United Kingdom House of Lords Decisions

HOUSE OF LORDS

SESSION 2007-08

[2008] UKHL 68

OPINIONS

OF THE LORDS OF APPEAL

FOR JUDGMENT IN THE CAUSE

R (on the application of JL) (Respondent) v Secretary of State for Justice (Appellant)

Appellate Committee

Lord Phillips of Worth Matravers

Lord Rodger of Earlsferry

Lord Walker of Gestingthorpe

Lord Brown of Eaton-under-Heywood

Lord Mance

Counsel

Appellants:

Nigel Giffin QC

Philip Sales QC

Cecilia Ivimy

(Instructed by Treasury Solicitors)

Respondents:

Ben Emmerson QC

Kristina Stern

(Instructed by Bindman & Partners)

Heather Williams QC

Raza Husain

(Instructed by the Equality and Human Rights Commission)

6 and 7 OCTOBER 2008

ON

WEDNESDAY 26 NOVEMBER 2008

HOUSE OF LORDS

OPINIONS OF THE LORDS OF APPEAL FOR JUDGMENT

IN THE CAUSE

R (on the application of JL) (Respondent) v Secretary of State for Justice (Appellant)

[2008] UKHL 68

LORD PHILLIPS OF WORTH MATRAVERS

My Lords,

1. This appeal raises the question of the nature of the investigation that must be carried out by the State whenever a prisoner in custody makes an attempt to commit suicide that nearly succeeds and which leaves him with serious injury.
2. The respondent, who has been referred to as JL, was born in Jamaica on 5 October 1981. He came to this country in May 2002 and, on 18 July 2002, was arrested and charged with possessing cocaine with intent to supply. He was remanded in custody to Feltham Young Offender Institution ("Feltham"). There, on 19 August 2002, he was found hanging from the bars of the window of his cell, having used a sheet to make a noose around his neck. He had stopped breathing, but was resuscitated. Deprivation of oxygen had resulted in serious brain damage. He has been left incompetent to conduct his own affairs.
3. The London Area Manager of the Prison Service initiated an investigation into what had occurred. He instructed Mr Sheikh, a retired Prison Governor acting as a Senior Investigating Officer to carry this out. Mr Sheikh submitted a written report to the Area Manager on 16 October 2002. No relative of JL or person representing his interests was involved in that investigation and Mr Sheikh's report was not published or disclosed until 26 January 2005. It was then disclosed to the Official Solicitor who, on behalf of JL, had written a letter before action to the Treasury Solicitor. Mr Sheikh's report summarised the facts that he had ascertained and set out conclusions, which included findings that the treatment and care provided to JL at Feltham was in line with the national as well as the local requirements and that the staff at Feltham had provided the "necessary required care and attention and support".
4. JL's claim for judicial review was issued on 21 September 2005. It was heard by Langstaff J [2006] EWHC 2558 (Admin). JL contended that article 2 of the European Convention on Human Rights ("article 2") imposed a duty on the Secretary of State to carry out an independent investigation into his attempted suicide, that this investigation had to satisfy a number of criteria, that Mr Sheikh's investigation did not satisfy those criteria but that it disclosed facts that raised the possibility that Feltham had failed to discharge the duty to safeguard JL's life imposed on the State by article 2. He sought a mandatory order requiring the Secretary of State to carry out an investigation that satisfied article 2, reserving the right, in the light of the findings of this investigation, to pursue a further claim for breach of the obligation to safeguard his life.

The issues

5. It is important to identify at the outset the issue that Langstaff J was asked to resolve and the premises upon which he was asked to do so. These appear from the following section at the beginning of his judgment:

"6. The issue for my determination is thus whether in the circumstances of the present case the defendant was, or should be, obliged to conduct an enquiry satisfying the minimum standards required by article 2.

7. Both Ms Stern for the claimant, and Mr Eadie, for the defendant, say that this particular issue has not been addressed in earlier cases. Those cases have dealt with the content of an investigation, it being accepted that article 2 required such an investigation to be held: they asked me to address whether the threshold requiring any 'article 2 investigation' has been crossed.

8. To explain the way in which this question arises, and how it is to be answered in the present case, it will be necessary to set out the relevant law, then the relevant facts, before addressing the appropriate answer. However, I must first observe that the importance of this case to the parties is a practical one. What the claimant seeks, and the defendant refuses is an enquiry by a person (or body) institutionally and practically independent from those implicated in the circumstances which led to the life-threatening injury, who (or which) takes steps to secure all relevant evidence in relation to them, open to public scrutiny, and involving the next of kin. So far as the investigation thus far conducted is concerned, it plainly did not have either of the latter two qualities, its independence is not clearly established, and the claimant makes points of detail which indicate the enquiry did not secure (and certainly did not reveal) some relevant evidence in relation to the near death. However, I am not asked to determine in these proceedings precisely what article 2 (if it applies) requires to be done in the present circumstances by way of enquiry. I am asked simply to decide whether it does necessitate an enquiry, it being assumed by the parties that any such enquiry must necessarily have the characteristics which I have identified, amongst others."

6. Langstaff J was thus asked to proceed on the premise (1) that where article 2 requires an investigation this will necessarily have certain specific characteristics and (2) that it is possible to specify the circumstances in which such an investigation will be required. Langstaff J was asked to restrict himself to identifying those circumstances and ruling whether they applied in the present case. He expressed reservations as to whether it was possible to draw a clear line between the question of when the requirement to hold an investigation is triggered and the question of the content of the investigation, but nonetheless set out, as requested, to answer the former question.
7. After considering both Strasbourg and domestic authority Langstaff J formed the following general conclusions. Article 2 requires an investigation where a State or its agents potentially bear responsibility for loss of life. An unexpected death or life-threatening injury in custody will usually, although not always, require an investigation sufficient to satisfy article 2 obligations.
8. Langstaff J then considered the facts that had been found by Mr Sheikh and concluded that, on the basis of these, it was arguable that the State was responsible for the injuries sustained by JL. There was thus an obligation to hold an investigation that complied with article 2. He granted a declaration to that effect but stressed that he had not been asked to determine what precise form the investigation should take in order discharge this obligation.
9. On 24 July 2007 the Court of Appeal dismissed the Secretary of State's appeal [2007] EWCA Civ 767. In giving the leading judgment Waller LJ also expressed anxiety about making an attempt at a definition that covered all cases of suicide or near-suicide in custody. In the event, however, he felt able to advance certain principles. He said:

"I am clear that the simple fact of a death or serious injury of a person in custody gives rise to an obligation on the State to conduct the enhanced type of investigation. The extent of the investigation will depend on the circumstances…As regards the nature of the investigation it seems to me that a death or near death in custody ipso facto means that the State must commence an investigation by a person independent of those implicated in the facts. The extent to which there must then be some further inquiry in the nature of a public hearing in which the next of kin or injured person can play a part will depend on the circumstances. In the case of a death there will be an inquest, and the coroner may have to decide whether the circumstances are such as to require something [further]. In cases of serious injury the nature of the further inquiry necessary will depend on the facts as discovered by the independent investigator." (paras 32, 33)

10. Waller LJ went on to indicate that unless from the independent investigation it is "plain that the State or its agents can bear no responsibility" a further investigation would be required with the ingredients identified by the Court of Appeal in R(D) v Secretary of State for the Home Department (INQUEST intervening) [2006] EWCA Civ 143; [2006] 3 All ER 946. I shall describe an investigation with these ingredients as a "D type investigation".
11. Waller LJ held that the requirement for an initial independent investigation had not been satisfied in this case, if only because Mr Sheikh did not have the degree of independence required. He went on to hold that, if one had regard to the facts found by Mr Sheikh, these led to the conclusion that a further D type investigation was necessary.
12. The respondent supports the findings of Waller LJ as to the nature of the investigation required by article 2 where a near-suicide occurs in prison custody. So too does the intervener, the Equality and Human Rights Commission.
13. While the Secretary of State does not accept the findings of the courts below, he no longer seeks to avoid a D type investigation into JL's near-suicide and preparations for this are in hand. He has sought and obtained permission to appeal because he is concerned by the resource implications if the principles identified by Waller LJ are applied generally. His submissions, as advanced by Mr Nigel Giffin QC, can be summarised as follows:

i)  The same principles apply where a suicide or a near-suicide takes place in prison. A near-suicide is one that nearly succeeds and leaves the prisoner with serious injuries.

ii)  Where a suicide or a near-suicide takes place in prison the relevant facts must first be considered by the prison authorities in order to determine whether there is an arguable case that there has been a breach of the substantive duty imposed on the State by article 2 to protect life.

iii)  If there is no such arguable case no further investigation is required.

iv)  If it is arguable that there has been a breach of the duty imposed by article 2 to protect life, there must then be an independent investigation. The nature of that investigation will depend upon the particular facts. There will not necessarily be the need for a D type investigation. In all but exceptional cases an independent investigation into the circumstances in which the suicide or near-suicide took place, which is prompt and effective and involves to an appropriate extent the relatives of the prisoner in the case of suicide, or the prisoner and his representatives in the case of a near-suicide, with the results made known to them, will be sufficient to comply with article 2.

14. It is thus common ground between the parties that, where a suicide or near-suicide takes place in prison, there must be an initial investigation of the facts and that this may give rise to the requirement for a further investigation. The issues between the parties are as follows:

i)  Must the initial investigation be independent or can it be carried out by the prison authorities themselves?

ii)  Must a further investigation be held whenever it is not plain from the initial investigation that the State or its agents bear no responsibility for the near-suicide or only where the initial investigation demonstrates that there is an arguable case that the State was at fault?

iii)  Where a further investigation is required, must this necessarily be a D type investigation.

15. My Lords, I share the reservations of the courts below as to how far it is possible to give definitive guidance that will apply to every case of near-suicide in prison. The resource implications of the issues are, however, considerable and I believe that it is possible to identify certain principles that will normally apply to such cases. I propose to confine my remarks to the situation where a prisoner's attempt at suicide (i) comes close to success and (ii) leaves the prisoner with the possibility of serious long term injury. Thus I shall be considering the case where there is a victim whose interests have to be considered.

