England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE SPECIAL IMMIGRATION APPEALS COMMISSION
The Hon Mr Justice Ouseley, Senior Immigration Judge Jordan and
Mr J Mitchell
SC/42 & 50/2005

		Royal Courts of Justice Strand, London, WC2A 2LL
		09/04/2008

Sir Anthony Clarke:

This is the judgment of the court to which each member of the court has contributed.

Introduction

1. On 3 October and 14 December 2005 respectively AS and DD were each served with notice of a decision to make a deportation order against him on the grounds that his presence in the United Kingdom was not conducive to the public good because he was a danger to national security. Each was detained on the grounds of national security. They each appealed on various grounds and, although their cases are entirely separate, both their appeals were heard by SIAC at the same time from 30 October to 10 November and on 16 and 17 November 2007. The members of SIAC who heard their appeals were the chairman of SIAC, Ouseley J, Senior Immigration Judge Jordan and Mr J Mitchell.
2. Before SIAC there were a plethora of issues, which are discussed in great detail in a very lengthy judgment containing 433 paragraphs which was handed down on 27 April 2007. In the result the appeals against the deportation orders were allowed on a single ground, namely that there were substantial grounds for believing that the respondents faced a real risk of suffering treatment contrary to article 3 of the European Convention on Human Rights ('the Convention') if they were returned to Libya. The Secretary of State for the Home Department ('the SSHD') sought permission to appeal from SIAC but permission was refused by Mitting J. In refusing permission, he acknowledged that the issues raised in these cases were important but said that they were factual. However, permission was subsequently granted by Hallett LJ. The respondents are at present on bail subject to stringent conditions.
3. The statutory scheme is described at [2, 9 and 93] in our judgment in MT, RB and U (Algeria) v SSHD [2007] EWCA Civ 808, [2008] 2 WLR 159 and we need not repeat it here. It is sufficient to note that an appeal from SIAC to this court lies only on a point of law. We consider below the correct approach in a case of this kind in the light of AH (Sudan) v SSHD [2007] EWHL 49, [2007] 3 WLR 832, especially per Baroness Hale at [30].
4. The appeal is concerned solely with the issue of safety on return to Libya. The striking aspect of the case is that the SSHD's case depends entirely upon a memorandum of understanding ('MOU') between Libya and the United Kingdom signed on 18 October 2005. SIAC exhibited the MOU to its judgment. It is sufficient for present purposes to say that it provides the United Kingdom with assurances that anyone deported from the United Kingdom will be properly treated. It is not necessary to go further because, on the one hand it is accepted by the SSHD that, in the absence of the MOU, there would be substantial grounds for believing that there is a real risk of the respondents being tortured on their return to Libya, whereas, on the other hand, it is accepted on behalf of the respondents that if Libya complies with the MOU there is no such risk. The sole issue on the facts before SIAC on this part of the case was whether the MOU reduced the risk to an acceptable level. SIAC accepted that, motivated by self-interest and pragmatic reasons, the Government of Libya had provided the MOU in good faith, intending to honour it. However, for reasons which are discussed in some detail at [70-80] below, SIAC held that Libya's motivation and reasoning might change, that it might not honour the MOU and that, in consequence, there was a real risk of the respondents being tortured on return. The case for the SSHD in this appeal is that SIAC erred in law in reaching that conclusion.
5. The SSHD now has four grounds of appeal, although what is now the first ground was added by amendment. In summary they are these:

i) SIAC failed to give sufficient weight to the evidence of the FCO witness Mr Layden and/or wrongly substituted its own assessment for his and/or failed to give sufficient reasons for rejecting his evidence.

ii) SIAC failed to direct itself as to the correct test to measure the degree of risk that the respondents would suffer ill-treatment contrary to article 3 on return.

iii) SIAC erred in lowering the test for risk on return to take account of the unpredictability of future events in Libya.

iv) SIAC's findings of fact do not warrant, and are not capable of supporting, a conclusion that substantial grounds have been shown for believing that the respondents face a real risk of suffering treatment contrary to article 3 on return.

6. SIAC handed down both open and closed judgments and we have considered both, in the latter case with the assistance of the special advocates, but the SSHD's appeal is advanced without reference to the closed material. We return to the relevance of the closed material below.

Background to issues in the appeal

7. SIAC held that both AS and DD were a threat to national security. Since that is accepted in this appeal, it is not necessary to refer to this part of the case, save in order to give an outline of the facts which are relevant to risk on return. Both respondents are Libyan nationals. SIAC described DD and the threat he posed in some detail at [71-75] as being a real and direct threat to the national security of the United Kingdom. Those paragraphs include these findings which are relevant for present purposes:

"71. We are entirely satisfied that DD is a real and direct threat to the national security of the UK. He is an Islamist extremist. He is a member of the LIFG [ie the Libyan Islamic Fighting Group] and at least within the UK is a figure of some importance and influence. He has close links with a number of senior LIFG members. …

72. DD is a global jihadist with links to the Taleban and Al Qa'eda. Such differences as exist between those two groups have no relevance to the danger he poses. He left Libya earlier than he admits, and has travelled significantly. We are quite satisfied that the more sinister interpretations of his so called "family" website are correct and show his support for suicide operations. …

73. The evidence strongly supports the conclusions that he has probably been involved in the procurement and production of false documentation for use by LIFG members. …

74. DD is also a threat to the UK's national security because his opposition to the Qadhafi regime is a major aspect of his global jihadist outlook. Opposition to the Qadhafi regime, including opposition from an Islamist perspective i.e. from the viewpoint that the regime is anti-Islamic according to their particular strand of religious belief, is not of itself a threat to the UK's national security. It is the extremist Islamist opposition, which countenances and supports the use of violence against the regime, which is a threat; and particularly so where it is part of a wider jihadist outlook. These activities cannot sensibly be regarded as legitimate self-defence.

8. SIAC's conclusions with regard to AS are summarised at [104]:

"104 The SSHD alleged that AS was a committed Islamist extremist who had been actively involved in providing logistic support to individuals linked to Al Qa'eda, and was linked to a terrorist cell based in Europe which was involved in raising funds, procuring forged documents and in facilitating the travel of recruits to terrorist training camps. He had links to individuals who were involved in attack planning in Europe, and himself had received terrorist training in Afghanistan. It was not alleged against him that he was a member of the LIFG, although the Libyans had accused him of being a member. …"

9. Those conclusions focus on the risk posed by the respondents to the national security of the United Kingdom. In part drawing on those conclusions, SIAC found that the respondents would each be seen in Libya as senior terrorist suspects associated with LIFG and Al Qaeda, determined to use violence for extremist ends against the Libyan regime, which SIAC accepted would place them in a particularly vulnerable position if they were returned to Libya. At [128-138] SIAC made findings as to the risks on return faced by the respondents.
10. Those findings, which are in the open judgment and thus available to Libya, may be summarised in this way. In various notes verbales in 2006 the Libyan government made it clear that it regarded both respondents as members of LIFG and that on return both would be tried under article 206 of the Penal Code, which provides for the death penalty, subject to mitigating factors: see [128 and 129]. Both respondents are supporters of global jihad. They are both seen by Libya as Islamist extremists hostile to the Qadhafi regime, both would be detained and questioned on return to Libya and both would be put on trial, perhaps under the more serious article 207 of the Penal Code: see [130 and 131].
11. At [133] SIAC stated that the witnesses on behalf of the SSHD accepted that, but for the MOU, the United Kingdom would have serious concerns about the real risks faced by the appellants as extreme Islamist opponents of the Qadhafi regime and their alleged membership of the LIFG. It identified the risks as being torture or other ill-treatment, incommunicado detention without trial, an unfair trial, imprisonment and torture as political prisoners, the imposition of the death penalty and perhaps the death penalty being carried out for an article 207 offence.
12. Mr Anthony Layden, who was the UK Ambassador to Libya from October 2002 to April 2006 and is now a retired diplomat and appointed to the FCO as Special Representative for Deportation with Assurances, was the principal witness for the SSHD. He expressed not only his own views but also those of the FCO. He made it clear that it was not the case for the SSHD that there were changes in Libyan society or politics which were sufficient in themselves to provide protection from the risks identified above to people like the respondents who were regarded as enemies of the state: see [134-5].
13. In October 2006 the Home Office had issued an Operational Guidance Note on Libya in these terms (quoted by SIAC at [137]):

"The following human rights problems were reported in 2005: inability of citizens to change the government; torture; poor prison conditions; impunity; arbitrary arrest and incommunicado detention; lengthy political detention; denial of fair public trial; infringement of privacy rights; severe restriction of civil liberties-freedom of speech, press, assembly, and association; restriction of freedom of religion; corruption and lack of government transparency; societal discrimination against women, ethnic minorities, and foreign workers; trafficking in persons and restriction of labour rights.

