England and Wales High Court (Administrative Court) Decisions

R (ON THE APPLICATION OF WIDGERY SOLDIERS) v. MEMBERS OF THE TRIBUNAL SITTING AS THE BLOODY SUNDAY INQUIRY [2001] EWHC Admin 888 (16th November, 2001)

LORD JUSTICE ROSE :

This is the judgment of the court.

1. There is before the court, as a matter of urgency and with the permission of Sullivan J, a challenge by judicial review to a ruling of the Bloody Sunday Inquiry made on 1st August and published on 2nd August 2001. It ruled that the soldiers and former soldiers, whom, for convenience, we shall refer to as “the soldier witnesses”, between two hundred and four hundred in number, who are to give oral evidence before the Tribunal over a period of some six months or more starting in the Autumn of next year, should do so in Londonderry Guildhall where, hitherto, for the most part the Tribunal has sat, rather than in London or some other part of Great Britain.
2. The circumstances in which the Tribunal was established under the Tribunals of Inquiry (Evidence) Act 1921 are well known. They are set out, together with the history of the Tribunal and the nature and course of two previous applications for Judicial Review, in paragraphs 2 – 28 of the Court of Appeal’s judgment in R v Lord Saville of Newdigate ex parte A [2000] 1 WLR 1855. These matters need not be repeated here. The Court of Appeal held that the soldier witnesses who were alleged to have fired rounds were entitled to anonymity. Subsequently, the Tribunal granted anonymity to all the soldier witnesses, (save four senior officers who were well known), whether they were said to have fired rounds or not. Since the Court of Appeal decision in July 1999 there has been one change in the constitution of the Tribunal. The Right Honourable Sir Edward Somers has retired and has been replaced by the Honourable Mr John Toohey.
3. The Tribunal has been sitting for the best part of 3½ years. It is therefore exceptionally well qualified to make case management decisions as to how its proceedings should be conducted. Furthermore, its members are of the highest judicial standing. There is accordingly, at first blush, a degree of unreality in this court being invited to quash one of its decisions by means of a remedy developed, historically, for the control of inferior courts and tribunals. That said, save for the written submissions by Sir Louis Blom-Cooper QC on behalf of Northern Ireland’s Civil Rights Association, to which in a moment we shall come, it is common ground before us, as it was in the earlier proceedings culminating in the Court of Appeal’s decision in ex parte A, that this court has, properly, a reviewing jurisdiction in relation to decisions of the Tribunal.
4. Sir Louis Blom-Cooper’s submission was that only in exceptional circumstances will the court exercise its supervisory role over Public Inquiries, which are of an inquisitorial nature. He referred, among other authorities, to Notts CC v S of S for the Environment [1986] AC 240 per Lord Scarman at 250 to 251 and Sir Richard Scott’s report on the Arms to Iraq Inquiry, Volume IV Section 1.5. Tribunal witnesses, as such, have no rights or interests that require legal protection (Lawlor v Flood [1999] 3 IR 107 per Murphy J at 138 to 144). He accepted, however, that a tribunal under the 1921 Act has an overriding duty to act fairly (NSW v Canellis [1994] 181 CLR 309 at 330). The anonymity issue considered in ex parte A was an exceptional circumstance justifying the courts’ jurisdiction by way of Judicial Review. He also submitted that the tribunal is not a public authority within the meaning of s6 of the Human Rights Act 1998, being neither a court or tribunal within ss6(3) and 21, as legal proceedings may not be brought before it, nor a public authority, as it possesses no powers to determine how others should act ( see Aston Cantlow PCC v Wallbank [2001] 3 AER 393 at paragraphs 29, 35 and 36). The Tribunal exercises functions in connection with proceedings in Parliament and is therefore within the exemption in s6. Accordingly, none of the articles of ECHR has direct applicability. In any event, Article 2 of the Convention envisages operational not procedural safeguards – see Osman v UK 22 EHRR CD 137.
5. We reject these submissions. A possible threat to life which arises from a Tribunal decision is an exceptional circumstance requiring, when appropriate, the court’s intervention. The Tribunal is master of its own procedure but the requirements of fairness are for determination by the courts (see ex parte A at 1868B) and procedural fairness involves an obligation to be fair to witnesses (ex parte A at 1868E). The Tribunal’s preliminary decisions can be quashed if they cause real injustice (ex parte A at 1870C). The Tribunal is, plainly in our view, a public authority within s6(3)(b) of the Human Rights Act. We accept Mr Lloyd Jones QC’s written submission that the Tribunal’s functions are those of public not mutual governance, its relationship with witnesses is created by rules of law independently of the volition of the Tribunal or the witnesses, and the Tribunal possesses powers to determine how others should act (see Aston Cantlow PCC v Wallbank). Furthermore, the Tribunal was not created by Parliament under the 1921 Act but by the Secretary of State for Northern Ireland. The fact that it reports to Parliament does not mean it is exercising functions in connection with proceedings in Parliament: there are no proceedings in Parliament in connection with which the Tribunal exercises functions. A similar argument was rejected by Roch LJ in the Divisional Court in ex parte A on 17th June 1999 (transcript 47H-48D).
6. Accordingly, the Tribunal has to comply with the Human Rights Act and this court has jurisdiction to entertain the present application.
7. Before turning to the rival submissions, it is convenient to refer to the terms of the Tribunal’s ruling. In paragraph 5 they affirm the correctness of their preliminary view “that the natural place to hold an inquiry of the present kind was where the events in question occurred”. They go on:

“events of that day although of great national and international concern have undoubtedly had their most serious and lasting effects on the people of that city. It is there that the grief and outrage that the events occasioned are centred. It seems to us that the chances of this inquiry restoring public confidence in general and that of the people most affected in particular (which is the object of public inquiries of this kind) will be very seriously diminished (if not destroyed) by holding the inquiry or a major part of the inquiry far away and across the Irish sea, unless there were compelling reasons to do so. It is for similar reasons that public inquiries generally are held in or near to the place where the events to be investigated occurred."

