R (Iran) & Ors v Secretary of State for the Home Department

Lord Justice Brooke: This is the judgment of the court. Part 1 Introductory: the powers of the IAT under the 2002 Act

1. These five appeals were listed before the same division of the court in a single week in order to enable us to address a number of questions that have arisen since the jurisdiction of the former Immigration Appeal Tribunal (“IAT”) was restricted to appeals on a point or points of law. Although the IAT has now been abolished and replaced by the single tier Asylum and Immigration Tribunal (“AIT”), much of what we say will be equally relevant to the new statutory regime, in which the power to reconsider an original decision in the AIT will only arise in relation to issues of law. We are setting out the detailed facts of these appeals in the Appendix to this judgment. In this main judgment we will be concerned to explain the relevant principles of law, and then to state briefly how they are to be applied to the individual cases whose details are to be found in the Appendix.

2. The jurisdiction of the adjudicator in all these cases was conferred by s 82(1) of the Nationality, Immigration and Asylum Act 2002 (“ the 2002 Act ”) because all the impugned decisions of the Secretary of State postdated 1 st April 2003. Section 84(1) (g) of the 2002 Act provided that an appeal against an immigration decision (a phrase which includes a decision to remove) might be brought on the ground, among others: “(g) that [the appellant’s] removal from the United Kingdom in consequence of the immigration decision would breach the United Kingdom’s obligations under the Refugee Convention or would be unlawful under section 6 of the Human Rights Act 1998 as being incompatible with the appellant’s Convention rights.”

3. Section 86(3) provided that the adjudicator must allow the appeal in so far as he thought that: “(a) a decision against which the appeal is brought was not in accordance with the law (including immigration rules) or (b) a discretion exercised in making a decision against which the appeal is brought … should have been exercised differently.”

4. In each of the present cases the decision of the adjudicator postdated 9 th June 2003, so that appeal to the IAT lay pursuant to s 101(1) of the 2002 Act , which provided that: "(1) A party to an appeal to an adjudicator under section 82 … may, with the permission of the Immigration Appeal Tribunal, appeal to the Tribunal against the adjudicator's determination on a point of law."

5. Section 102, for its part, provided that: "(1) On an appeal under section 101 the Immigration Appeal Tribunal may - (a) affirm the adjudicator's decision; (b) make any decision which the adjudicator could have made; (c) remit the appeal to an adjudicator; (d) affirm a direction given by the adjudicator under section 87; (e) vary a direction given by the adjudicator under that section; (f) give any direction which the adjudicator could have given under that section. (2) In reaching their decision on an appeal under section 101 the Tribunal may consider evidence about any matter which they think relevant to the adjudicator's decision, including evidence which concerns a matter arising after the adjudicator's decision. … (4) In remitting an appeal to an adjudicator under subsection (1)(c) the Tribunal may, in particular – (a) require the adjudicator to determine the appeal in accordance with directions of the tribunal; (b) require the adjudicator to take additional evidence with a view to the appeal being determined by a Tribunal.” Part 2 Recent authoritative guidance: the power of the IAT to correct errors of fact

6. During the course of the last two years the House of Lords and this court have been resolving issues of substantive law relating to the interface between the Human Rights Act 1998 and the immigration and asylum legislation. This court has also been resolving issues of procedural law and the law of evidence in relation to the appellate jurisdiction of both the IAT and of this court (to whom appeals have always lain from the IAT on a point of law only from the time that such a direct route of appeal was first created). Some of the decisions relating to procedure and evidence, however, were concerned with pre-June 2003 cases in which appeals on both law and fact lay to the IAT. In the course of this judgment we will have to review the relevance of some of these decisions in relation to the period between June 2003 and April 2005 when appeal only lay to the IAT on a point of law.

7. The judgments of this court in Indrakumar v SSHD [2003] EWCA Civ 1677 ; [2004] Imm AR 76 and Subesh v SSHD [2004] EWCA Civ 56 ; [2004] Imm AR 112 provided general advice to the IAT which related to the days when appeal lay to the IAT against the findings of fact by an adjudicator. So far as findings of fact were concerned, the burden lay on an appellant to persuade the IAT not merely that a different version of the facts was reasonable and possible, but that there were objective grounds upon which the IAT ought to conclude that this different version was the right one.

8. In reviewing the findings of fact made by an adjudicator, Laws LJ made it clear in Subesh that the IAT were to follow these guidelines: i) It would only very rarely be able to overturn a finding of fact based on oral evidence and the assessment of credibility; ii) It could more readily overturn a finding of fact based on documentary evidence specific to the individual case (because the IAT was in just as good a position to assess such evidence), but great caution would be required in those cases where there might be an important relationship between the assessment of the person involved and the assessment of those documents; iii) The IAT would be at least as well placed as the adjudicator to assess findings as to the general conditions, or the backdrop, in the country concerned which would be based on the objective country evidence; the more so if the adjudicator had departed without solid justification from a relevant IAT country guidance decision; iv) The IAT would be entitled to draw its own inferences as to the application of those general country conditions to the facts of the particular case. Part 3 The jurisdiction to correct errors of law: examples of errors of law commonly encountered

9. When the court gave this guidance in Subesh , it was aware that it would not be of any relevance to an appellate regime in which appeals were restricted to points of law. It may be convenient to give a brief summary of the points of law that will most frequently be encountered in practice: i) Making perverse or irrational findings on a matter or matters that were material to the outcome (“material matters”); ii) Failing to give reasons or any adequate reasons for findings on material matters; iii) Failing to take into account and/or resolve conflicts of fact or opinion on material matters; iv) Giving weight to immaterial matters; v) Making a material misdirection of law on any material matter; vi) Committing or permitting a procedural or other irregularity capable of making a material difference to the outcome or the fairness of the proceedings; vii) Making a mistake as to a material fact which could be established by objective and uncontentious evidence, where the appellant and/or his advisers were not responsible for the mistake, and where unfairness resulted from the fact that a mistake was made.

