in deciding whether the new facts had a decisive influence on the original judgment, the Court had to make its assessment in the light of the case law on article 3 as it stood at the time of the original judgment.

It is likely that the deplorable treatment to which the Hooded Men were subjected at the hands of the security forces would be characterised today, applying the standards of 2021, as torture. There is a growing body of high judicial authority in support of this view. In 2005 in A (No 2) (above) Lord Bingham, pointing out that it would be wrong to regard as immutable the standard of what amounts to torture, observed (at para 53) that it may well be that the conduct complained of in Ireland v United Kingdom would now be held to fall within the definition of torture in article 1 of the UN Convention against Torture. Lord Hoffmann agreed (para 97):

This view was also shared by the majority of the Court of Appeal in the present proceedings.

On the basis that the treatment to which the Hooded Men were subjected is indeed to be characterised as torture (at least by the standards applicable in 2021), Miss Quinlivan is able to advance a powerful argument that the Convention values test is satisfied in the present case. Not only does she point to the special status of the prohibition on torture in international law, but there are other features which support her submission that this involved a fundamental violation of Convention values repugnant to the rule of law. This treatment was administered as a matter of deliberate policy by the law enforcement agencies of the state. Those who administered it were acting under orders and were trained as to how it should be inflicted. It was authorised at a very high level including ministerial authorisation and was, therefore, an administrative practice of the state.

However, in considering whether the Convention values test is satisfied so as to revive an investigative obligation in respect of conduct 50 years ago, the contemporary standards of 1971 have a critical role to play. The issue is whether, exceptionally, the temporal scope of the Convention, or, as in this case, domestic implementing legislation, may be extended so as to give rise to an investigative obligation into events which occurred before the relevant jurisdiction was accepted. The jurisprudence of the Strasbourg Court, which applies by analogy in relation to the HRA, establishes that “the required connection may be found to exist if the triggering event was of a larger dimension than an ordinary criminal offence and amounted to the negation of the very foundations of the Convention” (Janowiec, para 150). That question must be answered by reference to legal standards as they existed at the time of the triggering event. Any other approach would open up the prospect of potentially unlimited extensions of temporal scope in order to take account of subsequent legal developments. Accordingly, we consider that, in the present case, the Convention values test must be applied on the basis of the law as it stood in 1971 and in the years immediately following during which the original inquiries and investigations took place.

Ms Quinlivan’s submission in relation to the Convention values test therefore encounters the difficulty that in 1978 the Strasbourg Court held that the treatment to which the Hooded Men were subjected in 1971 was not to be characterised as torture. Whether it would be characterised as torture by the standards of 2021 is, in our view, strictly irrelevant to the application of the Convention values test.

Mr Southey sought to persuade us that, even if the relevant conduct is not to be characterised as torture, the fact that it undoubtedly constituted inhuman and degrading treatment is sufficient to satisfy the requirements of the Convention values test. There are, however, difficulties in the path of this submission. First, it is clear from the Strasbourg Court’s exposition of the Convention values test in Janowiec that it is intended to apply only to “extraordinary situations” which do not satisfy the genuine connection test (para 149). The required connection may be found to exist “if the triggering event was of a larger dimension than an ordinary criminal offence and amounted to the negation of the very foundations of the Convention” (para 150). The examples which the Court provides of such events - “serious crimes under international law, such as war crimes, genocide or crimes against humanity” - indicate that it had in mind the most extreme violations. Secondly, although the precise boundary between torture and lesser forms of abusive conduct may be difficult to draw, international law undoubtedly accords to torture a special status which is not granted to lesser infringements. This is reflected, for example, in the speech of Lord Bingham in A (No 2) (above) (at para 53). Having referred to the distinction between torture and inhuman or degrading treatment under article 3 of the Convention and to that between torture and “other acts of cruel, inhuman or degrading treatment or punishment which do not amount to torture as defined in article 1” of the Torture Convention (article 16(1)), he rejected an assimilation of these concepts:

It is, however, not necessary to resolve this issue in order to decide the appeal in the cases of Mr McGuigan and Ms McKenna. In the light of the clear conclusion to which we have come in relation to the Brecknell test (see paras 119-132 above), it is not necessary to express a concluded view in relation to the application of the Convention values test to the particular circumstances of the case of the Hooded Men.

As we have stated in para 109(vii) above, the Strasbourg Court has recognised the requirement of independence as comprising two elements: (i) a lack of hierarchical or institutional connection between the investigating organisation and those implicated in the events under investigation and (ii) practical independence in the conduct of the investigation itself. We have also pointed out (para 111) that the Strasbourg Court does not require absolute independence but mandates that the persons and bodies responsible for the investigation are sufficiently independent of the persons and structures who may be responsible for the death or inhuman or degrading treatment which is the subject of the investigation. Thus, in Ramsahai (above) the fact that the public prosecutors relied on the police for information and support was not enough of itself to support the conclusion that they lacked sufficient independence from the police. The adequacy of the degree of independence falls to be assessed in the light of the circumstances of the specific case.

Tunç v Turkey (above) is a decision of central importance in this aspect of these appeals. This is because in its judgment the Grand Chamber of the Strasbourg Court undertook to give general guidance on the requirement that an investigation be independent. It explained that the procedural issues under articles 2 and 3 of the Convention are not assessed in the same manner as the requirements of a fair hearing under article 6. In para 221 the Strasbourg Court explained that the criteria for article 6 independence are in the main statutory criteria as to the manner of appointment of members of a court and the duration of their term of office, adding “[t]he question whether the body presents an appearance of independence is also of relevance”.