In paragraph 8, they refer to the central importance to local people of the Inquiry coming to where Bloody Sunday took place and, at paragraph 9, they say

“in our judgment, since the oral evidence of the soldiers will form a major part of the inquiry the starting point is that this evidence should be given at the Guildhall where all or virtually all the other oral evidence will be heard, unless indeed there are compelling reasons to take a different course”.

They refer to Article 2 of the European Convention and the dictum of Lord Phillips MR in R (Mahmood) v Secretary of State for the Home Department [2001] 1 WLR 840 at 857:

“interference with human rights can only be justified to the extent permitted by the Convention itself. Some articles of the Convention brook no interference with the rights enshrined within them.”

At paragraph 15, they speak of the Tribunal making a decision “on venue by reference to the events which may or may not occur though of course the past may throw light on the likelihood of such events taking place in the future”. They point out that the safety of the soldiers is in the hands of state agencies rather than the Tribunal. “However that does not relieve the Tribunal of the obligation to assess as best it can on the material available to it the risk to the soldiers in hearing their evidence in one place rather than another”. At paragraph 16 they say

“the most helpful judgment is that of Osman v United Kingdom”

They cite from paragraph 116 of the judgment of the European Court of Human Rights in that case and refer also to Ergi v Turkey ECtHR 28th July 1998 and Chahal v United Kingdom 23 EHRR 413. They refer at paragraph 21 to the assessments, provided to the Tribunal by the RUC and the Security Service, of the threat to soldier witnesses giving oral evidence to the Inquiry.

“The threat was assessed as moderate both in Northern Ireland and Great Britain. However as the Security Service pointed out the threat is only one of the two factors to be taken into account when considering the risk to a person, the other being the vulnerability of the person in question.”

They then refer to a meeting on 18th June 2001, to which later we shall return, at which senior representatives of all the relevant security agencies gave their views to the Tribunal on the security of soldier witnesses. A summary of that meeting was distributed to the parties and further submissions were made to the Tribunal.

It is necessary to set out paragraphs 23, 24 and 25 of the Tribunal’s ruling in full:

“23. As will be seen from the summary, it is the view of the concerned security agencies that the risk to soldier witnesses of terrorist reprisals would be higher in Northern Ireland than in Great Britain. The soldiers submit that accordingly it would be an infringement of their rights for the Tribunal to require them to give evidence at the Guildhall rather than in Great Britain. Their case is that the Tribunal is bound to take all feasible precautions to avoid or minimise, to the greatest extent possible, any risk to the life of the individual; and that this can only be done by hearing the evidence of the soldiers in Great Britain, where the risk is lower.

24. It seems to us that the fact that the risk is greater in the one place rather than the other is not of itself determinative of the matter. On the basis of the Osman decision, it is incumbent on the authorities (which in the present case include both the Tribunal and the agencies responsible for the protection of witnesses) to do all that can reasonably be expected of them to avoid a real and immediate risk to life of which they ought to have knowledge. We are satisfied on the basis of the security advice that we have received, that the security authorities in Northern Ireland can provide, for soldiers giving evidence at the Guildhall, a level of protection sufficient to avoid any such risk. In such circumstances we consider that the Tribunal would not be acting incompatibly with the rights of the soldiers by requiring them to give evidence at the Guildhall rather than in London, for in neither place would there be a real and immediate risk to them. Neither the MoD nor the RUC (the state authorities who have a duty to protect the soldiers while giving evidence and who must accordingly deploy the proper resources to do so) have advised us that, notwithstanding the security precautions that they could and would put in place for soldiers giving evidence at the Guildhall, the level of risk would be so high that it could be described as real and immediate or in terms to the same or similar effect.

25. We have of course borne in mind the history of terrorist attacks on military and other targets in Northern Ireland, particularly those that have recently taken place in and around the city, and the present situation with regard to terrorist organisations. We have also borne in mind that over at least the last thirty years, it appears that the protection afforded by state authorities to those required to attend courts in Northern Ireland, often in circumstances of the greatest controversy where the risk of terrorist attacks has clearly been high, has been sufficient to avert any loss of life or injury from terrorist organisations. In our judgment the authorities will have done all that could reasonably be expected of them by providing, as they say that they can, a level of security commensurate with what has regularly (and successfully) been provided for trials in Northern Ireland where persons at risk such as soldiers, security officers, informers and others have been required to attend.”

The Tribunal went on in paragraph 26 to consider the procedural aspects of Article 2. They referred to Jordan v UK Application No 246/94 4th May 2001 and concluded that, as the Inquiry was concerned with the use of lethal force by state authorities, it should be conducted where the events in question occurred and that the soldiers’ rights would not be infringed by requiring them to give evidence at the Guildhall with the protection that can be given. They went on:

“it seems to us that it would be unreasonable and indeed in contravention of the Article 2 procedural requirements for the Tribunal to conduct a central part of the inquiry at somewhere other than the natural and proper place for it”

Having referred to R v Governor of Pentonville Prison ex parte Fernandez [1971] 1 WLR 987, R v Lord Saville ex parte A and Mahmood the Tribunal concluded in paragraph 34 that there was no difference between the common law and Convention obligations, referred again to Osman “ and the need for awareness actual or imputed of the existence of a real and immediate risk to life” and said

