AM & Ors, R (on the application of) v Secretary of State for the Home Department & Ors

Article 3 of the European Convention on Human Rights provides:

Article 2 of the Convention begins:

A body of European and domestic case-law has established that, when there is credible evidence of a breach of art. 2, the state has an obligation to provide or to institute an effective official investigation. The purposes of such an investigation were described by Lord Bingham in R (Amin) v Home Secretary [2004] 1 AC 653, §31:

“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 learnt from his death may save the lives of others.”

It is also well established that an analogous duty is created by art. 3 where credible evidence suggests that one or more individuals have been subjected by or with the connivance of the state to treatment sufficiently grave to come within the article. The issue in this appeal is whether, on the evidence before the court about the disturbances at Harmondsworth Immigration Detention Centre in late November 2006, this obligation was engaged and, if engaged, discharged. A major aspect of this issue is whether there are differences of kind as well as of degree between the art. 2 obligation and the art. 3 obligation.

In a carefully reasoned but not unproblematical judgment, [2008] EWHC 11598 (Admin), Mitting J held that, but for the fact that the state had been alerted too late, the duty would have been engaged. The claimants appeal against the dismissal of their claim; the Home Secretary cross-appeals against the finding that the duty of investigation was in any event not fulfilled.

Kalyx Ltd, the second respondent, manages the Centre on behalf of the Home Office. It is accepted by both respondents that in doing so it is deploying the state’s powers and discharging the state’s obligations. It is also common ground that those held at the Centre are not serving custodial sentences but are in administrative detention pending a lawful disposal. The establishment is required by the Detention Centre Rules 2001 to “provide for the secure but humane accommodation of detained persons in a relaxed regime”.

The facts which are assumed for present purposes (because they are untested allegations) include the following:

7.1

In the period before the disturbances of 28-29 November 2006, there was a culture of oppression, bullying, violence and neglect, much of it confirmed by a report of the Inspector of Prisons, Anne Owers, compiled in July 2006 and published, coincidentally, on the first day of these disturbances.

7.2

In consequence of a disturbance which broke out in B wing during the evening of 28 November 2006, uninvolved detainees (including the claimant AM) were first kept in confinement while water from the sprinkler system entered their cells, then ordered out into the exercise yard in the cold while many of them were still wet, then readmitted and locked into cells. There AM and others were affected by smoke from a fire started by other inmates in an adjacent room; others were soaked by the sprinklers; there was reduced ventilation and, for many, a complete absence of toilet facilities. Some inmates spent well over 12 hours in these conditions without food or water. Two of the claimants, HM and LM, were assaulted by detention officers or rapid response personnel.

7.3

The dispersal of detainees which followed was in many cases carried out callously; some were transported long distances without their belongings.

It was not suggested on behalf of the Home Secretary that these events did not need to be taken seriously. Indeed it was a major element of the case advanced by James Eadie QC on her behalf that the Home Office had done exactly that by appointing a retired civil servant, Robert Whalley CB, to inquire into the disturbance and to report. It will be necessary to consider to what extent, if at all, his status and remit met the requirements of art.3. Mitting J (§30) took the view that the remit was insufficiently claim-specific; but in my judgment this was not the major problem with the Whalley investigation.

Before Mitting J, however, and again before us, Jim Sturman QC for Kalyx – but not Mr Eadie for the Home Secretary - has contended that none of the evidential material is capable, even if accepted, of crossing the threshold of art. 3. For this reason he submits not only that the present claim for an inquiry fails but that the individual claims for damages which by order of Black J stand adjourned should also be dismissed. His grounds are summarised in his skeleton argument in this way:

I will come in due course to the problem of demanding an investigative sledgehammer to crack a possibly modest nut. But Mr Sturman’s oral submissions served if anything to illustrate how much there was, in principle, to inquire into. He pointed to the undisputed evidence of his witness Mr Gomersall  that no detainee had taken the opportunity to give evidence to the Whalley inquiry; but, as Jessica Simor for the appellants pointed out in response, what Mr Gomersall had asked the Home Office to do was “to randomly select 5 or 6 detainees rather than advertising to all if you would find that easier”.

Mr Sturman went on to point to Mr Gomersall’s evidence that detainees were “not normally locked in”; but had at once to accept that this was a contested question, as well as that the rapid response unit had locked up uninvolved detainees. If there was a fire, as there was, then there was necessarily going to be water: the claimants, submitted Mr Sturman, could not complain of both. But there were very plainly complaints, which were not visibly inconsistent, about failures to protect detainees adequately from either.

All of this, and much else besides (including the report of the Chief Inspector of Prisons), far from supporting Mr Sturman’s claim to have an irrefutable answer to any allegation that art. 3 had been violated, showed how real the issues were. We did not find it necessary to hear Ms Simor in response.

For the rest, Mitting J would evidently have granted a mandatory order for an inquiry had it not been that, in his view, by the date (21 May 2007) when the letter before action was sent by Liberty’s in-house solicitor to the Home Office it was too late for a useful inquiry to take place. This was not a point which had been taken by the Home Secretary.

It is common ground that the state’s investigative obligation under art. 3 is not triggered until and unless the state knows of the facts or allegations warranting investigation. But although Mr Eadie had submitted that, at least, a police investigation capable of satisfying art.3 was still (in mid-2008) possible, the judge considered this unrealistic. Before this court, however, now in early 2009, Mr Eadie submits that it really is too late for any meaningful inquiry to take place; so that even if there was an earlier failure to implement the Convention obligation, no substantive relief should now be granted.

I have found some analytical difficulty with Mitting J’s approach to the question of time. At §32 he said:

32.

The problem which has arisen in this case has not, as far as I can tell, arisen in any other case; it is: what is the state's obligation of investigation where an allegation of a breach of Article 3 rights is communicated at a time when it would be difficult or impossible to undertake a worthwhile investigation? In this case, that circumstance has arisen because none of the claimants made complaints to the relevant authorities: to the police, to the operators of the detention centre, or to the Ombudsman, at a time when events were fresh in mind. No inquiry is now practicable with a view to fulfilling the primary obligation of identifying and punishing those responsible for breaches of Article 3, whether by criminal sanction or disciplinary sanction. All that can now in practice be achieved is redress by civil proceedings under section 8 of the Human Rights Act 1998.

The simple answer now advanced by Ms Simor is that as early as 1 February 2007 enough was manifestly  known to the Home Office to warrant an inquiry, because on that date the Home Secretary, acting by the Chief Executive of the Border and Immigration Agency, appointed Mr Whalley with the following terms of reference:

In his skeleton argument Mr Eadie suggests that, insofar as the allegations made by the claimants disclose “unintentional ill-treatment arising from the way in which the disturbance was dealt with”, the Home Secretary’s knowledge did not amount to awareness of any allegation of a breach of art.3. I do not accept this. While causing accidental or incidental suffering may well be less culpable than deliberate ill-treatment, it is still plainly capable of amounting, in a sufficiently serious case, to inhuman or degrading treatment – for example in a riot situation in which uninvolved prisoners have been locked up and forgotten or ignored while their cells filled with smoke or water. While it is not argued on the claimants’ behalf that the Home Secretary had imputed knowledge of such an allegation from the start, she incontestably had knowledge of it once she had received the request for an inquiry.

Ms Simor draws attention to the absence of the Home Office from the judge’s list of “relevant authorities” in §32. She also points out that the police investigation into criminal responsibility for the disturbance was still in its early stages in the spring of 2007, and that in April 2007 the police had received and looked into a complaint by a detainee about his treatment by a Kalyx officer.

In these circumstances it was hardly open to the Home Secretary to contend that by May 2007, when the letter before action came, it was not practicable to investigate the claimants’ allegations; and Mr Eadie on her behalf did not do so. It was Mitting J who took the point in his judgment, and it is, with respect, a proper criticism of him that he did so without alerting the parties to it. Canvassed as it has now been before us, it was not a sound point.

If nevertheless the judge was to take the view that by June 2008, when he gave his judgment, too much time had gone by to permit the grant of effective relief which his other findings warranted, this might have been a proper ground for the exercise of discretion to withhold at least mandatory relief, provided again that the question had first been canvassed with the parties. Although discretion was not the basis on which Mitting J dealt with the question of time, it will be an inescapable question for this court if an unmet art. 3 obligation is made out, as Mitting J held it was.