By contrast, as the Strasbourg Court stated in para 222, the requirements of article 2 (and article 3) call for “a concrete examination of the independence of the investigation in its entirety, rather than an abstract assessment”. Case law illustrated the elements which the Strasbourg Court had taken into account, including (i) the fact that the investigators were potential suspects, (ii) that they were, or were likely to be, direct colleagues of the persons subject to investigation; (iii) that they were in a hierarchical relationship with the potential suspects; or (iv) that the specific conduct of the investigative bodies indicated a lack of independence. The Strasbourg Court gave as examples of such specific conduct under heading (iv) the failure to carry out measures that were necessary to elucidate the case and punish those responsible, the giving of excessive weight to the statements of the suspects, the failure to explore lines of enquiry which were clearly required, or excessive inertia. It appears to us that elements (i)-(iii) may properly be regarded as examples of unacceptable hierarchical or institutional connection while element (iv) focusses on the methods adopted by and actual behaviour of the investigating officers in conducting the investigation, thereby addressing the question of the practical independence of the particular investigation. As we have said, the Strasbourg Court went on to state that absolute independence of those conducting an investigation is not required; what is required is “that they are sufficiently independent of the persons and structures whose responsibility is likely to be engaged (see Ramsahai, cited above, paras 343 and 344)”. The adequacy of the degree of independence is to be assessed in all the circumstances of the specific case.

It is appropriate to quote in full paras 224 and 225 of the Tunç judgment:

Mr Southey sought to downplay the significance of the judgment in Tunç on the basis that it concerned a death in which the deceased’s family suggested that an army conscript was implicated rather than the security forces as an institution (as para 255 of the judgment reveals). But it is clear that the Strasbourg Court, when it set out the principles relating to the independence of investigations in paras 217-225, was addressing the article 2/3 investigative obligation both where there was a suggestion of direct state responsibility for a death or inhuman or degrading treatment and where there was not. See, for example, para 223 in which the Strasbourg Court spoke of the persons and bodies responsible for the investigation being independent of “the persons and structures whose responsibility is likely to be engaged” and in its reference to paras 343 and 344 of the Ramsahai judgment, a case which involved policemen shooting the deceased. It is an error to argue that the guidance in Tunç is not directed to circumstances such as those which have arisen in the current appeals.

It is only fair to acknowledge that the important guidance in Tunç, which so clearly distinguishes the approaches under articles 2 and 3 on the one hand and article 6 on the other, and which ties the question of the independence and impartiality of an article 2 or 3 investigation into the overarching question of its effectiveness, was not placed before Maguire J or the Court of Appeal and so was not available to inform their reasoning.

In a domestic case which predated Tunç by over a decade, In re Kelly’s Application for Judicial Review [2004] NIQB 72, Kerr LCJ, who later as Lord Kerr of Tonaghmore was a member of this court, adopted the approach of requiring the completion of an investigation before forming a judgment on its independence and effectiveness, where he was satisfied that it had the capacity to be effective. In that case, Mr Kelly had been murdered in July 1974, there were suspicions that Ulster Defence Regiment soldiers were involved in his murder, and there were allegations of collusion by RUC officers. There were serious concerns about the adequacy of previous enquiries into Mr Kelly’s death and it was conceded that those enquiries had been inadequate. It was known to the court both that the Chief Constable of the PSNI had appointed a senior police officer, who had been seconded from an English police force, to lead a further investigation and that the officers of the PSNI who were engaged in the investigation had not been in service in the RUC at the time of Mr Kelly’s death, and had no connection with the area or any involvement in the events surrounding his death. Kerr LCJ referred to several decisions of the Strasbourg Court, including Shanaghan v United Kingdom (above), McKerr v United Kingdom (above), and Jordan (above), in support of the view that the court should conduct a post hoc examination of the actual investigation procedures and the outcome of the investigation. Being satisfied that the investigation had the capacity to fulfil the procedural requirements of article 2, he dismissed the application for judicial review which was brought before the investigation had been completed. He stated (para 32):

In the McQuillan case the Court of Appeal in part six of its judgment analysed the legal principles as to investigatory independence. We do not agree with some of that analysis, and in particular the use made in this context of the domestic test of apparent bias in Porter v Magill [2001] UKHL 67; [2002] 2 AC 357, as discussed in the succeeding section of this judgment. But we share the Court of Appeal’s view that it is correct as a general rule to adopt the approach in Kelly of awaiting the outcome of an investigation before ruling on its effectiveness, provided it has the capacity to fulfil the procedural requirement of independence. We agree with the Court of Appeal’s statement (para 173): “The approach in Kelly allows an assessment of the practical independence of the investigation informed by the outcome and an ability to assess the degree of rigour and fairness in the investigation”. The Court of Appeal recorded in its summary of applicable legal principles (in para 176(d) and (f)) that there should be a strong presumption against a judicial review application challenging the practical independence of a police investigation before the conclusion of the investigation. We agree. The court should generally intervene before the conclusion of an investigation only if it can be shown that the arrangements for the investigation as envisaged and any arrangements which could or might sensibly be put in place as the investigation proceeds would not have the capacity to fulfil the article 2/3 investigative obligation by being an effective investigation, thereby giving rise to the need for a fresh investigation.

In his first judgment in the McQuillan case, which we summarised in para 56 above, Maguire J emphasised the paramount importance of having regard to public perception and concluded that the LIB lacked the requisite independence to perform an investigation into Ms Smyth’s death which complied with article 2 of the Convention. He did not draw any distinction in his judgment between hierarchical or institutional independence on the one hand and practical independence on the other, as the Strasbourg Court has done. In its analysis of his judgment, the Court of Appeal, whose judgment we have summarised in paras 61-69 above, interpreted his conclusion as being that an investigation by the PSNI would lack practical independence. The Court of Appeal gave weight to the reforms of the police service since the Patten report and concluded (i) that the PSNI was institutionally distinct from the RUC, (ii) that the PSNI did not have a hierarchical or institutional connection with the military or security forces, and (iii) that where it was necessary to achieve independence from the PSNI itself, measures could be put in place to ensure sufficient hierarchical disconnection with the Chief Constable, given the oversight of the Policing Board (paras 193-196).