10. Each of these grounds for detecting an error of law contain the word “material” (or “immaterial”). Errors of law of which it can be said that they would have made no difference to the outcome do not matter. This need to identify an error of law which would have made a material difference to the outcome was at the relevant time underscored by Rule 17(3) of the Immigration and Asylum Appeals (Procedure) Rules 2003, which provided that: “(3) The grounds of appeal must – (a) identify the alleged errors of law in the adjudicator’s determination; and (b) explain why such errors made a material difference to the outcome.” Part 4 Perversity, the failure to give reasons, and proportionality

11. It may be helpful to comment quite briefly on three matters first of all. It is well known that “perversity” represents a very high hurdle. In Miftari v SSHD [2005] EWCA Civ 481 , the whole court agreed that the word meant what it said: it was a demanding concept. The majority of the court (Keene and Maurice Kay LJJ) said that it embraced decisions that were irrational or unreasonable in the Wednesbury sense (even if there was no wilful or conscious departure from the rational), but it also included a finding of fact that was wholly unsupported by the evidence, provided always that this was a finding as to a material matter.

12. We mention this because far too often practitioners use the word “irrational” or “perverse” when these epithets are completely inappropriate. If there is no chance that an appellate tribunal will categorise the matter of which they make complaint as irrational or perverse, they are simply wasting time – and, all too often, the taxpayer’s resources – by suggesting that it was.

13. The second preliminary matter is this. Adjudicators were under an obligation to give reasons for their decisions (see reg 53 of the Immigration and Asylum Appeals (Procedure) Regulations 2003), so that a breach of that obligation may amount to an error of law. However, unjustified complaints by practitioners that are based on an alleged failure to give reasons, or adequate reasons, are seen far too often. The leading decisions of this court on this topic are now Eagil Trust Co Ltd v Pigott-Brown [1985] 3 All ER 119 and English v Emery Reimbold & Strick Ltd [2002] EWCA Civ 605 , [2002] 1 WLR 2409 . We will adapt what was said in those two cases for the purposes of illustrating the relationship between an adjudicator and the IAT. In the former Griffiths LJ said at p 122: “[An adjudicator] should give his reasons in sufficient detail to show the [IAT] the principles on which he has acted and the reasons that have led him to his decision. They need not be elaborate. I cannot stress too strongly that there is no duty on [an adjudicator], in giving his reasons, to deal with every argument presented by [an advocate] in support of his case. It is sufficient if what he says shows the parties and, if need be, the [IAT], the basis on which he has acted, and if it be that the [adjudicator] has not dealt with some particular argument but it can be seen that there are grounds on which he would have been entitled to reject it, [the IAT] should assume that he acted on those grounds unless the appellant can point to convincing reasons leading to a contrary conclusion.”

14. In English Lord Phillips MR said at para 19: “[I]f the appellate process is to work satisfactorily, the judgment must enable the [IAT] to understand why the [adjudicator] reached his decision. This does not mean that every factor which weighed with the [adjudicator] in his appraisal of the evidence has to be identified and explained. But the issues the resolution of which were vital to the [adjudicator]’s conclusion should be identified and the manner in which he resolved them explained. It is not possible to provide a template for this process. It need not involve a lengthy judgment. It does require the [adjudicator] to identify and record those matters which were critical to his decision. If the critical issue was one of fact, it may be enough to say that one witness was preferred to another because the one manifestly had a clearer recollection of the material facts or the other gave answers which demonstrated that his recollection could not be relied upon.”

15. It will be noticed that the Master of the Rolls used the words “vital” and “critical” as synonyms of the word “material” which we have used above. The whole of his judgment warrants attention, because it reveals the anxiety of an appellate court not to overturn a judgment at first instance unless it really cannot understand the original judge’s thought processes when he/she was making material findings.

16. What we have said does not absolve an adjudicator of his/her duty of devoting the intense scrutiny to the appellant’s case that is required of a decision of such importance. What we wish to make clear, however, is that the practice of bringing appeals because the adjudicator or immigration judge has not made reasoned findings on matters of peripheral importance must now come to an end.

17. As to the third preliminary matter, when an issue arises on an appeal that there is a risk that one of the appellant’s ECHR rights would be violated if he/she were removed to another country, it is now well established that the adjudicator must make the decision himself, following the guidance given by this court in Huang v SSHD [2005] EWCA Civ 105 . One of the cases with which we are concerned raises an issue under Article 8 of the ECHR, of a kind that often arises in asylum cases, and it will be useful to describe the processes that should be followed both at adjudicator and IAT level in relation to such a claim.

18. The relevant substantive law is now reasonably clear following the decision of the House of Lords in R (Ullah) v SSHD [2004] UKHL 26 ; [2004] 2 AC 323 and R (Razgar) v SSHD [2004] UKHL 27 ; [2004] 2 AC 368 , and the decision of this court in Huang v SSHD [2005] EWCA Civ 105 . In Razgar Lord Bingham said at paras 17-20: "17. In considering whether a challenge to the Secretary of State's decision to remove a person must clearly fail, the reviewing court must, as it seems to me, consider how an appeal would be likely to fare before an adjudicator, as the tribunal responsible for deciding the appeal if there were an appeal. This means that the reviewing court must ask itself essentially the questions which would have to be answered by an adjudicator. In a case where removal is resisted in reliance on article 8, these questions are likely to be: (1) Will the proposed removal be an interference by a public authority with the exercise of the applicant's right to respect for his private or (as the case may be) family life? (2) If so, will such interference have consequences of such gravity as potentially to engage the operation of article 8? (3) If so, is such interference in accordance with the law? (4) If so, is such interference necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others? (5) If so, is such interference proportionate to the legitimate public end sought to be achieved?