…

The Libyan government continues to be repressive of any dissent and opposition political activists and opposition Islamic activities are generally not allowed to operate on any substantial scale within the country. If it is accepted that the claimant has in the past been involved in opposition political activity or is a radical Islamic activist for one of the opposition political or Islamic groups mentioned above then there is a real risk they will encounter state-sponsored ill-treatment amounting to persecution within the terms of the 1951 Convention. The grant of asylum in such cases is therefore likely to be appropriate."

14. SIAC added at [138] that it could safely be assumed that that account represents the UK government's views on the state of affairs in Libya. SIAC also said at [161] that conditions in Libyan prisons were such that they were likely to reach the article 3 threshold in the case of political prisoners.

Approach of the Court of Appeal

15. As already stated an appeal lies only on a point of law. It is common ground that SIAC is a specialist tribunal. The approach to an appeal from such a body on a point of law has recently been summarised by Baroness Hale in AH (Sudan) v SSHD at [30]:

"30. … This is an expert tribunal charged with administering a complex area of law in challenging circumstances. To paraphrase a view I have expressed about such expert tribunals in another context, the ordinary courts should approach appeals from them with an appropriate degree of caution; it is probable that in understanding and applying the law in their specialised field the tribunal will have got it right: see Cooke v Secretary of State for Social Security [2001] EWCA Civ 734, [2002] 3 All ER 279, para 16. They and they alone are the judges of the facts. It is not enough that their decision on those facts may seem harsh to people who have not heard and read the evidence and arguments which they have heard and read. Their decisions should be respected unless it is quite clear that they have misdirected themselves in law. Appellate courts should not rush to find such misdirections simply because they might have reached a different conclusion on the facts or expressed themselves differently. I cannot believe that this eminent Tribunal had indeed confused the three tests or neglected to apply the correct relocation test. The structure of their determination can be explained by the fact that this was a "country guidance" case: but that makes it all the more important that the proper approach to the internal relocation alternative, as explained by the House in this case, is followed in future."

It will be noted, relevantly to the task of this court of considering only points of law, that Baroness Hale's guidance as to the limited role of this court extends to decisions by a specialist tribunal on points of law as well as to the lower court's assessment of the facts.

16. All counsel accepted that that passage set out the principle relevant to an appeal to this court from SIAC. This is perhaps not surprising in a case where this appeal has been heard with that in Othman v SSHD and counsel for the respondents in this case are counsel for the appellants in Othman and counsel for the SSHD in both cases are the same, whereas the SSHD is the appellant here but the respondent in Othman. However that may be, counsel were correct to treat [30] in AH (Sudan) as authoritative guidance in appeals from SIAC, although it was itself an appeal from the AIT. Although counsel suggested various glosses on Baroness Hale's statement, we think, with respect, that the passage is clear and well able to stand for itself.
17. As to the other members of the House of Lords in AH (Sudan), Lord Hope at [19] agreed in terms with Baroness Hale's statement; Lord Brown of Eaton-under-Heywood agreed with the general approach both of Baroness Hale and of Lord Bingham; and Lord Bingham, with whom Lord Hoffmann agreed in full, stressed, in particular at [11], that the judgment of a specialist tribunal (perhaps like any judgment) must be read as a whole and that one would not expect such a tribunal to make what Lord Bingham described as an egregious and inexplicable error. That is an important point in the present appeal because, as appears below, the SSHD's ground of appeal which asserts that SIAC misstated the relevant test in our judgment involves the assertion that it made just that, an inexplicable error.
18. The nature of SIAC as a tribunal was considered by this court, both in M v SSHD [2004] EWCA Civ 324, [2004] 2 All ER 863 at [2] and [16] and in MT, RB and U (Algeria). In particular we note that in M at [2] Lord Woolf CJ, giving the judgment of the court, which also comprised Potter and Clarke LJJ, explained that each tribunal at SIAC must have three members, one of whom is a judge who holds or has held high judicial office, one of whom is the chief adjudicator or a member of the IAT (as it then was) and one of whom it is intended should have experience of national security matters, although Lord Woolf added that the last requirement is not a statutory requirement. There is no reason to think that Mr Mitchell did not have relevant experience of that kind.
19. In these circumstances the correct approach of this court is to consider the judgment of SIAC as a whole and only to hold that it has erred in law if it is quite clear that it has done so. That approach fully respects the guidance that is given in AH(Sudan). That is in a context where, as in many appeals of this kind, the appellant asserts that that test is satisfied, whereas the respondent asserts that there was no error of law and that the complaints amount only to challenges to SIAC's findings of fact. That is precisely what has happened in the case of AS and DD on the one hand and in the case of Othman on the other.
20. As to the identification of issues of law in a case such as the present, we venture to repeat what we said in MT, RB and U(Algeria) at [97] and [109]:

"97. In the domestic jurisdiction as previously understood the question of whether an applicant faces a real risk of being subjected to treatment contrary to article 3 (the issue … , taken from [80] of the judgment of the ECtHR in Chahal) is a mixed question of fact and law. That expression is not here used, as it sometimes is, as a way of dressing up an issue of fact as an issue of law. Rather, it indicates that there are two discrete issues involved, one of fact and one of law. As Donaldson MR put it in O'Kelly v Trusthouse Forte [1984] QB 90 at [122H]-[123A]:

"While it may be convenient for some purposes to refer to questions of "pure" law as contrasted with "mixed" questions of fact and law, the fact is that the appeal tribunal has no jurisdiction [under section 136(1) of the Employment Protection (Consolidation) Act 1978, which limited the jurisdiction of the Employment Appeal Tribunal to questions of law] to consider any question of mixed fact and law until it has purified or distilled the mixture and extracted a question of pure law."

In the present case the issue of fact that is distilled by proper analysis is the question of what treatment the applicant risks receiving when returned to Algeria. That is a pure issue of fact, no different from, for instance, the issue in a personal injury case of when the claimant will be free of disability. The second issue is, however, one of law: does the treatment found fall within the terms of article 3. That is to be decided according to legal rules, and in particular the jurisprudence of the ECtHR, as to the meaning of article 3.

…

109. … The issue of fact in this case is whether there is a sufficient risk of BB being tortured on return to Algeria. That is a single and undifferentiated question of fact, which it is for the fact-finding tribunal to determine. In making that determination the fact-finding court will no doubt assess the impact of other findings that it has made, just as it will assess the reliability of witnesses in deciding what findings to make. But that process is all part of the fact-finding process. That process of assessment is quite different from, and plays a quite different role from, for instance, the assessment that the court has to make, based on the found facts, of whether the defendant acted negligently: which is a matter of legal judgement, and not just a question of what is going to happen in certain circumstances in the future."

21. Those principles seem to us to apply here just as they did in that case. We shall therefore seek to apply them here.

The correct test

22. Article 3 of the Convention provides:

"No one shall be subjected to inhuman or degrading treatment or punishment."

Although the nature of the correct test is the subject of the SSHD's second ground of appeal, it is convenient to address it first because it provides the basis for the consideration of the remainder of the questions in the appeal. It is common ground that the correct test is whether substantial grounds were shown for believing that the respondents would face a real risk of suffering treatment contrary to article 3 of the Convention on return to Libya. It was common ground before SIAC that that is the correct test and it is common ground now.

23. Yet it is submitted on behalf of the SSHD that SIAC misdirected itself as to the test and applied the wrong test. The complaint is that, as evidenced by [299], [333], [344] and [345] and in particular by its summary at [371], SIAC repeatedly stated that the question was whether there was a real risk of the respondents suffering torture contrary to article 3 on return to Libya, whereas it should have asked itself whether there were substantial grounds for believing that the respondents would face such a real risk.
24. We should first note that, in our judgment, the requirement that there must be substantial grounds for believing that there would be a real risk of ill-treatment contrary to article 3 on return, means no more than that there must be a proper evidential basis for concluding that there was such a real risk. This is made clear in Saadi v Italy, which is a decision of the Grand Chamber of the ECtHR, application 37201/06, which was handed down on 28 February 2008, after we had heard most of the argument in this appeal, at [128-9]. We set out those paragraphs in their context below but [128] made it clear that, in considering whether there were substantial grounds, the ECtHR will consider all the material before it; and [129] added:

"It is in principle for the applicant to adduce evidence capable of proving that there are substantial grounds for believing that, if the measure complained of were to be implemented, he would be exposed to a real risk of being subjected to treatment contrary to Article 3. … Where such evidence is adduced, it is for the Government to dispel any doubts about it."

A considerable amount of evidence was put before SIAC, which considered it with care and, on the basis of it, concluded that there would be a real risk of the respondents being tortured on return to Libya. Thus SIAC applied the correct test.