“since none of the concerned agencies has suggested that such a real and immediate risk exists notwithstanding the precautions that would be put in place, it seems to us that to require the soldiers to give their oral evidence at the Guildhall would not offend their common law rights….We consider that we are justified in requiring of the soldiers no more than what has been required on many occasions of others who have had to give evidence of killings in Northern Ireland, namely to appear and testify where the events took place, with the security authorities doing all that can be reasonably expected of them to provide a safe environment. Clearly the soldiers would prefer to give their evidence in Great Britain but this does not demonstrate, nor do we accept, that they have reasonable fears for their safety while going to or from the Guildhall or actually giving their evidence there in view of the security precautions…The soldiers (with few exceptions) have the advantage of anonymity and all have the right to require the state security services to protect them. They have our assurance that they will not be required to give oral evidence here if anything occurs that means they cannot do so in proper calm and quiet conditions…we can see nothing unfair (let alone unlawful) in requiring the soldiers to give their oral evidence at the Guildhall.”

They went on to consider whether there was a real danger of public disorder while the soldiers gave evidence at the Guildhall, repeated that the Inquiry was conducted in a calm and quiet manner, and said that witnesses would have a proper and fair opportunity to be heard and any attack on those providing protection for witnesses would be treated as calculated to destroy the necessary environment for the Inquiry to be conducted at the Guildhall. They said in paragraph 40

“we have nothing which suggests that political expediency would lead to protests or the like disrupting the orderly progress of the Inquiry…We reject the suggestion…there is a serious risk of public disorder…which to our minds is neither supported by the May 2000 threat assessment made by the RUC nor by the views expressed at the meeting of 18th June nor indeed by anything else.”

As to procedural fairness to the soldiers, they accepted that the Inquiry was “likely to engender or re-kindle very strong feelings” and that the soldiers are “likely to find it an unpleasant and intimidating experience to give evidence” but this was unavoidable and the Guildhall was not “an especially hostile or intimidating environment for the soldiers”. There were no grounds for supposing that proceedings could not continue to be conducted in a quiet and calm manner. In paragraph 47 they concluded that none of the arguments for the soldiers “is sufficient to provide a compelling reason for not hearing the oral evidence of the soldiers at the Guildhall which we regard as the proper place for this Inquiry”.

The transcript of 18th June shows that the threat to the soldiers was assessed by the security agencies as moderate on the mainland and in Northern Ireland but the risk to the soldiers was agreed to be higher in Derry (93,94) where there was also a reduced or slightly reduced prospect of a successful outcome in relation to the protection of witnesses and security force personnel (96,97,98). There was the capability to protect soldiers giving evidence at the Guildhall comparable to that for others giving evidence in high profile cases for 30 years but a question as to maintaining the protection for 6 months or more (95,97). Lord Saville referred more than once to an “acceptable degree of security” and asked (99) “can you do the job or are you going to say no we can’t do the job, these people are likely to get shot if we try it. Now I gather the answer to both London and Northern Ireland is yes we think we can do the job”. But it would remain riskier in Northern Ireland (101).

We turn to the claimants’ grounds. Ground 9 has not been pursued and ground 8 is in our judgment, of comparatively little substance. Grounds 1 to 7, in different ways, advance 3 principal complaints. First, the Tribunal misdirected itself in law and applied the wrong test when assessing the risk to the soldiers if they gave evidence in Londonderry. Secondly, it was not open to the Tribunal to conclude that the soldiers have no reasonable fears for their own safety as witnesses in Northern Ireland because security precautions would be taken. Thirdly, the Tribunal’s approach was flawed in starting from a presumption that the soldiers should give evidence in Londonderry unless they produced a compelling reason to the contrary and in concluding that only if the soldiers did give evidence in Londonderry could the Tribunal restore public confidence.

As to the first complaint, Mr Lloyd Jones submitted that the Tribunal mis-directed itself, by reference to Osman v UK, 29 EHRR 245, that the test at common law as well as in relation to a breach of the right to life protected by law under Article 2 of the Convention, was whether there existed a real and immediate risk to the life of the soldiers. Osman was concerned with the limits of a public authority’s duty to act to protect individuals from a third party. But, in the present case, as it was the Tribunal’s own decision which exposed the soldiers to risk, the correct test, derived from Lord Diplock’s speech in ex parte Fernandez at 994, a case involving extradition, is whether there is a “reasonable chance”, “substantial grounds for thinking”, or “serious possibility” that the soldiers’ right to life would be put at risk by requiring them to testify in Londonderry (see per Lord Woolf in ex parte A at 1877B-H). The test adopted by Lord Woolf at 1877B in relation to anonymity, Mr Lloyd Jones submitted, is equally apt here: once it is accepted that the soldiers’ fears for their safety are based on reasonable grounds, is there any compelling justification for them to be required to give evidence in Londonderry, the evidence being that this would increase the risk to them? The Tribunal could only properly require the soldiers to give evidence in Londonderry if either (a) the contemplated security measures would so reduce the risk to them that there is no reasonable chance or serious possibility of life threatening attack on them or (b) there is a compelling justification for requiring the soldiers to run the additional risk in Londonderry. In relation to (a) the Tribunal wrongly applied the Osman test and, having concluded that there was no real and immediate risk, did not consider (b). Further, in referring to an acceptable degree of risk and to the state’s obligation to take care of people to a reasonable or proportionate degree (transcript 75) the Tribunal adopted the wrong approach and the security agencies were never asked if there was a real and immediate risk.