18. If the reviewing court is satisfied in any case, on consideration of all the materials which are before it and would be before an adjudicator, that the answer to question (1) clearly would or should be negative, there can be no ground at all for challenging the certificate of the Secretary of State. Question (2) reflects the consistent case law of the Strasbourg court, holding that conduct must attain a minimum level of severity to engage the operation of the Convention: see, for example, Costello-Roberts v United Kingdom (1993) 19 EHRR 112 . If the reviewing court is satisfied that the answer to this question clearly would or should be negative, there can again be no ground for challenging the certificate. If question (3) is reached, it is likely to permit of an affirmative answer only.

19. Where removal is proposed in pursuance of a lawful immigration policy, question (4) will almost always fall to be answered affirmatively. This is because the right of sovereign states, subject to treaty obligations, to regulate the entry and expulsion of aliens is recognised in the Strasbourg jurisprudence (see Ullah and Do, para 6) and implementation of a firm and orderly immigration policy is an important function of government in a modern democratic state. In the absence of bad faith, ulterior motive or deliberate abuse of power it is hard to imagine an adjudicator answering this question other than affirmatively.

20. The answering of question (5), where that question is reached, must always involve the striking of a fair balance between the rights of the individual and the interests of the community which is inherent in the whole of the Convention. The severity and consequences of the interference will call for careful assessment at this stage. The Secretary of State must exercise his judgment in the first instance. On appeal the adjudicator must exercise his or her own judgment, taking account of any material which may not have been before the Secretary of State. A reviewing court must assess the judgment which would or might be made by an adjudicator on appeal. In Secretary of State for the Home Department v Kacaj [2002] Imm AR 213, para 25, the [IAT] … observed that: 'although the [Convention] rights may be engaged, legitimate immigration control will almost certainly mean that derogation from the rights will be proper and will not be disproportionate.' In the present case, the Court of Appeal had no doubt … that this overstated the position. I respectfully consider the element of overstatement to be small. Decisions taken pursuant to the lawful operation of immigration control will be proportionate in all save a small minority of exceptional cases, identifiable only on a case by case basis." When he turned to Mr Razgar’s case, Lord Bingham said in para 24 that when an adjudicator reached Question (5), he might well decide it against Mr Razgar.

19. When he analysed the effect of this speech in para 40 of his judgment in Huang , Laws LJ said that it could be inferred from paras 20 and 24 that Lord Bingham contemplated that an adjudicator would properly have arrived at his own decision on the merits in Razgar , had he been called on to answer Question (5). He then held that the Court of Appeal must jettison its earlier case law and consider the situation afresh. He eventually concluded (at paras 59-60): “It might be said that the Immigration Rules constitute for all cases the balance to be struck between private right and public interest, and this is conclusive for any judgment in an Article 8 case as to whether removal or deportation is proportionate and so justified under Article 8(2). But the Secretary of State rightly does not so contend. If that were the law, our municipal statute need do no more than confer a right of appeal to allow the immigrant to contend that on the true facts he has a good claim under the Rules. However, whatever else may be said about the relation between s 65(1) and paragraph 21(1) of Schedule 4 to the 1999 Act, it is surely plain that the legislature contemplated appeals on Convention grounds, including Article 8, which might succeed even though the appellant had no good claim under the Rules. The true position in our judgment is that the HRA and s 65(1) require the adjudicator to allow an appeal against removal or deportation brought on Article 8 grounds if, but only if, he concludes that the case is so exceptional on its particular facts that the imperative of proportionality demands an outcome in the appellant's favour notwithstanding that he cannot succeed under the Rules. In such a case the adjudicator is not ignoring or overriding the Rules. On the contrary it is a signal feature of his task that he is bound to respect the balance between public interest and private right struck by the Rules with Parliament's approval. That is why he is only entitled on Article 8 grounds to favour an appellant outside the Rules where the case is truly exceptional. This, not Wednesbury or any revision of Wednesbury , represents the real restriction which the law imposes on the scope of judgment allowed to the adjudicator. It is not a question of his deferring to the Secretary of State's judgment of proportionality in the individual case. The adjudicator's decision of the question whether the case is truly exceptional is entirely his own. He does defer to the Rules; for this approach recognises that the balance struck by the Rules will generally dispose of proportionality issues arising under Article 8; but they are not exhaustive of all cases. There will be a residue of truly exceptional instances. In our respectful view such an approach is also reflected in Lord Bingham's words in Razgar , which we have already cited: ‘Decisions taken pursuant to the lawful operation of immigration control will be proportionate in all save a small minority of exceptional cases, identifiable only on a case by case basis.’” In Huang the adjudicator was operating under the different regime set out in the Immigration and Asylum Act 1999 (“ the 1999 Act ”). There is, however, no material difference in the language of the 2002 Act .