The nature of a D type investigation

16. The facts in D bear similarity to those of the present case in that the applicant sustained severe brain damage as the result of a near-successful attempt to hang himself in his cell. As in this case a Senior Investigating Officer in the Prison Service carried out an investigation. The Secretary of State had accepted that a further investigation was required and had instructed Mr Stephen Shaw, the Prisons and Probation Ombudsman to conduct this. The applicant contended, however, that the procedure that it was proposed that Mr Shaw should follow would not satisfy the requirements of article 2 and brought judicial review proceedings in which he claimed "a full and effective investigation into the circumstances of" his attempted suicide.
17. D was dissatisfied with the aspects of the proposed investigation which included the following:

i)  It would not be held in public, although the report would be made public when completed;

ii)  Mr Shaw would not have the power to compel witnesses;

iii)  D's representatives would not be able to attend the questioning of witnesses or to require questions to be put to witnesses.

Munby J granted the application and made the following declaration:

"(i) The inquiry must be held in public, save where there are convention-compatible reasons to hear the evidence of a particular witness, or other parts of the hearing, in private. (ii) The inquiry must be capable of exercising a power to compel the attendance of witnesses, if this becomes necessary for the inquiry to be effective, and this power must be capable of being exercised without undue delay. (iii) Subject to (i) above, D's representative must be able to attend at public hearings of the inquiry and put questions to witnesses in person. (iv) D's representative must be given reasonable access to all relevant evidence in advance. (v) Adequate funding for D's representative must be made available"

The Court of Appeal upheld Munby J's order, save that they held that the judge went too far in holding that D's representatives should be entitled to cross-examine witnesses at the public inquiry. It sufficed that they should be entitled to attend the inquiry and make representations as to the matters about which the witnesses should be examined.

The regime where there is a suicide in prison

18. Where a death occurs in prison section 8(3) of the Coroners Act 1988 requires the coroner to conduct an inquest with a jury. It is also the practice of the Prisons and Probation Ombudsman for England and Wales to carry out an investigation into the death. The Coroner will consider his report in order to assist him to decide whether there are issues in relation to the conduct of the prison authorities that he will wish to be covered by the jury's verdict in accordance with the procedure laid down by your Lordship's House in R (Middleton) v West Somerset Coroner [2004] UKHL 10; [2004] 2 AC 182. The Coroner has the power to summon witnesses. Under the Coroners Rules 1984 notice of the inquest must be given to the next of kin who are entitled to examine witnesses. Inquests are open to the public and the verdict is given in public.
19. It is common ground that this regime satisfies the obligations imposed by article 2 where a suicide takes place in prison. It has also been common ground that article 2 imposes no more stringent obligations in relation to the investigation of a near-suicide than it imposes in relation to the investigation of a successful suicide. No one has submitted that a near-suicide necessarily requires an investigation that has all the attributes of an inquest. It follows that it is implicit in the submissions made to us, and Mr Giffin confirmed this, that the investigation that takes place in the case of a suicide will, in some cases at least, do more than is necessary to satisfy the requirements of article 2.
20. I am not persuaded that it is correct to proceed on the premise that the requirements of article 2 in respect of investigation are identical in the case of a suicide and a near-suicide. In this jurisdiction the law has always treated death as a matter of particularly grave concern. There is, I believe, justification for a regime that imposes requirements as to investigation where a death occurs that do not apply automatically in other circumstances. At all events it is not helpful to approach the requirements of article 2 in relation to a near-suicide by reference to the existing domestic requirements where an actual suicide occurs.

The reasons why article 2 requires an investigation

21. It is fundamental to the Secretary of State's case that the reason why article 2 requires an investigation into a near-suicide in prison is to secure the accountability of agents of the State in respect of possible breaches of the substantive obligations imposed by that article. Thus, so Mr Giffin argues, if the State can show that there is no arguable case of such a breach, there is no requirement for an investigation. These submissions receive some support from the decided cases, both at Strasbourg and in this jurisdiction, and it is time to consider these insofar as they bear on this question.
22. Article 2(1) provides:

"Everyone's right to life shall be protected by law. No-one shall be deprived of his life intentionally…"

This article imposes (1) a duty to refrain from the intentional and unlawful taking of life and (2) an obligation to take positive steps to protect the right to life of those living within the jurisdiction of the State - see Osman v United Kingdom (2000) 29 EHRR 245. In McCann v United Kingdom (1995) 21 EHRR 97, the Strasbourg Court recognised for the first time that article 2 imposed by implication a third obligation, namely a duty to carry out an effective official investigation, held in that case to apply "when individuals have been killed as a result of the use of force by, inter alios, agents of the State" (para 161).

23. Since then the Strasbourg Court has, on numerous occasions, considered the obligation imposed by article 2 to hold an investigation, which has sometimes been described as a procedural obligation. It has, however, always done so in circumstances where the applicants' primary complaint has been of a substantive breach of the Convention. The allegation of a failure to investigate has always been an ancillary allegation of the alleged substantive breach.
24. In Jordan v United Kingdom (2001) 37 EHRR 52 the applicant complained that his son had been unjustifiably shot and killed by a police officer. He also claimed that there had been no effective investigation into or redress for his death. In considering the latter claim, which succeeded, the Court said this about the purpose of such an investigation:

"The essential purpose of such investigation is to secure the effective implementation of the domestic laws which protect the right to life and, in those cases involving State agents or bodies, to ensure their accountability for deaths occurring under their responsibility." (para 105)

The words that I have emphasised demonstrate that the objects of an investigation go beyond ensuring accountability of State agents.

25. In Edwards v United Kingdom (2002) 35 EHRR 487 the applicant, whose son had been killed by a fellow prisoner, contended that the authorities had failed to protect his life and had further failed to carry out an investigation into his death that was effective. The Court used the same language in this context to describe the objects of the necessary investigation.
26. The duty to investigate imposed by article 2 can arise even where there is no question of any direct involvement of a State agent. In Menson v United Kingdom (2003) 37 EHRR CD220 a black man was killed as a result of being set on fire by assailants during a racist attack. The Court held that in such circumstances there was an obligation for "some form of effective official investigation", adding:

"Where death results, as in Michael Menson's case, the investigation assumes even greater importance, having regard to the fact that the essential purpose of such an investigation is to secure the effective implementation of the domestic laws which protect the right to life."

It seems to me that the obligation to have an investigation in circumstances such as these is not so much a secondary procedural obligation but rather part of the positive obligation, also noted by the Court, to have in place

"effective criminal law provisions to deter the commission of offences against the person, backed up by law enforcement machinery for the prevention, suppression and punishment of breaches of such provisions."

27. Menson was cited by Lord Bingham of Cornhill in R (Amin) v Secretary of State for the Home Department [2003] UKHL 51; [2004] 1 AC 653. He expressed the following conclusion at para 31:

"The state's duty to investigate is secondary to the duties not to take life unlawfully and to protect life, in the sense that it only arises where a death has occurred or life-threatening injuries have occurred….The purposes of such an investigation are clear: to ensure so far as possible that the full facts are brought to light; that culpable and discreditable conduct is exposed and brought to public notice; that suspicion of deliberate wrongdoing (if unjustified) is allayed; that dangerous practices and procedures are rectified; and that those who have lost their relative may at least have the satisfaction of knowing that lessons learned from his death may save the lives of others".

28. Lord Bingham returned to this theme in R (Gentle) v Prime Minister [2008] UKHL 20; [2008] 2 WLR 879. He first referred to the substantive obligations imposed by article 2: the obligation not to take life without justification and the obligation to take measures to protect life. Lord Bingham then proceeded to consider the nature of what he described as a procedural obligation that supplemented the substantive obligations

"'to initiate an effective public investigation by an independent official body into any death occurring in circumstances in which it appears that one or other of the foregoing substantive obligations has been, or may have been, violated and it appears that agents of the state are, or may be, in some way implicated' [see Middleton para 3]…This procedural duty does not derive from the express terms of article 2, but was no doubt implied in order to make sure that the substantive right was effective in practice…It is clear … that the procedural obligation under article 2 is parasitic upon the existence of the substantive right, and cannot exist independently. Thus to make good their procedural right to the inquiry they seek the claimants must show, as they accept, at least an arguable case that the substantive right arises on the facts…article 2 not only prohibits the unjustified taking of life by the state and its agents, but also requires a framework of laws, precautions, procedures and means of enforcement which will, to the greatest extent reasonably practicable, protect life. In either case the question whether the state unjustifiably took life or failed to protect it will arise in respect of a particular deceased person."

29. These observations were directed to the obligation imposed by article 2 to hold a public investigation. They should not be read as suggesting that the State never has a duty to carry out an investigation into a life-threatening incident unless there is reason to believe that it may demonstrate that State agents have failed to perform the substantive obligations imposed by article 2. Still less do they support an argument that the only object of such an investigation is to ascertain whether or not State agents have been in breach of duty. The investigation will be concerned to see what lessons can be learned for the future, whether or not there has been fault in the particular case. In Amin Lord Slynn of Hadley remarked:

"The result of 'an incident waiting to happen' may just as much as an actual killing require detailed and profound investigation, though in some cases the procedure to be adopted may be justifiably different." (para 41).

Many activities today carry with them so great a risk to life that the duty of the State to put in place "a framework of laws, procedures and means of enforcement" will include a duty to require investigations of one form or another to be carried out in the event of a mishap, even if this does not actually result in loss of life. The investigations will not necessarily be independent or held in public. Requirements for such investigations can readily be found in the regulations governing carriage by rail, sea and air and in regulations governing health and safety at work. The primary purpose of such investigations is to learn lessons for the future.

30. It was for this reason that, in the present case, Langstaff J rejected the Secretary of State's submission that the function of the investigative obligation imposed by article 2 was simply to secure the accountability of those agents of the State who might be said to be at fault. He said:

"So far as accountability is concerned, where a person is compelled by the coercive power of the State to be and remain in prison there is a duty to account for his physical integrity which rests not simply upon the civil or criminal law, nor just upon State agents, but upon the State itself. Where the complaint may be made that a person knew or ought to have known of a potential risk to life, it is easy to hold him or her accountable. Where, however, the system itself holds risks which are not apparent (and which may be revealed for the first time by a life threatening injury), no one person may be held accountable. However, the lessons of history must be learned. The State needs not simply to hold individuals accountable, but to learn of potential systemic problems."