25. As to the formulation of the test, it would be very surprising indeed if SIAC misunderstood or misstated it. This is for two reasons. First, Ouseley J wrote the judgment, not only in this case and Othman, but also in MT, RB and U (Algeria), where there was no suggestion that SIAC misstated the test. If SIAC was mistaken in this case, it made precisely the kind of egregious and inexplicable error to which Lord Bingham referred in AH (Sudan). It is in our judgment inconceivable that it made such a mistake.
26. The second reason is that, as Mr Garnham QC (one of the special advocates for AS) submits, not only was the test common ground before SIAC but the closing oral submissions on behalf of the SSHD described the test in just the way that SIAC did in its judgment. In their opening submissions, counsel for the SSHD asserted that the applicable test was first set out by the ECtHR in Soering v United Kingdom (1989) 11 EHRR 439 at [91] and that it was applied to deportation cases in Chahal v United Kingdom (1996) 23 EHRR 413 at [80]. It was stated thus in Chahal at [80]:

"The prohibition provided by Article 3 against ill-treatment is equally absolute in expulsion cases. Thus, whenever substantial grounds have been shown for believing that an individual would face a real risk of being subject to treatment contrary to Article 3 if removed to another State, the responsibility of the contracting State to safeguard him or her against such treatment is engaged in the event of expulsion. In these circumstances, the activities in question, however undesirable or dangerous, cannot be a material consideration. …"

That test was repeated in the written closing submissions on behalf of the SSHD and no-one, including SIAC, suggested any different test.

27. In their oral closing submissions, the special advocates for both AS and DD repeated the same test and, in written submissions to SIAC, counsel for the SSHD made it clear that there was no dispute about the test, except that the case for the SSHD was that by 'substantial grounds' was meant 'strong grounds'. Thus all parties accepted before SIAC that the test was not simply whether there was a real risk of being exposed to torture but whether there were substantial (or strong) grounds for believing that there was such a risk.
28. In these circumstances, we accept the respondents' submission that, when SIAC referred to the test of real risk without adding that there must be substantial (or strong) grounds for believing that there was such a risk, it was simply using the reference to 'real risk' as shorthand for the full test. Counsel for the SSHD unsurprisingly did the same in their open closing submissions. At paragraph 8.1 they set out [390] of SIAC's judgment in Y v SSHD (which was subsequently called MT v SSHD in this court), which began:

"390. … What matters is whether such assurances in any individual case signify that there is no real risk that the individual would be subject to treatment breaching Article 3…".

As in the present case, the full test is not set out. We would add that we also accept the respondents' submission that the importance of the words 'substantial grounds for believing' is to stress that the decision as to the existence of a real risk must be based on evidence and not on mere assertion or speculation.

29. In these circumstances we are unable to accept that SIAC made the elementary error suggested on behalf of the SSHD. We accept the submission made on behalf of the respondents that there was a particular reason why the parties all readily accepted the shorthand. As stated above, the SSHD had conceded that, absent the assurances contained in the MOU, there would be substantial grounds for believing that there would have been a real risk of the respondents being tortured on their return to Libya. The question was whether the assurances reduced the risks to a level which would not infringe article 3.
30. For these reasons, we reject the second ground of appeal. We return below to the correct approach to the application of the test in the light of the Strasbourg jurisprudence, including in particular Saadi v Italy. Before considering Saadi in detail, it is convenient to turn to the first ground of appeal, which focuses on the correct approach to the evidence of Mr Layden.

The correct approach to the evidence of Mr Layden

31. SIAC found that, if Libya abided by the terms of the MOU, there would be no violation of article 3 in returning the respondents there. The critical question was whether Libya would abide by the terms of the MOU and the extent of any risk that it would not. The SSHD placed considerable reliance on the evidence of Mr Layden, who (as indicated above) has very considerable experience of Libya and, in particular of diplomatic relations with Libya. In this connection, by contrast with the issue of what is the correct test, we accept that Mr Layden had experience which none of the members of the tribunal had and moreover that he was giving not only his own view but the collective view of the FCO. As formulated in the first ground of appeal referred to above, it is the case for the SSHD that SIAC failed to give sufficient weight to his evidence and/or wrongly substituted its own assessment for his and/or failed to give sufficient reasons for rejecting his evidence.
32. In its judgment SIAC set out Mr Layden's evidence in some detail. Having done so, it said this at [320-324]:

"Conclusions: the FCO evidence and deference

320. The SSHD has relied on the evidence of Mr Layden as to the reliability of the assurances which have been negotiated, and has also submitted that we should accord his views deference. Mr Layden was an impressive witness – forthright, completely honest, realistic, with a commitment to truth and fairness, and to the upholding of the UK's international human rights obligations. He had the advantages of long experience of diplomacy in the Middle East and of being an Arabic speaker. Above all, he had been the British Ambassador in Libya when the MOU was contemplated and negotiated; he was a participant in the negotiations. True it is that he is not an independent expert witness in the conventional forensic sense, and his roles in the process could suggest that he could be an enthusiast for the work which he had done. But he is retired and only fills his particular post because he has been asked to, and because he believes that the agreements which he has negotiated would assist the security of the UK without breaching the ECHR. He has been frank about why he would not have adopted that stance earlier in relation to Libya. His expertise is particularly relevant to the assessment of the significance of the course of negotiations, to the domestic political situation, to the relationship between the various personalities who feature in this case, and to the assessment of why the interests of a diplomatic relationship between Libya and the UK would suffice to prevent a breach of the MOU. Of course, when he says that a breach would be "well nigh unthinkable", that view commands considerable respect, although it is very strong indeed.

321. Nonetheless, for the reasons which we have given in Othman, in paragraphs 339-340, adopting [MT] in paragraphs 324-326, we do not treat his views with deference on those matters. They are entitled to weight according to the expertise, experience and cogency with which they were expressed and with which the difficulties were considered and dealt with. We have set out the areas in which he has particular expertise and experience.

322. This is not to downplay the value of the views of Professor El-Kikhia (the respondents' expert witness), and we do not do so. Although the disappearance of his cousin, probably at the hands of the Libyan authorities, might be capable of impairing his objectivity, we saw little evidence of that. Much of what he says accords with the evidence of Mr Layden. But as we have already said, there is a significant difference between them on the question of whether the Libyans can be trusted as the result of a number of incidents which arose during the initial stages of the rapprochement, including over the abandonment of WMD (weapons of mass destruction). On these, we conclude that Mr Layden has by a long way the greater immediate and direct knowledge; part of it was dealt with only in the closed material. We accept Mr Layden's evidence that the Libyans have proved in the end completely trustworthy in the way in which they have dealt with some very difficult issues, and have kept to what they said they would do, even if there have been some uncertainties and surprises along the way. It is in the very nature of their roles that Mr Layden's knowledge would be greater.

323. Mr Layden is also able to speak with a greater understanding than Professor El-Kikhia could have of the diplomatic relationship between the UK and Libya, its origin and its important components, the incentives on Libya to adhere to the obligations, and the sanctions open to the UK in the event of a reported breach of the MOU. There can be an advantage in the distance which Professor El-Kikhia has from Libya, but there is a much greater advantage in the personal immediacy and recentness of knowledge of people and events as they affect this relationship.

324. We also take Mr Layden as representing the considered and collegiate views and wisdom of the FCO, and as not pursuing some personal cause which the FCO has been pleased to see him promote if he felt able to do so."

33. It would be difficult to imagine a more handsome tribute to a witness. In our opinion SIAC fully described the experience and expertise of Mr Layden and approached his evidence entirely properly and appropriately. Indeed, it gave his evidence very considerable weight. However, SIAC was right to say, as it did at [321], that his views should not be treated with deference but that his views were entitled to weight according to the expertise, experience and cogency with which they were expressed and with which the difficulties were considered and dealt with. It follows that SIAC was not bound to accept every statement of opinion expressed by Mr Layden. On the contrary, it was SIAC's responsibility, not that of Mr Layden, to determine the facts, including the key question of fact, namely what risks the respondents would be exposed to on return to Libyan in circumstances in which there is in place an MOU. Given that it was accepted on the one hand that, but for the MOU, the respondents would be exposed to a real risk of torture and, on the other hand, that if Libya complied with the MOU there would be no such risk, the question became whether Libya would comply with the provisions of the MOU.
34. In the end the question is a narrow one. SIAC accepted the great majority of Mr Layden's evidence. It did not however accept his view that a breach would be well nigh unthinkable. The question is whether it was entitled to reach its own conclusion. In support of the SSHD's submission that SIAC erred in principle in its approach to Mr Layden's evidence, Mr Sales QC relies upon the approach identified in three cases in particular.
35. The first is R v Governor of Brixton Prison ex p Kotronis [1971] AC 250, where the House of Lords allowed an appeal against an order of habeas corpus made by the Divisional Court in which it had set aside an order committing Mr Kotronis to prison to await extradition to Greece. The basis of the extradition sought was a conviction in Greece. Mr Kotronis sought to challenge his conviction but the House of Lords held that, while it would be open to him to challenge the fact of his conviction, it was not open to him to contend that the conviction was a nullity as being contrary to the rules of natural justice. That was a matter for the Secretary of State. The House further rejected a submission that the Greek Government was acting in bad faith on the ground that it in truth wanted his extradition for political reasons. In that connection Lord Reid said this at 278H-279A:

"So it would be a clear breach of faith on the part of the Greek Government if he were detained in Greece otherwise than for the purpose of serving his sentence, and it appears to me to be impossible for our courts or for your Lordships sitting judicially to assume that any foreign Government with which Her Majesty's Government had diplomatic relations may act in such a manner."