20. It follows that if an adjudicator correctly directed himself as to his duty under the law, and clearly adopted the approach prescribed in Razgar and Huang , then the IAT could not as a matter of law interfere with his decision, which will usually involve a judgment on proportionality, except on traditional public law lines (for which see para 9 above). He is the judicial decision-maker, and his is the obligation of intense scrutiny described by Lord Steyn in R (Daly) v SSHD [2001] UKHL 26 at [27]; [2001] 2 AC 532 . Part 5 Country Guidance cases

21. Three matters require more detailed treatment. The first relates to the practice of the IAT of giving “country guidance” (“CG”) decisions. This practice has proved to be so useful that it is now firmly embedded in the recently published Practice Directions of the AIT in these terms: “18.2 A reported determination of the Tribunal or of the IAT bearing the letters ‘CG’ shall be treated as an authoritative finding on the country guidance issue identified in the determination, based upon the evidence before the members of the Tribunal or the IAT that determined the appeal. As a result, unless it has been expressly superseded or replaced by any later ‘CG’ determination, or is inconsistent with other authority that is binding on the Tribunal, such a country guidance case is authoritative in any subsequent appeal, so far as that appeal: a) relates to the country guidance issue in question; and b) depends upon the same or similar evidence. 18.3 A list of current ‘CG’ cases will be maintained on the Tribunal website. Both the respondent and any representative of the appellant in an appeal concerning a particular country will be expected to be conversant with the current ‘CG’ determinations relating to that country. 18.4 Because of the principle that like cases should be treated in like manner, any failure to follow a clear, apparently applicable country guidance case or to show why it does not apply to the case in question is likely to be regarded as grounds for review or appeal on a point of law.”

22. The principle that like cases should be treated in a like manner is another way of describing what Lord Hoffmann described in Arthur J S Hall & Co v Simons [2002] 1 AC 615 , 688H as “the fundamental principle of justice which requires that people should be treated equally and like cases treated alike.” See also Sedley LJ in Shirazi v SSHD [2003] EWCA Civ 1562 at [29] and [31]; [2004] INLR 92 when he described as “inimical to justice” the inconsistency that was evident when different decisions were taken by different panels of the same appeal tribunal on very similar facts.

23. Although AIT Practice Direction 18 was not of course published during the period with which we are concerned in these appeals, and the arrangements for publishing CG determinations have recently become more sophisticated, it sets out usefully what has been the purpose of issuing these determinations ever since such they were first introduced about four years ago. They represent an institutional response by the IAT to the need identified by Lord Woolf MR and Brooke LJ in their judgments in Manzeke [1997] Imm AR 524. Lord Woolf said: “It will be beneficial to the general administration of asylum appeals for Special Adjudicators to have the benefit of the views of a Tribunal in other cases on the general situation in a particular part of the world, as long as that situation has not changed in the meantime. Consistency in the treatment of asylum-seekers is important in so far as objective considerations, not directly affected by the circumstances of the individual asylum-seeker, are involved.” Brooke LJ added: “It often occurs in asylum appeals that Special Adjudicators are asked to consider reports about conditions in the different countries to which asylum-seekers may return. Sometimes different Special Adjudicators reach different conclusions on the same, or much the same, evidence. This is an unfortunate fact which has led appeals and applications in such cases to be pursued right up to this court in recent months. In those circumstances the Tribunal may perform a valuable function if it decides in any given case to review all the reports available to it relating to a particular country over a particular period of time, so as to give helpful guidance to Special Adjudicators as to how they should approach that evidence in a future case.”

24. In S and Others v SSHD [2002] EWCA Civ 539 ; [2002] INLR 416 Laws LJ described the IAT’s evolving practice of issuing CG decisions. He described them as “factual precedents” (an expression which the IAT has subsequently been at pains to say was a misrepresentation of their purpose: see para 141 of Ouseley J’s decision in NM (Somalia) which is cited in para 26 below), and praised their usefulness. He called them “benign and practical”. In particular he said: “28. Refugee claims vis-à-vis any particular State are inevitably made against a political backdrop which over a period of time, however long or short, is, if not constant, at any rate identifiable. Of course the impact of the prevailing political reality may vary as between one claimant and another, and it is always the appellate authorities’ duty to examine the facts of individual cases. But there is no public interest, nor any legitimate individual interest, in multiple examinations of the state of the backdrop at any particular time. Such revisits give rise to the risk, perhaps the likelihood, of inconsistent results; and the likelihood, perhaps the certainty, of repeated and therefore wasted expenditure of judicial and financial resources upon the same issues and the same evidence.”

25. It is unnecessary for the purposes of these appeals to quote more than two sentences from the very long description of the function of CG determinations which is contained in the determination of the IAT in BD (Croatia CG) [2004] UKIAT 00032. They said at para 58: “ SK was intended to and did give authoritative guidance as to the approach to be adopted towards the position of ethnic Serbs returning to Croatia. Giving such guidance is an important part of the Tribunal’s function, enabling the parties to know where they stand and assisting the achievement of consistent decision-making throughout the asylum and immigration process.”

26. Instead, we can go straight to NM and Others (Lone women – Ashraf) Somalia CG [2005] UKIAT 00076, in a determination Ouseley J delivered at the very end of his term of office as President of the IAT, when he suggested that in S and Others Laws LJ had misunderstood the nature of a CG decision. He said: “139. Decisions of the Tribunal to that end had been made for a number of years. They were to be applied by the Tribunal itself and by Adjudicators unless there was good reason, explicitly stated, for not doing so. Failure to adopt that approach was an error of law in that a material consideration had been ignored or legally inadequate reasons for the decision had been given. The inconsistency itself with authoritative cases would be regarded by higher authority than the Tribunal as an error of law. There was a need to formalise that system so that parties knew where they stood, at least as the starting point for consideration of their circumstances, and for the Tribunal itself to bring forward those decisions which it had made, which it thought were representative and useful still, as a guide to country conditions.

140. These decisions are now denoted as ‘CG’. They are not starred decisions. Those latter are decisions which are binding on points of law. The requirement to apply CG cases is rather different: they should be applied except where they do not apply to the particular facts which an Adjudicator or the Tribunal faces and can properly be held inapplicable for legally adequate reasons; there may be evidence that circumstances have changed in a material way which requires a different decision, again on the basis that proper reasons for that view are given; there may be significant new evidence which shows that the views originally expressed require consideration for revision or refinement, even without any material change in circumstances. It may be that the passage of time itself or substantial new evidence itself warrants a re-examination of the position, even though the outcome may be unchanged. It is a misunderstanding of their nature, therefore, to see these cases as equivalent to starred cases. The system does not have the rigidity of the legally binding precedent but has instead the flexibility to accommodate individual cases, changes, fresh evidence and the other circumstances which we have set out.