Before I turn to this, however, it is relevant to note two missed opportunities to comply with art. 3 which have become apparent in the present case and which hold very clear lessons for the future.

In spite of his undoubted experience and integrity, Mr Whalley manifestly lacked both the independence and the remit required for an investigation that would satisfy art. 3.  In Bati v Turkey (3.6.2004) the European Court of Human Rights held (§135) that if an investigation into art. 3 violations is to be effective

“the general rule is that the persons responsible for the inquiries and those conducting the investigation should be independent of anyone implicated in the events…. This means not only that there should be no hierarchical or institutional connections but also that the investigators should be independent in practice….”

Mr Whalley had spent the majority of his 36 years of public service in the Home Office, 7 of them in the Immigration and Nationality Department, and much of the rest as an adviser on security. He was clearly chosen because he was a government and Home Office insider, and he was given a remit which sought to address the management of detention centres rather than the protection of human rights either generally or (importantly) those of the claimants and other detainees whose entitlement not to be treated inhumanly or degradingly is in issue.

I have set out Mr Whalley’s terms of reference at §16 above. His report, thorough and conscientious as it was, can be seen by its opening summary of findings to have been almost entirely management-oriented:

Key Message: Continuing Risk of Disturbances:

6.

The Harmondsworth and Campsfield House disturbances were very different, both in causation and in how they unfolded. Both occurred at a time when recent population pressures, falling heavily on vulnerable fabric in a hard-pressed detention estate, were accompanied by dislocation in casework handling, especially in the case of Foreign National Prisoners, which caused a build up of latent tensions. (Foreign National Prisoners are defined for this purpose as ex-foreign national prisoner detainees.)

7.

It did not take much to trigger these events. When they started, they soon escalated, despite best efforts to prevent this happening. The underlying causes are still there and, without any changes, the same thing could happen again at either establishment.

8.

Action is proposed in respect of

•

procedures in both centres

•

improvements to the fabric of immigration removal centres

•

training for staff both in the centres and in caseworking units

•

oversight by BIA of the conduct of operations in centres

•

improved management of casework within BIA

•

better information for detainees and their legal representatives

Yet this was an inquiry which, if it had been put in the hands of someone not only suitably qualified but visibly independent of the Home Office, could have been required in the course of addressing the causes and handling of the disturbance to address the art.3 concerns: whether individual detainees were inhumanly or degradingly treated in dealing with it and, if they were, why it had happened and what could be done to prevent it happening again. It was a deplorably missed opportunity.

The Prison and Probation Ombudsman

The office of Prisons Ombudsman was created in 1999. When the National Probation Service was set up, probation was added to its remit, which was further enlarged in 2004 to include fatal incidents not only in prisons and probation hostels but in immigration removal centres. Then in October 2006, a month before the Harmondsworth disturbances, the Prisons and Probation Ombudsman (the PPO) was given responsibility (though apparently without additional resources) for investigating complaints by immigration detainees. The PPO had before then conducted and reported on two special investigations into the treatment of immigration detainees, one concerning Oakington Reception Centre and the other concerning Yarl’s Wood Immigration Removal Centre.

The PPO can become involved either at the request of the Home Secretary or on an individual complaint. A complaint, however, will ordinarily not be entertained by the PPO unless it has first been through the internal complaints procedure and is made within a month of management’s response or within six weeks of the complaint if there has been no response.

Without embarking on the question whether the availability of the PPO will nowadays always satisfy the state’s investigative obligation under art. 3, it is sufficient to record that the claimants have accepted that their complaints could have been dealt in conformity with art. 3 by an ombudsman investigation. This concession seems to me wisely made in the circumstances of the present case. It recognises the necessarily personal character of any violation of art. 3 without forfeiting the contention that a proper investigation of the violations alleged to have occurred in the course of the Harmondsworth disturbances would need to go into questions of behaviour, practice and culture of the kind highlighted in the report of the Chief Inspector of Prisons.

Nobody, however, appears to have thought of this at the time. Ms Simor tells us that, although the PPO’s enlarged remit was made public on the first day of the Harmondsworth disturbances, his website continued up to the date of the High Court hearing not to reflect it. Neither the Home Office nor Kalyx relied on, or even mentioned, the availability of the PPO in response to Liberty’s letter before action. It was the judge who first raised it at the hearing. No doubt for this reason the point has not been taken against the claimants either at first instance or on appeal.

This was nevertheless, in my view, another missed opportunity. I do not mean, by saying this, to allocate blame; only to record that an independent investigatory limb of the state, capable of fulfilling the art. 3 obligation in the present case, was at all material times available and, if used, would have made these proceedings unnecessary.

I turn therefore to the central issue canvassed on this appeal: do the availability of tort proceedings, the possibility of a criminal investigation and the actuality of the Whalley inquiry and report, either singly or in combination, amount in the present case to a fulfilment of the state’s obligation summarised in §2 above? Mitting J held that they did not. The Home Secretary contends that they do.

There is at least agreement between Ms Simor and Mr Eadie that the ECtHR jurisprudence establishes the following requirements (as formulated in Mr Eadie’s skeleton argument) of any art. 3 investigation:

41.1

The investigation should be capable of leading to the identification and punishment of those responsible;

41.2

It may generally be regarded as necessary for the persons responsible for and carrying out the investigation 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;

41.3

It must be effective in the sense that it is capable of leading to a determination of whether the force used was or was not justified in the circumstances;

41.4

It must be thorough, in that the authorities must make a serious attempt to find out what happened and should not rely on hasty or ill-founded conclusions to close their investigation or as the basis of their decisions. They must take all reasonable steps available to them to secure the evidence concerning the incident; and

42.5

It must permit effective access for the complainant to the investigatory procedure.

What remains in issue is the content of such an investigation.

Before embarking on the respective arguments, it is necessary to establish what they need to be directed to. There is no breach of art. 3 unless an individual is sufficiently ill-treated by or with the connivance of the state. But the nature of the state’s obligation to inquire into such possible breaches is case-specific. What will suffice for an isolated instance of inhuman or degrading treatment (which may be a prosecution or a civil action, at least if one is brought) will not necessarily suffice for systemic and multiple breaches of art. 3 such as are alleged here. The reason is obvious: litigation is designed to secure individual redress, prosecution to establish individual culpability. Neither is in the ordinary way equipped to make the appraisal of culture and system which the Inspector of Prisons and, to a limited extent, Mr Whalley undertook and which would be an essential part of any such inquiry as Mitting J was asked to order here. It is essential because the art. 3 case advanced on behalf of the three claimants is not simply about what happened to them: it is about why it happened, and about why what happened was in their submission not accidental – a contention to which the inspection report gives substance. These issues, all of which lie potentially within the investigative ambit of art. 3, were in my judgment brought to life by the case put in Liberty’s letter before action. It has to be remembered too that, insofar as such a case raises questions of discrimination on grounds of race or nationality, art. 14 becomes engaged alongside art. 3.

Judicial review, for its part, is not concerned, except incidentally, with the vindication of private law rights. It is concerned with public law wrongs, in this case the asserted failure of the Home Secretary to set up an inquiry. The relevance of the claimants’ personal rights is not that without them there would be no claim for judicial review. Any person or body with a sufficient interest can seek judicial review. The claimants’ interest is that they say they were among those who suffered violations of their right not to be inhumanly or degradingly treated; but there is no requirement of law that their claim for judicial review must be confined to their private law or personal interests.

It follows, in my judgment, that arguments about the degree of involvement or suffering of the particular claimants – unless these were to found a submission that they are mere meddlers who lack standing – are largely irrelevant. What matters is whether the entirety of what they have now brought to the court’s attention requires, or at some point required, the Home Secretary to set up an inquiry.

The principal foundation of the Home Secretary’s position on the sufficiency of the extant procedures is Banks v United Kingdom (Application 21387/05; 6 February 2007). Mr Banks was one of 10 applicants whose claims were held inadmissible by a chamber of the European Court of Human Rights. Their application was argued without opposition, but failed. Ms Simor seeks to marginalise the decision as simply an admissibility decision. This may well be an appropriate caveat where a case has been declared admissible, since full argument will not yet have taken place. But where, as here, a claim is held to be manifestly ill-founded and so not to require an inter partes hearing, there is nothing provisional about the decision. What can, however, be cogently said is that in an adverse decision on admissibility the Court is highly unlikely to innovate. If such a decision appears to contain a proposition for which no previous decision of the Court provides support, it needs to be approached with great caution.