Lord Morris said much the same at 279C-D and 280F.

36. In our judgment, neither the decision in Kotronis nor those statements of Lord Reid and Lord Morris assists the SSHD here. The House of Lords was not considering what weight should be given to the evidence of a witness. Moreover, if the House was saying that the conduct of a foreign state with which the United Kingdom has diplomatic relations can never be examined in the English courts, whatever the position was in 1971, it is plain that the position now is different. There may well be areas into which the courts will not go but the conduct or likely conduct of Libya in the present context is not one of them.
37. Indeed, Mr Sales does not go that far. He correctly accepts that it was the duty of SIAC to consider whether there were substantial grounds for believing that the respondents would face a real risk of torture on return to Libya and that that involved consideration of the question whether Libya would or might break the assurances contained in the MOU. This is clear from a number of decisions of the ECtHR, culminating in Saadi v Italy at [148], to which we return below. In these circumstances we conclude that Kotronis does not bear on this ground of appeal.
38. The second case was R v SSHD ex p Launder [1997] 1 WLR 838. This was a judicial review of a decision of the SSHD to extradite Mr Launder to Hong Kong. It is important to note that this was a decision made before the Human Rights Act 1998 made the Convention part of English domestic law. At that time the only basis on which an individual could rely upon the Convention was to say that the decision of the SSHD was flawed because of a failure to have regard to it. The case for the applicant was that it would not be just to return him to Hong Kong because of the risks he would face after the transfer of sovereignty over Hong Kong to China in July 1997.
39. The only substantive speech was given by Lord Hope. In a passage in which he was considering the ordinary principles to be adopted in extradition cases, he considered the topics of procedural impropriety, illegality and irrationality at pages 856G-859. He first rejected procedural impropriety and illegality at pages 856G-857A and then said this with regard to irrationality:

"As for irrationality, which Mr Alun Jones said was the only real issue in the case, this also seems to me to be a complaint which is without any real substance. The question whether it is unjust or oppressive to order the applicant's return to Hong Kong must in the end depend upon whether the PRC can be trusted in implement of its treaty obligations to respect his fundamental human rights, allow him a fair trial and leave it to the courts, if he is convicted, to determine the appropriate punishment.

It cannot be stressed too strongly that the decision in this matter rests with the Secretary of State and not at all with the court. The function of the court in the exercise of its supervisory jurisdiction is that of review. This is not an appeal against the Secretary of State's decision on the facts. His decision has had to be taken amidst an atmosphere of mistrust and suspicion which a court is in no position to penetrate. The visible part is the framework of law which I have described. That part can be explained and analysed. The invisible part is about the hearts and minds of those who will be responsible for the administration of justice in Hong Kong after the handover. This is not capable of analysis. It depends in the end, upon the exercise of judgment of a kind which lies beyond the expertise of the court. That, no doubt, is why the decision whether or not to grant the warrant has been entrusted to the Secretary of State by Parliament."

40. Then after considering some of the factual considerations, Lord Hope said at page 859D-E:

"The emphasis that [the SSHD] gave to the legal framework was not just, as Mr Vaughan asked us to accept, an assertion that the law is the law. It was the basis of his decision, which I have said is not an irrational one, that the PRC – despite its actions elsewhere and in other circumstances – could be relied upon to respect the law in the applicant's case and not to interfere in the process of justice in bringing him to trial in Hong Kong and, if he is convicted, imposing and enforcing the appropriate penalties. If that assumption is made it provides a rational and complete answer to all the questions."

41. Those passages seem to us to be of no assistance here because they focus on whether the SSHD acted irrationally or not. The question for SIAC was whether there were substantial grounds for believing that the respondents would face the risk of torture in the absence of the assurances by Libya in the MOU. That is a question of fact which it was SIAC's responsibility to resolve. As we have already said, in resolving it SIAC was obliged to consider what the risks were of Libya failing to comply with the assurances. If that involved a consideration of the hearts and minds of Colonel Qadhafi and members of his regime, so be it. It might be said that such questions are not justiciable because, in the words of Lord Hope, their resolution depends upon the exercise of judgment of a kind which lies beyond the expertise of the court. That is, however, not the SSHD's case. Nor could it be in the light of the Strasbourg jurisprudence culminating in Saadi v Italy.
42. Mr Sales seeks to adapt the principle stated by Lord Hope in order to support his submission that SIAC should have accepted the evidence of Mr Layden. The part of the speech of Lord Hope set out above seems to us to give no support to that submission. Mr Sales however relies upon later passages in his speech in which he considers the Convention. At page 867C-F Lord Hope considered the relevance of the Convention as follows:

"It is often said that, while the Convention may influence the common law, it does not bind the executive. This view was reflected in the observation by Sir Thomas Bingham MR in Reg v Ministry of Defence, Ex parte Smith [1996] QB 517, 558E that exercising an administrative discretion is not of itself a ground for impugning that exercise. That is so; but the whole context of the dialogue between the Secretary of State and the applicant in this case was the risk of an interference with the applicant's human rights. That in itself is a ground for subjecting the decisions to the most anxious scrutiny, in accordance with the principles laid down by this House in Reg v Secretary of State for the Home Department, Ex parte Bugdaycay [1987] AC 514, as Sir Thomas Bingham MR also recognised in Ex parte Smith, at p. 554H. Then there is the question whether judicial review proceedings can provide the applicant with an effective remedy, as article 13 requires where complaints are raised under the Convention in extradition and deportation cases: see Soering v United Kingdom (1989) 11 EHHR 248; D v United Kingdom, The Times, 12 May 1997. If the applicant is to have an effective remedy against a decision which is flawed because the decision-maker has misdirected himself on the Convention which he himself says he took into account, it must surely be right to examine the substance of the argument. The ordinary principles of judicial review permit this approach because it was to the rationality and legality of the decisions, and not to some independent remedy, that Mr. Vaughan directed his argument.

43. The arguments arising out of the Convention which Lord Hope then considered were (1) whether the SSHD correctly took into account the scope and content of the Convention; (2) whether the SSHD had personally and properly considered the applicant's individual case and circumstances; and (3) the rationality of the SSHD's decisions: see pages 867G-868A. Lord Hope then considered the first two arguments and continued at pages 868E-F:

"The remaining part of Mr. Vaughan's argument can be dealt with under the heading of irrationality. He maintained that the Convention required detailed scrutiny and evaluation of the facts in an extradition case where the person's human rights were at risk. He referred to Soering v. United Kingdom, 11 EHRR 439 to illustrate his point that confidence in the legal system was not enough. There had to be an examination of the domestic law and practice as it was applied in reality. In that case, on the facts, practice in the United States was at serious risk of failing to conform to the standards of the Convention, so the decision to order the detainee's surrender to that country was held to involve a breach of the Convention. Mr. Vaughan said that the Secretary of State had based his decision on a formal interpretation of the Joint Declaration, asserting that the law was the law, rather than an analysis of practice in Hong Kong."

44. Mr Sales submits that that passage shows that the House of Lords was giving direct consideration to questions which arose under the Convention. We do not agree. In our opinion the remaining passage in this part of Lord Hope's judgment shows that he was treating the argument as one of irrationality. He said at pages 868G-869B:

"Here again, however, the argument raises points which I have already dealt with under previous headings, and in particular in my examination of the question whether the decisions could be said to be irrational. I do not think that it is necessary to go over these points again. It is enough to say that the argument which Mr Vaughan presented under this heading seemed to me to be inextricably linked to those which he had already presented under the heading of irrationality. No new points of substance were raised in this branch of his argument. The decision which he says should have been taken would have had to have been based on the conclusion that, despite the provisions of the Joint Declaration and the Basic Law, the practice of the PRC in the field of human rights to date within its own territory showed that there was a serious risk that the provisions of these instruments would be departed from in Hong Kong SAR. But these arguments are not all one way on this point, as I have already sought to demonstrate. A reasonable Secretary of State could, on the material available to him, have concluded that the concerns were indicated by the PRC's actions in other places and in other circumstances were not so serious as to give rise to a serious risk of injustice or oppression in the applicant's case. The human rights context has not been overlooked in this assessment. On the contrary, it lies at the heart of the whole argument. It is precisely because it was not irrational for the Secretary of State to say that he was not persuaded that there was a case on human rights grounds for refusing extradition to Hong Kong that his decisions stand up to the required degree of scrutiny."