141. The comments of the Court of Appeal in S and Others [2002] INLR 416, while recognising the role of the Tribunal in giving guidance, appear to have thought that the Tribunal guidance on country conditions was binding as a legal precedent in the way in which a starred conclusion on a point of law would be. But S and Others was not starred for that guidance, but for the point of law which arose. We have pointed this out in DK and elsewhere and we hope that it is now widely understood. It led to a standard being proposed for such cases, which may be ideal but cannot always be necessary in order for guidance to be given and for the injustice which inconsistency itself can bring to be avoided. Unlike starred decisions, it is always possible for further evidence to show that the original decision was wrong or to expose other issues which require examination. They are not accurately understood or described as ‘factual precedents ’.

142. The system enables the parties and the judiciary to know where to look for what the Tribunal sees as the relevant guidance, the parties to know what they have to deal with, and, if they wish to take issue with it, what it is that has to be the target of their evidence or argument. It enables parties to rely on the material which others have had accepted without reproducing or repeating it every time, or if it has been rejected, to know that there is no point in repeating it. Consistency and the justice which that brings can be provided for, even though differing and perhaps reasonable views can be taken of a wide variety of material. It also has the advantage of enabling the understanding of country conditions to be refined as successive decisions may lead to the identification of consequential issues to be grappled with which had hitherto been unrecognised; the Turkey cases are an example. There is recognised scope for improvement and parties can focus their evidence and arguments upon the aspect with which they take issue. The IAA website has made these cases publicly accessible.”

27. It will have been noticed that Ouseley J said that any failure to apply a CG decision unless there was good reason, explicitly stated, for not doing so would constitute an error of law in that a material consideration had been ignored or legally inadequate reasons for the decision had been given. This suggestion has now been repeated and adopted in para 18.4 of the AIT Practice Direction. We have no hesitation in endorsing that approach. It would represent a failure to take a material matter into account, which is the third of the generic errors of law we have identified in para 9 above. We should add, by way of completeness, that in S and Others Laws LJ was considering the status of CG determinations at a time when the IAT’s jurisdiction was not limited to appeals on a point of law. We will be considering in para 73 below the circumstances in which the IAT might continue to give CG decisions after its jurisdiction was limited to the consideration of errors of law. Part 6 Error of law: unfairness resulting from a mistake of fact

28. The next matter we must address relates to the circumstances in which an appellate body like the IAT, whose primary role during the relevant period was restricted to identifying and correcting errors of law, could entertain an argument to the effect that the outcome in the lower court was unfair as a result of a mistake of fact, and that this constituted an error of law which entitled it to interfere.

29. In E and R v Home Secretary [2004] EWCA Civ 49 ; [2004] QB 1044 this court was concerned to provide a principled explanation of the reasons why a court whose jurisdiction is limited to the correction of errors of law is occasionally able to intervene, when fairness demands it, when a minister or an inferior body or tribunal has taken a decision on the basis of a foundation of fact which was demonstrably wrong. Carnwath LJ gave at least eight examples in his review of the case law. Contrary to the basis on which the original decision was reached: (i) There was in fact contemporary documentary evidence of the injuries sustained by a claimant for compensation from the Criminal Injuries Compensation Board (para 45); (ii) There was in fact, contrary to a minister’s belief, adequate school accommodation in a local education authority’s area for the pupils to be educated (para 54); (iii) The land in question had in fact once been part of the Green Belt (para 58); (iv) The proposed building extension would in fact obstruct a particular aspect (para 58); (v) The restructuring of a building was in fact viable (para 58); (vi) A study by a local council did not in fact relate to the inclusion of a particular site within the Green Belt (para 59); (vii) A critical witness was in fact a member of a totally different political party in Ethiopia to that which he was believed to support (paras 60, 78-79); (viii) The appellant had in fact been tried and convicted in his absence in his home country and sentenced to ten years’ imprisonment, a matter which cast an entirely new light on the risks he faced if he were returned there (paras 60, 87).

30. At para 64 Carnwath LJ said that there was a common feature of all these cases, even where the procedure was adversarial, in that the Secretary of State or the particular statutory authority had a shared interest with both the particular appellant and with any tribunal or other decision-maker that might be involved in the case in ensuring that decisions were taken on the best information and on the correct factual basis. At para 66 he identified asylum law as representing a statutory context in which the parties shared an interest in co-operating to achieve a correct result. He went on to suggest that the ordinary requirements for a finding of unfairness which amounted to an error of law were that: (i) there must have been a mistake as to an existing fact, including a mistake as to the availability of evidence on a particular matter; (ii) it must be possible to categorise the relevant fact or evidence as “established” in the sense that it was uncontentious and objectively verifiable; (iii) the appellant (or his advisers) must not have been responsible for the mistake; (iv) the mistake must have played a material (not necessarily decisive) part in the tribunal’s reasoning. He made it clear that he was not seeking to lay down a precise code.

31. Needless to say, such a mistake could not be identified by the supervising or appellate court unless it was willing to admit new evidence in order to identify it. Paragraphs 68 to 89 of the judgment in E and R contain an analysis of relevant case law on the power to admit new evidence. It concluded with the observation that the case of Khan v SSHD [2003] EWCA Civ 530 that gave rise to the problem summarised in (viii) above was a good example of the need for a residual ground of review for unfairness arising from a simple mistake of fact and that it illustrated the intrinsic difficulty in many asylum cases of obtaining reliable evidence of the facts that gave rise to the fear of persecution and the need for some flexibility in the application of Ladd v Marshall principles.