As to the second complaint, Mr Lloyd Jones submitted that the existence of even the highest security precautions would not render the soldiers’ fears for their safety unreasonable. The Tribunal accepted in paragraphs 23 and 24 of its ruling that the risk to the soldiers in Northern Ireland is higher than in Great Britain, but concluded that security measures would avoid a real and immediate risk to them. That conclusion, submitted Mr Lloyd Jones, was contrary to the evidence of the security agencies that complete protection could not be provided against a real and unpredictable risk of attack. And, in any event, in the light of the material at the 18th June meeting, it could not be said that security would eliminate a reasonable chance or serious possibility of attack. Soldiers are regarded as a legitimate target by terrorists and 46 terrorist incidents, which the security services were powerless to prevent, took place between February 2000 and 20th July 2001, including a bomb at Shackleton Barracks Ballykelly in February 2000 and a mortar attack on Ebrington Barracks, Londonderry, in January 2001, and showed a specific terrorist intention and ability to kill soldiers. The soldiers involved in Bloody Sunday are particularly attractive targets to terrorists, many of them having already been categorised as murderers. Londonderry is a small city where secure accommodation and the scope for variation of routes to the Guildhall is limited. The Guildhall is situated on the west bank of the River Foyle and approached from the east by only two bridges. Close quarters assassination, remote control bombs and sniper attacks are possible and the core vulnerability of the soldiers is irreducible (Ministry of Defence assessment core bundle 493). There is significant vulnerability to witnesses and those seeking to safeguard the witnesses in the Guildhall (transcript page 56). In these circumstances, Mr Lloyd Jones submitted, it is impossible to say that the soldiers’ fear for their safety is other than reasonable. These fears would necessarily be increased by the requirement for all witnesses to be in Londonderry for at least two days, for many who were likely to be questioned for many days and for some who would have to make more than one visit to Northern Ireland or stay for an extended period. They will require accommodating and escorting under armed guard with armoured transport and armed uniformed RUC officers would be needed in the Guildhall during their evidence. Military support would be necessary. There would be collateral risk to those protecting them. Giving evidence in such circumstances would be prejudicial to them and therefore procedurally unfair. The Tribunal did not take into account the soldiers’ perception of and confidence in the Inquiry.

As to the third complaint, Mr Lloyd Jones submitted that the Tribunal erred, first, in starting from the proposition, in paragraphs 5 and 9, that the soldiers needed to establish a compelling reason why their evidence should not be heard in Londonderry and in concluding, in paragraph 47, that they had failed; secondly, in concluding that public confidence in the Tribunal was dependent on the soldiers giving evidence in Londonderry. Although Londonderry, as the scene of the events and the home of the victims and their families, may be the natural place to hold much of the Inquiry and to hear evidence from the families and other witnesses living there, it does not follow that it is the natural place to hear the evidence of hundreds of witnesses living on the mainland. And the Tribunal did not consider whether any loss of confidence arising from where the soldiers gave evidence would be reasonable or would exist in the minds of responsible people, (see ex parte A Divisional Court 17th June 1999 per Roch LJ at 37A and Maurice Kay J at 47B), bearing in mind that the crucial purpose of the Inquiry was to establish the truth, that such a result is most likely if proceedings are fair to all witnesses, that the families would be legally represented at temporary mainland hearings of the soldiers’ evidence which would be in public and relayed to Londonderry and followed by final submissions in Londonderry. None of these matters was properly taken into account by the Tribunal. Only two reasons appear to have been given. First, the importance of the Inquiry to local people: this could properly be no more than a factor to be considered together with others and symbolism is not a justification for exposing the soldiers to risk of death. Secondly, to conduct a central part of the Inquiry elsewhere would be a breach of Article 2 procedural requirements: nothing in Jordan v UK or McCann v UK 21 EHRR 1997 suggests that Article 2 requires an Inquiry to be held where fatal events occurred. What is required is an independent, effective investigation, securing evidence with reasonable expedition, subject to public scrutiny and involving the next-of-kin to the extent necessary to safeguard their legitimate interests (see Jordan paragraphs 105 to 109). None of these features of open justice would be undermined by the soldiers giving evidence in Great Britain. The Tribunal had a duty to be fair to the soldier witnesses as well as to the families.

For the Ministry of Defence, Mr Burnett QC in submissions which echoed and in some respects adopted Mr Lloyd Jones’s argument submitted first that the Tribunal misdirected itself in applying the higher Osman test rather than the lower ex parte A test and in holding that the procedural obligations under Article 2 told in favour of the soldier witnesses giving evidence in Londonderry. Secondly, there was no compelling reason why the soldiers should face the higher risk in Londonderry even after all precautions had been taken. Thirdly, it could not properly be concluded that the soldiers’ fears if they gave evidence in Londonderry were not reasonable: elaborate security has not been and will not be sufficient to protect soldiers from terrorist attack and the real risk to soldiers only becomes immediate when an attack is implemented. In any event the position of the Bloody Sunday soldiers is different from that of other soldier witnesses in ordinary trials (transcript 94).

On behalf of the defendants, Mr Christopher Clarke QC confirmed that the Tribunal’s ruling was based upon the proposition that the Osman test was applicable; Article 2 was not engaged unless it was demonstrated that the risk to life was “real and immediate”. He submitted that there was no difference in substance between this formulation and that set out by Lord Diplock in Fernandez.