45. In these circumstances we are not persuaded that the decision or reasoning in ex p Launder assists the SSHD to sustain her submission that SIAC erred in principle in its approach to the evidence of Mr Layden.
46. The same is true of the third case upon which Mr Sales relies in this context, namely SSHD v Rehman [2003] 1 AC 153. Mr Sales submits that, in considering the weight due to the evidence of Mr Layden, SIAC should have adopted the same approach as the House of Lords held that it was required to adopt in relation to informed and considered assessments by the government about national security issues, where the practical experience and expertise available to the government was greater than that available to SIAC. He refers in particular to Lord Slynn at [22 and 26], Lord Steyn at [31] and Lord Hoffmann at [49, 53, 57-8, and 62].
47. We do not think that the statements in those paragraphs support Mr Sales' submission. The appeal to SIAC in Rehman was an appeal against a deportation order which had been made by the SSHD on the ground that it would be conducive to the public good in the interests of national security. The appeal was a review of the SSHD's assessment based upon the interests of national security. Lord Slynn put it thus at [26]:

"In conclusion even though the Commission has powers of review both of fact and of the exercise of the discretion, the Commission must give due weight to the assessment and conclusions of the Secretary of State in the light at any particular time of his responsibilities, or of government policy and the means at his disposal of being informed of and understanding the problems involved. He is undoubtedly in the best position to judge what national security requires even if his decision is open to review. The assessment of what is needed in changing circumstances is primarily for him."

Although Lord Steyn noted at [31] that issues of national security do not fall outside the competence of the courts, he added that it is self-evidently right that national courts must give great weight to the views of the executive on matters of national security.

48. The most detailed analysis is that of Lord Hoffmann at [54]:

This does not mean that the whole decision on whether deportation would be in the interests of national security is surrendered to the Home Secretary, so as to "defeat the purpose for which the Commission was set up": see the Commission's decision. It is important neither to blur nor to exaggerate the area of responsibility entrusted to the executive. The precise boundaries were analysed by Lord Scarman, by reference to Chandler v Director of Public Prosecutions [1964] AC 763 in his speech in Council of Civil Service Unions v Minister for the Civil Service [1985] AC 374, 406. His analysis shows that the Commission serves at least three important functions which were shown to be necessary by the decision in Chahal. First, the factual basis for the executive's opinion that deportation would be in the interests of national security must be established by evidence. It is therefore open to the Commission to say that there was no factual basis for the Home Secretary's opinion that Mr Rehman was actively supporting terrorism in Kashmir. In this respect the Commission's ability to differ from the Home Secretary's evaluation may be limited, as I shall explain, by considerations inherent in the appellate process but not by the principle of the separation of powers. The effect of the latter principle is only, subject to the next point, to prevent the Commission from saying that although the Home Secretary's opinion that Mr Rehman was actively supporting terrorism in Kashmir had proper factual basis, it does not accept that this was contrary to the interests of national security. Secondly the Commission may reject the Home Secretary's opinion on the ground that it was "one which no reasonable minister advising the Crown could in the circumstances reasonably have held". Thirdly, an appeal to the Commission may turn upon issues which at no point lie within the exclusive province of the executive. A good example is the question, which arose in Chahal itself, as to whether deporting someone would infringe his human rights under article 3 of the Convention because there was a substantial risk that he would suffer torture or inhuman or degrading treatment. The European jurisprudence makes it clear that whether deportation is in the interests of national security is irrelevant to rights under article 3. If there is a danger of torture, the Government must find some other way of dealing with a threat to national security. Whether a sufficient risk exists is a question of evaluation and prediction based on evidence. In answering such a question, the executive enjoys no constitutional prerogative."

49. The critical part of that analysis for present purposes is Lord Hoffmann's view that a good example of an issue which at no point lies within the exclusive province of the executive is the question whether deporting someone would infringe his rights under article 3 of the Convention. See also, to much the same effect, per Lord Hoffmann at [57].
50. The question what, if any, risks a deportee would face on return is a question of fact for SIAC. In considering that question SIAC must consider all the relevant evidence, including of course the evidence of an expert witness like Mr Layden. It must have regard to his particular experience and expertise and to the fact that his evidence reflects the views of the FCO but SIAC is in no way bound to accept every part of his evidence, provided that it gives rational reasons for not doing so.
51. In our judgment SIAC approached the evidence of Mr Layden on this part of the case correctly. It did so in the context of its approach to the assurances given by Libya, which it summarised at [319], immediately before the paragraphs relating to Mr Layden which we have set out above:

"The Commission has also been clear that the assessment of the value and effectiveness of assurances is less a matter of their text, though that can be relevant in showing what issues have been considered and what room may exist for a government to take a strictly legalistic view of what it has undertaken, and more a matter of the domestic political forces which animate a government and of the diplomatic and other pressures which may impel its performance of its obligations, or lead to quick discovery and redress for any breach."

That was a correct approach to the assurances and it is clear from SIAC's judgment that it appreciated that Mr Layden had particular expertise in identifying the forces and pressures which would or might govern the way in which Colonel Qadhafi responded to them.

52. Ultimately the question is whether SIAC erred in law in not accepting Mr Layden's view that it was well-nigh unthinkable that Libya would act in breach of the MOU. We do not accept Mr Sales' submission that, when SIAC said at the end of [320] that Mr Layden's view commands considerable respect "although it is very strong indeed" it discounted his evidence in some way. On the contrary, in our judgment, it is clear that when SIAC's reasoning is taken as a whole it had due regard to the strong views expressed by Mr Layden. We return below to SIAC's findings in this regard, which are at [345-371].

Failure to apply the proper test

53. Under this compendious heading in their skeleton argument counsel for the SSHD collected together the issues in the second, third and fourth grounds of appeal other than the issue whether SIAC correctly directed itself as to the test, which we have already discussed. In short the submission is that SIAC failed to apply the proper test under article 3. In particular it applied a test which was too undemanding and a threshold of such risk which was too low, in an attempt to take account of what it found to be the unpredictability of future events in Libya. Finally, it is submitted that SIAC's findings of fact are not capable of supporting a conclusion that the correct threshold had been passed.
54. In the course of the argument for the SSHD during the first two days of argument in this appeal, and before the judgment in Saadi v Italy was handed down, it was submitted that, when the Strasbourg jurisprudence was considered as a whole and, given in particular the stringent nature of the test as explained in the cases, in order to satisfy the test of real risk of treatment contrary to article 3 the respondents had to show that such treatment was more probable than not.
55. Once Saadi v Italy was available, the SSHD had no alternative but to abandon that submission because the ECtHR expressly rejected the argument of the United Kingdom, which had intervened in Saadi, that the test was one of the balance of probabilities: see [140]. The ECtHR also expressly rejected a further argument which had been advanced on behalf of the UK that, before holding that an individual could not be deported on article 3 grounds, the deporting state was entitled to balance the risk posed to its national security by the deportee against the risk to the deportee of treatment contrary to article 3: see [137-139].
56. Before Saadi v Italy was available we were naturally referred to a large number of Strasbourg decisions. It is now necessary for us to refer separately to very few of them because they are clearly summarised in Saadi. So far as relevant, the facts of Saadi were shortly these. An order was made against Mr Saadi in Italy for his deportation to Tunisia. Sometime later the ECtHR requested a stay of execution of his deportation on account of a real risk that he would be subject to treatment contrary to article 3 of the Convention: see [50] of the judgment of the ECtHR. On 29 May 2007 Italy sent Tunisia a note verbale requesting assurances that, if the applicant were to be deported to Tunisia he would not be subject to treatment contrary to article 3 and would not suffer a flagrant denial of justice. On 4 July 2007 Tunisia replied simply saying that it was prepared to accept the transfer of Tunisians imprisoned abroad once their identity had been confirmed "in strict conformity with the national legislation in force and under the sole safeguard of the relevant Tunisian authorities". In a second note verbale dated 10 July 2007 Tunisia added:

"The Minister of Foreign Affairs confirms that the Tunisian laws in force guarantee and protect the rights of prisoners in Tunisia and secure to them the right to a fair trial. The Minister would point out that Tunisia has voluntarily acceded to the relevant international treaties and conventions."

57. At [71] the ECtHR set out part of a report of Amnesty International expressing concern that Mr Saadi would be at risk of torture if he were returned to Tunisia and also stating that people who had recently been returned to Tunisia had been held in incommunicado detention, during which time they had been subjected to torture and other ill-treatment. The court also referred to reports of other bodies to similar effect.
58. The applicant's case was that the enforcement of his deportation would expose him to the risk of treatment contrary to article 3. After setting out the arguments of the parties and of the interveners, including the United Kingdom, the court summarised the relevant principles at [124-133] as follows, omitting some of the references:

"Responsibility of Contracting States in the event of expulsion

124. It is the Court's settled case-law that as a matter of well-established international law, and subject to their treaty obligations, including those arising from the Convention, Contracting States have the right to control the entry, residence and removal of aliens …. In addition, neither the Convention nor its Protocols confer the right to political asylum ….