Not only are there serious questions about what Banks decides: if the Home Secretary is right in at least one regard, the decision may conflict with the domestic law set out in Amin (see §3 above). If so, our obligation under s.2(1) of the Human Rights Act 1998 to “take into account” the Strasbourg jurisprudence will have to give way to the discipline of domestic law: see Kay v Lambeth LBC [2006] UKHL 10, 2 AC 465, §43.

Nine of the applicants in Banks were former prisoners in HMP Wormwood Scrubs who credibly asserted that, in the late 1990s and thus before the coming into effect of the Human Rights Act, they had been violently and degradingly treated by prison officers. The tenth was the brother of a prisoner who had been found hanged in circumstances which the applicant contended were suspicious. The Chief Inspector of Prisons had subsequently issued a highly critical report on the prison, and two police investigations into allegations made by 13 Wormwood Scrubs prisoners had resulted in the prosecution of 27 officers, 6 of whom were convicted. There had also been an internal inquiry, of which more below.

Of the nine prisoners in Banks, three officers who had assaulted the first had been given substantial gaol terms; those alleged to have assaulted the third had been tried and acquitted; and the trials of those alleged to have assaulted the sixth and eighth had been discontinued. Eight had brought civil actions for damages which were settled for very substantial sums by the Home Office without going to trial.

The tenth applicant had no civil claim arising out of his brother’s death. The police were still considering whether there was sufficient evidence arising out of the compromised civil actions – including evidence of mock hangings -to found a criminal prosecution. There had been an inquest which resulted in an open verdict.

In this situation all ten applicants were seeking a finding that art. 3, with or without art. 13, and in the tenth applicant’s case art. 2, required the United Kingdom to hold an independent public inquiry into the treatment of prisoners in Wormwood Scrubs in the material period.

The Court did not demur to the proposition that what was asserted amounted to individual and systemic breaches of art. 3, but it pointed out that, in contrast to art. 2 which almost by definition dealt with cases in which the immediate victim was no longer alive, art. 3 generally affected individuals who were still able to act on their own behalf. It followed, first, that in some cases art. 13 – the right to an effective remedy - afforded all the required protection. The Court explained:

“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 [viz procedural complaints] at the relevant time (see Khashiyev v Russia …§178 …Ilhan v Turkey [GC] … §89-92).

In the present case the Court observes that the applicants do not raise any substantive complaints of ill-treatment under Article 3 of the Convention. If they had, issues would have arisen as to whether they could still claim to be victims or had in fact exhausted domestic remedies since six [sic] applicants settled the civil proceedings which they had brought alleging assault by prison officers and systemic negligence on the part of the prison service, while the other two [sic] did not bring any such proceedings … Further, while the applicants raise complaints as to whether the investigations into their allegations complied with the standard required by the procedural obligation, this is not a case where there was a lack of any investigation capable of establishing the facts and attributing responsibility (see Assenov v Bulgaria …§99-106).”

Pausing here, it is evident that, precisely because the wrongs done to them had been the subject both of criminal prosecution and of civil actions, the majority of the applicants in Banks were seeking only an inquiry into the situation which had enabled these things to happen: “identifying the root causes of the culture of abuse that existed,” as the Court put it, “and the means to ensure their eradication”. Having noted this, the Court went on to say – in a passage which I admit I do not find at all easy to understand – that it was “[a]ccordingly … not persuaded that this is a case in which issues arise under the procedural head of Article 3 of the Convention and … that the applicants’ complaints fall rather to be considered under Article 13”. It would have seemed to me, with great respect, that the reverse was the case.

Next, however, the Court made that very assumption and on that basis went on, as it put it, to make three points. To an English lawyer at least (and the United Kingdom judge, Sir Nicolas Bratza, was of course a member of the chamber) what follows will be obiter dicta.

First, the Court said, the strongest guarantee of effective fact-finding and allocation of criminal responsibility is criminal process. Secondly, where intentional violence is not the issue, a civil or disciplinary remedy may “be sufficient to provide protection” under art.2 or, by parity of reasoning, art. 3, especially where, as here, the claims were able to include systemic failings. Thirdly, the minimum procedural requirement of art. 3 was

“that where a State or its agents potentially bear responsibility for serious ill-treatment, the events in question should be subject to an effective investigation or scrutiny which enables the facts to become known.”

There was, said the Court, no indication that in the cases before the Court this objective had not already been achieved in terms both of investigation and of responsibility. But it continued:

“The wider questions raised by the case as to the background of the assaults and the remedial measures apt to prevent any recurrence in a prison in the future are, in the Court’s opinion, matters for public and political debate which fall outside the scope of Article 3 of the Convention (see mutatis mutandis, Taylor family and others v United Kingdom, no.23412/94, Comm. Dec. 30.8.94, D.R.79, p.127).”

I am bound to say, with respect, that I find this a surprising proposition, both in the light of the Court’s own jurisprudence and in the light of United Kingdom authority derived from it. Taking the approach I have suggested in §32 above, one needs to look first at the single decision cited by the Court itself.

Taylor v United Kingdom was a 1994 decision of the full Commission. It concerned the victims of a trainee nurse, Beverley Allitt, who had killed four children, had attempted to murder three others and had seriously injured another six. There had been an official inquiry which made its report public but which sat in private and had no powers of compulsion. The claim before the Commission was of a breach of art. 2, notwithstanding that the lives of six of the victims had not been either taken or threatened; but there is no reason to think that an art. 3 claim in relation to these would have fared differently.

The Commission concluded that the prosecution of Beverley Allitt, the inquiry by an independent panel (which had had the willing cooperation of everyone it wanted to hear from) and the availability of civil proceedings for negligence on the part of the health authority together amounted to compliance on the part of the state with its investigative duty under art. 2. This much can be regarded as a purely factual determination. But the Commission in its penultimate paragraph went on to say this:

“The Commission acknowledges that neither the criminal proceedings nor the Inquiry addressed the wider issues relating to the organization and funding of the National Health Service as a whole or the pressures which might have led to a ward being run subject to the shortcomings apparent on Ward Four.  The procedural element contained in Article 2 (Art. 2) of the Convention however imposes the minimum requirement that where a State or its agents potentially bear responsibility for loss of life the events in question should be subject to an effective investigation or scrutiny which enables the facts to become known to the public, and in particular to the relatives of any victims. The Commission finds no indication that the facts of this case have not been sufficiently investigated and disclosed, or that there has been any failure to provide a mechanism whereby those with criminal or civil responsibility may be held answerable. The wider questions raised by the case are within the public domain and any doubts which may consequently arise as to policies adopted in the field of public health are, in the Commission's opinion, matters for public and political debate which fall outside the scope of Article 2 (Art. 2) and the other provisions of the Convention.

If this is a definitive statement of the law, it appears to draw a bright line between the immediate circumstances of a violation of art. 2 or art. 3 and the systemic or larger issues which may underlie and explain the violation.  Banks in that case represents no more than a reiteration of this distinction.

One turns therefore to decisions which appear to controvert or at least not to support the dichotomy suggested by Banks and Taylor. Ms Simor has furnished us with a schedule of relevant Strasbourg decisions. A number of them hold that the right given by art. 13 to an effective remedy goes beyond compensation to “a thorough and effective investigation” of the material events, but that the purpose of such an investigation is “the identification and punishment of those responsible”, not – or not necessarily - a wider inquiry: see Aksoy v Turkey (1996) 23 EHRR 553, §98; Keenan v United Kingdom (2001)  33 EHRR 913, §131; Khashiyev v Russia (2005) 42 EHRR 397, §177.

As to the ambit of art. 3, some guidance is afforded by the decision in Bati v Turkey (3.6.2004). The Court (§123) found the Turkish state to have been complicit in torture and hence in breach of the substantive limb of art. 3. It then, however (§126-7), held that the complaint of a failure to investigate was more appropriately considered under art. 13 than under art. 3. Even under this rubric, it nevertheless concluded (§137) that:

“…. 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. The degree of public scrutiny required may well vary from case to case. In all cases, however, the complainant must be afforded effective access to the investigatory procedure…”

In Šecic v Croatia (31.5.2007), §59, the Court’s conclusion was that it was “the failure of the State authorities to further the case or obtain any tangible evidence with a view to identifying the attackers” that put the state in breach of the procedural limb of art. 3. That this is, however, not a comprehensive account of the procedural requirement is shown by the Court’s proposition in Khashiyev (above) §178 that

“The procedural limb of art. 3 is invoked, in particular, where the Court is unable to reach any conclusions as to whether there has been treatment prohibited by art. 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.”