31. The duty to investigate imposed by article 2 covers a very wide spectrum. Different circumstances will trigger the need for different types of investigation with different characteristics. The Strasbourg court has emphasised the need for flexibility and the fact that it is for the individual State to decide how to give effect to the positive obligations imposed by article 2. In this jurisdiction every death calls for a certificate of the cause of death from a doctor or a coroner. In specified circumstances an inquest is required. These include where there is reasonable cause to suspect that the deceased died a violent or unnatural death, that the death was sudden and the cause unknown, or where the death occurred in prison. In further specified circumstances the inquest must be conducted with a jury. I have already described the nature of such an inquest where the death in prison was caused by suicide. Thus death requires a spectrum of different types of investigation, depending upon the circumstances of the particular case. This regime is part of the way in which the United Kingdom gives effect to the obligations of article 2. The regime makes no provision for a near-death from suicide. This appeal raises the question of how such an event is to be accommodated within the spectrum.

The nature of the initial investigation

32. It is common ground, and obviously correct, that where a prisoner attempts to commit suicide in prison, nearly succeeds and causes himself serious injury in the attempt, some investigation of the surrounding facts is necessary. The Secretary of State contends that the initial investigation can be internal and that, unless it shows that there is an arguable case that the prison authorities were at fault in permitting the suicide attempt to occur, there will be no need for any further investigation. JL and the intervener contend that article 2 requires that, from the outset, the investigation must be carried out by a person independent of the prison authorities.
33. Waller LJ was in no doubt that a near-death in custody ipso facto meant that the State was obliged to conduct an "enhanced type of investigation" and that this called for the commencement of the investigation "by a person independent of those implicated in the facts". His conclusion was based in part upon an analysis of the Strasbourg and domestic authorities, to a number of which I have already referred. So far as the former were concerned, he referred to:

i)  McCann where the Court said that there must be some form of effective investigation when individuals have been killed as a result of the use of force by inter alios agents of the State;

ii)  Jordan where the Court repeated this, but without reference to killing by agents of the State;

iii)  Edwards where the obligation was said to arise because the deceased was "a prisoner under the care and responsibility of the authorities" when he was killed by acts of violence by another prisoner;

iv)  Menson, where it was not suggested that agents of the State were directly at fault but where the obligation arose because there was "reason to believe that an individual has sustained life-threatening injuries in suspicious circumstances";

iv)  Salman v Turkey (2000) 34 EHRR 425, where the Court said in relation to a death in police custody:

"…the mere fact that the authorities were informed of the death in custody of Agit Salman gave rise ipso facto to an obligation under article 2 to carry out an effective investigation into the circumstances surrounding the death."

34. So far as domestic authorities are concerned, Waller LJ relied particularly on Amin as establishing that the mere fact of a death in custody gave rise to an obligation on the part of the State to account for that death. He observed that the decision of your Lordships' House in that case gave no support for the proposition that there was an additional requirement for an arguable case of fault on the part of the State before the obligation to put in train an investigation arose.
35. Waller LJ considered that these decisions were applicable to a near-suicide of a prisoner in custody, so as to impose an automatic requirement to initiate an "enhanced investigation" into such an event. The Strasbourg Court has repeatedly made plain the essential ingredients of such an investigation. In Amin at para 22 Lord Bingham cited those ingredients as spelt out by the Strasbourg Court in Edwards. I can summarise them as follows:

i)  The investigation must be initiated by the State itself;

ii)  The investigation must be prompt and carried out with reasonable expedition;

iii)  The investigation must be effective;

iv)  The investigation must be carried out by a person who is independent of those implicated in the events being investigated;

v)  There must be a sufficient element of public scrutiny of the investigation or its results;

vi)  The next of kin of the victim must be involved in the procedure to the extent necessary to safeguard his or her legitimate interests.

36. Waller LJ had regard to the fourth of these ingredients when he postulated that the commencement of the investigation had to be carried out by a person independent of those implicated in the facts. He then moved to consider the circumstances in which there would be a need for "some further inquiry in the nature of a public hearing in which the next of kin or injured party can play a part". Earlier he had spoken of the possibility of there being a series of triggers of the duty to investigate.
37. My conclusions in relation to the initial investigation that must follow a near-suicide in custody are as follows:

i)  A near-suicide of a prisoner in custody that leaves the prisoner with the possibility of a serious long term injury automatically triggers an obligation on the State under article 2 to institute an enhanced investigation.

ii)  That obligation cannot be discharged, or removed, by an internal investigation of the facts.

iii)  In some circumstances an initial investigation will satisfy the requirements of article 2. In others a further investigation will be necessary, which may well require to be a D type public inquiry.

38. The following are my reasons for concluding that a near-suicide that results in serious injury triggers the requirement for an enhanced investigation. The positive duty on the State to protect life has particular application in relation to the risk of suicide by prisoners. Those who are imprisoned pose a particularly high suicide risk. Grim statistics of suicide in prison were quoted by Lord Bingham in Middleton at para 5. Your Lordships were informed by the Secretary of State that it is estimated that there are up to about 150 cases each year where resuscitation may have been necessary following serious suicide attempts which do not in fact succeed. In 2006 there were 874 incidents of self-harm which were recorded as involving one or more of the following factors: hanging, resuscitation and hospitalisation with life-support. These unhappy statistics reflect the mental stresses that result from being placed in custody and the fact that the majority of those who are imprisoned suffer from some form of mental disorder or disability.
39. Article 2 places on the prison authorities a positive duty to take reasonable care for the safety of those in custody and, in particular, to take reasonable steps to ensure that they do not commit suicide - see Reeves v Commissioner of Police of the Metropolis [2000] 1 AC 360. Discharge of this duty requires the putting in place of systemic precautions against suicide in prison. A description of this country's response to this need was given by Lord Hope of Craighead in R (Sacker) v West Yorkshire Coroner [2004] UKHL 11; [2004] 1WLR 796 at paras 4 to 12.
40. Where, despite these precautions, a suicide takes place an investigation is required for the reasons given in that case by Lord Hope at para 11:

"There is a high level of awareness, and much effort has been devoted to improving the system for the prevention of suicides. But every time one occurs in a prison the effectiveness of the system is called into question. So all the facts surrounding every suicide must be thoroughly, impartially and carefully investigated. The purpose of the investigation is to open up the circumstances of the death to public scrutiny. This ensures that those who were at fault will be made accountable for their actions. But it also has a vital part to play in the correction of mistakes and the search for improvements. There must be a rigorous examination in public of the operation at every level of the systems and procedures which are designed to prevent self-harm and to save lives."

41. A suicide attempt that does not succeed but that results in serious injury is a matter of public concern, albeit not usually of such serious concern as where the attempt succeeds. The reasons why a successful suicide requires an investigation also apply in this situation. They require an enhanced investigation, albeit not necessarily a public inquiry.
42. There are two reasons why an internal investigation that does not disclose an arguable case of fault on the part of State authorities will not preclude the need for an enhanced investigation. The first is that, as I have shown, the object of the investigation goes beyond determination of whether or not the State authorities were at fault. The second reason is as follows. The scope of an investigation into a near-suicide will normally be considerable. It will involve consideration of what was known, or should have been known, of the risk that the prisoner might commit suicide and an investigation of whether the prison procedures against suicide risk were appropriate and properly implemented. One object of the investigation will be to call on the prison service to account for something that appears to have gone seriously wrong. If the investigation is to be and to be seen to be impartial, it is essential that it should be carried out by a person who is independent of those involved.

The need for a further investigation

43. Whether or not a further investigation is necessary will depend not merely upon whether the initial investigation is independent, but upon whether it satisfies all the requirements of an enhanced investigation. The initial investigation should be prompt, so that the facts are investigated while the evidence is still fresh and the material witnesses are readily available to be questioned. If all such witnesses give their evidence readily, the course of events appears clear and the circumstances in which the attempted suicide took place are shown to involve neither a possible defect in the system for preventing suicide nor a possible shortcoming on the part of any one in operating that system, the initial investigation may satisfy the requirement of efficacy without the need for further inquiry. In that event, if the prisoner who attempted to commit suicide or his representatives are appropriately involved in the investigation and a report of the investigation is published, the other requirements of an enhanced investigation may be satisfied.
44. It is desirable that the initial investigation should be sufficiently rigorous to satisfy the requirements of an enhanced investigation where this is possible. A D type investigation will necessarily be more protracted and expensive. In Edwards, where the Court held that the requirements of expedition were satisfied, the death in prison occurred in November 1994, the decision to hold an Inquiry was taken in July 1995, proceedings opened in May 1996 and the report of the Inquiry was published in June 1998.
45. There will, however, be circumstances in which the initial investigation will not be adequate to satisfy article 2 and where a D type investigation is required. The public interest may itself require this. In Edwards the Court remarked that the manner in which the deceased lost his life was so horrendous that the public interest in the issues thrown up called for the widest exposure possible. The need for an efficacious investigation may require this. If witnesses refuse to give evidence then it may be necessary to request the Minister to convert the investigation into a public inquiry under the Inquiries Act 2005. Where the initial investigation discloses serious conflicts of evidence a D type investigation may be called for. There will be other circumstances in which the person carrying out the initial investigation will decide to recommend a D type investigation. It is also possible to conceive of circumstances in which the independent investigator identifies some area that requires further investigation without the need for a full D type public inquiry. I do not believe that it would be appropriate for your Lordships to attempt to prescribe the circumstances in which a D type investigation will be necessary to satisfy article 2.

Deficiencies

46. As one would expect, the Prison Service has put in place in the form of Prison Service Orders directions in relation to the investigations that should be carried out into untoward incidents. PSO 1300, issued on 19 June 1993 and updated on 25 July 2005 provides:

"1.6 Formal Investigation.