Article 2.(1) places three obligations on a public authority: a negative obligation to refrain from intentionally depriving a person of their life; a positive obligation to protect the right to life, and a procedural obligation effectively to investigate killings resulting from state action. The negative obligation to refrain from taking life intentionally is absolute, and is a fundamental provision of the Convention which admits of no derogation in peacetime. The only circumstances in which a public authority may intentionally take a life are those set out in Article 2(2) which are exhaustive, and must be narrowly interpreted: Stewart v UK [1984] 7 EHRR 453, para. 13. However, the ambit of this absolute prohibition is relatively narrow. The word “intentionally” in Article 2.(1) must be given its natural and ordinary meaning: the purpose of the prohibited action must be to cause death: In re A (conjoined Twins: surgical separation) [2001] 2 WLR 480, NHS Trust A v M [2001] 2 WLR 942. This absolute negative obligation has no application to the present case. The positive obligation is not absolute. It requires state authorities to do all that can reasonably be expected of them to avoid a real and immediate risk to life of which they have or ought to have knowledge, whether this risk arises from criminal acts of third parties, or any other factor: Osman at pp. 277-8, paragraphs. 89 and 91, and p. 306, paragraphs 115-116. He said this means that Article 2 permits the state deliberately to expose an individual to any risk to life provided that risk cannot be demonstrated to be “real and immediate”. Article 2 is not engaged unless the “real and immediate” threshold is crossed. He suggested that a decision by the state to expose an individual to a lesser degree of risk to life might well be reviewable on rationality grounds.

Chahal and Soering are concerned with the state’s non-derogable obligations (in relation to torture, etc.) under Article 3. They are of no assistance in determining the ambit of the qualified, positive, obligation under Article 2(1). Ergi, at paragraphs 79-81, is an application of the Osman duty - to do all that can reasonably be expected - to particularly dramatic circumstances. Depending on the circumstances, the Osman duty may place a heavy burden on the state; but the underlying test of “real or immediate risk” does not alter merely because the state authorities are deliberately exposing the individual to a risk to life.

The state cannot reasonably be expected to do everything that can possibly be done to avoid exposing an individual to any risk, however small or insignificant, to life. To impose such an obligation would be a disproportionate response in the context of a qualified duty: see the balancing exercise carried out by the court in Van Mechelen v. Netherlands 25 EHRR 647, at page 674, paragraphs 56 & 57.

He submitted that the Tribunal was right to conclude that the “real and immediate risk” test is, for all practical purposes, the same as the approach adopted in Fernandez, namely, are there “substantial grounds for thinking”, is there a “reasonable chance” or “a serious possibility”. In Soering the court referred in paragraph 88 of its judgment both to “substantial grounds for thinking he would be in danger...”, and to “a real risk of exposure to inhuman or degrading treatment”. In Bensaid v. UK 6 February 2001 the court noted, with approval, the test applied by Simon Brown L.J. in R v. Secretary of State for the Home Dept ex parte Turgut [2001] 1 All ER 719: “the right not to be exposed to a real risk of article 3 treatment”. A similar approach was found in HLR v. France 26 EHRR 29, page 50, paragraph 40. In Secretary of State for the Home Department v. Rehman [2001] 3 WLR 877, Lord Hoffmann referred in paragraph 54 to “a substantial risk”, and pointed out, in paragraph 56, that the concept of a standard of proof is not particularly helpful when assessing the extent of a future risk. The question is one of evaluation and judgment. In ex parte A it was common ground that the soldiers’ fears (if anonymity was withheld) were based on reasonable grounds. It is not enough for the claimants to assert that there is some risk. Whether the level of risk is sufficient to bring Article 2 into play is a question of degree, for the evaluation and judgement of the Tribunal: see Rehman, paragraph 56.

Turning to the implied procedural obligation under Article 2, Mr Clarke referred to Jordan v. UK 4 May 2001, paragraphs 105-109, and to McKerr v UK 4 May 2001, paragraphs 159-161. The Tribunal correctly concluded that it would be in contravention of the Article 2 procedural requirement to conduct a central part of the Inquiry anywhere other than in Londonderry. Remaining in Londonderry would retain public confidence in the state’s willingness to maintain the rule of law, prevent any appearance of collusion in the unlawful use of force, ensure that the investigation was transparent and accessible to the families of the deceased and the local community and enable the next of kin of the victims to be involved in the proceedings to the extent necessary to safeguard their legitimate interests. There was a particular need not to repeat the errors of the past. The fact that the Widgery inquiry had been held in Coleraine, rather than in Londonderry, had led to the widespread belief that the venue had been chosen for the convenience of army witnesses and to disadvantage local people.

In deciding whether the soldiers’ fears for their safety were reasonable, the Tribunal was required to assess, on an objective basis, whether the available evidence led to the conclusion that there was a sufficient likelihood of the risk occurring. The Tribunal was not dealing with the soldiers’ subjective fears. On an objective basis, a real and immediate risk was not established, so the soldiers’ fears could not be regarded as reasonable in the context of Article 2; which was simply not engaged. Since Article 2 was not engaged it had not been necessary for the Tribunal to carry out a balancing exercise to decide whether there was any “compelling justification” (see ex parte A at p.1877 A) for exposing the soldiers to the increased risk. In deciding that the witnesses should be heard in Londonderry unless there were “compelling reasons” not to do so, the Tribunal was not placing a burden of proof on the soldiers. It was merely following “a logical progression of thought”. The end result would be the same if one approached the question on the basis: if Article 2 is not engaged, Londonderry is the natural place to hold the Inquiry. The Tribunal’s conclusion that the soldiers’ fears were not reasonable, given the level of protection that would be provided by the state security services, was a finding of fact, arrived at by a uniquely experienced tribunal following detailed enquiries of all the relevant services. All those at the 18th June meeting were well aware of its purpose. The Tribunal was concerned to ascertain not merely the means of protection that would be employed, but their outcome, namely, what degree of safety would be achievable in practice? Because of the inherent lack of precision in such concepts as “moderate risk”, “increased vulnerability”, “secure environment”, the Tribunal looked for a practical, objective, standard, namely the level of protection that the security services had been able to provide for vulnerable witnesses in Northern Ireland over many years. If, as a result of that protection, the risk to the soldiers could not be described as “real and immediate” it is irrelevant that it was higher in one place, Londonderry, than in another.