If there is a focus in these decisions on finding and punishing the guilty party – something which a well-conducted criminal process can be expected best to achieve – it is because in many of these cases there has been an associated art. 13 issue arising from the failure of the prosecuting authorities to act at all. But the more weight is placed on art. 13 as a primary investigative requirement, the less need there is to deduce such a requirement from art. 3.  If there is not in at least some instances a distinct obligation to investigate the larger circumstances surrounding a violation of art. 3, it is difficult to see what the procedural aspect of that article, or of art. 2, adds to art. 13.

This thinking is clearly reflected in the United Kingdom decisions which have sought to give effect to the Strasbourg jurisprudence. Some of these concern art. 2, some art. 3, some both.

It has been proposed in argument that the jurisprudence governing the investigative duty under art. 2 should not, or not necessarily, govern the same duty under art. 3. The principal reason is that in the art. 2 situation the immediate victim is by definition unable to advance his or her own case, while in the art. 3 situation direct recourse to law is generally open. I believe this to be a false dichotomy. There are cases, such as Keenan v United Kingdom, in which a death has resulted but the causative breach has been of art. 3, and others in which the victim has survived (for example some of the children in Taylor v United  Kingdom) but the claim has lain under art. 2. It is  therefore understandable that in no decision that we have been shown does the Court make any such formal distinction. What in my judgment will always matter, whether the issue arises under art. 2 or art. 3, is the practical ability, whether of the victim or of others, to secure adequate investigation, retribution and redress without an ad hoc inquiry instituted by the state itself.

What then makes investigation adequate? So far as domestic law is concerned I do not think one can improve upon what Jackson J said in R (Wright)v Home Secretary [2001] EWHC Admin 520, [2001] UKHRR 1399: that an art. 2 or art. 3 investigation “is required in order to maximise future compliance with those articles.” The purpose, in other words, is neither purely compensatory nor purely retributive; nor is it necessarily restricted to what has happened to the particular victim. Nor, however, is it to usurp the role of government. It is to inform the public and its government about what may have gone wrong in relation to an important civic and international obligation and about what can be done to stop it happening again.

This was unmistakably the view of the House of Lords in R (Amin) v Home Secretary [2003] UKHL 51, [2004] 1 AC 653. The claim before the House was for an art. 2-compliant inquiry into the beating to death of a young Asian offender by a cellmate who was known to be a violent racist. Lord Bingham, giving the leading speech, accepted that there was no single model of Convention-compliant investigation but accepted too that there were minimum standards of compliance. He endorsed what Jackson J had said in Wright and formulated his conclusion, derived from the Strasbourg jurisprudence, in the terms set out in §3 of this judgment.

It is significant, in the light of the foregoing discussion, that that formulation includes ensuring, so far as possible, that the full facts are brought to light and that lessons will be learned and implemented. Both of these objectives go markedly beyond the identification and punishment of those responsible. They are reflected both in Lord Bingham’s explicit reference in §21 to the particular anxiety which may attend a systemic failure to protect life and in Hooper J’s reasons (endorsed by the House) for directing the holding of an inquiry:

“It is accepted that [the victim] was put in the same cell as his killer because of ‘systemic failures’. Established procedures were not followed and there is an appalling history at Feltham of failure to comply with earlier recommendations. It seems likely (and is certainly arguable) that there were serious human failings … which have not been publicly identified. On the facts of this case the obligation to hold an effective and thorough investigation can, in my judgment, only be met by holding a public and independent investigation …”

For reasons I have given, there is no reason in principle to draw a line in this regard between art. 2 and art. 3. So long as the minimum requirements are met, the distinction between a need for an independent ad hoc inquiry and the satisfaction of the investigative obligation through existing procedures is a fact-sensitive and pragmatic one. But our domestic jurisprudence, including the binding decision of the House of Lords in Amin, makes it clear that the investigative obligation of the state may – depending on what facts are at issue – go well beyond the ascertainment of individual fault and reach questions of system, management and institutional culture. Insofar as this goes beyond the jurisprudence of the Strasbourg court (and I am not persuaded that it does), it is domestic authority which we are bound to follow: see Kay (above), §43.

In this light there will nevertheless be a good many cases in which, although they involve allegations of inhuman or degrading treatment at the hands of the state, a civil action for damages, especially if it includes aggravated or even exemplary damages, or a properly conducted prosecution or disciplining of the offender, will achieve as much as art. 3 procedurally requires. It is likely that the average wrongful arrest, assault or false imprisonment claim against individual police officers will fall into this class. It may also be that where for good reason the view is taken that such a claim will fail, there will be no case for an inquiry in lieu; but the reasons for taking this view will matter. Where a civil claim is made and settled, much may turn on the terms of settlement, which may range from tokenism to full acknowledgment. I mention these random possibilities to make it clear that the state’s investigatory function will often be discharged by the ordinary processes of law, and that a claimant who does not attempt to use these may be held not to have exhausted his domestic remedies.

This leaves at least one unanswered question: is a civil court before which proceedings of this character are brought required by s.6(1) and (3)(a) of the Human Rights Act to adapt its procedures and to enlarge its inquiry so as to fulfil the state’s procedural duty under art. 3, even if both tort law and case management point against it? If not, then in any case where a larger inquiry is necessary, civil proceedings cannot discharge the state’s art. 3 function.

Mitting J, who will have been conscious of these issues, considered in sequence the three forms of recourse on which, singly and in combination, the Home Secretary relied as constituting compliance with art.3:

27.

There can be no doubt that a fully contested civil claim, by a court having available to it the right to make a declaration as well as to award monetary compensation, and having power to order disclosure of documents for the provision of further information, and which will hear evidence from witnesses who will be cross-examined, can determine the facts of an individual claim and grant relief sufficient to satisfy the substantive Article 3 rights. I have no doubt that a full hearing of these claimants claims could achieve that, but it will still not satisfy the state's investigative obligation, and I do not understand the Strasbourg Court to have watered down the requirement for such an investigation by the use of the word "scrutiny" in Banks. It is therefore necessary to examine what investigation was carried out into what by the police and by Mr Whalley.

28.

The police investigated possible criminal offences committed by detainees. As far as can be ascertained, no complaints were made to the police by detainees of criminal conduct by the prison or detention centre officers. Certainly, no inquiry was conducted by the police into such claims.

29.

Mr Whalley was appointed-

"to investigate the circumstances of the disturbance at Harmondsworth Immigration Removal Centre on 28/29 November 2006;

to establish the lessons to be learnt from this event for the management of immigration detainees and for the immigration detention estate; and

to report to you accordingly."

I see little reason to disagree with any of this. To a considerable extent it represents fact-finding, with which we should not interfere without solid cause. But it also seems to me to reflect the realities of the case.

Can two or more of these forms of intervention then be combined so as to satisfy art. 3? I do not doubt that in some cases this is possible, but I can see no way in which it can be done here. The police have, perhaps understandably, shown little interest in prosecuting anyone other than the detainees who were responsible for the disturbance. The civil actions of the claimants seek compensation for harm done to each of them personally. They may possibly shed some light on what was going on at Harmondsworth, but that is not their purpose. Mr Whalley’s inquiry, for the reasons the judge gives, had a purpose which went wide of art. 3.

The upshot is that the issues raised by the claimants about the culture and conduct of the Harmondsworth management and staff, as manifested both in the handling of the disturbance and in the run-up to it, have not been and are highly unlikely now to be addressed in any conventional forum to which the claimants have access.

In my judgment Mitting J was right to find that the issues raised by the claimants, especially in the context of what was by then known from the report of the Inspector of Prisons, were such as to trigger the state’s obligation under art.3 to investigate what had arguably been inhuman or degrading treatment, both reactive and systemic, in a custodial institution. He was entitled also to find that neither the possibility of a criminal investigation, nor the bringing of civil proceedings, nor the conduct of an internal Home Office inquiry, nor any combination of these, had satisfied the state’s duty. He was wrong, in the circumstances I have described, to hold that the assertions had reached the Home Office too late to be given effect.