The Chief Constable also applies to introduce affidavit evidence from Detective Superintendent Stephen Wright, the deputy head of the LIB, on current LIB practice. D Supt Wright’s affidavit includes information about (i) the current composition of the LIB, including the continuing reduction in the number of officers who had previously been officers in the RUC, (ii) the role of the Departmental Review Officer in the LIB in addressing the criticisms of the HET by HMIC concerning the verification of intelligence reports and the verification of conflict of interest declarations, and (iii) the development by the LIB of revised policy documents on conflicts of interest, on the strategy for engaging with families and the prioritization of cases awaiting investigation. We are aware that the statements in the affidavit are not agreed and are satisfied that it would not be appropriate that we make factual findings on the matters which are addressed in the affidavit. We have had regard to the affidavit in so far as it shows that the LIB has continued to adapt its procedures to enhance its independence, but the task of this court is to form a view on the correctness of the judgment of the Court of Appeal, without regard to subsequent events.

In our view the Court of Appeal was correct to conclude that the LIB did not lack hierarchical and institutional independence from the military and the RUC. There is no evidence to support the view that the LIB has unacceptable connections with the perpetrators of the events which are the subject matter of these appeals and which occurred almost 50 years ago. This is in marked contrast to the Turkish cases referred to in Nachova (above) which concerned investigations that took place shortly after the relevant deaths. We are not persuaded that there is any reason for concluding in advance of an investigation that, as a matter of generality, the PSNI cannot carry out an effective investigation of a death or maltreatment in which the RUC, the military or the security services were implicated.

Mr Southey criticised the Court of Appeal for drawing a strict divide between hierarchical and institutional independence on the one hand and practical independence on the other. He argued that the proper approach should be holistic. He defended the Court of Appeal’s invocation of Porter v Magill and pointed out that one could have an investigation which was adequately carried out but was still found to be ineffective under article 2 because the investigators lacked independence and could not secure public confidence. In his submission the key issue was the appearance of bias. However, we are not persuaded either that it is helpful to elide the two tests of lack of hierarchical and institutional connection and the practical independence of the investigation in question or that prominence should be given to the appearance of bias so as to undermine the approach which the Strasbourg Court has set out in Tunç.

The Court of Appeal used the test of apparent bias in Porter v Magill (above), namely whether the fair-minded and informed observer, having considered the facts, would conclude that there was a real possibility that the tribunal was biased, as an aid in applying the Strasbourg Court test of lack of hierarchical or institutional connection and practical independence (paras 151-157 and 176(g)). It justified that use on the basis that the underlying aim of the Strasbourg Court test for the independence of an article 2 or 3 investigation and the aim of the domestic test for apparent bias were the same, that is the maintenance of public confidence.

We are not persuaded that there is a sufficiently close analogy between the circumstance of an investigation under article 2 or 3 of the Convention and the application of the rule against apparent bias in relation to adjudicative bodies to support the use of the Porter v Magill test as an aid in applying the tests set out by the Strasbourg Court. In our view its use has the danger of muddying the waters and leading to error because it risks departing from the approach laid down by the Grand Chamber in Tunç and by Kerr LCJ in Kelly. The Porter v Magill test is designed to address the role of a person or body performing an adjudicative role, such as a judge, arbitrator or arbitral tribunal, or, as in the Porter case itself, the auditor under the Local Government Finance Act 1982 whose role included acting as a judge, or, at least in the view of one of their Lordships in In re Duffy [2008] UKHL 4; [2008] NI 152, the Parades Commission for Northern Ireland which acted as a mediator and decision-maker on the contentious issue of parades and marches. A challenge on the ground of apparent bias can result in a judge or arbitrator recusing himself or herself at a preliminary stage in proceedings or in any event long before a dispute has been determined or, in the case of arbitration, in an application at such an early stage for the removal of an arbitrator under section 24 of the Arbitration Act 1996. In Tunç, as we have said, the Strasbourg Court at para 221 recognised the relevance of the appearance of independence in challenges under article 6 of the Convention. By contrast, in relation to investigations under articles 2 and 3 the Strasbourg Court held that article 2 required a concrete examination of the investigation in its entirety (para 222) and that the various parameters were to be taken jointly in assessing the effectiveness of the investigation (para 225 which we have quoted in para 196 above). While there is no question that one of the aims of the article 2 procedural duty is the maintenance of public confidence and that public perception is an important consideration in the maintenance of public confidence, it is clear from the jurisprudence of the Strasbourg Court that this aim is sufficiently met when the requirements of institutional, hierarchical and practical independence as explained by the Grand Chamber in Tunç are satisfied. Where they are satisfied, it can be seen that there has been an effective investigation and it is the effectiveness of the investigation which maintains public confidence in the rule of law. By contrast, a different standard is applicable in relation to a body which adjudicates on legal rights: for such a body, it will not be sufficient to show that it in fact conducted a rigorous examination of the case before it to conclude that it has been effective in applying the law. In addition, having regard to the function it performs in adjudicating on legal rights, it has to be free of any taint of lack of impartiality, judged to a particularly stringent standard.