1.6.1 A formal investigation will be necessary if, from the findings of a simple investigation or from the outset, it appears that any of the following apply:-

General

•  The incident has major consequences (disorder, damage, injury etc).

•  There was serious harm to any person.

. . .

Specialist

In addition to the above:

•  Where a formal investigation is made mandatory by another instruction e g PSO 1301 Investigating Deaths in Custody.

1.6.3 Normally the investigation will be carried out by a local team, except where the Commissioning Authority judges that a greater level of independence is needed. When an incident prompts high levels of public concern or there is potential to cause embarrassment to Ministers or the Service, an investigation might well need to be independent of the establishment or group in which it is conducted. The Commissioning Authority may also need to bring in outside investigators where specialist skills or team members are not available locally.

1.6.4 In exceptional circumstances, such as major and/or simultaneous incidents, an independent external investigation, from outside the Prison Service may be commissioned by Ministers or the Director General. Such investigations are beyond the scope of this Order."

This direction requires amending so as to require, at the least, an independent investigation in the case of a near-suicide that results in serious injury.

47. The investigation into the case of D, which followed the decision of the Court of Appeal, was conducted by Mr Stephen Shaw, the Prisons and Probation Ombudsman for England and Wales. His Report, published in May of this year, included the following recommendation:

"Until such time as the jurisprudence is clarified, I recommend that the Prison Service requires all prisons to carry out investigations into attempted suicides, incidents of serious self-harm and other near deaths. These should include an independent element, and engage the person who has been harmed and/or their family."

This recommendation applies to more than near-suicides resulting in serious injury and probably to circumstances that would not engage article 2, but I consider that it makes good sense nonetheless.

48. Mr Sheikh's investigation accorded with PSO 1300 but it did not satisfy the requirements of article 2. Mr Sheikh appears to have held the position of Senior Investigating Officer in the London Area Litigation Unit of the Prison Service and did not have the requisite independence. Neither JL nor anyone representing his interests took any part in the investigation or was even made aware that it was taking or had taken place. Mr Sheikh's report was not published.
49. Langstaff J and the Court of Appeal identified aspects of Mr Sheikh's report that called for further investigation and that raised the possibility that there had been shortcomings in the way in which the prison staff had carried out their obligations to safeguard JL against suicide risk. A self harm risk form F2052SH was opened in respect of JL but was subsequently closed again. There is concern that not all the appropriate persons, including the chaplaincy, were consulted prior to its closure. On one occasion a noose made out of bed sheets was found in JL's cell. On another he was seen to have a short wide piece of bed sheet around his neck. Neither of these incidents was recorded. The implications of some of the comments made by JL about his state of mind should arguably have received more detailed consideration.
50. After the decision of the Court of Appeal the Secretary of State decided to carry out a D type investigation without awaiting the result of this appeal. I consider that he was right to do so. There was no valid ground for challenging the order made by Langstaff J and confirmed by the Court of Appeal.
51. For these reasons I would dismiss this appeal.

LORD RODGER OF EARLSFERRY

My Lords,

52. While in custody in Feltham Young Offender Institution, JL attempted to commit suicide by hanging himself. He was left with permanent and serious brain damage which means that he is unable to conduct his own affairs. Unfortunately, what happened to him is only part of a larger pattern of suicides, near-suicides and incidents of self-harm among people held in custody. The occurrence of these incidents does not, of itself, point to any failure on the part of those having custody of the prisoners: there can be all kinds of motives for what the prisoners do and all kinds of reasons for the incidence of such behaviour being higher among prisoners than among the corresponding population at large. That said, there is no room for complacency. The prison authorities are, of course, well aware that, in addition to any purely humanitarian considerations, they have legal obligations to try to prevent the individuals in their custody coming to harm. These obligations derive from at least two sources.
53. In his classic speech in Tomlinson v Congleton Borough Council [2004] 1 AC 46 Lord Hoffmann expounded a ruggedly individualist version of the common law of tort, which would not impose on a local authority a duty of care to fit young men to prevent them from taking risks which could, and in Mr Tomlinson's case did, result in catastrophic injury. Some years before, in Reeves v Commissioner of Police of the Metropolis [2000] 1 AC 360, a case where a man aged 29 hanged himself in a police cell, Lord Hoffmann had indeed affirmed, at p 368B-D, that the intuition of the common law was sound when it distinguished "between protecting people against harm caused to them by third parties and protecting them against harm which they inflict upon themselves." Nevertheless he held, at pp 368H-369A, that the Commissioner was correct to accept that the police owed those in their custody an "unusual" duty of care to prevent them from committing suicide. The House did not need to explore exactly what this duty entailed, but Lord Hoffmann gave some indication of its contours when he said, at p 369A-D:

"Autonomy means that every individual is sovereign over himself and cannot be denied the right to certain kinds of behaviour, even if intended to cause his own death. On this principle, if Mr Lynch had decided to go on hunger strike, the police would not have been entitled to administer forcible feeding. But autonomy does not mean that he would have been entitled to demand to be given poison, or that the police would not have been entitled to control his environment in non-invasive ways calculated to make suicide more difficult. If this would not infringe the principle of autonomy, it cannot be infringed by the police being under a duty to take such steps."

54. In the present case the House is concerned with the obligations on the Prison Service which derive from article 2 of the European Convention. Like the common law, Convention law draws a distinction between prisoners and individuals who are at liberty: Pretty v United Kingdom (2002) 35 EHRR 1, 29-30, para 41. In Edwards v United Kingdom (2002) 35 EHRR 487, where the applicants' son had died following an attack by a fellow prisoner, the European Court noted, at p 507, para 56, that "persons in custody are in a vulnerable position and that the authorities are under a duty to protect them." The Court went on to say, at p 511, para 69, that State agents or bodies must be held accountable "for deaths occurring under their responsibility". As is its wont, the Court has reiterated this line of thinking, in much the same words, in numerous other cases, the most recent being Renolde v France (Application No 5608/05, BAILII: [2008] ECHR 1085 ) 16 October 2008, para 83.
55. At its most basic, article 2 requires the prison authorities not to harm those in their custody. But it goes further. In particular, the European Court has recognised that they are under various positive obligations to protect the lives of prisoners. While the authorities are not obliged to regard all prisoners as potential suicide risks (Younger v United Kingdom (2003) 36 EHRR CD252 , 268), they do have to proceed on the footing that prisoners as a class present a particular risk of suicide. Indeed in Tanribilir v Turkey (Application No 21422/93, BAILII: [2000] ECHR 612 ) 16 November 2000, the European Court observed, at para 74, that, by its very nature, any deprivation of physical liberty carries with it a risk of suicide, against which the authorities must take general precautions. See also Akdogdu v Turkey (Application No 46747/99 BAILII: [2005] ECHR 932 ) 18 October 2005, para 47. For this reason the prison authorities must take general (i e systemic) measures and precautions to diminish the opportunities for prisoners to harm themselves, without, however, infringing their personal autonomy: Keenan v United Kingdom (2001) 33 EHRR 913, 958, para 91.
56. The authorities' obligation is not, of course, absolute: it is not indeed to be interpreted as imposing an impossible or disproportionate burden, bearing in mind, among other factors, the unpredictability of human conduct and the operational choices which must be made in terms of priorities and resources: Keenan v United Kingdom 33 EHRR 913, 957-958, para 89; Akdogdu v Turkey, para 45; Renolde v France, para 82.
57. In addition to the obligation to take these general measures, the prison authorities are also under an "operational" obligation, in certain well-defined circumstances, to protect a particular prisoner from the risk that he will kill himself: Keenan v United Kingdom 33 EHRR 913, 958, paras 90-92; Tanribilir v Turkey, para 70; Akdogdu v Turkey, para 44. This obligation arises when the authorities know, or ought to know, that there is a real and immediate risk that the prisoner in question will commit suicide: Keenan v United Kingdom, 33 EHRR 913, 958, para 92. The authorities must then do everything that can reasonably be expected of them in the circumstances to prevent him from doing so.
58. Precisely because the obligation on the prison authorities to protect a prisoner from himself is not absolute and depends on the particular circumstances, a suicide can occur without there having been any violation of the prison authorities' obligations under article 2 to protect the prisoner. Focusing on that point, Mr Giffin QC argued on behalf of the Secretary of State that article 2 did not require an independent investigation to be held unless there was some positive reason to believe that the authorities had indeed been in breach of their obligation to protect the prisoner.
59. That argument is mistaken. Whenever a prisoner kills himself, it is at least possible that the prison authorities, who are responsible for the prisoner, have failed, either in their obligation to take general measures to diminish the opportunities for prisoners to harm themselves, or in their operational obligation to try to prevent the particular prisoner from committing suicide. Given the closed nature of the prison world, without an independent investigation you might never know. So there must be an investigation of that kind to find out whether something did indeed go wrong. In this respect a suicide is like any other violent death in custody. In affirming the need for an effective form of investigation in a case involving the suicide of a man in police custody, the European Court held that such an investigation should be held "when a resort to force has resulted in a person's death": Akdogdu v Turkey, para 52.
60. In R (Middleton) v West Somerset Coroner [2004] 2 AC 182, another case of a suicide in custody, at p 191, para 3, Lord Bingham of Cornhill summarised the jurisprudence of the European Court as imposing an obligation to hold an independent investigation if "it appears that one or other of the … substantive obligations has been, or may have been, violated and it appears that agents of the state are, or may be, in some way, implicated." Mr Giffin suggested that Lord Bingham's formulation was inconsistent with there being a requirement for an independent investigation in all cases of suicide in custody. I do not agree. In summarising the case law, Lord Bingham was recognising that, where the circumstances of a prisoner's death in custody indicate that the substantive obligations of the State may have been violated, any violation, whether due to a systemic or operational failure, will necessarily have involved members of the prison service in one capacity or another. An independent investigation is therefore required to see whether there was, in fact, a violation.
61. If, then, an independent investigation is required in all cases of suicide, does the same apply where a prisoner has attempted to commit suicide but has failed? The Secretary of State now accepts that JL's case is one where article 2 required him to set up an independent investigation. This concession meant that the discussion at the hearing tended to focus on the form which any inquiry should take, rather than on precisely when an attempted suicide would trigger an article 2 obligation to hold an independent investigation.
62. This is an area into which it would be unwise to venture far without the benefit of full argument. At the most basic, what brings an attempted suicide within the scope of article 2 is that, even though the prisoner has not died, his life has been in danger. But, starting from there, cases falling within the description of attempted suicide vary enormously. A prisoner may climb on to a ledge, be seized by a prison officer at the last moment before he can jump to his death, and emerge unharmed. Or else a prisoner may cut his wrists, but end up with only a few scars to show what happened. A prisoner may attempt suicide by jumping from a ledge and survive with his mental capacity unimpaired, but with injuries that mean he can never walk again. Or, as here, a prisoner may attempt to hang himself and be left with brain injuries which mean that he is unable to look after his own affairs.
63. Responsible prison authorities would wish to conduct an inquiry of some kind into all of these incidents. But it does not necessarily follow that the prisoners who attempted to commit suicide would all have an article 2 Convention right to have the Secretary of State set up an independent investigation into the circumstances and that he would act unlawfully if he refused to hold one.
64. Your Lordships have referred to the situation where "a prisoner in custody makes an attempt to commit suicide that nearly succeeds and which leaves him with serious injury" and to near suicides, "genuine suicide attempts resulting in lasting serious injury". Either description could be applied to the case of JL - or, for example, to the case of the prisoner who jumps off the roof and sustains an injury which means he will never walk again.
65. Certainly, they are both the primary victims of any breach of the prison authorities' article 2 obligation to protect them from themselves. To that extent their positions are the same. The difference is that, while the prisoner who jumped from the ledge is unable to walk, he has his mental faculties intact, knows what happened and is therefore, prima facie, in a position to take the appropriate civil proceedings afforded by English law in respect of any perceived violation of his article 2 Convention right. By contrast, a prisoner in JL's position is incapable of looking after his own interests: he may well be quite unable to recall or to explain what happened and he certainly cannot take proceedings by himself on the basis of any recollection he may have. The situation in his case is more like the position where a prisoner has succeeded in committing suicide. When the prisoner is dead, it is likely that no-one but the prison authorities will know much about what happened in the period before he killed himself. Moreover, he may have had no relative or friends interested enough to ask questions or to raise proceedings after his death. The potential for concealing errors or misconduct by the authorities is all too obvious. So article 2 imposes an obligation on the State spontaneously to hold an independent investigation to establish the facts and discover whether anything went wrong and, if so, who was responsible. Important lessons for the future may also be learned from the results of the investigation.
66. It is not hard to see that, by a similar process of reasoning, article 2 should be interpreted as imposing an obligation on the State to hold an independent inquiry in a case, like the present, where a prisoner's life is put at risk and, because of his injuries, he cannot take steps by himself to hold the authorities responsible for any failures on their part which led to his attempted suicide.
67. In Banks v United Kingdom (2007) 45 EHRR SE2, on the other hand, all but one of the applicants were alive and were complaining of maltreatment by prison officers in breach of article 3. In fact, the matter had been investigated by the Crown Prosecution Service which had decided not to prosecute. Civil proceedings had been raised and settled. The applicants contended that, even after all that, an independent public inquiry should be held, since it was the only means of ensuring compliance with the article 3 procedural obligation. The Fourth Section of the Court dismissed their complaint on the view that the ordinary mechanisms of civil and criminal justice had provided for an adequate scrutiny of the incident itself and there was no need to have a public inquiry into the general background.
68. The Court explained the difference between the procedural obligation under articles 2 and 3 in this way, at para 2:

"Procedural obligations have been implied in varying contexts under the Convention, where this has been perceived as necessary to ensure that the rights guaranteed under the Convention are not theoretical or illusory but practical and effective. Such obligations requiring an effective investigation into allegations of unlawful use of force and serious ill-treatment have been interpreted as arising under Articles 2 and 3 of the Convention respectively….

The Court would emphasise that these obligations are not identical either in content or as regards their applicability. In the context of Article 2 of the Convention, the obligation to conduct an effective investigation into allegations of the unlawful use of force attracts particular stringency in situations where the victim is deceased and the only persons with knowledge of the circumstances are officers of the State. It is important, with a view to ensuring respect for the rule of law and confidence of the public, that the facts, and any unlawfulness, are properly and swiftly established. In the context of Article 3, where the victim of any alleged ill-treatment is, generally, able to act on his own behalf and give evidence as to what occurred, there is a different emphasis and … since Article 13 of the Convention requires an effective remedy to be provided for arguable breaches of Article 3, it will not always be necessary, or appropriate, to examine the procedural complaints under the latter provision. The procedural limb of Article 3 principally comes into play where the Court is unable to reach any conclusions as to whether there has been treatment prohibited by Article 3 of the Convention, deriving, at least in part, from the failure of the authorities to react effectively to such complaints at the relevant time" (internal citations omitted).

69. The European Court envisages that, where the victim is able to act on his own behalf and to give evidence about what happened, so far as any spontaneous independent investigation under article 3 is concerned, "there is a different emphasis" and the absence of any prompt independent investigation will mainly come into play if the Court is unable to determine, as a matter of fact, whether or not there has been any treatment prohibited by article 3.
70. Mutatis mutandis, the same may apply in the case of article 2, where the prisoner is in a position to act on his own behalf, raise the appropriate civil proceedings, and give evidence about what happened when he attempted to commit suicide. The obligation on the prison authorities to hold a spontaneous and prompt independent investigation into the incident remains the same. The existence of a right to bring civil proceedings does not satisfy the procedural obligation under article 2. Suppose, however, that the prisoner does indeed take civil proceedings relating to the incident. In that event, any failure to hold a prompt investigation might mainly come into play where, because of it, the judge in the civil proceedings was unable to determine, as a matter of fact, whether or not the prison authorities had violated any of their substantive obligations to the prisoner under article 2.
71. That was indeed the position in Makaratzis v Greece (2004) 41 EHRR 1092 where police officers had shot at the applicant's car which had driven through a series of road blocks. As a result of the incident the applicant sustained certain injuries and was kept in hospital for nine days. An administrative investigation was opened and seven police officers were prosecuted but acquitted. The European Court found that there had been striking omissions in the conduct of the investigation which had prevented the national court from making as full a finding of fact in the criminal trial as it might otherwise have done. The Court accordingly concluded that the authorities had failed to carry out an effective investigation of the incident and that there had been a violation of article 2 in that respect.
72. The Court explained its approach in this way, at p 1125, para 73:

"The obligation to protect the right to life under Article 2 of the Convention, read in conjunction with the state's general duty under Article 1 to 'secure to everyone within [its] jurisdiction the rights and freedoms defined in [the] Convention', requires by implication that there should be some form of effective official investigation when individuals have been killed as a result of the use of force. The essential purpose of such an investigation is to secure the effective implementation of the domestic laws safeguarding the right to life and, in those cases involving state agents or bodies, to ensure their accountability for deaths occurring under their responsibility. Since often, in practice, the true circumstances of the death in such cases are largely confined within the knowledge of state officials or authorities, the bringing of appropriate domestic proceedings, such as a criminal prosecution, disciplinary proceedings and proceedings for the exercise of remedies available to victims and their families, will be conditioned by an adequate official investigation, which must be independent and impartial. The same reasoning applies in the case under consideration, where the Court has found that the force used by the police against the applicant endangered his life."

73. I have thought it right to touch on these questions, even though they do not arise for determination and, having heard no detailed submissions, I have formed no concluded view whatever on them. I would only say that, in my opinion, the Secretary of State was right to concede that article 2 required an independent investigation in the present case. The same would apply to R (on the application of D) v Secretary of State for the Home Department [2006] 3 All ER 946, where, too, the prisoner who attempted suicide was left with brain injuries which meant that he could not act on his own behalf. It is not a case where a criminal investigation would be in prospect. And even supposing that, despite the difficulties, civil proceedings in the name of JL were to be contemplated, this is the kind of case envisaged by the Court in Makaratzis where the availability of such proceedings would, in all likelihood, be conditioned on an adequate independent investigation.
74. One thing is clear from the case law: if there has to be an independent investigation, then the sooner it starts work the better. This is just common sense, but the European Court emphasised the point in Edwards v United Kingdom (2002) 35 EHRR 487, 515, para 86:

"The Court reiterates that it is crucial in cases of deaths in contentious situations for the investigation to be prompt. The passage of time will inevitably erode the amount and quality of the evidence available and the appearance of a lack of diligence [will] cast doubt on the good faith of the investigative efforts, as well as drag out the ordeal for the members of the family."

Similarly, in Trubnikov v Russia (Application No 49790/99, BAILII: [2005] ECHR 462 ) 5 July 2005, at para 88, the Court said:

"the competent authorities must act with exemplary diligence and promptness and must of their own motion initiate investigations which would be capable of, firstly, ascertaining the circumstances in which the incident took place and any shortcomings in the operation of the regulatory system and, secondly, identifying the State officials or authorities involved."