But the question of time is still a real one for this court. It is, moreover, not only a question of time. A full independent inquiry has major resource implications for any state. Although it is not suggested either that the role of the Inspector of Prisons is a surrogate for such an inquiry, or that the missed opportunity of alerting the Prisons and Probation Ombudsman makes the claim for an inquiry untenable, it remains a relevant fact that the functioning of Harmondsworth had just been rigorously reported on by the former, and a relevant consideration that, albeit by misfortune, the potential role of the latter in fulfilment of the state’s art. 3 duty was overlooked.

In my judgment the right course, in the fresh exercise by this court of the High Court’s discretion, is not to make the mandatory order which is sought, and to limit the relief it grants to a declaration that the Home Secretary (since the common law does not know an entity called the state) has failed to meet the United Kingdom’s obligation under article 3 of the ECHR to institute an independent inquiry, to which the claimants would have full access and which would make its findings public, into the handling of the disturbance which took place at Harmondsworth Immigration Detention Centre on 28-29 November 2006.

The law about the power (and indeed the duty) of the court to order the Home Secretary to set up an effective official investigation into alleged breaches of Articles 2 and 3 of the European Convention of Human Rights has developed rapidly since the United Kingdom enacted the Human Rights Act 1998. It may be said to have begun with the judgment of Jackson J in R (Margaret Wright) v SSHD [2001] EWHC Admin 520 in which he set out the state’s obligation in the following words (para. 43)

“1.

Articles 2 and 3 enshrine fundamental human rights. When it is arguable that there has been a breach of either article, the state has an obligation to procure an effective official investigation.

2.

The obligation to procure an effective official investigation arises by necessary implication in articles 2 and 3. Such investigation is required, in order to maximise future compliance with those articles.

3.

There is no universal set of rules for the form which an effective official investigation must take. The form which the investigation takes will depend on the facts of the case and the procedures available in the particular state.”

In R (Amin) v Home Secretary [2004] 1 AC 653, 668 Lord Bingham of Cornhill described Jackson J’s judgment as succinct and accurate and approved Jackson J’s description of the features which an investigation had to have “to satisfy Article 2” namely that the investigation must be independent, effective, reasonably prompt, have a sufficient element of public scrutiny and appropriately involve the next-of-kin. This was followed by R (Green) v Police Complaints Authority [2004] 1 WLR 725 where the allegation was that a police officer had attempted to kill or cause grievous bodily harm to the complainant.

Liberty has collected evidence from 3 detainees at the time of the Harmondsworth riot in November 2006 who, it is said, on the one hand were incarcerated for too long a period in their cells while fire was raging and water, used to quench the fire, was seeping into their cells or, on the other hand, locked out of their cells for too long a period on a cold night in the open air and in both cases having insufficient food and water. There are two alleged instances of inmates being assaulted with no justification. The Secretary of State accepts that these two alleged assaults constitute an arguable breach of Article 3, although he asserts that the other complaints do not cross the Article 3 threshold. Mr Sturman QC’s attempt for Kalyx to argue that the Secretary of State’s concession in relation to the assaults was wrong cannot, in my view, be correct.

Building on the case law, Liberty wrote a long letter to the Home Secretary on 21^st May 2007 calling on him to conduct a “full public inquiry” into the underlying cause of the disturbance and the treatment of detainees while it took place. They suggested the following terms of reference:-

“1.

To inquire into the circumstances leading up to the disturbance at Harmondsworth Immigration Removal Centre that begun on 28^th November 2006, including the conditions of detention and treatment of detainees by immigration staff, with the aim of establishing the underlying reasons why the disturbance took place.

2.

To investigate the manner in which the disturbance was managed including the treatment of detainees by immigration officers, contract personnel, prison service officers, police and any other non-detained persons during the disturbance and its aftermath.

3.

To investigate into the adequacy of the systems and procedures put in place by the Home Office to deal with disturbances in immigration detention centres.

4.

To recommend what steps should be taken to prevent such a disturbance happening again and to ensure that the safety of immigration detainees is not compromised in the future, and to report its findings as soon as possible. The public inquiry should be chaired by a senior member of the judiciary”

Speaking only for myself I am left somewhat uneasy. Can it really be the law that it is the duty of the state to conduct or the court to order “a full public inquiry” (whatever that may precisely mean) whenever an allegation of a breach of Article 3 is made? I cannot bring myself to believe that it is.

In the first place it must be remembered that the trigger for the duty and/or power to order an effective official investigation is the particular alleged breach of Article 2 or 3 as the case may be. The court has no general power to order an inquiry merely because the court may consider it appropriate in all the circumstances. The maximum, therefore, that the court could do in the present case is, in my view, to order an inquiry into the particular treatment suffered by the complainants which is alleged to constitute a breach of Article 3; without the peg of particular breaches in respect of particular victims, there would be no power to order an inquiry “into the circumstances leading up to the disturbance” or an inquiry intended to establish “the underlying reasons why the disturbance took place”.

Secondly the authorities do not seem to me to go as far as Liberty suggests. Both Wright and Amin were cases of death in custody which merited a full independent inquiry into the circumstances of the death. Green was a case of alleged attempted murder or grievous bodily harm. It is not difficult to imagine cases of alleged Article 3 mistreatment (such as torture or the infliction of serious harm) which would merit a full independent inquiry. But every allegation of a breach of Article 3? There must surely be some sense of proportion. If Jackson J intended to say that any alleged breach of Article 3 required an effective official investigation, he went further than he needed to and I would respectfully disagree. It is noteworthy that when Lord Bingham approved Jackson J’s judgment he specifically enumerated the features which the investigation must have “to satisfy Article 2”. I do not read his speech as saying that the same features must apply to Article 3. Green proceeded on the assumption that Article 3 was engaged without any exploration of its precise ambit, see para 58 of the speech of Lord Rodger of Earlsferry.

Nor do later cases support Liberty’s contentions. In Banks v UK (Application No. 21387/05, decided 10^th June 2005) a number of prisoners incarcerated in Wormwood Scrubs made allegations of assaults and ill-treatment by prison officers. They complained that there had been a failure to provide an adequate investigation into their allegations and that the only means of ensuring compliance with the procedural obligation of Article 3 was for there to be an independent public inquiry

“to establish the factual background, the full nature and extent of the culture of violence at Wormwood Scrubs in the 1990’s how this took root and prospered and the extent to which it continues and establishing responsibility for the above.” (page 9)

The European Court of Human Rights (4^th Section) decided that the application was inadmissible. They considered that the complaint related more properly to Article 13 (requiring the provision of an effective remedy for breach of Article 3) rather than to the procedural obligation of Article 3 itself, but they decided that, even if the procedural aspect of Article 3 was engaged, the complaint was inadmissible because

The application in the present case is remarkably similar to that in Banks. Ms Simor for the appellants submitted that decisions on admissibility do not change the established principles laid down by European Court and that cases such as Jordan and Edwards followed by the House of Lords in Amin required a public inquiry to be held. But Jordan and Edwards, like Amin, were Article 2 cases. The European Court in Banks considered that those cases were inapplicable to Article 3 cases and that established principle meant that a combination of the availability of a criminal trial and civil proceedings would normally be adequate for Article 3. It is noteworthy that Lord Rodger of Earlsferry contemplated the possibility that this might even be an appropriate approach in an Article 2 case where the applicant had survived life threatening injuries but was in a position to give proper instructions to his legal representatives, see R (L) v Secretary of State for Justice [2008] UKHL 68, 3 WLR 1325, paras 67-73.

It seems to me that in the present case the allegations of breach of Article 3 can be properly dealt with by the combination of the availability of criminal proceedings and civil proceedings, just as the allegations in Banks could be properly dealt with. The availability of those proceedings thus constitutes compliance with the procedural obligation of Article 3 on the facts of this case. The focus of any inquiry which the court is empowered to order has to be on the alleged breach of Article 3. The wider inquiry which Liberty wants is no doubt “a matter for public and political debate” but does not fall within Article 3.