125. However, expulsion by a Contracting State may give rise to an issue under Article 3, and hence engage the responsibility of that State under the Convention, where substantial grounds have been shown for believing that the person concerned, if deported, faces a real risk of being subjected to treatment contrary to Article 3. In such a case Article 3 implies an obligation not to deport the person in question to that country (see Soering v. the United Kingdom, … §§ 90-91; Vilvarajah [(1989) 14 EHRR 248] …, § 103; Ahmed …, § 39; H.L.R. v. France, judgment of 29 April 1997, Reports 1997-III, § 34; Jabari v. Turkey, no. 40035/98, § 38, ECHR 2000-VIII; and Salah Sheekh v. the Netherlands, no. 1948/04, § 135, 11 January 2007).

126. In this type of case the Court is therefore called upon to assess the situation in the receiving country in the light of the requirements of Article 3. Nonetheless, there is no question of adjudicating on or establishing the responsibility of the receiving country, whether under general international law, under the Convention or otherwise. In so far as any liability under the Convention is or may be incurred, it is liability incurred by the Contracting State, by reason of its having taken action which has as a direct consequence the exposure of an individual to the risk of proscribed ill-treatment (see Mamatkulov and Askarov v. Turkey [(2005) 41 EHRR 25] § 67, …

127. Article 3, which prohibits in absolute terms torture and inhuman or degrading treatment or punishment, enshrines one of the fundamental values of democratic societies. Unlike most of the substantive clauses of the Convention and of Protocols Nos. 1 and 4, Article 3 makes no provision for exceptions and no derogation from it is permissible under Article 15, even in the event of a public emergency threatening the life of the nation (see Ireland v. the United Kingdom, [(1978) 2 EHRR 25, § 163; Chahal, … § 79; Selmouni v. France [GC], no. 25803/94, § 95, ECHR 1999-V; Al-Adsani v. the United Kingdom [GC], no. 35763/97, § 59, ECHR 2001-XI; and Shamayev and Others v. Georgia and Russia, no. 36378/02, § 335, ECHR 2005-III). As the prohibition of torture and of inhuman or degrading treatment or punishment is absolute, irrespective of the victim's conduct (see Chahal, … § 79), the nature of the offence allegedly committed by the applicant is therefore irrelevant for the purposes of Article 3 (see Indelicato v. Italy, no. 31143/96, § 30, 18 October 2001, and Ramirez Sanchez v. France [GC], no. 59450/00, §§ 115-116, 4 July 2006).

Material used to assess the risk of exposure to treatment contrary to Article 3 of the Convention

128. In determining whether substantial grounds have been shown for believing that there is a real risk of treatment incompatible with Article 3, the Court will take as its basis all the material placed before it or, if necessary, material obtained proprio motu (see HLR v. France [(1997) 26 EHRR 29] § 37, and Hilal v. the United Kingdom, no. 45276/99, § 60, ECHR 2001-II). In cases such as the present the Court's examination of the existence of a real risk must necessarily be a rigorous one (see Chahal …, § 96).

129. It is in principle for the applicant to adduce evidence capable of proving that there are substantial grounds for believing that, if the measure complained of were to be implemented, he would be exposed to a real risk of being subjected to treatment contrary to Article 3 (see N. v. Finland, no. 38885/02, § 167, 26 July 2005). Where such evidence is adduced, it is for the Government to dispel any doubts about it.

130. In order to determine whether there is a risk of ill-treatment, the Court must examine the foreseeable consequences of sending the applicant to the receiving country, bearing in mind the general situation there and his personal circumstances (see Vilvarajah and Others …, § 108 in fine).

131. To that end, as regards the general situation in a particular country, the Court has often attached importance to the information contained in recent reports from independent international human-rights-protection associations such as Amnesty International, or governmental sources, including the US State Department (see, for example, Chahal, … §§ 99-100; Müslim v. Turkey, no.^o53566/99, § 67, 26 April 2005; Said v. the Netherlands, no. 2345/02, § 54, 5 July 2005; and Al-Moayad v. Germany (dec.), no.^o35865/03, §§ 65-66, 20 February 2007). At the same time, it has held that the mere possibility of ill-treatment on account of an unsettled situation in the receiving country does not in itself give rise to a breach of Article 3 (see Vilvarajah and Others, … § 111, and Fatgan Katani and Others v. Germany (dec.), no. 67679/01, 31 May 2001) and that, where the sources available to it describe a general situation, an applicant's specific allegations in a particular case require corroboration by other evidence (see Mamatkulov and Askarov, … § 73, and Müslim, cited above, § 68).

132. In cases where an applicant alleges that he or she is a member of a group systematically exposed to a practice of ill-treatment, the Court considers that the protection of Article 3 of the Convention enters into play when the applicant establishes, where necessary on the basis of the sources mentioned in the previous paragraph, that there are serious reasons to believe in the existence of the practice in question and his or her membership of the group concerned (see, mutatis mutandis, Salah Sheekh, … §§ 138-149).

133. With regard to the material date, the existence of the risk must be assessed primarily with reference to those facts which were known or ought to have been known to the Contracting State at the time of expulsion. However, if the applicant has not yet been extradited or deported when the Court examines the case, the relevant time will be that of the proceedings before the Court (see Chahal, …, §§ 85 and 86, and Venkadajalasarma v. the Netherlands, no. 58510/00, § 63, 17 February 2004). This situation typically arises when, as in the present case, deportation or extradition is delayed as a result of an indication by the Court of an interim measure under Rule 39 of the Rules of Court (see Mamatkulov and Askarov, … § 69). Accordingly, while it is true that historical facts are of interest in so far as they shed light on the current situation and the way it is likely to develop, the present circumstances are decisive."

59. As indicated earlier, at [137-140] the ECtHR rejected the submissions that the test should be that of the balance of probabilities and that there should be some kind of balance struck between the risk posed by the deportee to the United Kingdom and the risk posed to the deportee on return. It stressed the absolute nature of the right enshrined in article 3. At [139] the court stated that either the evidence reveals that there is a "substantial risk" if the person is sent back or it does not and observed that any threat posed by the deportee does not affect the risk faced by the deportee on return. These are two entirely different questions. At [140] the court rejected the "more likely than not" test and stressed the Chahal test in these words:

"On the contrary, [the court] reaffirms that for a planned forcible expulsion to be in breach of the Convention it is necessary – and sufficient – for substantial grounds to have been shown for believing that there is a real risk that the person concerned will be subjected in the receiving country to treatment prohibited by Article 3 (see paragraphs 125 and 132 above and the case-law cited in those paragraphs)."

60. Mr Sales submits that the use of the expression "substantial risk" in [139] emphasises that the test of real risk is a high test. However, as we see it, the effect of Saadi is to stress that the test is that identified in Soering and Chahal and other cases, namely that of real risk. A real risk is more than a mere possibility but something less than a balance of probabilities or more likely than not. We do not think that it is helpful further to elaborate the test beyond that stated in Saadi and the cases referred to in it.
61. There was some debate in the course of the argument as to what, if any, role the concept of proof beyond reasonable doubt plays in a case of this kind. In our judgment, it must follow from the rejection of the 'more likely than not' test in Saadi that the test cannot require the deportee to show that there are substantial grounds for believing that it is beyond reasonable doubt that he will be tortured or subject to other ill-treatment contrary to article 3. However, in so far as some of the cases refer to proof beyond reasonable doubt, with one possible exception, they were not saying that it must be beyond reasonable doubt that the deportee will be tortured on return; only that, where it is alleged that there has been an historical breach of article 3 by a deporting state, in assessing the evidence relating to the proof of that breach, the court applies the standard of proof "beyond reasonable doubt" (which the court then defined): Shamayev v Georgia and Russia at [338], citing Ireland v United Kingdom (1978) 2 EHRR 25 at [161], which in turn referred to The Greek Case, 1969 Yearbook of the Convention at [30], and Anguelova v Bulgaria no 38361/97, ECHR 2002-IV, at [111].
62. However, in Shamayev, after saying at [352] that a mere possibility of ill-treatment was not of itself enough to give rise to a breach of article 3 the ECtHR added at [353]:

"In consequence, the court considers that, in the light of the evidence in its possession, the facts of the case do not support "beyond any reasonable doubt" the assertion that, at the time when the Georgian authorities took the decision, there were no real or well-founded grounds to believe that extradition would expose the applicants to a real and personal risk of inhuman or degrading treatment, within the meaning of Article 3 of the Convention. There has accordingly been no violation of that provision by Georgia."

We do not think that in that passage the court can have intended to alter the principles which it had previously identified in a number of cases. Read in its context, we think that the reference to "beyond reasonable doubt" was a reference back to the discussion at [338], where (as just stated) the court was focusing on circumstances where it was alleged that there had been breaches of article 3. Historical breaches, which may of course be relevant to the assessment of risk, must be proved beyond reasonable doubt as defined by the ECtHR.