Although the Tribunal had not reached the stage of being required to consider whether there was a compelling justification for requiring the soldiers to accept an increased degree of risk, such justification could be found in the Tribunal’s conclusions in paragraphs 5 and 26: moving a major part of the inquiry out of Londonderry would seriously damage public confidence, and be in breach of the Article 2 procedural requirements. The Tribunal was particularly well qualified to assess the former. They were well aware of the fact that restoring public confidence was not the sole object of the Inquiry, but it was a vital part of the Tribunal’s role, given the lack of confidence, particularly amongst those most affected by Bloody Sunday, as to the willingness of the British state to carry out a genuinely thorough and impartial investigation. To render the hearing inaccessible to those most affected, the families of the victims, and other members of a tightly knit local community, would shatter their confidence. Their position, as victims in the context of Article 2, should not be equated with the position of ordinary members of the public having an interest in, and wishing to attend, an ordinary public inquiry.

Substantial security precautions would be required at any venue on the mainland. The risk is greater in Londonderry, but the Tribunal had been advised by the security agencies that an acceptable level of security could be provided in both places. Against this background the balance tipped decisively in favour of a hearing in Londonderry.

The Tribunal had been entitled to reject, in paragraph 40, the suggestion that there would be a serious risk of public disorder should the soldiers give oral evidence at the Guildhall. The Tribunal was advised on 18th June that with such measures as segregation, public order could be maintained (transcript pp. 35-40). The Tribunal took account of the past incidents relied upon by the soldiers (paragraph 38) but some of these incidents were relatively old, and many had been associated with the Marching Season and other sectarian activities, all of which had no connection with the conduct of the Inquiry.

In terms of procedural fairness, security arrangements will have to be made wherever the soldiers give evidence. Those likely to be provided in Londonderry would effectively insulate the witnesses, so they will be well able to do justice to themselves. There are collateral risks, albeit of a different character, whether the Inquiry is held in Londonderry or on the mainland. Since the Tribunal has not accepted that the soldiers’ fears for themselves are reasonable, concern for the safety of those protecting them cannot result in any procedural unfairness. The job of the security forces necessarily involves their being required to take risks.

For the families we received written and oral submissions from four leading counsel. Mr Lavery QC stressed that the families are acting in good faith and want no more deaths. Their wish is to see the Inquiry finished not jeopardised. The Tribunal has gained the trust and confidence of the people of Northern Ireland but discovery of the truth will be a waste of time if the Tribunal does not continue to enjoy that confidence. The families do not believe there is a risk to the soldier witnesses in Londonderry and for the Tribunal to hear their evidence in England would be perceived as a step to protect the soldiers. There is no evidence that any particular group from those involved in the Inquiry has been singled out by terrorists because of the role it plays. It is purely speculative and dependent on a reading of the Republican psyche to say the risk to soldiers is greater in Northern Ireland than on the mainland. The Ministry of Defence views are not detached. The Tribunal is in as a good position as any observer to assess the risk. It is not open to this court to read the transcript of 18th June and take a different view. He referred to Bensaid 33 EHRR 205 and the approval in paragraphs 55 and 56 of the English judicial review procedures and, in particular, (paragraph 28) of the judgment of Simon Brown LJ in ex parte Turgut [2001] 1 All ER 719 that an irrationality challenge under Article 3 will only succeed if consideration of the underlying factual material compels a court to a different conclusion.

Lord Gifford QC adopted Mr Clarke’s analysis of the authorities and expressed mystification at Mr Lloyd Jones’s criticism of the Tribunal’s approach to the law. He submitted that the law before the Tribunal was agreed, namely that Article 2 would only be engaged if the authorities could not eliminate a real risk. The present circumstances are to be distinguished from those considered in ex parte A, particularly as absence of anonymity would remove protection forever whereas the risk arising from giving evidence in Londonderry would be of only short duration. Londonderry should be regarded as the safest, not the least safe, place for the soldiers to give evidence. He relied on the statement of Liam Wray explaining his family’s wish to be present at the hearing, to see the witnesses, which would be impossible if they are heard on the mainland. The process by which the truth is reached is of great importance and full healing and reconciliation will not take place if the witnesses are not seen and heard by the families. On 18th June, the Tribunal had taken great care to obtain appropriate advice and had taken all appropriate matters into account.

Miss McDermott QC adopted the submissions of Mr Clarke and Lord Gifford. Hearing soldiers’ evidence on the mainland would render the Inquiry inaccessible to the Docherty family, which includes six children. Hearing evidence in Londonderry would cause the Tribunal to continue being seen as independent, transparent and unbiased. It presently enjoys the full confidence of the people most affected.

Mr Mansfield QC submitted that there would be no real risk to soldiers in Londonderry whether the risk was described as immediate or a serious possibility: for all practical purposes the tests are synonymous. Neither threshold is crossed because of the thirty-year history of non-interference with witnesses and because of the measures which can be put into place. The Tribunal's crucial conclusions, in paragraphs 23 to 25, were within the range of options open to them. The security agencies provided no concrete material showing real risk. The threat assessments, which were originally made in relation to anonymity, had not addressed the thirty-year history (transcript page 94) and no soldier had been attacked when giving evidence at the 1972 Inquiry.

In addressing these rival submissions it is as well to begin by setting out the approach which we have adopted. This is in accordance with the judgment of Lord Woolf in ex parte A at 1865H to 1868H.