My Lord takes the view that criminal and civil court proceedings will not ordinarily suffice when allegations of systemic and multiple breaches of Article 3 are made. I think that this puts the matter too widely. If a particular individual or, as in this case, three individuals make one or more allegations of conduct amounting to a breach of Article 3, I see no reason why they cannot be investigated by the police and the courts in the ordinary way. It cannot be right, in my view, that merely by adding an allegation that the conduct is systemic one can be entitled to a public inquiry. There can hardly be a requirement for a public inquiry every time somebody plausibly alleges institutional violence or institutional racism on the part of the authorities. Unless the state’s recognised ways of investigating such allegations by the use of legal proceedings or the Ombudsman are appropriate, there will be a risk that there will be considerable public expenditure to little purpose.

It is, of course, true that the lapse of time since November 2006 may make it difficult for criminal and civil proceedings now to be effective. But that lapse of time is in no way the fault of the state. One recognises that some people may be reluctant (for a range of reasons) to make a complaint to the police or to institute civil proceedings but for such persons a complaint to the Ombudsman is an available recourse. When the state provides proper procedures for the investigation of breach of Article 3, but those procedures are not used promptly and time then passes, that cannot then itself be a reason to have a public inquiry.

I have no quarrel with the proposition in paragraph 60 of my Lord’s judgment that the investigative obligation of the state may – depending on what facts are at issue – go beyond the ascertainment of individual fault and reach questions of system management and institutional culture especially in cases of death or severe physical injury while in the custody of the state. But I regret I cannot follow him to the conclusion in para 67 that the issues raised by the claimants were such – on the facts of this particular case – as to trigger the state’s obligation under Article 3 to investigate what they assert to be inhuman or degrading treatment at Harmondsworth. It is not difficult to predict that, if this court were to order an inquiry, the individual grievance of the 3 claimants would quickly be submerged in a welter of allegations about systemic ill-treatment of numerous other persons and its causes, together with arguments about the appropriate way of treating people subject to immigration detention and indeed whether such detention should be used at all for what can sometimes be lengthy periods of time. These are, of course, all legitimate subjects of debate but they are a long way from the specific allegations of ill-treatment of the three individual claimants on the particular night of the riot in November 2006. Judges should, in my view, be wary of becoming embroiled in arguments about the appropriateness of public inquiries which are liable to go well beyond the investigation of particular breaches of Article 3.

There must also be a margin of appreciation for the Secretary of State to decide when to hold and when not to hold a public inquiry. The resource implications can be considerable. The Secretary of State’s decision in the present case seems to me to be within the margin she must undoubtedly have.

I would therefore dismiss this appeal not (as the judge thought) because any inquiry would now come too late, although I can readily understand that conclusion, but rather because I see no reason why the legitimate Article 3 complaints could not be dealt with by recourse to the ordinary processes of law available in the United Kingdom. It follows that, if left to myself, I would not make the declaration proposed by my Lord in his final paragraph.

I am grateful to Lord Justice Sedley for setting out the legal and factual background, which I adopt. The question in this case is whether the Secretary of State was required by law to set up an independent investigation into the alleged breaches of Article 3 which occurred during the disturbance at the Harmondsworth Immigration Detention Centre on 28-29 November 2006. A secondary question is whether, even assuming that she was, it is now too late for her to do so. Mitting J held that the Secretary of State ought to have set up such an investigation, but that it was plainly too late for it to serve any useful purpose at least by the time of his judgment in June 2008. The appellants claim that he ought not to have concluded that it was too late, and that the judge erred in other, more peripheral, ways. The respondents cross appeal on the grounds that the judge ought not to have found that any independent investigation was required at all and that the civil proceedings which the appellants are proposing to take, particularly if combined with police investigations into possible criminal charges, were sufficient to satisfy the Article 3 procedural obligations.

Initially Liberty had asked the Secretary of State to conduct a very wide ranging public inquiry under section 10 of the Public Inquiries Act 2005. The nature and purpose of the inquiry they sought is set out in the judgment of Longmore LJ at paragraph 4 and I will not repeat it. Suffice it to say that even if the appellants are correct to say that some form of investigation should have been, or should now be, initiated, this would go far beyond anything that Article 3 would require. Any Article 3 investigation would, however, have to meet the minimum requirements identified by Lord Bingham of Cornhill in R (Amin) v Secretary of Statefor the Home Department [2003] UKHL 55 [2004] 1 AC 653, para 22. In essence these are that it should be initiated by the state promptly; should be effective, in the sense that it is thorough and capable of leading to the identification of those responsible for any Article 3 breaches; should be carried out by someone independent of those implicated; should involve the victims to the extent necessary to involve their interests; and should have a sufficient element of public scrutiny. Amin was an Article 2 case, but the same essential ingredients apply to an Article 3 investigation: see R (Green) v Police Complaints Authority [2004] UKHL 6; [2004] 1 WLR 725 per Lord Rodger of Earslferry at paragraph 59 and Lord Carswell at paragraph 84. It may be that during the course of any such investigation it would become apparent that something more was required. However, at this stage I consider that the only issue we should determine is whether in principle an independent investigation meeting those minimum standards should have been set up, and if it should, whether it is now too late to do so.

There are two features about the case that I would wish to emphasise. First, the appellants were at the time of the disturbance in detention. It is not a prison, and the objective is to provide a relaxed regime with as much freedom of movement and association as is compatible with the need to maintain a safe and secure environment. Sadly, this has not been achieved. The reports from the Inspector of Prisons have demonstrated that in practice standards have fallen lamentably short of that objective. A report made following an inspection shortly before the disturbance noted that officers ran the Centre as if it were a high security prison and that there were high levels of oppression and bullying. The inmates are in a subordinate and difficult situation - there was much evidence, for example, that many were afraid to complain of ill treatment - and they face greater difficulties in pursuing grievances effectively than would potential Article 3 victims who were not so detained.

Second, the nature of the Article 3 complaints is not merely that there was specific deliberate ill treatment such as physical assaults. There are also allegations of ill treatment amounting to a breach of Article 3 arising from the manner in which the disturbance was managed and controlled. The allegations include being locked in a cell for hours without food or access to a toilet; being kept outside in the freezing cold for many hours; and lack of food or changes of clothing. In part the allegations are that some of this treatment was deliberate and intended to degrade and humiliate. But it is also alleged that whether that was the intention or not, it was the effect of what was done and therefore still infringed Article 3. Kalyx contended that this treatment did not cross the threshold of establishing even an arguable breach of Article 3. The Secretary of State agreed that it did -although it was not clear how far that concession extended beyond the actual allegations of intentional physical assault. In my judgment, for the reasons given by Sedley LJ, the potential breach of Article 3 extends to the whole range of the alleged treatment in issue. That seems to have been the premise on which the judge below determined the case, and I think that he was right to do so. Of course, as the judge also noted, it may well be that on investigation it would be demonstrated that this treatment could be justified given the difficult situation facing the officers, and would not in fact pass the Article 3 threshold. But the only question at this stage, when determining whether an investigation was required, is whether the alleged breaches arguably constituted Article 3 breaches.

There are two important preliminary questions which are the subject of appeal. The first is this: what is the range of alleged ill treatment which the Secretary of State should have considered when determining whether to hold an investigation? There are only the three appellants who have in fact pursued a case before the courts that their Article 3 rights were infringed by the failure to hold the investigation, and Mitting J focused on the wrongs done to them.

Ms Simor submitted that there were many others whose rights had been equally infringed and that the whole range of conduct should be taken into account when analysing whether an investigation was required. It was an arbitrary decision merely to identify three particular claimants; many more could have been added to the proceedings. The judge was wrong to limit the potentially relevant evidence in the way that he did. Moreover, he failed to identify other potentially relevant breaches, including allegations of discriminatory conduct contrary to Article 14, and incidents pre-dating and post-dating the disturbance itself.

I do not accept the full scope of that submission. The obligation to carry out an investigation is a procedural one which is parasitic on alleged substantive breaches of the Article: see the observations of Lord Bingham of Cornhill R (Gentle) v Prime Minister [2008] UKHL 20;[2008] 1 AC 1356 para 6. The nature of that obligation is inextricably linked to the specific nature of the alleged breaches. Lord Justice Sedley is plainly right, with respect, to say that public law rights can be vindicated by anyone with a sufficient interest, whether their private law rights are affected or not. But the question whether there is a duty to set up an investigation under Article 3, and if so what form it should take, can only be answered in the context of specific and particularised complaints. I do not intend to suggest that the Secretary of State should have considered the allegations relating to each claimant in isolation, nor that he should have ignored the wider context in which the alleged mistreatment occurred. I also accept that he should have had regard to other particularised claims even though the accounts were given by anonymous witnesses. But he would have to determine the issue on the basis of specific allegations of mistreatment of particular individuals, not generalised statements to that effect. In this case particular weight in the Liberty letter to the Secretary of State was placed upon the complaint of these three complainants. For this reason, I do not share Sedley LJ’s observation that the particular suffering of the particular appellants is “largely irrelevant”. They were the focus of Liberty’s complaint. Moreover, I suspect that they were the focus of the case below - that was plainly the judge’s understanding - in which case it is unfair to criticise the judge on this score.