The court in assessing the effectiveness of an investigation must make an objective assessment; and evidence that a particular investigator assigned to an investigation has shown partiality towards a person or organisation thought to be involved in a death would militate strongly against a conclusion that an investigation by that investigator was or could be effective, because of the importance of impartiality and the appearance of impartiality. To that extent there is some similarity to Porter v Magill. But as discussed above (paras 195-200), the norm, which the Court of Appeal recognised and which is consistent with the Grand Chamber’s approach in Tunç, is that, where there is a challenge on the ground of a lack of practical independence, the investigation should be allowed to proceed to its conclusion and its effectiveness can then be assessed by analysing what has been done and what is its product. In order for an earlier challenge to an investigation under articles 2 or 3 to succeed, it must be shown that the investigation could not have the capacity to be effective, having regard to the elements which the Strasbourg Court discussed in Tunç and despite the possibility that the investigator could be required to carry out further and better inquiries (as happened in Tunç itself: see paras 63-74, 183 and 206) or that other investigators could be brought in to make good any deficiencies in the independence of the investigation. To the extent that the courts below relied on Porter v Magill as a test to determine the practical independence of an investigation at a stage before the investigation has been conducted, where there was no finding of any unacceptable hierarchical or institutional connection between the investigating organisation and the alleged perpetrators, they erred in law.

But that does not mean that on the particular facts of the investigation into the death of Ms Smyth, the Court of Appeal reached the wrong conclusion. As the Strasbourg Court has emphasised in many cases, the question of effectiveness of an investigation and the question of independence as an element of the assessment of that effectiveness must be assessed by reference to the particular circumstances of each case: see for example Brecknell, para 72. It is to be borne in mind that the Strasbourg Court has emphasised that the criteria or parameters which determine whether an investigation is effective are inter-related and are to be taken jointly in the court’s assessment: Tunç, para 225. In this case the Court of Appeal correctly took into account the inadequacy of the initial investigation by the RUC in the 1970s and the shortcomings of the HET review between 2006 and 2008, shortcomings which may or may not have had a benign explanation. The Court of Appeal took into account the delay in establishing the HIU which appears to be a result of continuing political disagreements. The Court of Appeal also took into account the fact that the LIB did not have some of the attributes of the HET which had distanced it from the hierarchy of the PSNI; but, as the Court concluded that the PSNI was hierarchically and institutionally distinct from the RUC and the military, it is our view that distance from the PSNI hierarchy is not a significant element in the assessment. Those elements, including the recognised shortcomings of the HET review, did not of themselves mean that a further investigation by the PSNI could not be independent. But they created doubts in the family’s minds and in the wider community which had an objective basis. Further, an important criterion of effectiveness of an article 2 review or investigation is the involvement of the deceased person’s family in that investigation or review. This involvement, as the Strasbourg Court has held, does not entitle the family to have access to the material available to the investigators during the review or investigation. In cases of preliminary investigation the involvement of the family can be achieved by informing them of at least the gist of the findings and recommendations in due course: Hackett v United Kingdom (Application No 34698/04) 10 May 2005. But the degree of family involvement and public scrutiny that is required depends upon the circumstances of the particular case.

In this case the Court of Appeal (para 199) criticised the Chief Constable for failing to give any indication of the arrangements which he would put in place for the investigation of Ms Smyth’s death. It concluded that in the absence of any such statement on behalf of the Chief Constable the LIB had not demonstrated that it had the capacity to satisfy the article 2 requirement of practical independence.

The Chief Constable argues that it would be premature to inform families about a proposed investigation before an investigation had been assigned and refers the court to the LIB Family Engagement Strategy, which envisages initial contact with the family at the outset of the review. He argues that such a practice would encourage families to mount challenges in an attempt to jump the queue in a context in which there is a very significant backlog of investigations and only limited resources to undertake them.

We are not persuaded that that is a satisfactory reason for declining to provide information in the circumstances of Ms McQuillan’s case. The relevant background was the inadequacies of the previous investigations into Ms Smyth’s death and especially that of the HET for which the PSNI is responsible, the concerns expressed by HMIC about certain elements of the practice of the HET, including the concern that there had been a culture of preferential treatment for soldiers who were under investigation, and the problems which the PSNI has faced of inadequate resources to progress the investigations in relation to the Troubles while awaiting the establishment of the HIU. Against that background, it is readily understandable that families, victims and members of the public had concerns about whether the further inquiry into Ms Smyth’s death would be effective. This is because nobody had given the family and others an explanation as to how it was proposed that the investigation would be conducted to achieve practical independence. In the circumstances of this investigation there was no reason why the Chief Constable could not have given an indication, as was done in Kelly (para 199 above), as to the nature of the measures that would be taken to enhance the prospect of achieving the practical independence of the investigation, such as by undertaking to make sure that none of the LIB investigation team to be established for the investigation had RUC experience. Had the Chief Constable given an indication of the measures which were likely to be adopted, that might not have satisfied Ms Smyth’s family or all members of the public, who might have to await the outcome of the investigation to judge its effectiveness, including its practical independence. But the circumstances of Ms McQuillan’s application called for a greater degree of engagement and candour on the part of the PSNI in order to engage with the family and enable public scrutiny of the PSNI’s proposed investigative processes. As the Strasbourg Court explained in Tunç (para 225) the elements which achieve compliance with the article 2 procedural requirement are inter-related. Once sufficient hierarchical and institutional disconnection has been achieved and recognised, the practical independence of a future investigation is not the only element in play and a lack of engagement with the family and the public may also compromise the effectiveness of an investigation. While we are satisfied that it was not correct to conclude that the investigation which was then proposed would inevitably lack practical independence, we are persuaded that the lack of such engagement did compromise the effectiveness of that investigation in the particular circumstances of this case. In our view, absent such an explanation in the case concerning Ms Smyth’s death, the Court of Appeal was entitled to conclude that the investigation was not being conducted in an effective manner.