“31. The Tribunal is subject to the supervisory role of the courts. The courts have to perform that role even though they are naturally loath to do anything which could in any way interfere with or complicate the extraordinary difficulty task of the Tribunal. In exercising their role the courts have to bear in mind at all times that the members of the Tribunal have a much greater understanding of their task than the courts. However, subject to the courts confining themselves to their well recognised role on applications for judicial review, it is essential that they should be prepared to exercise their role regardless of the distinction of the body concerned and the sensitivity of the issues involved. The court must also bear in mind that it exercises a discretionary jurisdiction and where this is inconsistent with its performance of its duty it should avoid interfering with the activities of a Tribunal of this nature to any greater extent than upholding the rule of law requires.”

When reviewing the reasonableness of a Tribunal’s decision:

“33….If a decision could affect an individual’s safety then obviously there needs to be a greater justification for taking that decision than if it does not have such grave consequences.

34. The appropriate test…is the test adopted by Sir Thomas Bingham MR in the Court of Appeal Civil Division in Reg v Ministry of Defence ex parte Smith [1996] QB 517, 554E-F.” The test was based on submissions of Mr David Pannick in that case. They were in these terms:

“The court may not interfere with the exercise of an administrative discretion on substantive grounds save where the court is satisfied that the decision is unreasonable in the sense that it is beyond the range of responses open to a reasonable decision-maker. But in judging whether the decision-maker has exceeded this margin of appreciation the human rights context is important. The more substantial the interference with human rights, the more the court will require by way of justification before it is satisfied that the decision is reasonable in the sense outlined above. ”

“35…Sir Thomas Bingham MR indicated that he regarded this statement as “an accurate distillation of the principles laid down by the House of Lords in Reg v Secretary of State for the Home Department ex parte Bugdaycay [1987] AC 514 and Reg v Secretary of State for the Home Department ex parte Brind [1991] 1 AC 696” and we would respectfully agree with him….

“37…When a fundamental right such as the right to life is engaged the options available to the reasonable decision maker are curtailed. They are curtailed because it is unreasonable to reach a decision which contravenes or could contravene human rights unless there are sufficiently significant countervailing considerations. In other words it is not open to the decision-maker to risk interfering with fundamental rights in the absence of compelling justification. Even the broadest discretion is constrained by the need for there to be countervailing circumstances justifying interference with human rights. The courts will anxiously scrutinise the strength of the countervailing circumstances and the degree of the interference with the human right involved and then apply the test accepted by Sir Thomas Bingham MR in Reg v Ministry of Defence ex parte Smith.

“…38 Turning to the role of the courts on judicial review to ensure procedural fairness…While [this Tribunal] is master of its own procedure and has considerable discretion as to what procedure it wishes to adopt, it must still be fair. Whether a decision reached in the exercise of its discretion is fair or not is ultimately one which will be determined by the courts. …The requirements of procedural fairness for witnesses is well recognised in the courts…At this Inquiry where there are no defendants the requirement of procedural fairness surely involves an obligation to be fair to witnesses…

40..Procedural fairness must be viewed in the round and fairness to the former military witnesses [is] only one dimension of the question posed: it [is] also necessary to consider the interests of the dead and injured and the public interest. ”

In our judgment the Tribunal misdirected itself in law as to the test to be applied when assessing the threshold of risk to soldier witnesses from terrorist reprisals. In paragraph 23 the Tribunal accepted that that risk would be higher in Northern Ireland than in Great Britain. They were right to say in paragraph 24 that the fact that the risk was greater in one place rather than the other was not of itself determinative. But in proceeding to apply the Osman test the Tribunal in our judgment fell into error. The Osman and ex parte A tests are, as it seems to us, conspicuously different in purpose and effect. In Osman the European Court of Human Rights limited the obligation of the state to intervene to protect against the activities of third parties to those circumstances in which there is a real and immediate risk to life. In ex parte Fernandez the House of Lords and in ex parte A the Court of Appeal defined the obligation of a public authority more broadly as being not to make a decision exposing anyone to the real possibility of a risk to life in the future. This misdirection fundamentally flaws the Tribunal’s decision Whether their decision would have been the same or different if the correct test had been applied is a matter for the Tribunal rather than this court to determine. They will, no doubt, wish to bear in mind: that the need to seek advice from all the security agencies on 18th June demonstrated that, whatever the degree of risk, it was by no means fanciful; that likelihood of being shot was not, as we have sought to explain, the correct approach; that the security agencies were not asked whether there was a real possibility of risk; and that the question of whether the necessary security could be maintained for 6 months or more was never resolved.

Furthermore, the Tribunal’s conclusion that the soldiers had no reasonable fears for their own safety, in the light of the protection the security services would afford, was erroneous and gave rise to procedural unfairness in relation to the soldier witnesses. As we have said, the correct test to be applied was whether there was a real possibility of risk. Also, as it seems to us, it was not reasonably open to the Tribunal to conclude that the soldier’s fears were not reasonable. The security agencies will of course do their best to ensure an adequate level of protection. But it does not follow that the people to be protected do not have reasonable fears for their safety notwithstanding the existence of that protection. The recent history of events in the province, including the attacks on barracks at Ballykelly and Ebrington demonstrate that, despite intensive security precautions, terrorist activity puts soldiers’ lives at risk. And the Tribunal’s emphasis on the quiet and calm manner of proceedings at the Guildhall ignores the impact on middle aged witnesses, many of whom have been civilians for many years, of the extensive security measures required in relation to accommodation, transport and close protection. The Tribunal said that the Guildhall itself is not an especially hostile or intimidating environment but this, in our judgment, ignores both the security precautions in relation to soldier witnesses before reaching the Guildhall and after leaving it and the necessity, indicated by the security agencies on 18th June, for uniformed armed personnel in the Guildhall while evidence is being given. Accordingly, when re-considering the matter, the Tribunal must in our judgment take into account that the soldier witnesses’ fears for their own safety must properly be characterised as reasonable.