75. As my noble and learned friend, Lord Walker of Gestingthorpe, points out, it is unavoidable that the very first steps in investigating an incident, immediately after it has occurred, will be internal to the Prison Service. In a case like the present, however, the need to set up an independent investigation in compliance with article 2 will be apparent, at latest, as soon as the prison authorities become aware of circumstances which suggest that the prisoner attempted to hang himself and that he is going to be incapacitated. At that point the Prison Service should take the necessary steps to have the independent investigation established and to provide it with the results of any internal inquiry which officials have already conducted. It has been suggested that the Prisons and Probation Ombudsman's organisation might carry out such investigations. That may be so. But it is not for this House to prescribe who should carry out the independent investigation. Whoever it is must be independent of the Ministry of Justice and in a position to set to work and complete the investigation reasonably quickly: Bakorzina v Russia (Application No 69481/01, BAILII: [2006] ECHR 751 ) 27 July 2006, para 119.
76. Once the independent investigation has been established with the powers and resources it needs, it is very much up to the investigator to decide how to proceed in order to achieve the objectives for which it was set up. But, presumably, the first steps will involve assembling relevant material, in the form of records and reports, and taking statements from officials and others who can speak to the circumstances. All this can be done in private, but it is essential that the prisoner's relatives and his representative, if there is one, should be told that the investigation is under way. They should also be given an opportunity to participate. In due course they must be informed of the investigator's conclusions. It is not, however, necessary for the relatives to be granted access to all aspects of a current investigation if this might prejudice private individuals or other investigations: Ramsahai v The Netherlands (Application No 52391/99, BAILII: [2007] ECHR 393 ) 15 May 2007, para 347. Sometimes relatives will be in a position to contribute information about the prisoner's state of mind in the period before the incident. They may be able to suggest lines of inquiry. Being independent, the investigator is free to reject the suggestions if he considers that the inquiries would not be useful. Where the relatives have had little contact with the prisoner and so have no relevant knowledge of the circumstances, the investigator's main duty will be to keep them informed of the progress of the investigation and to tell them his conclusions.
77. The Secretary of State is concerned about the financial implications of having to hold an independent investigation in cases of attempted suicide. His concern is entirely proper, as the European Court has recognised in the judgments cited in para 56 above. His anxieties may have been fuelled, however, by an impression that, whenever article 2 requires an independent investigation to be set up, that investigation has to have all the bells and whistles of the full-blown public inquiry described by the Court of Appeal in R (on the application of D) v Secretary of State for the Home Department [2006] 3 All ER 946 - sometimes called a "type D inquiry". Nothing could be further from the truth. I respectfully endorse what my noble and learned friend, Lord Brown of Eaton-under-Heywood, says on this matter in paras 107 and 108 of his speech.
78. The principal hallmark of an article 2-compliant inquiry is that it is "effective". The Grand Chamber explained what this means in Ramsahai v The Netherlands, a case where the police had shot someone suspected of stealing a scooter. The court said, at paras 324-325:

"324. In order to be 'effective' as this expression is to be understood in the context of Article 2 of the Convention, an investigation into a death that engages the responsibility of a Contracting Party under that Article must firstly be adequate. That is, it must be capable of leading to the identification and punishment of those responsible. This is not an obligation of result, but one of means. The authorities must have taken the reasonable steps available to them to secure the evidence concerning the incident. Any deficiency in the investigation which undermines its ability to identify the perpetrator or perpetrators will risk falling foul of this standard.

325. Secondly, for the investigation to be 'effective' in this sense it may generally be regarded as necessary for the persons responsible for it and carrying it out to be independent from those implicated in the events. This means not only a lack of hierarchical or institutional connection but also a practical independence" (internal citations omitted).

The Grand Chamber stresses that those carrying out the investigation should be entirely independent of those who may have been implicated in the events. So, in a case like the present, the independent investigator could not use Prison Service officials to carry out inquiries on his behalf. But, beyond that, what matters is that the investigator should take all reasonable steps to secure the evidence concerning the incident and to find out, if possible, what happened and what, if anything, went wrong. The steps which the investigator needs to take to fulfil these requirements will inevitably depend on the circumstances of the particular case. There neither is, nor can be, any single off-the-peg model that is suitable for use in all cases.

79. When the investigator embarks on his inquiries, he does not know what they will reveal. So he will usually be in no position to say how elaborate they will have to be or what form they will eventually take. He may have a better idea once he has studied the material produced by his initial inquiries. Sometimes this material will reveal a clear picture and the investigator will be able to complete his investigation without going further. Mr Emmerson QC was right to emphasise that, when this happens, the investigator does not break off, or cut short, the investigation: he does all that is required to complete an effective investigation of the incident. In a passage quoted by Lord Bingham in R (Amin) v Secretary of State for the Home Department [2004] 1 AC 653, 664, para 19, the European Commission said, in McCann v United Kingdom (1995) 21 EHRR 97, 140, para 193, that, even in the case of a death:

"there may be cases where the facts surrounding a deprivation of life are clear and undisputed and the subsequent inquisitorial examination may legitimately be reduced to a minimum formality."

In such a case the investigator may well conclude that the public interest will be met by his proceeding straightaway to publish a report, explaining the circumstances and, where appropriate, identifying what went wrong and who was responsible. Both the public interest and the requirements of article 2 are satisfied because the independent investigator has looked at the circumstances, has made them public and considers that nothing would be gained by taking the matter any further. It is a fully compliant article 2 investigation.

80. Similarly, because the investigator is independent, his investigation may well be effective, and so fulfil the requirements of article 2, even though no part of it is conducted in public. Again, it depends on the particular case. In Anguelova v Bulgaria (2004) 38 EHRR 31, where a youth had died in police custody a few hours after being arrested for attempted theft, the European Court said, at p 686, para 140:

"There must be a sufficient element of public scrutiny of the investigation or its results to secure accountability in practice as well as in theory, maintain public confidence in the authorities' adherence to the rule of law and prevent any appearance of collusion in or tolerance of unlawful acts. The degree of public scrutiny required may well vary from case to case. In all cases, however, the next of kin of the victim must be involved in the procedure to the extent necessary to safeguard his or her legitimate interests…."

The Grand Chamber adopted that approach in Ramsahai v The Netherlands 15 May 2007. It said, at para 353:

"Article 2 does not go so far as to require all proceedings following an inquiry into a violent death to be public. As stated in, for example Anguelova …, the test is whether there is a sufficient element of public scrutiny in respect of the investigation or its results to secure accountability in practice as well as in theory, maintain public confidence in the authorities' adherence to the rule of law and prevent any appearance of collusion in or tolerance of unlawful acts. It must be accepted in this connection that the degree of public scrutiny required may well vary from case to case."

81. Indeed, applying that approach, the Grand Chamber held, at para 354, that in the circumstances of that case there had been no need for the Dutch Court of Appeal to carry out its review, of the decision not to prosecute, in public - or even to publish its findings. It was enough that the deceased's relatives had had access to the investigation and had been provided with the Court of Appeal's reasoned decision which they themselves could then have published if they had wished to do so:

"Turning to the facts of the present case, the Court agrees with the Chamber that the Court of Appeal's proceedings did not have to be open to the public. Unlike the Chamber, however, the Court takes the view that the Court of Appeal's decision was not required to be made public either. The applicants were allowed full access to the investigation file and were enabled to participate effectively in the Court of Appeal's hearing; they were provided with a reasoned decision. There was thus little likelihood that any authority involved in the case might have concealed relevant information from the Court of Appeal or the applicants. In addition, given that the applicants were not prevented from making the decision public themselves, the Court takes the view that the requirement of publicity was satisfied to an extent sufficient to obviate the danger of any improper cover-up by the Netherlands authorities.

82. Rightly, the Grand Chamber has made no attempt to specify types of cases in which a public hearing will be needed. The House should follow that example. But it is worth stressing that, whatever the steps the investigator takes from the time of his appointment until he finishes, they are all part of the single independent investigation which is required by article 2. That investigation may stop once the initial material is assembled. Alternatively, it may continue with witnesses being heard in private, or in public - or some in private and some in public, depending on what is needed for an effective investigation. If the shorthand expression, "a type D inquiry", fosters an idea that, when an investigator decides to hear some evidence in public, he has to conform to a set model, it is potentially misleading. In reality, whatever its form, if the investigation is independent and effective, it will fulfil the requirements of article 2.
83. So, in the present case, as in any other, it is for the independent investigator to decide, once he has become familiar with the issues, whether there needs to be a public hearing and, if so, what shape it should take. In reaching that decision, he will have in mind both what his initial enquiries have revealed and the factors which the Grand Chamber said in Ramsahai should determine the degree of public scrutiny that is required.
84. With these observations and for substantially the same reasons as all of your Lordships, I would dismiss the appeal.

LORD WALKER OF GESTINGTHORPE

My Lords,

85. In agreement with all your Lordships, I would dismiss this appeal. The formal disposal of the appeal is of limited practical importance, since the Prison Service has already agreed to commission a further independent inquiry into Mr L's attempted suicide, the inquiry to be conducted as what has been called a D inquiry (see R(D) v Secretary of State for the Home Department [2006] 3 All ER 946) and the costs will in any event fall on public funds. I am also in general agreement with your Lordships' reasons. But because of the importance of this appeal I add some observations of my own.
86. Between 1996 and 2007 the mean annual average of suicides in prison was about 82. The worst period was from 2002 to 2004, with an average of 95 suicides (including a disproportionately high average of 12 women). There was an improvement during 2005 and 2006 but in 2007 there were 92 suicides (84 men and 8 women). In R(Middleton) v West Somerset Coroner [2004] 2 AC 182, para 5, Lord Bingham of Cornhill (reviewing the position down to 2003) referred to these figures as "grim", but added the observation quoted in the first paragraph of the opinion of my noble and learned friend Lord Brown of Eaton-under-Heywood. Every suicide is a tragedy, in or out of prison, and nonetheless so because it is practically inevitable that there will be prison suicides. For too long our prisons have been seriously overcrowded, and the sheer volume of numbers of prisoners often makes it difficult for prison officers to perform the duties of care which they owe to those who are in the custody of the state.
87. As Lord Bingham observed in Middleton (para 5),

"These statistics, grim though they are, do not of themselves point towards any dereliction of duty on the part of the authorities (which have given much attention to the problem) or any individual official. But they do highlight the need for an investigative regime which will not only expose any past violation of the state's substantive obligations [under article 2 of the Convention] but also, within the bounds of what is practicable, promote measures to prevent or minimise the risk of future violations."

Lord Bingham did not therefore restrict the purpose of an independent investigation to establishing a past violation of the state's substantive obligations. He included the wider purpose of learning from experience, whether or not there have been identifiable failures, systemic or operational, for which the management of the Prison Service as a whole, or particular individuals within the Prison Service, must take responsibility.