32. The reference to the Ladd v Marshall principles is a reference to that part of the judgment of Denning LJ in Ladd v Marshall [1954] 1 WLR 1489 when he said at p 1491 that where there had been a trial or hearing on the merits, the decision of the judge could only be overturned by the use of further evidence if it could be shown that: (1) the new evidence could not with reasonable diligence have been obtained for use at the trial (or hearing); (2) the new evidence must be such that, if given, it would probably have had an important influence on the result of the case (though it need not be decisive); (3) the new evidence was apparently credible although it need not be incontrovertible.

33. By way of a final summary of the position, Carnwath LJ said in E and R at para 91 that an appeal on a question of law might now be made on the basis of unfairness resulting from “misunderstanding or ignorance of an established and relevant fact” and that the admission of new evidence on such an appeal was subject to Ladd v Marshall principles, which might be departed from in exceptional circumstances where the interests of justice required. Part 7 Evidence of a change of circumstances since the original decision

34. In the ordinary run of litigation in the courts the legal rights of the parties fall to be decided in accordance with the facts as they appear to the first instance judge. There is little room for the admission of evidence of changed circumstances at the hearing of an appeal. From time to time, however, such evidence was admitted. Case law reveals the following examples under the pre-CPR regime: i) Where there has been a change of circumstances after the granting of an interlocutory injunction such that if the new circumstances had been before the judge they would have justified the variation of the injunction ( Hadmor Productions Ltd v Hamilton [1983] 1 AC 191 , 220D); ii) More generally, where a change of circumstances since the trial has falsified the basis on which discretionary relief was granted ( Attorney General v Birmingham, Tame and Rees District Drainage Board [1912] AC 788 , 802); iii) Where the passage of time since a trial has falsified a conclusion of the trial court based on complaints of delay ( EMA v ACAS (No 2) [1980] 1 WLR 302 , 320F); iv) From time to time, on the basis that the court should not speculate where it knows, damages will be assessed on the facts as they appear at the date of the appeal hearing ( Curwen v Jones [1963] 1 WLR 748 , 753; Lim Poh Choo v Camden and Islington AHA [1980] AC 174 , 194E).

35. In Murphy v Stone-Wallwork (Charlton) Ltd [1969] 1 WLR 1023 HL it was said that the power to admit fresh evidence which showed that damages had been assessed on a false basis should be exercised very sparingly in view of the importance of the principle of finality in litigation (see Lord Pearce at p 1028B and Lord Upjohn at p 1031A).

36. RSC Order 59 Rule 10, which was the rule in force in the Court of Appeal prior to 2 nd May 2000, provided that: “(2) The Court of Appeal shall have power to receive further evidence on questions of fact … but in the case of an appeal from a judgment after trial or hearing of any cause or matter on the merits, no such further evidence ( other than evidence as to matters which have occurred since the date of the trial or hearing ) shall be admitted except on special grounds. (emphasis added) (3) The Court of Appeal shall have power to draw inferences of fact and to give any judgment and make any order which ought to have been given or made, and to make such further or other order as the case may require.”

37. It is noticeable that the court’s willingness to admit evidence of changed circumstances was exercised in particular when it was determining whether a restraint as to future conduct should be retained in force or lifted, or whether an award of damages which contained a significant element of crystal-gazing should be altered when more of the material facts had become known. In all these cases the court possessed an appellate jurisdiction over both fact and law. Part 8 The willingness of the IAT to admit evidence of changed circumstances

38. Ten years ago this court considered issues relating to the scope of the appellate jurisdiction of the IAT on two different occasions. In those days section 19 (1) of the Immigration Act 1971 created the right of appeal to an adjudicator in most immigration matters, whereas it was section 8(1) of the Asylum and Immigration Act 1993 that created for the first time an express right of appeal in an asylum case. It was framed in these terms: “8. (1) A person who is refused leave to enter the United Kingdom under the 1971 Act may appeal against the refusal to a special adjudicator on the ground that his removal in consequence of the refusal would be contrary to the United Kingdom’s obligation under the Convention.” (emphasis added)

39. Section 20 (1) of the Immigration Act 1971 created the right of appeal to the IAT, and the IAT’s powers on such an appeal were expressed in these terms: “20. (1) Subject to any requirement of rules of procedure as to leave to appeal, any party to an appeal to an adjudicator may, if dissatisfied with his determination thereon, appeal to the Appeal Tribunal, and the Tribunal may affirm the determination or make any other determination which could have been made by the adjudicator.” For the effect of this provision, see Borrisov v SSHD [1996] Imm AR 524, per Hirst LJ at p 535.

40. The IAT’s power to admit new evidence on the hearing of an appeal was governed at that time by Rule 27 of the Asylum Appeals (Procedure) Rules 1996. This contained in Rule 27(2) a requirement that notice should be given of a wish to adduce new evidence, and Rule 27(3)(a) provided that: “…the Tribunal may, in its discretion, receive or decline to receive further evidence of which notice has been given in accordance with paragraph (2).”

41. In contrast to what happens in the general run of civil litigation, the practice of asylum law is complicated by the fact that it is all about future risk, and on many occasions there are relevant changes of circumstances between the time of the original refusal of asylum and the time of the IAT’s decision. In Ravichandran [1996] Imm AR 97, for instance, Colombo was generally regarded as a much safer place for returning Tamils in 1995 (when the IAT heard their appeals) than it had been in 1993 (when the Secretary of State made his decision to refuse asylum). The question therefore arose whether the IAT should continue to follow the course that had always pertained to immigration appeals generally (that is, to assess the facts as at the date of the original decision by the Secretary of State), or whether there should be some special rule for asylum appeals.