For completeness, we observe that the Chief Constable also argues that the Court of Appeal erred in asserting at para 175 of its judgment that the Belfast Agreement created a legal obligation on the Chief Constable at the request of the families to demonstrate in advance the arrangements to achieve the practical independence of a proposed investigation. We do not read that paragraph of the Court of Appeal’s judgment as asserting that the Belfast Agreement conferred legal rights on individuals, but the point is of no significance in this appeal.

We turn, then, to the Hooded Men case. As the Court of Appeal explained in para 47 of its judgment in the challenges by Mr McGuigan and Ms McKenna, when Maguire J addressed the question of independence in that litigation, he had the ruling of the Court of Appeal in McQuillan that the arrangements in the LIB for carrying out such investigations lacked the practical independence necessary to satisfy the procedural obligation under article 2. The Court of Appeal itself in para 71 of its judgment in the Hooded Men case considered that it was unnecessary to address the submissions on the independence of the PSNI because of its prior decision in McQuillan. Thus, the reasoning of the Court of Appeal in McQuillan, which we have not accepted, governed the decision on independence in the Hooded Men appeal.

In our view, it has not been established that the LIB is not capable of carrying out an effective investigation on the basis either of institutional or hierarchical connection or that it is not capable of conducting an investigation with practical independence. Our decision in upholding the result of the Court of Appeal’s decision on effectiveness in the McQuillan application rests essentially on the failure of the Chief Constable of the PSNI in the particular circumstances of that case to explain to Ms Smyth’s family, the wider public, and when faced with the judicial review challenge, the court, what he intended to do. That reasoning cannot be read across into the circumstances of the challenge relating to the Hooded Men. We address below (paras 223-252), in the context of the common law challenges, the research into the documents at the National Archive at Kew carried out by a member of HET and conclude that the decision of the PSNI on 17 October 2014 not to investigate further in the light of that research was irrational. However, there is nothing to suggest that it would not be possible to assign appropriate officers of the PSNI to carry out any further investigations to a proper standard. We agree with Sir Donnell Deeny that the McQuillan case is to be distinguished on its facts.

As an alternative to the claim that the defendants in these cases had a duty of investigation pursuant to article 2/3 and section 6(1) of the HRA, in their written submissions the claimants argued that the defendants were subject to a duty of investigation arising at common law, including an equivalent duty of independence. In the event, Mr Southey, who was dealing with this part of the case for the claimants, did not press this point in his oral submissions. The submission that there is an equivalent common law investigative obligation was decisively rejected by the House of Lords in In re McKerr and again by this court in Keyu. In the present appeals the court has not been taken to any authority to support the submission; nor has any substantive argument been addressed to us which could call in question the correctness of those decisions on this point. We are not persuaded that there is any reason to revisit those previous rulings, let alone overrule them pursuant to the 1966 Practice Statement (Practice Statement (Judicial Precedent) [1966] 1 WLR 1234). It is not necessary to say anything more about this part of the case.

Mr Southey and Ms Quinlivan also submit that section 32 of the 2000 Act imposes an equivalent duty of independence of investigation to that which arises under an article 2/3 investigative obligation. Section 32 has the heading, “General functions of the police”. So far as is relevant, section 32 provides:

Neither Maguire J nor the Court of Appeal considered that section 32 imposes an obligation of independence equivalent to that which arises under the article 2/3 investigative obligation with respect to any investigation the PSNI might undertake. We agree. Section 32 imposes a general obligation on the officers of the PSNI to undertake the usual functions to be expected of a police force across the whole range of their activities in investigating crime. It is not drafted so as to refer to any distinct standard of independence in relation to investigations they might undertake, let alone to the particular standard of independence which arises under the article 2/3 investigative obligation. Nor can such an obligation of independence be implied. As explained above, no such obligation exists at common law, so one cannot say that such an obligation is implicitly imported into section 32 for that reason; there is no other foundation for identifying such an implied obligation.

The claimants’ case based on legitimate expectation arises out of events at the Northern Ireland Policing Board on 3 July 2014 (para 99 above). The legitimate expectation is said to arise out of the Chief Constable’s written response to a written question asking what actions he had taken in light of the allegation in the RTÉ documentary that Lord Carrington had sanctioned the use of torture. We repeat the written response here:

Maguire J considered that the written response amounted to no more than a statement that the Chief Constable would do his duty, ie in accordance with section 32 of the 2000 Act (paras 312-314). That was also the view of Sir Donnell Deeny in the Court of Appeal (para 25). However, a majority of the Court of Appeal held that the written response gave rise to a legitimate expectation in law “as to the nature of the investigation that should be carried out” and characterised it as “a legitimate expectation of a procedural kind to the public at large” (para 113).

The comments by the majority of the Court of Appeal at paras 114 and 115 (referred to at para 108 above) in the context of their analysis of the claim based on legitimate expectation, to the effect that an issue could arise “about whether there is likely to be any public confidence in an investigation without practical independence from the PSNI” (para 114) and that the judge was right to quash the decision of the PSNI of 17 October 2014 “so that a proper investigation in which the public could have confidence could proceed” (para 115), have led both sides to assume that they took the legitimate expectation to be an expectation of an investigation conducted according to the requirement of independence applicable in the context of an article 2/3 investigative obligation or an equivalent requirement at common law. The PSNI understood this to be by reference to article 2/3 itself; Ms Quinlivan for Ms McKenna (whose submissions on this part of the case were adopted by Mr Southey) understood it to be by reference to an equivalent requirement of independence under the common law. It should be noted, however, that the majority of the Court of Appeal did not in terms say this. In our view, the parties have perhaps read too much into their judgment: the only relevant order which the Court of Appeal made was to dismiss the appeal against that part of the order made by Maguire J quashing the PSNI’s decision of 17 October 2014. Be that as it may, the Chief Constable denies that any legitimate expectation arose and supports the reasoning of the judge and of Sir Donnell Deeny on this point.