Nor in my judgment is there anything in the point that the judge did not identify the full range of alleged breaches in his judgment. It was unnecessary for him to do so since he found that there was a breach of the duty to hold an investigation in any event. Equally, however, it is plain that he would still have concluded that any investigation would have been to no purpose even had he specifically addressed the full range of alleged wrongdoing.

As to the failure to have regard to other incidents which occurred before and after the disturbance, these again were not in my judgment material to the decision which the judge had to make. The scope of the case before him was whether an investigation should have been made with respect to the breaches occurring during the disturbance and that is essentially what he considered. Ms Simor also submitted that these wider allegations were material to the investigation in order for it to be effective and achieve its legal purpose. It was necessary to identify the cause of the disturbance and to learn lessons from it. I do not accept that submission, for reasons I develop below.

A second preliminary issue which arises is this: at what point ought the Secretary of State to have considered whether or not to set up the investigation? When was the duty - if there was a duty - triggered? The judge did not consider that there was any obligation to consider setting up an inquiry until 21 May 2007 when Liberty first provided its detailed dossier to the Home Secretary setting out these complaints. One of the appellants’ grounds of appeal is that this was unjustified and rested on the mistaken assumption that no duty to set up an Article 3 investigation arose until complaints were made. They say that the duty is one cast upon the state and applies whether or not any specific complaints have been received.

I do not accept that the judge did make the error attributed to him. It is, I accept, an important element in the procedural protection afforded by Articles 2 or 3 that the state should of its own motion instigate an independent investigation, and should do so promptly. They thereby demonstrate that they are not in any sense complicit in, or acquiescing in, the alleged unlawful conduct. But the obligation to consider whether or not to set up an investigation cannot in my view be triggered until the authorities knew or ought to have known that the alleged breaches of Article 3 had occurred. On the facts of this case the judge held that this was not before 21 May.

The appellants submit that there was clear evidence that the Secretary of State must have known about this at a much earlier stage. In particular reliance is placed on the fact that an independent departmental inquiry was set up, with Mr Whalley CB to conduct it, in February. I do not accept that this demonstrates that it must have been appreciated by the Secretary of State that there were potential Article 3 breaches. The purpose of that inquiry was not to look into complaints by inmates but rather so that lessons which could be learnt about how disturbances of this kind could be better managed. I do not think it legitimate to infer that the Secretary of State must have had knowledge of Article 3 breaches at that time. The report does indeed refer, under the heading “welfare matters”, to some of the concerns identified by the detainees (para 93). But it is plain that much of the information was obtained as a result of Liberty’s letter to the Secretary of State, and the report was not produced until July.

Again, Ms Simor referred to the fact that there was evidence that the police had looked into an allegation of ill treatment in April 2006. However, that does not in my view assist these appellants, not least because it does not follow that an investigation of the kind now sought would have been appropriate to deal with the allegations then made. In my view the judge was plainly entitled to conclude that the Secretary of State did not have the requisite knowledge before 21 May. On any view it seems to me that he did not until then have knowledge of the detailed range of allegations which are now being relied upon by the appellants, and yet it is precisely the nature and extent of the alleged breaches which the appellants are now relying upon to support their case.

The principal issue that requires to be addressed is whether the judge was right in concluding that in principle an independent investigation should have been required. What investigations were necessary to satisfy the State’s procedural obligations with respect to Article 3? Mr Eadie QC, counsel for the Secretary of State, submits that in the circumstances of this case civil proceedings, together with the possibility of further criminal investigations and the Whalley report, was enough. Existing legal procedures constituted sufficient compliance. It is well established that in determining whether the obligation is satisfied it is necessary to look the totality of the procedures employed, as the Court of Appeal stated in R (Takoushis) v Inner North London Coroner [2006] 1 WLR 461 para 105. Ms Simor accepts that what is appropriate will always depend on the particular circumstances of the case. Sometimes the combination of criminal and civil proceedings will indeed be enough. However, she contends that this is not such a case.

I respectfully agree with Sedley LJ that the Whalley inquiry is of no relevance when considering the Article 3 obligation. It was a departmental investigation which was not focused on the alleged ill treatment of inmates at all. I do not doubt that Mr Whalley was eminently suited to the task allotted to him, but it had nothing to do with these complaints. I am minded to agree with Sedley LJ that had the inquiry been extended to cover these complaints, then Mr Whalley would not have been sufficiently independent to satisfy the procedural requirements of Article 3, essentially for the reasons which my Lord has given. But the issue does not directly arise in this case.

The essential issue, therefore, is whether civil proceedings, together with possible criminal investigations, would have sufficed.

There has been extensive case law on what the procedural obligations require. I find the authorities difficult to reconcile. An important starting point in my view, emphasised regularly in the authorities, is that the form of investigation to be adopted must depend on the particular circumstances. Even where Article 2 investigations are concerned, it is not possible to say that “one size fits all”. As Lord Phillips of Worth Maltravers observed in R (L) v Secretary of State for Justice [2008] UKHL 68; [2008] 3 WLR 1325 (para 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 the Takoushis case, for example, the court cited the decision in Vo vFrance 40 EHRR 259 para 90 where the ECHR had said that if the infringement of the right to life or to physical integrity is not caused deliberately, for example where it results from negligence in a hospital, the procedural obligation under Article 2 may be satisfied by an action in the civil courts alone. That may be enough to enable the liability of the wrongdoer to be established and provide for appropriate civil redress.

In my judgment the procedural requirement of an Article 3 investigation will often be less onerous than an Article 2 investigation. I would accept the observation of Sedley LJ that there is not a formal distinction between the requirements for the two Articles, and that in cases of near death, for example, the protection afforded by the two Articles may merge into one another. But what is required depends on the circumstances, and there are typically four significant differences between Article 2 and Article 3 cases which will be likely to be reflected in the appropriate procedures.

First, as Mitting J observed, the duty under Article 2 arises from the fact of a death in which the state may in some way be implicated, whereas the duty under Article 3 arises only when there is an arguable breach of the substantive rights. Second, as Lord Phillips pointed out in L (para 20) death is always treated as a matter of particularly grave concern and the need for a very full investigation into a death, whether state agents are suspected of being at fault or not, is particularly important. Third, as Lord Rodger observed in the same case, one of the differences between a death and a near death is that in the latter situation a prisoner who has his mental faculties intact, is “ 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.” A fortiori is that the case where infringements of Article 3 are alleged. Fourth, there are likely to be far fewer Article 3 breaches resulting from systemic wrongdoing. The combination of civil and, if necessary, criminal or disciplinary proceedings will often suffice in those circumstances to meet the Article 3 requirements.

In my judgment an important factor in helping to determine what investigation is required is to ask for what purpose it is being conducted. The appellants say that the purpose is precisely as expressed by the judge, Mitting J, who simply took the analysis adopted by Lord Bingham of Cornhill in the Amin case at para.31 and applied it with suitable modifications to take account of the specific interests which Article 3 is designed to protect. The resulting formulation is as follows:

“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[been ill treated] may at least have the satisfaction of knowing that [their ill treatment has been acknowledged , and may save others from ill treatment in the future.

The importance of this formulation is that the purpose is not limited, as it is in many of the ECHR cases, to finding the facts so that potential wrongdoers may be brought to book, and so that appropriate redress might be provided. (Sedley LJ has identified some of these cases at para. 51; and a more recent case of Makaratzis v Greece (2005) 41 EHRR 49 [GC], para. 74 is to the same effect.) Later decisions of their Lordships’ House have also emphasised the importance that may be attached to the need to learn lessons for the future which a careful examination of the facts will disclose. Indeed, in L, which concerned a case of near suicide in prison, the Secretary of State contended that the obligation to conduct an inquiry arose only if there was evidence to suggest that state agents had been at fault. Their Lordships did not accept this submission principally because holding officers to account would not be the only purpose of an investigation.