42. Simon Brown LJ answered this question by deciding that in asylum cases the appellate structure created by the 1993 Act was to be regarded as an extension of the decision-making process rather than as a process for enabling the original administrative decision to be reviewed. He said: “I am, I think, entitled to reach that conclusion as a matter of construction on the basis that the prospective nature of the question posed by section 8 of the 1993 Act overrides the retrospective approach ordinarily required (implicitly) on a section 19 appeal. Section 8, after all, could but does not, identify the ground of appeal as being that the appellant’s removal ‘would have been’ (rather than ‘would be’) contrary to the United Kingdom’s Convention obligations. Moreover, section 8(1) refers to a particular class of appeals and section 19 to appeals in general. It would be a strong thing to say that the general was to override the particular.”

43. After commenting on the way in which the situation in a particular foreign country might change for better or for worse during the period that intervened before an appeal was heard, Simon Brown LJ said: “In either event, if the appellate authorities were bound to ignore such changes, it would render their decisions substantially less valuable. If the situation had improved but, because the appellate authorities had to ignore such improvement, the appeal succeeded, the Secretary of State might nevertheless, in reliance upon article 1C(5) of the 1951 Convention refuse the appellant refugee status. Article 1C(5) provides that the Convention ceases to apply if: ‘(5) he (the refugee) can no longer, because the circumstances in connection with which he has been recognised as a refugee have ceased to exist, continue to refuse to avail himself of the protection of the country of his nationality.’ Equally, had the situation deteriorated but, because this had to be ignored, the appeal failed, the claimant could put a fresh case to the Secretary of State. In either event, the appeal process could well have to start all over again.”

44. Staughton LJ, agreeing, observed that the language of section 8(1) of the 1993 Act manifestly looked to the future at the date of the appeal. It reflected Article 33 of the 1951 Geneva Convention. He said: “In those circumstances the procedural provisions of section 19 of the Immigration Act 1971 , as incorporated by paragraph 4(2) of schedule 2 of the 1993 Act, must yield if and insofar as they would otherwise confine the appellate authorities to the facts that existed at the time of the Secretary of State’s decision.” Part 9 Examples of the admission of new evidence or changed circumstances in asylum appeals

45. Prior to the judgment of this court in E and R , it was the regular practice in asylum appeals to admit evidence of material facts that post-dated the adjudicator’s decision without any reservations, and to apply a slightly relaxed version of the first rule in Ladd v Marshall to the admission of evidence of such facts if they pre-dated the adjudicator’s decision, so long as the evidence was credible and likely to affect the final outcome. See for instance the judgments of Latham J in R v IAT ex p Aziz [1999] Imm AR 476, 482 and of this court in R v IAT ex p Askhosravi [2001] EWCA Civ 977 at [23].

46. Rule 18(11) of the Immigration and Asylum Appeals (Procedure) Rules 2000 represented an attempt by the rule-maker to control the flow of fresh evidence to the IAT: “Subject to section 77, where evidence which was not submitted to the adjudicator is relied upon in an application for leave to appeal [to the IAT], the Tribunal shall not be required to consider that evidence in deciding whether to grant leave to appeal, unless it is satisfied that there were good reasons why it was not submitted to the adjudicator.”

47. So long as the IAT’s jurisdiction contained the power to correct errors of fact, once the admission of the new evidence revealed that the IAT had to reassess the risk on return on a new factual basis (whether it was now revealed that the adjudicator had got some of the material facts wrong or that circumstances had now changed in some material respect), it did not much matter whether the IAT’s intervention was properly categorised as being due to an error of law or an error of fact, or simply to the changed circumstances: in all these cases it was entitled to intervene. The position was different when new evidence came to light between the IAT’s decision and a subsequent hearing in the Court of Appeal because the Court of Appeal’s jurisdiction has always been limited in this jurisdiction to the correction of errors of law. This was why in E and R the Court of Appeal had to identify an error of law in order to be able to put things right.

48. In his judgment in E and R Carnwath LJ cited two recent asylum appeals: R (Haile) v IAT [2001] EWCA Civ 663 ; [2002] INLR 283 and Khan v SSHD [2003] EWCA Civ 530 . In both these cases this court intervened on the basis that new evidence before it had altered the position significantly, but on neither occasion was the court concerned to identify the precise jurisprudential basis on which it was entitled to intervene.

49. We have described in para 30 above the solution which this court propounded in E and R to the problem which was left unsolved in Haile and Khan. In Cabo Verde v SSHD [2004] EWCA Civ 1726 the Secretary of State was the appellant who wished new evidence to be introduced. Although his asylum claim had been dismissed, the asylum-seeker had succeeded before the IAT on Article 3 of the ECHR on the basis that there were thought to be strong grounds for believing that he would be subjected to inhuman treatment if he were returned to Angola, given his history of being tortured in a prison there for four months in early 2002. A year later an extradition request from Portugal contained a contention which was said to show without a shadow of a doubt that he was committing crimes and/or in prison in Portugal at that time and not in Angola at all (although the asylum-seeker denied these assertions).