The law on legitimate expectations was reviewed recently in this court in Re Finucane and the principles set out by Lord Kerr in that case are common ground on these appeals. For a legitimate expectation to arise, an undertaking must be given which is clear, unambiguous and devoid of relevant qualification: see R v Inland Revenue Comrs, Ex p MFK Underwriting Agents Ltd [1990] 1 WLR 1545, 1569 per Bingham LJ; R (Davies) v Revenue and Customs Comrs [2011] UKSC 47; [2011] 1 WLR 2625, paras 28 and 29 per Lord Wilson of Culworth, delivering the judgment with which the majority of the court agreed, and para 70 per Lord Mance; Re Finucane, para 62; R (Friends of the Earth Ltd) v Heathrow Airport Ltd [2020] UKSC 52; [2021] PTSR 190 (sub nom R (Spurrier) v Secretary of State for Transport), para 106.

In our view, applying this standard, the written response did not give rise to a legitimate expectation enforceable at law. The Chief Constable spoke on behalf of the PSNI. As explained above, the PSNI is subject to a general duty in relation to the investigation of crime under section 32(1)(d) of the Police (Northern Ireland) Act 2000. Against that background, the written response could not be construed as saying anything other than that an investigation would take place in accordance with the PSNI’s general statutory duty. There was no clear and unambiguous statement to do anything other than that. In so far as it is relevant to deal with this, still less was there any clear and unambiguous statement that such investigation would comply with Convention standards in respect of an article 2/3 investigation or some equivalent requirement of independence, either at all or more particularly if neither those standards nor any such requirement are applicable in law. The written response said nothing at all about article 2/3 investigatory standards or what standards of independence might be applied.

Although Maguire J did not accept that any legitimate expectation arose in relation to the investigation undertaken by the PSNI, he nevertheless concluded that the decision of the PSNI on 17 October 2014 to take no further action was irrational and should be quashed. A majority of the Court of Appeal upheld that order. The Chief Constable contends that they were wrong to do so. In considering this challenge it is necessary to describe in more detail the investigation which led to the PSNI’s decision.

As mentioned in para 99 above, at the meeting of the Northern Ireland Policing Board on 3 July 2014 ACC Harris stated that the PSNI was aware from the RTÉ documentary of the existence of the Rees Memo and that the PSNI wished to source the original document and other materials in order to confirm or further clarify its contents.

Following the meeting, an investigating officer, who was a temporary worker with the HET, was sent to carry out research in the National Archives at Kew. His instructions were to verify the existence of the Rees Memo and any other documentation which may explain the context in which it was written and whether or not the use of torture had been authorised.

The Rees Memo was written by the then Home Secretary, Mr Merlyn Rees, to the Prime Minister on 31 March 1977 after reading a briefing note prepared by the UK Attorney General after a meeting with his Irish counterpart. The meeting was held in the hope of reaching a friendly settlement of the proceedings brought by Ireland against the United Kingdom which were then pending before the Strasbourg Court. In the briefing note the UK Attorney General had expressed the view that the British and Irish Governments were too far apart on the issue of prosecutions or disciplinary action against those who had carried out the interrogations of the Hooded Men for a settlement to be reached.

In the Rees Memo the Home Secretary commented upon the possibility of prosecutions or disciplinary action in the following terms:

The investigating officer produced two reports. His first report summarised his initial findings; the second was a more detailed report.

In his first report, dated 10 July 2014, the investigating officer explained that he had located two relevant files in the National Archives but not the Rees Memo, which appeared to have been returned to the original department and replaced by a dummy sheet. He said that the two files contained an abundance of information which added significant clarity to the Rees Memo and the situation at the time. In relation to the Rees Memo, he commented:

The investigating officer stated that he had taken the matter as far as he could within a limited time and suggested that further research on the subject “would involve the possible reading of no less than one hundred and fifty individual files and may never answer the question posed”. He suggested areas that could be focussed on if it were deemed necessary to conduct further research. They included limited research into Lord Carrington and Government papers during the 1971/72 period and applying for the Rees Memo.

In his second report, dated 18 August 2014, the investigating officer suggested that the reporter who presented the RTÉ documentary could not have inspected the Rees Memo in the National Archives, contrary to her assertion in a press report, because “this document has, in effect, been retained under secure conditions since 1978”. This was said to raise “interesting questions” about how the reporter had had access to the document. An affidavit made by the reporter in these proceedings states that the officer’s supposition was wrong and that she had indeed found the original Rees Memo in the National Archives in April 2014 (in a different file from the two files located by the investigating officer). The reporter has exhibited to her affidavit a copy of a photograph she took of the Rees Memo, which was shown on screen in the documentary.

The investigating officer went on to discuss some of the documents contained in the two files which he had inspected at the National Archives and annexed to his report copies of these documents. They included a letter dated 12 April 1977 from the Secretary of State for Defence, Mr Fred Mulley, to the Prime Minister, copied to the Home Secretary, in response to the Rees Memo. In this letter Mr Mulley expressed his strong agreement that there should be no question of either prosecuting or taking disciplinary action against those responsible for “deep interrogation” in 1971. He stated that he was, however, “a little surprised by the statement that our predecessors, and particularly Lord Carrington, took a ‘decision to use methods of torture in Northern Ireland’”. He said that “the published records do suggest that this is perhaps a rather hard way of putting the decision to use deep interrogation”. He referred to the statement by Lord Balniel in the House of Commons on 9 December 1971 that the use of the five techniques had been authorised by the Northern Ireland Government with the knowledge and concurrence of the UK Government, but that the rules then in force to govern the conduct of interrogation expressly prohibited torture, brutality and humiliating or degrading treatment. He pointed out that the UK Government had rested publicly on this statement and had not expressly accepted the subsequent finding of the European Commission that the five techniques constituted torture (although it had not contested the finding). He suggested that Mr Rees had “compressed the record rather too starkly and in a way which goes beyond any public position”.