63. Subject to that, the principles are those summarised in Saadi, which contains no reference to proof beyond reasonable doubt. As we see it, it is not necessary further to elaborate the test or the principles beyond those set out in Saadi, which were the principles applied by SIAC in this case.
64. In this regard we note that there is nothing in Saadi to suggest that the test should be that of a real and immediate risk. That test has been developed in cases where the question is whether a state is under a positive obligation under article 2 of the Convention to take steps for or towards the preservation of life at the hands of others than the state. The leading case in the ECtHR is Osman v United Kingdom (1998) 29 EHRR 245, where the principles are stated at [115-116]. There have now been a number of decisions on this topic in these courts, including the comparatively recent decision of the House of Lords in In re Officer L [2007] UKHL 36, [2007] 1 WLR 2135. We do not think that that jurisprudence is of any real relevance here. It is no doubt for that reason that it is not discussed in Saadi.
65. However, we accept Mr Sales' submission that the principles in Saadi include the proposition that the applicable test is a stringent test which it is not easy to satisfy. This can be seen both in Saadi and many of the cases to which it refers and in cases in England and Wales. In the latter category it is sufficient to refer to R (Ullah) v Special Adjudicator [2004] UKHL 26, [2004] 2 AC 323, where, after reviewing the Strasbourg jurisprudence, Lord Bingham said at [24]:

"In relation to article 3, it is necessary to show strong grounds for believing that the person, if returned, faces a real risk of being subjected to torture or to inhuman or degrading treatment or punishment".

Lord Bingham then referred to a number of Strasbourg cases including Soering. In the jurisprudence of the ECtHR, as we have seen, the test is whether there are 'substantial' grounds for believing that there is a real risk of torture. We do not think that there is any difference between 'strong grounds' and 'substantial grounds'.

66. Saadi makes it clear that the test is rigorous. Thus in [142] the court said:

"Furthermore, the Court has frequently indicated that it applies rigorous criteria and exercises close scrutiny when assessing the existence of a real risk of ill-treatment (see Jabari, cited above, § 39) in the event of a person being removed from the territory of the respondent State by extradition, expulsion or any other measure pursuing that aim. Although assessment of that risk is to some degree speculative, the Court has always been very cautious, examining carefully the material placed before it in the light of the requisite standard of proof (see paragraphs 128 and 132 above) before indicating an interim measure under Rule 39 or finding that the enforcement of removal from the territory would be contrary to Article 3 of the Convention. As a result, since adopting the Chahal judgment it has only rarely reached such a conclusion."

67. There was a suggestion in the course of the argument that this principle was simply a reference to the anxious scrutiny that all cases of this kind require. We are not however able to accept that that is the case. The reference to rigorous criteria is to underline the fact that there must be strong or substantial grounds, based on evidence, for believing that there is a real risk of torture or other ill-treatment contrary to article 3. This can be seen from the reference to [128] and [132] of Saadi which we have quoted above. Thus the examination of the evidence must be rigorous and there must be serious reasons (motifs serieux et avères) to believe in the risk of ill-treatment.
68. As already stated, it was conceded on behalf of the SSHD that the test, however rigorous or difficult to surmount, would have been satisfied but for the assurances. In Saadi the ECtHR described its own role in a case where assurances are relied on in this way at [148]:

"Furthermore, it should be pointed out that even if, as they did not do in the present case, the Tunisian authorities had given the diplomatic assurances requested by Italy, that would not have absolved the Court from the obligation to examine whether such assurances provided, in their practical application, a sufficient guarantee that the applicant would be protected against the risk of treatment prohibited by the Convention (see Chahal, cited above, § 105). The weight to be given to assurances from the receiving State depends, in each case, on the circumstances obtaining at the material time."

69. Consistently with that approach, it was for SIAC to examine whether the assurances given by Libya, in their practical application, were a sufficient guarantee that the respondents would be protected against torture. The weight to be given to the assurances depended upon the facts of this particular case. It can thus be seen that the exercise upon which SIAC was embarking was an investigation of fact, leading to a conclusion of fact. In our judgment, if SIAC made any error (and we do not divine one), it was an error of fact and not an error of law.
70. We entirely accept that SIAC found that Libya signed the MOU in good faith (see [336-340]) and that if it and various subsequent notes verbales were adhered to, there would be no basis for concluding that there would be a risk of a breach of article 3 on the respondents' return to Libya: see [297]. SIAC analysed the future risks at [345-371], where it considered the evidence of Mr Layden in considerable detail. It would not be appropriate to quote the whole of that analysis here. In summary, SIAC held that Libya would probably keep its word but that there was a risk that it would not. The reason why the assurances would be honoured was not because Colonel Qadhafi or his regime were men who could be trusted to keep their word as men of integrity and honour but because it would be in their interests to do so. It was in the interests of Libya to honour the assurances but there remained an important element of unpredictability as to what Colonel Qadhafi might do.
71. SIAC identified these problems at [347-349]:

"347. First, this combination of pragmatism with a mercurial personality, which we take to mean for these purposes changeability of view and unpredictability of reaction, highlights a contrast and even a conflict between a short term reaction and a longer term course to which the regime would revert after the short term reaction had worn off. This may mean that for a short period, but of uncertain duration, some course divergent from the longer term course could be followed. The regime may see that longer term course as continuing so far as it is concerned, possibly at some short term cost. It might see no incompatibility between its longer and shorter term courses.

348. Second, the way in which Colonel Qadhafi sees his pragmatic interest in his survival may itself be unpredictable and need not to western eyes, be rational or in his self-interest. That has been so in the past, and indeed there is plenty of evidence over many years of Colonel Qadhafi adopting an approach which must have reflected his assumed pragmatism for his regime's survival, but which Mr Layden saw as counter-productive, contrary to Colonel Qadhafi's own best interests. This had happened in the short term as well. Colonel Qadhafi may well see a course of conduct as necessary for his survival which Mr Layden would regard as unlikely to be conducive to that end.

349. Third, if Colonel Qadhafi has adopted a pragmatic approach in the past, based on his perception of what would preserve and enhance the regime's long term prospects of survival, that pragmatism is also compatible with torture, incommunicado detention, and unfair trials. Indeed, much of the purpose of those abuses would have been to enable the regime to remain in power. It would be possible to give a "pragmatic" interpretation to any of those, especially if reprisals are allowed as a rational form of policy. Such an interpretation could be attempted for the Lockerbie bombing."

There followed a consideration at [350] as to whether the Lockerbie attack had the blessing of Colonel Qadhafi. Mr Layden thought not but SIAC observed that that would mean that others could misunderstand what he intended.

72. SIAC then considered the trial of the Bulgarian nurses in Benghazi, which SIAC described in detail at [153-160]. In brief, in 1999 a group of Bulgarian nurses were arrested on charges of deliberately infecting Libyan patients with the HIV virus. In 2004, they were convicted and sentenced to death after a trial based on evidence obtained by torture. The view of the diplomatic community was that the trial had been unfair and the outcome unjust. There was evidence that the outbreak of HIV had been due to poor hygiene in the hospital concerned. On appeal in 2005, a retrial was ordered. The diplomatic community, including Mr Layden, confidently expected that the retrial would result in acquittal. Mr Layden attended the trial and was deeply dismayed both at its unfairness and at the result, which was the reconviction of the defendants and the reimposition of the death sentence. Mr Layden's reaction was shared by the diplomatic community (and the world press). Diplomatic activity increased and, in due course, the nurses were repatriated to Bulgaria, where they were pardoned. In respect of this history, SIAC said at [351-2]:

"351. The willingness of the regime to endure international opprobrium and diplomatic pressure, whether in pragmatic mode or in short term changeable mode, in a way which cannot be explained other than by the vital importance of maintaining a particular domestic posture, can be seen in the way in which the Bulgarian medics have been dealt with. It is a dismal story of injustice, despite the high level, persistent diplomatic pressure and adverse publicity which the trial and detention have attracted. If it is the position, as we accept, that Colonel Qadhafi could determine the outcome of the trial, he plainly has not done so in response to that very considerable external pressure. There must be overriding considerations of domestic politics which outweigh in his mind all other considerations.

352. These may illustrate the difficulty which Colonel Qadhafi has in seeing a trial lead to acquittals when he has pronounced on the defendants' guilt, the difficulty he has in accepting that the medical system run by the state might have been to blame for the outbreak of HIV/AIDS, the effect of local pressure from the families for a guilty verdict and the counter-productive effect of public external pressure and publicity. We would accept that the outcome also suggests that Mr Layden has been over-optimistic in his assessment of when the ordeal would be over."

73. SIAC was in our judgment entitled to form the view that, with respect to the Bulgarian nurses, Mr Layden had indeed expressed an over-optimistic view and that the trial was an example of the risks which would or might face the respondents if returned to Libya. As SIAC put it at [353], this illustrated the limitations of judging reliability by the pursuit of self-interest as assessed from the outside. SIAC was entitled to find that that what is pragmatic to Colonel Qadhafi may not be so to western eyes.
74. At [354] SIAC expressed this conclusion:

"We conclude that the pragmatism of the regime in its own self-interest is not sufficient itself to exclude a real risk that it would act discordantly with that long term course, temporarily or on occasions, whilst still taking the view that they were acting pragmatically to ensure its survival. The question is whether such temporary or occasional acts would lead the regime leaders or others to breach the MOU, particularly with regard to the way in which the Appellants would be treated in detention or during questioning. Certainly, the past and current practices of the regime and its security organisations show that violence and human rights abuses are regarded as legitimate, even necessary, weapons to be deployed to protect the regime or to punish opponents. There is no institutional or personal rejection of such acts when used to those ends."