88. The distinction between systemic and operational failures is central to the Osman principle (Osman v United Kingdom (2000) 29 EHRR 245 ) as it has developed in relation to the state's duty to protect those in its custody against self-harm (or harm from others in custody). Sometimes the systemic duty is expressed at a high level of generality, as Lord Bingham of Cornhill put it in Van Colle v Chief Constable of the Hertfordshire Police [2008] 3 WLR 593, para 28:

"The state's . . . primary duty to secure the right to life by putting in place effective criminal law provisions to deter the commission of offences against the person backed up by law enforcement machinery for the prevention, suppression and sanctioning of breaches of such provisions."

But at a lower level there is often no clear dividing line between systemic and operational failures, and the distinction is one of degree only. Moreover an excessively elaborate and prescriptive system may carry the seeds of its own failure, in that it may be predictable that overworked and under-motivated staff in the lower levels of the system will be unable or unwilling to comply fully with it. In the Prison Service, as in branches of the social services dealing with vulnerable old people and children, there is almost inevitably a gulf between policy-makers' vision of informed, collective, multi-disciplinary decision-taking and the incidents of missing files, missed appointments, misunderstandings and muddling through which regrettably occur among those on the ground.

89. Your Lordships have been able to read some 400 pages of official documents in exhibits SC1 and SC2 to the affidavit of Saimo Chahal, a partner in the firm of solicitors acting for JL. Some of these (notably the report of Mr Sheikh, some incident reports and some medical reports) post-date Mr L's attempted suicide. But the bulk of the documentation came into existence before the suicide attempt. It includes the official Local Suicide Prevention Policy for Feltham YOI published in January 2002, the official explanation of the F2052 SH system, and the official records of Mr L's medical care and daily supervision and support, together with "multi-disciplinary continuation sheets" and logs of the periods for which Mr L was under specially close observation. Study of these documents leaves me with the clear impression that the Prison Service managers have worked very hard to put in place a system for identifying, monitoring and counselling prisoners at risk of suicide or self-harm. It may have tended towards a level of elaboration which might leave experienced and conscientious prison officers with restricted opportunities for relying on their own intuitive knowledge of human nature. But your Lordships did not hear any argument on this point, and it would be wrong to express any definite opinion about it. It is sufficient to say that if there were any serious errors in the treatment of Mr L—an issue which still remains to be considered at the forthcoming independent investigation—it seems likely that they will be found (if one must choose an alternative) to have been operational rather than systemic.
90. In bringing this appeal (while conceding that there is to be a full independent inquiry) the Secretary of State is for practical purposes seeking guidance from your Lordships as to future policy and procedures. The Secretary of State is particularly concerned, understandably, about resource implications. I can sympathise with this concern but there are obvious limits to how far it is for your Lordships, or any court, to give anything like detailed guidance on these matters. Even if a majority of your Lordships were to concur in my views, they would be no more than expressions of opinion. But with that caveat I will express some tentative views.
91. As I see it there are three interwoven strands to the problem: identifying which incidents of non-fatal self-harm by prisoners may call for independent investigation; working out the best procedure for decision-making about setting up an independent investigation; and determining the form of any independent investigation which is necessary.
92. On the first of these points, Prison Service Order 2700 (issued in October 2007) deals at length with different aspects of suicide prevention and self-harm management. Chapter 13 (actions following an incident of self-harm) contains the following paragraph (13.4.1):

"It is strongly recommended that following incidents of serious self-harm an investigation is carried out into the circumstances of the incident ('serious incidents' are defined as those which mean the prisoner involved required resuscitation and/or transfer to an outside hospital as a result of their harming themselves). As each individual incident will differ in level of severity, Governors/Directors will need to judge when such investigations are appropriate. Cases where the injury was life threatening, the person required hospitalisation and it is likely that they will be (sic) sustain permanent injuries as a result of the self-harm incident, are examples of where Governors/Directors are likely to consider an investigation into the incident to be imperative. Wherever possible the family ought to be included in such investigations. Care needs to be taken to retain required documentation."

93. The need for resuscitation and/or transfer to an outside hospital is no doubt a sensible starting point but it cannot (as the paragraph itself recognises) be exhaustive. Even in cases where there is no serious permanent injury, investigation may be needed if the self-harmer was a known suicide risk, or if the means of self-harm (whether a ligature, a sharp instrument or some harmful substance) suggested a failure in the system of searches (ineffective searches are mentioned in a recent IPCC publication, "Near Misses in Police Custody", as the most common causative factor in near-miss incidents). No doubt there are other special cases calling for the exercise of judgment by Governors and Directors.
94. As to the procedure to be adopted, it seems unavoidable, because of time constraints, that the first steps of any investigation should be internal to the Prison Service (and therefore lacking in independence). Senior staff at the prison may need to take immediate action to impound files and records, and make written records (however brief) of what they are told by prison officers or prisoners who were on the spot at the time. But an independent element should be added to the preliminary investigation as soon as possible. This point was emphasised on behalf of the intervener, the Equality and Human Rights Commission, by Mr John Wadham, its Group Legal Director, in his witness statement:

"Whenever a death or near death occurs it is very important that a decision as to what kind of investigation is necessary is made quickly, before any evidence is disturbed or lost and before those who witnessed the events forget the details or have their accounts contaminated by the accounts of others. Obviously the more likely it is that there may be some culpability or some systemic failure associated with those that can give evidence the more important that it is that those investigating are independent of the people that they are investigating."

The point was also made by counsel for the intervener in their supplementary written submissions, citing the decision of the Grand Chamber in Ramsahai v The Netherlands 15 May 2007. That was a case on very different facts: a man suspected of stealing a motor-scooter at gunpoint was shot by the police in Amsterdam, and there was a delay of some 15 hours before the State Criminal Investigation Department became involved in the investigation. In principle, however, it is clearly right that any preliminary investigation should acquire the character of an independent investigation at the earliest possible time.

95. In practice this means that the Prison Service should hand over the investigation (with all relevant documentary material, including official records and any preliminary written statements) to an independent investigator, or a team of independent investigators. Mr Wadham suggests that the Prisons and Probation Ombudsman would be an appropriate investigator. Your Lordships were not told much about the resources available to the Ombudsman. But he already carries out a similar function in relation to fatal incidents, and it seems to me that it would be appropriate, if the Ombudsman has or can be furnished with adequate human and other resources, for his oversight to be extended to non-fatal but serious incidents that may call for an article 2 investigation. The Ombudsman would form an independent view as to what further investigations were required; whether any such investigations ought to meet the requirements of article 2; and if so, whether there should be a full D-type inquiry. It seems likely that in the majority of cases some further investigation would be necessary, but that relatively few non-fatal incidents would call for a lengthy and expensive procedure involving taking oral evidence in public.
96. In agreement (as I understand it) with all your Lordships I agree that not every investigation is required, in order to comply with article 2, to amount to what has been referred to as a D-type inquiry. The essential requirements of article 2 are set out in para 107 of Lord Brown's opinion. I respectfully agree with all that he says in that and the following paragraph, and with what my noble and learned friend Lord Rodger of Earlsferry says in paras 73 to 80 of his opinion. Since it is for the State itself to initiate an article 2 investigation, the Ombudsman would (under the procedure suggested above) have to make a recommendation to the Prison Service; but he would do so in the confident expectation that his recommendation would be accepted. The precise procedure to be followed at a formal public inquiry, in the (perhaps comparatively rare) cases where it was called for, would be for the individual or tribunal conducting the inquiry.

LORD BROWN OF EATON-UNDER-HEYWOOD

My Lords,

97. The problem of suicides in custody is all too well-known. Although fewer such deaths occurred in 2005 and 2006 than in the previous three years, in 2007 numbers were tragically back to where they had been, almost two a week. The number of near-suicides in custody (genuine suicide attempts resulting in lasting serious injury) is greater still. As Lord Bingham of Cornhill pointed out in R (Middleton) v West Somerset Coroner [2004] 2 AC 182, 192, para 5:

"Many of those in prison are vulnerable, inadequate or mentally disturbed; many have drug problems; and imprisonment is inevitably, for some, a very traumatic experience."