This prompted a further memorandum from Mr Rees to the Prime Minister dated 18 April 1977 in which he stated:

On the basis of these (and two other) documents which he identified as placing the Rees Memo in context, the investigating officer wrote:

The investigating officer referred to a number of other documents which he had reviewed and also outlined the findings of the Compton and Parker Committees and materials relating to the Ireland v United Kingdom case in the Strasbourg Court. He concluded that no useful purpose would be achieved by taking the matter further and said of his research:

He also reported that he had found a “further RTÉ internet release” bringing attention to the programme broadcast on 4 June 2014. This showed a picture of what purported to be secret and confidential Government documents. He also annexed an “internet release” by Gerry Adams uploaded on 5 June 2014 which referred to research conducted by the Pat Finucane Centre. The investigating officer said he had no doubt that the RTÉ programme makers had used photographs of Government documents taken by researchers from the Pat Finucane Centre who had attended the National Archives to create “an entirely misleading pastiche” giving the false impression that they possessed the documents depicted on their webpage. He continued:

This appears to be a suggestion of some form of collusion involving the makers of the RTÉ programme and Gerry Adams (and possibly also researchers from the Pat Finucane Centre). It was a baseless speculation but gives further insight into the mind-set of the investigating officer.

In an email sent to Detective Chief Superintendent Hanna on 21 August 2014 attaching his second report, the investigating officer went so far as to state that it was “abundantly clear that the use of torture was never authorised at any level and that there is absolutely no documentation which suggests that it was”.

His report appears to have been considered at a senior level within the PSNI and, as explained above, on 17 October 2014 a decision was taken at the level of Assistant Chief Constable not to take the matter any further on the ground that the research carried out had identified no evidence which supported the allegation that the British Government authorised the use of torture in Northern Ireland. As Maguire J records (at para 135 of his judgment), it was this decision which appears to have triggered the present litigation.

Maguire J concluded that the decision to end the police investigation in reliance on the investigating officer’s report was seriously flawed, in particular because the inquiry was too narrowly based. Instead of investigating whether there was evidence of official authorisation by Lord Carrington or others of unlawful methods of deep interrogation which involved criminal assaults, the inquiry had erroneously treated the use of the word “torture” as its guiding light.

While a majority of the Court of Appeal affirmed the judge’s conclusion, Sir Donnell Deeny dissented. He described the decision to end the investigation (at para 31 of the judgment) as “a common sense decision” and said:

In performing their general duty, embodied in section 32(1)(d) of the 2000 Act, to bring offenders to justice, officers of the PSNI - like all police officers in the United Kingdom - have a very broad discretion. It is for them - and not any other public official or private individual - to judge whether an allegation that an offence has or may have been committed warrants investigation, if so what investigative steps to take, whether to continue or discontinue an investigation at any stage and whether sufficient evidence has been obtained to charge a suspect or refer a case to a prosecutor to consider whether a person should be charged with a criminal offence. In making such decisions, the police officers concerned are entitled to take into account a wide variety of factors and it is they - and not the courts - who have the constitutional responsibility and the practical competence to evaluate and decide what weight to give to those factors.

Given the nature of this discretion, a decision taken by a police officer to close a criminal investigation is seldom susceptible to legal challenge. Many cases confirm this. The leading authority is the decision of the House of Lords in R (Corner House Research) v Director of the Serious Fraud Office (JUSTICE intervening) [2008] UKHL 60; [2009] AC 756, paras 30-32. At the same time such a decision is not immune from judicial review, as the case law also confirms. Three recent examples of cases in which a Divisional Court in England and Wales quashed a decision not to prosecute on grounds of irrationality are: R (Purvis) v Director of Public Prosecutions [2018] EWHC 1844 (Admin); [2018] 4 WLR 118; R (Torpey) v Director of Public Prosecutions [2019] EWHC 1804 (Admin); and R (L) v Director of Public Prosecutions [2020] EWHC 1815 (Admin). A decision to end an investigation is similar in nature.

A challenge to the rationality of a decision made by a public official may take one of two forms. It may be argued that the overall outcome of the decision-making process lies beyond the outer bounds of reasonableness; or it may be argued that, even if the outcome could not be said to be irrational in itself, a critical step in the process of reasoning by which the decision was reached was irrational so that the decision must be taken afresh.

In the present case it could not be said that the decision of the PSNI made on 17 October 2014 not to take the matter further was, in itself, irrational. Given the passage of time since the ill-treatment of the Hooded Men in 1971, the fact that those who authorised the use of the five techniques were either dead or very elderly, our conclusion in this judgment that the new material publicised by the RTÉ documentary did not add to a significant extent to what was known already at the time of the previous investigation in 1978, and the many competing demands on police resources, a decision could rationally have been made not to undertake a further investigation. The decision to take no further action was not based, however, on any of the matters just mentioned. Its basis was stated to be that the investigation by members of the HET had not identified any evidence to support the allegation that the British Government authorised the use of torture in Northern Ireland.

In our view, the judge and the majority of the Court of Appeal were right to conclude that the reliance placed on the investigation and report of the HET officer was irrational and vitiated the decision based upon it. One reason was that given by the judge: the investigation was defective because it was unreasonably narrow. It is apparent from his final report that the investigating officer interpreted his task as being to see whether there were documents on file which used the word “torture” in conjunction with Lord Carrington or other members of the UK Government. To expect the authorisation of torture to be described so explicitly, however, is both unrealistic and a misdirection.