Mr Sales submits that here, as in other parts of the judgment, SIAC erred in principle in asking itself whether the pragmatism of the regime was sufficient to exclude a real risk of torture. He submits that that is to put the test the wrong way round.

75. We do not accept that submission. It was correctly accepted that (however the test is precisely formulated or defined), but for the assurances in the MOU, there were substantial grounds for believing that the respondents would face a real risk of torture on return. In these circumstances the question was indeed whether the pragmatism of the regime was sufficient to exclude that risk, or at least to reduce it to an acceptable level. This case is quite different from cases like Mamatkulov and Askarov v Turkey and Shamyev v Georgia. In any event each case must be decided on its own facts.
76. At [355] SIAC recognised that the particular events which might lead Colonel Qadhafi or the heads of the security organisations to take action contrary to what might objectively be thought to be the pragmatic course were unpredictable. They might include the recrudescence of LIFG in Libya which might tempt the regime to interrogate members of LIFG including the respondents by the use of torture: see [356]. Colonel Qadhafi might make a statement which was hostile to Islamist terrorists and the respondents might be tortured by over-enthusiastic hardliners in the regime: see [357]. The same might happen if the UK or others were perceived to be responsible for a sleight against Libya or Colonel Qadhafi: see [358].
77. SIAC said at [360] that it would expect all to be well initially and accepted at [361] that there was an element of speculation about the risks. It said this:

"361. Indeed, we would accept that there is an element of speculation about how any change of approach might occur in what we have set out. That is inevitable in this case for what we are satisfied about is that there is a considerable element of unpredictability which we do have to consider. That is where the risk first arises and it could result from a number of actions. We have to do what we can to assess its degree, causes and impacts. We are satisfied that there are real risks of such events occurring, which could lead to acts which diverge from the pragmatic course as Mr Layden would see it, even though the divergence would be occasional, responding to events, or temporary. These are not in our judgment unrealistic scenarios."

We reject Mr Sales' submission that SIAC there misapplied the test. It correctly said that it was for it to assess the degree, causes and impacts of the risk and said that it was satisfied that there were real risks of the contemplated events occurring. That was a correct statement of the test, with the nature of which SIAC was very familiar. Moreover the statement that those were 'not unrealistic scenarios' seems to us, when read in the context of the paragraph as a whole, to mean that they were realistic scenarios.

78. SIAC then identified the risks which it had in mind, seen in the context of what it called widespread and grave human rights abuses of political opponents in detention. They identified five risks in [362]:

"First, while those returned under the MOU might well be spared any simple if widespread reprisal in the event of a violent attack against the regime, they could well be subjected to treatment which breached Article 3 during the course of interrogation as part of the investigation into such an attack. Second, any one of the three intelligence services could conclude that it wanted more information from the Appellants which it believed they had, whether for a trial of some other defendant or for intelligence purposes. These would be newcomers to them and could have information which they felt had not been divulged. An absence of co-operation could be resented if there were a growing body of LIFG members in Libya returning and rebuilding its infrastructure there, and especially so if the Libyans believed that the UK was unable or unwilling to obtain or provide the information which they felt they needed. The Libyan intelligence or security organisations may not always operate in harmony rather than in competition. Third, if some grievance or slight were felt against the UK, the reaction could be to place an Appellant in a political prison, and in the case of a prison run by the Judicial Police but which had a political wing, a transfer would not be difficult to arrange, nor a return. But life in the political wings or prisons would involve a probable breach of Article 3. Either of those last two reactions could occur at any stage during detention, before or after trial and conviction. Fourth, although it would not necessarily lead to a breach of Article 3, Colonel Qadhafi could give instructions, or be interpreted as having done so, for the conviction of the Appellants and for the sentence, whether as a long term of imprisonment or as the death penalty. Appeals and any commutation of the death penalty could then be long delayed or used as a bargaining counter with the UK. The judicial and the political part of the commutation process in the HJC could be delayed for any number of reasons. Fifth, any desire to obtain a conviction could be reinforced by interrogation in breach of Article 3, to obtain a confession."

79. We detect no error of principle in the approach adopted by SIAC in identifying these risks, either in its approach to the evidence of Mr Layden or otherwise. SIAC then considered whether there were factors which would deter such breaches of the MOU. In particular, it considered the role of the Qadhafi Development Foundation (QDF) which was the body jointly appointed, by the parties to the MOU, for the purpose of providing independent monitoring of the implementation of the assurances. QDF was the only monitoring body available but the problem was (and is) that it is not independent of the regime, because its head is one of Colonel Qadafi's sons. As SIAC put it at [364], the problem with QDF was that it would be least effective when most needed. In these circumstances SIAC concluded at [365] that there was a very real prospect that any breach of the MOU could go undetected or undetected for a long time.
80. At [368] SIAC reiterated that the trial of the Bulgarian nurses showed that the regime could be impervious to international pressure and at [369-370] it noted that the regime did not respond promptly to private pressure, as for example from the United Kingdom. Finally at [371] SIAC set out its conclusion in these terms:

"We have accordingly come to the conclusion that although it is probable that Mr Layden's judgment as to how the Libyans would observe the MOU in relation to the physical treatment of the Appellants is sound, and that they would not be ill-treated in a way which breached Article 3, we cannot adopt his conclusion that that would be well-nigh unthinkable. Instead we think that there is a real risk that that would happen. The need in this case to make a large allowance for the unpredictable reaction, which in the short term or occasionally diverges from the pragmatic path upon which the Libyans are set means that we cannot eliminate the real risk which we have identified. The fact that the direction of Libyan foreign relations would largely remain the same does not remove the risk. There are no domestic changes, institutions or considerations which would assist. Above all the risk is not reduced sufficiently by the monitoring system because it is at these times that its limitations would be most evident and felt. We have to bear in mind that the monitoring system is intended to deter and check on potential breaches which can occur quite quickly, and to alert the UK's diplomats to the problem rapidly. The diplomatic pressure which the UK could bring to bear and the responses adverse to Libya's interests which it could deploy, would not be engaged if the monitoring were ineffective to report on possible abuse. We do not therefore have the confidence which we need to have, for the return of the Appellants not to breach the UK's international obligations. In short there is too much scope for something to go wrong, and too little in place to deter ill-treatment or to bring breaches of the MOU to the UK's attention."

81. In conclusion we do not accept that SIAC misunderstood or misapplied the test or that it failed to have proper regard to the evidence of Mr Layden. Nor are we persuaded that SIAC lowered the test to take account of the unpredictability of future events. SIAC was bound to have regard to the unpredictability of the future. It understood that the risk of mistreatment was contingent upon other events occurring; but that will often be the case. As we see it, SIAC fully understood and sought to apply the correct test. Its responsibility was to consider the many pieces of evidence in a complex picture and to decide whether there were substantial grounds for believing that there was a real risk that the respondents would be tortured some time after their return to Libya, notwithstanding the terms of the MOU. We are not persuaded that SIAC erred in principle by either misstating or misapplying the test.
82. Nor are we persuaded that SIAC's findings of fact are not capable of supporting the conclusion that the correct threshold was passed. This seems to us in effect to be a rationality challenge. On the assumption, which we have held to be correct, that SIAC identified the correct test and sought to apply it to the facts found, we can see no basis for holding that the findings of fact, namely that there were substantial grounds for believing there was a real risk of the respondents being tortured, despite the terms of the MOU, were not capable of satisfying the test or were otherwise irrational.

Conclusion

83. For these reasons the appeal of the SSHD must be dismissed. We would only add that, in reaching that conclusion, we have tried, as AH (Sudan) says we must, to consider the very lengthy reasons in SIAC's judgment as a whole. In the end we agree with Mitting J that the issues raised by this case before SIAC were important but they were issues of fact.
84. Finally, we should note that the appeal of the SSHD was conducted entirely on the open material and by an analysis of the open judgment. The SSHD has not sought to rely upon the closed material or the closed judgment. However, our attention has been drawn by the special advocates to some aspects of the closed judgment. We do not think that it is necessary for us to prepare a closed judgment or to analyse SIAC's closed reasoning in any way apart from saying that it confirms the conclusions which we have expressed in this judgment.

Article 6

85. It was submitted before SIAC that if the respondents were returned to Libya they would be denied a fair trial and that for that reason the appeal against deportation should be allowed. SIAC gave some consideration to this issue and the issue would be relevant in this court if the appeal of the SSHD on the article 3 point succeeded. However, since it has failed, there is no need for us to consider article 6 in this context. The correct approach to the risks of a trial in a receiving state which is not a party to the Convention is central to the issues in the appeal in the case of Othman, which forms a separate judgment to be handed down on the same day as this. In these circumstances we say nothing further about it here.