However, in my judgment it is important to consider these formulations of purpose in their context. Where the alleged breach occurs in custody and raises serious questions as to whether the state properly protected those in its charge from the risk of death, it is not difficult to see that a careful scrutiny of all the facts, identifying what went wrong, and why, and how matters might be put right for the future, is an important element in providing the protection which Article 2 requires. But the same is not necessarily required, even in an Article 2 case, where death results from negligent action of someone not in custody, as the Vo case shows. Indeed, in my judgment it would be wholly inconsistent with the fact sensitive nature of these cases to suggest that the procedural obligations arising in this area must always achieve the full panoply of the objectives identified in Amin. It would also impose an impossibly onerous financial burden on the state if that were to be the case.

Even where the purpose is to learn lessons, there are still significant limits to the scope of any investigation. The focal point must still be the acts which allegedly gave rise to a breach of Article 2 or 3. In an Article 2 case the focus is on the death and the circumstances surrounding it, and in Article 3 on the particular acts alleged to infringe that Article. It is not the purpose of an inquiry to engage in wider issues of a political nature. This was made clear by the House of Lords in the Gentle case in which their Lordships held that there was no obligation to set up a wide ranging inquiry pursuant to Article 2 into the circumstances in which the UK went to war with Iraq, at the behest of the mothers of two servicemen killed there. The principal reason for that conclusion was that there was no arguable substantive breach of Article 2. But in any event, Lord Bingham (with whose judgement Lords Hoffman, Hope, Rodger and Brown agreed) added this (para 9):

“Even if, contrary to my conclusion, the appellants were able to establish an arguable substantive right under article 2, they would still fail to establish a right to a wide-ranging enquiry such as they seek. Nothing in the Strasbourg case-law on article 2 appears to contemplate such an enquiry: Jordan v United Kingdom, above, para 128; Bubbins v United Kingdom(2005) 41 EHRR 458, para 153; Taylor v United Kingdom (1994) 79-A DR 127, 137; McShane v United Kingdom (2002) 35 EHRR 593, para 122; Banks v United Kingdom (Appn no 21387/05, 6 February 2007, unreported, BAILII [2007] ECHR 177, pp 12-13; McBride v United Kingdom (2006) 43 EHRR SE 102, para 1, pp 109-110. In Scholes v Secretary of State for the Home Department[2006] EWCA Civ 1343, para 67, Pill LJ threw some doubt on the current applicability of the ruling in Taylor, but I do not think the authorities justify his doubt and Arden LJ, in paras 82-83, applied what I respectfully think is the correct approach.”

It is important to read the Amin and L cases against that principle. Even given the wider purpose of the investigation required in such cases, it is still focused on the immediate circumstances surrounding the death or ill treatment, as the case may be. For this reason, the inquiry sought by Liberty plainly went well beyond anything that Article 3 could require.

The Secretary of State has relied heavily upon the Banks decision, which Sedley LJ has comprehensively analysed. I agree with his Lordship that parts of the decision in Banks are problematic. However, much of it is repeating established jurisprudence. In particular, in so far as the case suggests that civil or criminal proceedings will sometimes - indeed will generally - be sufficient to satisfy an Article 3 procedural obligation, it is in my view fully in line with established authorities. Similarly, the case confirms that Article 2 procedural obligations will generally be more stringent than those under Article 3, not least because the victims in the latter case are alive and can pursue their own claims. Finally, it confirms that the scope of an Article 3 investigation is limited in the manner I have indicated. It will be remembered that an important feature of the case was that the claimants were seeking to have an investigation into the whole culture of violence in Wormwood Scrubs prison which it was alleged (on powerful evidence) had existed throughout the 1990s. That certainly raises issues going well beyond the circumstances of the particular allegations of ill treatment, and I do not find it surprising that the court felt that this lay outside the scope of an Article 3 investigation. (That is not to say that there may not have been a strong case for an inquiry, but not as an element of an Article 3 obligation.)

In my judgment Banks, and the other cases referred to by Lord Bingham in Gentle, strongly support the observations of Lord Justice Longmore that it would not be appropriate for any investigation to consider the circumstances leading up to the disturbance, nor the reasons why it took place. It was, as I understood Ms Simor’s argument, principally to answer those questions that the appellants submitted that incidents both pre-dating and post-dating the disturbance had to be considered. Since those questions are not for an investigation of this nature, these matters are not material to the decision to hold the investigation.

Should a more limited independent investigation have been instigated? In my judgment the principles derived from the authorities clearly do not require the state to have to set up independent inquiry whenever anyone in custody made allegations that there had been a breach of Article 3. The financial cost would be wholly disproportionate to the benefits. Furthermore, in my opinion that is not what the law in this area requires even where the claimant links such allegations of breach to the existence of alleged cultural or institutional practices or arrangements.

The Secretary of State submits that Banks dictates that no investigation at all was required in this case. The ECHR was laying down the principle that criminal and civil proceedings will usually suffice to satisfy the Article 3 procedural obligation, and there were no features to displace that general principle here.

I do not accept that submission. The facts in Banks were quite different to the facts of this case. In Banks there had been a range of incidents extending over many years, and in many of the incidents there had been careful criminal investigations (and in one case, convictions). As the court noted, the purpose of the inquiry which the applicants sought was not particularly geared to finding the alleged wrongdoers but to focusing on the alleged culture of abuse that existed, and that is not the purpose of an Article 3 investigation.

Whilst in many, perhaps most, Article 3 complaints the combination of civil and criminal procedures will be enough to satisfy the Article 3 procedural obligations, I do not think that was the position here. In my judgment there are features of this case which required the Secretary of State to set up an independent investigation in May 2007, even though the alleged breaches are of Article 3 rather than Article 2. Its focus would, however, have had to be the alleged ill treatment and not the wider cultural or institutional difficulties which brought the problems to a head in the first place.

I would identify three factors in particular which, considered cumulatively, lead me to that conclusion. The first is the fact that the appellants were in custody. It is particularly important that the authorities must be alive to the very real potential for abusive behaviour towards such inmates. This is especially so given the reports of the Prison Inspector which have highlighted many defects in the way in which Harmondsworth is run and so would have given greater credence to these complaints than might otherwise have been the case.

Second, the allegations are not just of ill treatment by specific officers but include complaints of systemic ill treatment arising from the methods of managing the disturbance. Moreover, the allegations include complaints that these methods involved a breach of Article 3 even if there was no intent to debase or humiliate the claimants. It would be necessary to consider established policies and procedures and to assess to what extent, if at all, individual officers were departing from them.

Third, there were many officers brought into the prison on that night to deal with the disruption. It would be particularly difficult for an inmate to be able to identify who had taken the action alleged to constitute a breach of Article 3. The Strasbourg authorities have often emphasised –and did so in Banks itself- that one of the most important functions of an Article 3 investigation is to enable potential individual wrongdoers to be identified.

It is not entirely clear to me whether Mitting J was concluding that by the time the authorities were alerted to the wrongdoing, it was already too late to conduct an effective investigation, or whether he was saying that it was too late to do so by the date of the trial. The main ground of appeal is directed to that finding that it was too late. In so far as he found the former, I think he erred. I do not see why an effective investigation could not have been conducted at that stage, which was only six months after the incident, and indeed the Secretary of State was conceding that it could. In so far as he was exercising a discretion not to require an investigation at the date of his judgment, it was in my view a legitimate exercise of discretion.

In any event, I entirely agree with Lord Justice Sedley that it is certainly now too late to carry out an effective investigation, and that the court should not make any mandatory order to that effect. Nor do I think that there is any realistic prospect at this stage of police investigations leading to the identification of wrongdoers or any criminal charges. The appellants are pursuing their civil claims, and that is likely to be as good a means as any to establish whether there were breaches of Article 3 for which the state is liable, even if the particular transgressors cannot now be identified.

I would be willing to make a declaration that in the particular and somewhat unusual circumstances of this case, the Secretary of State ought to have conducted an independent investigation when, in May 2007, he was alerted to the possibility that the appellants may have been the subject of infringements of their Article 3 rights. To that extent, I would allow this appeal and dismiss the cross appeal.