In referring to the findings of the Parker Committee, the investigating officer quoted the following statement from Lord Gardiner’s minority report:

In fact, the Rees Memo did use the word “torture”. The statement of the investigating officer quoted at para 234 above that Mr Rees “felt he unwittingly” used that word is not substantiated. What the subsequent correspondence showed is that Mr Rees acknowledged that it was preferable to avoid referring to the use of the five techniques as “torture” because to do so contradicted the UK Government’s publicly stated position. There is nothing in the later documents identified by the investigating officer to suggest that Mr Rees did not believe his original description to be accurate, as opposed to impolitic. The important point, however, is that the investigating officer addressed the wrong question. To conclude that, because he had not come across other documents which described the use of the five techniques as “torture”, it was “abundantly clear that the use of torture was never authorised at any level and that there is absolutely no documentation which suggests that it was” was jejune. Whether torture was authorised could not on any rational view depend on whether the word “torture” was used.

There is a second reason why the investigation report and the reliance placed on it were seriously flawed. Not only did the report address the wrong question; we also accept the submission made by Ms Quinlivan on behalf of Ms McKenna and adopted on behalf of Mr McGuigan that the report shows on its face a lack of fairness in the officer’s approach and a willingness to base conclusions on partisan assumptions rather than evidence.

This is epitomised by the approach taken to the RTÉ documentary which had brought the Rees Memo to public attention. As mentioned, the investigating officer felt able to make outspoken criticisms of the RTÉ documentary in his report. No reasonable or responsible person would describe a documentary made by a public service broadcaster in an official report as “a particularly biased, badly researched and misleading programme”, if at all, without having watched the programme and carefully considered its contents. Given that it was the source of the allegation that he was asked to investigate, watching the RTÉ documentary was anyway the obvious place to begin the investigation. The investigating officer referred in his report and annexed to it a printed copy of a page on the Northern Ireland News section of the BBC website which contains a hyperlink to the documentary. He even examined (though apparently only “at the very last moment”) a page on the RTÉ website which, from his description of it, is the webpage on which the documentary was (and still is) available to view online. Yet nowhere in his reports does the investigating officer state that he had actually watched the programme; and various statements, including his incorrect assumption that the Rees Memo was the only archive material featured in the documentary, indicate that he had not. Such a lack of professionalism and impartiality discredited the work done.

The upshot is that the decision taken by the PSNI on 17 October 2014 on the basis of such a seriously flawed report was irrational and should not be allowed to stand.

Counsel for the Chief Constable submitted that it was unnecessary to quash the PSNI’s decision, as the Chief Constable had already decided voluntarily to conduct a further investigation. The judge cannot be faulted, however, for making a quashing order which gave formal expression to his conclusion that the decision challenged lacked a rational basis and that an entirely fresh consideration of the issues was required.

In the case of Mr McGuigan and Ms McKenna, the Secretary of State for Northern Ireland seeks to raise a limitation argument based on section 7(5) of the HRA which was not addressed by the Court of Appeal. Section 7(5) provides that proceedings are to be brought before the end of “(a) the period of one year beginning with the date on which the act complained of took place; or (b) such longer period as the court or tribunal considers equitable having regard to all the circumstances.” In essence, the Secretary of State argues that by 1978 the UK Government had made it clear that it did not intend any criminal or disciplinary investigation. The release of the new material in the National Archives in Kew from 2003 onwards did not create a new time limit and, in any event, the proceedings were commenced many years after that material had become available.

In response, Mr Southey points out that the Secretary of State had not raised the issue of delay at the stage of obtaining permission for the judicial review, when it should have been. In any event he submits that the claims were commenced in time as Mr McGuigan first learned that there was relevant material in the National Archives in August 2013 but was not aware of the fresh material until the RTÉ documentary was broadcast on 4 June 2014.

Because we have reached the conclusion that, except in relation to the decision on 17 October 2014 (paras 223-252 above), the applications for judicial review by Mr McGuigan and Ms McKenna must fail, we do not need to address these arguments.

For the reasons set out above, we have reached the following conclusions:

Applying the genuine connection test in relation to the death of Ms Smyth, we conclude that the PSNI was not under an obligation to investigate Ms Smyth’s death under article 2 of the Convention (paras 169-178 above).

As the Brecknell test is not satisfied in relation to the Hooded Men, the PSNI was not under an obligation to investigate the authorisation of the ill-treatment of the Hooded Men under article 3 of the Convention (paras 119-132 above).

If article 2 of the Convention had applied, the Court of Appeal would have been entitled to conclude that the then proposed investigation into Ms Smyth’s death would not have been effective in the particular circumstances of that case because the Chief Constable of the PSNI had failed to explain to her family and the public, and when faced with the judicial review challenge, the court, how he proposed to secure the practical independence of that investigation (paras 201-212).

If article 3 of the Convention had applied, the Court of Appeal erred in concluding that an enquiry by the PSNI into the ill-treatment of Mr McGuigan and Mr McKenna would lack practical independence (paras 213-214).

The PSNI was not under an obligation at common law or under section 32 of the Police (Northern Ireland) Act 2000 equivalent to the obligations in articles 2 and 3 of the Convention (paras 215-217).

The Chief Constable did not create a legitimate expectation that the PSNI would undertake an investigation of the persons responsible for authorising the ill-treatment of Mr McGuigan and Mr McKenna (paras 218-222).

The decision taken on 17 October 2014 not to investigate the allegation in the Rees Memo was based on a seriously flawed report, was therefore irrational, and falls to be quashed (paras 223-252).

We would therefore allow the appeals by the Chief Constable for Northern Ireland, the Secretary of State for Northern Ireland and the Northern Ireland Department of Justice, recall the orders of the courts below, uphold the decision of Maguire J and the majority of the Court of Appeal to quash the decision taken on 17 October 2014, but otherwise dismiss the applications for judicial review.