We also accept that Mr Lloyd Jones’s third principal criticism of the Tribunal is well founded. The Tribunal should not, as they did in paragraph 9, have started from the proposition that the soldier witnesses’ evidence should be given at the Guildhall unless they showed compelling reasons for a different course. It is common ground that the primary purpose of the Tribunal is to find the truth. It is also vital that the Inquiry commands public confidence. The confidence of the families in the Tribunal’s findings is obviously of great importance. So too, as it seems to us, is the confidence of the soldier witnesses, some of whom are accused of murder. Equally although the confidence of the people of Northern Ireland is of high importance, so too is the confidence of people in other parts of the United Kingdom. It is the Prime Minister of the United Kingdom whose announcement in Parliament gave rise to the setting up of the Tribunal. All of these matters must be taken into account by the Tribunal when carrying out the balancing exercise, properly for their determination, as to where the soldier witnesses should give evidence. In our judgment the Tribunal in its present ruling does not appear to have taken these matters into account. Accordingly, the Tribunal’s conclusion that the confidence of the people in Londonderry would be very seriously diminished if not destroyed if soldier witnesses gave evidence on the mainland should not properly have been regarded as determinative of the outcome without due consideration of wider public confidence including that of the soldier witnesses themselves. In any event, in our judgment, once the risk of death is a serious or real possibility it was for the Tribunal, as decision maker, to find some compelling justification for interference with the soldiers’ Article 2 rights rather than to require the soldiers to provide a compelling justification for giving their evidence elsewhere. (see Soering v UK 11 EHRR 439 paragraph 88, Chahal v United Kingdom 23 EHRR 413 at paragraph 80 and Ergi v Turkey European Court of Human Rights 28th July 1998 paragraph 79).

Finally, we do not accept the Tribunal’s conclusion, at the end of paragraph 26, that conducting a central part of the Inquiry somewhere other than Londonderry would contravene the Article 2 procedural requirements for the Tribunal, which we take to be a reference to the families’ Article 2 rights. The Tribunal sitting in Great Britain would be equally independent from those implicated in the events, transparent and subject to public scrutiny, non-collusive in unlawful acts and would involve the victims and next-of-kin (who would all continue to be legally represented) to the extent necessary to safeguard their legitimate interests (see Jordan v United Kingdom European Court of Human Rights 4th May 2001 paragraphs 105-109).

Accordingly, the Tribunal’s decision that the soldier witnesses must give evidence in Londonderry is quashed. We remit the matter to the Tribunal for reconsideration in the light of the terms of this judgment.

- - - - - - - - - -

THE ASSOCIATE: Judgment in The Queen on the application of A and Others v Members of the Tribunal sitting as the Bloody Sunday Inquiry.

LORD JUSTICE ROSE: For the reasons given in the judgment of the Court handed down, the Tribunal's decision in relation to the soldier witnesses is quashed and we remit the matter to the Tribunal for reconsideration in the light of the terms of this judgment.

MR BOOLS: My Lord, simply on the matter of costs, we ask that there be no order as to costs.

LORD JUSTICE ROSE: Does anybody resist that? Then there will be no order as to costs.

MISS ROSE: On behalf of the Tribunal we do apply for permission to appeal. As your Lordship is well aware, the test for permission is twofold under CPR part 52. First of all, whether or not there is a real prospect of the appeal succeeding and, secondly, regardless of your Lordships' view of the merits of the appeal, if there is in any event a compelling reason why the appeal should be heard. In our submission there are two compelling reasons why this matter ought to go to the Court of Appeal. The first is, as your Lordships have acknowledged in your judgment, that your Lordships are differing from the views of very senior members of the judiciary. The second is that it is patent that the matter is a matter is of very considerable public importance and warrants consideration at appellate level.

In any event we submit that there is a real prospect of the appeal succeeding. In particular your Lordships have ruled that the Tribunal misdirected themselves in law in relation to the question whether Osman and ex parte A set a different test. That is a matter on which there is currently no appellate authority and it is obviously a matter on which senior members of the judiciary may take different views, as is apparent from the history of these proceedings.

My Lords, finally we would submit that there are very good case management reasons why permission to appeal ought to be granted at this stage by this court. As your Lordships will be aware from our acknowledgement of service, we sought expedition at the outset. As your Lordships know, the inquiry has been sitting for three and a half years. The costs are enormous. Any delay is going to cause very significant inconvenience in costs as well as being adverse to the interests of justice. We would very much like, if it were possible for us to be accommodated, for any appeal to come on before Christmas. If permission is not granted at this stage, inevitably there will have to be a rolled-up hearing for permission and a substantive appeal which means there will be no saving in costs or time if permission is refused by this court, simply an added procedural complexity. I would ask your Lordships to avoid that by granting permission at this stage.

MR MANSFIELD: My Lord, as you may recall I represent in fact, for the purposes of this hearing, four families instead of three. But I would associate myself with those observations and also seek leave on behalf of those families. I will not repeat arguments I put forward on paper and orally before but your Lordships will recall that we have a specific point we wish to pursue in relation to the definition and threshold test.

MR HARVEY: Richard Harvey on behalf of the Wray family. I adopt my learned friends' arguments.

MR BOOLS: It is simply a matter for the Court. We have no submission.

(The Court conferred)

LORD JUSTICE ROSE: We shall grant permission, not because we think there is a real prospect of success, but because there is a compelling other reason, namely the public interest.