98. Unsurprisingly, therefore, the law imposes upon detaining authorities special duties with regard to safeguarding those (whether of sound or unsound mind) in their custody: a common law duty to take reasonable care for their safety (Reeves v Commissioner of Police of the Metropolis [2000] 1 AC 360 where, at p 369, Lord Hoffmann describes the duty as "a very unusual one, arising from the complete control which the police or prison authorities have over the prisoner, combined with the special danger of people in prison taking their own lives"), and a duty under article 2 of the European Convention on Human Rights "to protect them [and] to account for any injuries suffered in custody, which obligation is particularly stringent when the individual dies" (Keenan v United Kingdom (2001) 33 EHRR 913, para 90 where the Court emphasised, as it had in earlier cases, "that persons in custody are in a vulnerable position").
99. This appeal concerns the near-suicide of a 20 year old in youth custody awaiting trial. Initially he was assessed as a suicide risk but after 18 days his self-harm at risk form was closed. Eleven days later he was found hanging from his cell bars and, although resuscitated, sadly is left with serious brain damage.
100. Throughout these proceedings it has been common ground that the respondent's attempted suicide falls within the ambit of article 2. The appellant Secretary of State, however, contests the need in such cases for any article 2 investigation except where the State is in "arguable breach" of its substantive article 2 duty to protect life and this, he submits, will only be so when arguably the prison authorities knew or ought to have known of a real and immediate risk of the prisoner committing suicide and failed then to take reasonable preventive measures.
101. For the reasons given by my noble and learned friend Lord Phillips of Worth Matravers whose opinion I have had the advantage of reading in draft, I too would reject that argument and dismiss the appeal: as already explained the State bears a particular responsibility towards those detained and it is bound to carry out an article 2 investigation whenever its system for preventing suicide fails. In every such case it is for the State to account for the failure. Whether or not the investigation ultimately is likely to reveal a substantive breach of article 2—operationally (as in the Osman type case) or systemically (as to the procedures for detecting where such a real and immediate risk arises)—is really not material at the outset.
102. It by no means follows, however, that the article 2 investigation required in every case of near-suicide must match in all respects that achieved in this country in the case of all actual prison suicides: an initial investigation by the Prisons and Probation Ombudsman ("ensuring as far as possible that the full facts are brought to light and any relevant failing is exposed, any commendable action or practice is identified, and any lessons learned from the death", as the PPO's published terms of reference put it) followed by a Middleton type inquest.
103. It is really only on this aspect of the case that I wish to add a few paragraphs to what Lord Phillips has already said. I am concerned in particular with what my Lord at para 10 describes as a "D type investigation" (after R (D) v Secretary of State for the Home Department [2006] 3 All ER 946), which apparently is what the Secretary of State is now embarking upon in the present case. By a D type investigation I understand to be meant an independent public inquiry whereby whoever conducts the inquiry "would make the evidence and any written submissions public and take oral evidence in public, subject to the proviso . . . that there might be Convention-compatible reasons for not holding the whole investigation in public" (para 24 of D). That certainly was the nature of the investigation ordered by the Court of Appeal in D and subsequently carried out by the PPO (Mr Shaw) over some 20 months at the cost of about half a million pounds, leading to a comprehensive report in May 2008.
104. In my opinion a public inquiry of that sort (or, indeed, of any such sort) goes far beyond what can reasonably be judged necessary to satisfy the article 2 procedural duty arising in any save the most exceptional near-suicide case and for my part I regard D itself to have been wrongly decided. As I read the judgment it appears to have been founded principally upon two decisions in particular: that of the European Court of Human Rights in Edwards v United Kingdom (2002) 35 EHRR 487 and that of the House of Lords in R (Amin) v Secretary for the Home Department [2004] 1 AC 653. But those cases were to my mind critically different from suicide cases and still more so from near-suicide cases. Each of them involved the killing of a prisoner by his own cellmate. Mr Edwards was killed by a man suffering acute mental illness with a record of violence who plainly should have been recognised as a serious risk to others. Mr Mubarek (Mr Amin's nephew) was killed by a man known to be both very dangerous and a racist. Because there were criminal prosecutions, in neither case was there an inquest. And in each case a private rather than a public inquiry was held, with very little family involvement. Small wonder that in these circumstances the respective courts required open public inquiries to be held. The Strasbourg Court in Edwards (at para 83) said this:

"In the present case, where the deceased was a vulnerable individual who lost his life in a horrendous manner due to a series of failures by public bodies and servants who bore a responsibility to safeguard his welfare, the Court considers that the public interest attaching to the issues thrown up by the case was such as to call for the widest exposure possible."

And at para 31 of his speech in Amin, Lord Bingham, having noted (at para 21) the case's "strong similarities" to Edwards, set out the purposes which would ordinarily be served by an inquest (whereby deaths are usually "publicly investigated before an independent judicial tribunal with an opportunity for relatives of the deceased to participate"), namely:

"to ensure so far as possible that the full facts are brought to light; that culpable and discreditable conduct is exposed and brought to public notice; that suspicion of deliberate wrongdoing (if unjustified) is allayed; that dangerous practices and procedures are rectified; and that those who have lost their relative may at least have the satisfaction of knowing that lessons learned from his death may save the lives of others."

True it is that Lord Bingham was not there expressly confining his remarks to cellmate killings as opposed to deaths in custody generally. Plainly, however, cellmate killings (and the Strasbourg Court's judgment in Edwards) provided the immediate context for those remarks and I find it quite impossible to regard them as authority for the wider proposition that a D type investigation which "take[s] oral evidence in public" is ordinarily required in near-suicide cases. I recognise, of course, that, like cases of cellmate killings (although, of course, for a different reason) near-suicides are not the subject of inquests. But there can be no doubt that inquests themselves often go beyond the strict requirements of article 2 investigation.

105. There seem to me two very important distinctions between cases of cellmate killings and those of near-suicide. One is the difference between death and serious injury. Calamitous though near-suicide cases may be, death adds a further dimension of gravity—a point made by Lord Phillips at para 20 and recognised by the European Court of Human Rights in the passage cited above (para 98) from Keenan. Secondly, however deeply troubled one may be about cases of suicide and near-suicide in custody, they do not to my mind prompt quite the same feelings of horror and outrage as do cases where a prisoner is killed by the very person he is locked up with. Detaining authorities plainly owe a high duty to safeguard those in custody so far as possible from themselves. But the duty to protect them from others with whom they are confined must surely be higher still.
106. I accept unhesitatingly Mr Emmerson QC's submission that the costs of Mr Shaw's inquiry into D's attempted suicide were exceptional (indeed Mr Shaw's report itself says as much) and that any such inquiry in future might more reasonably be expected to cost nearer £65,000 (the Secretary of State's own working estimate for such an inquiry where evidence is heard in public). That itself, however, is substantially more than the estimated cost of £15,000 odd for the Ombudsman's routine (pre-inquest) investigation in all actual cases of suicide in custody. Of course the nature and extent of the state's article 2 obligation in cases of near-suicide cannot be measured in monetary terms. But it is idle to pretend that money is no object and certainly, the larger the category of near-suicide cases the Secretary of State (or, if he is challenged as to this, the court) recognises to attract the article 2 obligation, the more imperative it must be to keep the discharge of the obligation within manageable financial limits.
107. With regard to near-suicide cases resulting, as here, in lasting serious injury (I say nothing as to the substantially larger number of self-harm incidents recorded in the statistics), I have no doubt that Mr Shaw was right in his D report to recommend (as noted by Lord Phillips at para 47) investigations which "include an independent element, and engage the person who has been harmed and/or their family". If the Ombudsman himself, or perhaps a senior deputy, could carry them out, so much the better. Elementarily, to satisfy the basic requirements of any article 2 investigation, besides being independent and involving the family, they must in addition be initiated by the state, be promptly and reasonably expeditiously carried out, and provide for a sufficient element of public scrutiny. Beyond this, however, it is impossible to be prescriptive.
108. Generally speaking I can see no need for inquiries into near-suicides to take place in public although obviously the independent investigator's report would itself be made public. If, of course, any particular problems come to light during the investigation—if, say, witnesses prove uncooperative, or egregious failures become manifest (again one cannot be prescriptive about the circumstances which might occasion a change of course), the person conducting the investigation might feel it necessary to expand it into something akin to a D-type inquiry. For my part, however, I would expect that to be a comparatively rare event and, concerned though inevitably your Lordships must be about a number of apparently troubling features of the respondent's attempted suicide, I question whether this is itself such a case.
109. That, however, is really by the way. It is not suggested that your Lordships' opinions will affect the outcome of the present case. Rather it is hoped that they will illuminate the nature of the article 2 obligation arising in future such cases. It is not altogether easy to throw light on questions necessarily raised in somewhat general terms. The above represents the furthest I have felt able to go. I add only that I agree also with everything said by my noble and learned friend Lord Walker of Gestingthorpe.

LORD MANCE

My Lords,

110. I have had the benefit of reading in draft the judgments of all of your Lordships, with all of which I am in general agreement. As my noble and learned friend, Lord Phillips of Worth Matravers, states (paras 5 to 15), it is important to identify the issues with which this litigation is concerned. Langstaff J attempted carefully to confine them to the question whether in the present case the threshold was passed for an "enquiry satisfying the minimum standards" of article 2 of the Convention, without considering what those standards involved or what the content of such an enquiry should be in the particular circumstances.
111. Langstaff J's reservations about the feasibility of drawing this line are borne out by the subsequent history. In the Court of Appeal, Waller LJ (giving the main judgment with which the other two members of the court agreed) said (para 7) that "What each side is seeking primarily is the answer to the question whether the enhanced obligation to investigate and particularly to carry out an investigation with the features laid down in R(D) v Secretary of State for the Home Department [2006] EWCA Civ 143; [2006] 3 All ER 946 (a D-type investigation) has been triggered in this case". The answer that Waller LJ gave referred to a two-stage process, whereby, in cases of serious injury to prisoners, an independent investigator would first elicit sufficient facts to enable a decision to be made as to the nature of any further enquiry then required, and "the mere fact" of a death or serious injury in custody would not necessitate a second stage in the form of a full D-type inquiry (para 33). But Waller LJ concluded (paras 58 and 60) by expressing a clear view that a D-type investigation was required on the present facts.
112. The Statement of Facts and Issues before the House likewise identifies as the second of two issues "Whether and to what extent the [Secretary of State] was, in the circumstances of this case, obliged to conduct a further investigation into the attempted suicide …."; and both parties and the intervener accordingly addressed submissions on the form of any further investigation required.
113. In common, I understand, with all of your Lordships, I would reject the Secretary of State's submission that an article 2 investigation is only required where the State is in arguable breach of its substantive article 2 duty to protect life, in the sense that it ought arguably to have known of a real and immediate risk of a prisoner committing suicide and failed to take out reasonable preventive measures. While it is dangerous to generalise and I confine myself for the present to circumstances such as those of the present case, I agree that the relationship between the State and prisoners is such that the State is bound to conduct an article 2 compliant inquiry whenever its system for preventing suicide fails and as a result the prisoner suffers injuries in circumstances of near-suicide significantly affecting his or her ability to know, investigate, assess and/or take action by him or herself in relation to what has happened.
114. In general agreement with the reasons given by all of your Lordships, I therefore agree that an article 2 investigation was required in the present case. As to the nature of any inquiry required under article 2, and also as to the Court of Appeal's decision that a public hearing was required in the circumstances of R(D) v Secretary of State for the Home Department [2006] 3 All ER 946, I would also associate myself with the observations of my noble and learned friends, Lord Rodger of Earlsferry (in paras 69 to 79), Lord Walker of Gestingthorpe (in paras 91 to 96) and Lord Brown of Eaton-under-Heywood (in paras 103 to 109), which it would be unhelpful for me to seek to repeat or paraphrase.
115. Since the Secretary of State has actually undertaken to hold and is holding a D-type investigation into the circumstances of the present near-suicide in any event, it is immaterial that he may have been under no obligation to do so. On that basis and on the basis which I have indicated in the preceding paragraphs, I too would dismiss this appeal.