50. Buxton LJ said that this evidence showed that there was now material which suggested that the factual basis on which the IAT had proceeded was, through no fault of theirs, simply wrong, and that fairness, ie a proper and rational immigration policy, clearly demanded that all the facts of the matter should be before the IAT. Applying E and R , he said that it was plainly established as a fact that relevant evidence was not before the IAT, so that the four criteria for identifying a mistake of fact resulting in unfairness (see para 30 above) were all satisfied. The mistake of fact was a mistake as to the existence of evidence which put in issue the asylum-seeker’s claim to have been in Angola at the relevant time. The mistake was “established” within the second of the conditions in E and R ; and that mistake led to a finding of unfairness. It is important to bear in mind that a mistake as to where the asylum-seeker actually was (based on an evaluation of conflicting evidence) would not have done so. Part 10 The function of the IAT in its restricted appellate role

51. Against this background what we now have to decide is the function that the IAT was intended to perform in these matters in its new role once it had detected an error of law in the first instance decision such as to enable it to consider which of its powers under s 102(1) of the 2002 Act it should exercise: for these powers, see para 5 above. In this context there are three recent decisions of this court which we must consider.

52. In CA v SSHD [2004] EWCA Civ 1165 ; [2004] INLR 453 Laws LJ held at para 14 that under the new appellate regime the IAT’s function was plainly confined to appeals on law only: “14. …Accordingly, and this is I apprehend no more than elementary, an appeal cannot be allowed unless the Tribunal distinctly holds that the adjudicator has perpetrated a mistake of law. No doubt it must be a material error of law. If it could truly be shown that the result before the adjudicator must have been the same even if there had been no legal error, there would be scope for the Tribunal to dismiss the appeal despite the error.

15. That aside, once a material error of law is shown, I for my part would accept that the Tribunal must then decide what if any relief to grant in the light of the facts arising at the time it is considering the case. The appellant’s skeleton argument prepared for the permission application accepts as much. The judicial review court, and this court on appeal in cases where our jurisdiction goes to legal error only, proceeds in precisely the same manner. In this particular jurisdiction, this position is effected by s 102 of the 2002 Act , which I have read.”

53. He went on to reject (at para 18) a submission made on behalf of the Secretary of State to the effect that so long as the IAT had granted permission to appeal on a point of law it was then empowered to review the merits of the matter on an up-to-date basis even though it did not in fact identify an error of law at the substantive hearing. He added (at para 30): “… [T]he jurisdiction under s 101 of the 2002 Act forbids in effect the Tribunal deciding the merits itself unless at least it first concludes that the adjudicator’s decision cannot stand because it is marred by error of law.”

54. In B v SSHD [2005] EWCA Civ 61 this court ruled that the IAT was wrong to believe that once it had granted permission to appeal on a point of law, it had power ipso facto to review the adjudicator’s findings of fact. In that case the grounds for the Secretary of State’s appeal to the IAT had been ineptly drawn, but Lord Phillips MR said that this provided no excuse: “The grounds form the agenda on which the IAT considers the grant of permission and, if granted, conducts the appeal.”

55. More recently, in Miftari v SSHD [2005] EWCA Civ 481 this court allowed an appeal from the IAT because it could detect no error of law in the points raised by the Secretary of State on his appeal from the adjudicator. Buxton LJ set out three points of principle: i) The Vice-President of the IAT who considers an application for permission to appeal must determine whether the IAT has jurisdiction to entertain the appeal on the basis of the grounds of appeal that are before him (subject to any amendment of the grounds he may invite and allow if he discerns a point the parties have not taken). ii) At the substantive hearing of the appeal the IAT can only consider what is legitimately found in the actual or amended grounds, and has no jurisdiction to consider anything that is not there. iii) Unlike a court, the IAT has to consider its jurisdiction expressly because it has to pass on the grounds of appeal. It is very difficult to see how a decision as to jurisdiction can be saved by demonstrating that although the basis on which it was taken was unjustified, the Vice-President could have granted permission on a different basis that was not before the court. Although a Vice-President of the IAT granted permission to appeal in the Miftari case, Buxton LJ’s comments apply equally to any legally qualified chairman who grants permission to appeal.

56. Keene and Maurice Kay LJJ, agreeing, were both at pains to say that the principles set out by this court in R v SSHD ex p Robinson [1998] QB 929 remained good law even though the jurisdiction of the IAT was now restricted. Maurice Kay LJ observed (at para 39) that the decision in Robinson “enabled, indeed required, the immigration appellate authorities to consider an obvious point of Convention jurisprudence which may avail an appellant, even if it is not pleaded or otherwise advanced on his behalf. The rationale was that, if such a point were ignored on technical grounds, there will be a danger that this country will be in breach of its obligations under the Convention (per Lord Woolf MR, giving the judgment of the Court of Appeal, at p 946C).”

57. He added, however, that he was not convinced that the decision in Robinson , which enables this court to hold that there had been an obvious point of Convention jurisprudence favourable to the appellant not taken by the IAT (such as to entitle the Court of Appeal to interfere on grounds of error of law even though the point was not on the agenda propounded by the appellant in his grounds of appeal to the IAT), could ever avail the Secretary of State.

58. In Miftari it was sufficient for this court to decide the appeal on this basis. Because the Secretary of State had not included a meritorious ground of appeal in his notice of appeal to the IAT, the IAT had no jurisdiction to consider that ground at the substantive hearing of the appeal, and should not have decided the appeal in the Secretary of State’s favour on that ground. (No question appears to have arisen as to whether the IAT could have granted the Secretary of State permission to vary his grounds of appeal at the hearing itself pursuant to Rule 20(1) of the Immigration and Asylum Appeals (Procedure) Rules 2003, although no doubt the IAT would ordinarily have been very slow to exercise their power to permit such a very late variation.) Part 11 Buxton LJ’s obiter observations in Miftari

59. Buxton LJ, however, concluded his judgment in Miftari with some observations which, he recognised, were not necessary for the court’s decision. Subject to the reservations they entered in relation to other parts of his judgment, Keene and Maurice Kay LJJ agreed with him. These observations, which we will now quote in full, are not binding upon us in this appeal; although we recognise, of course, the weight that should be given to the considered view of this court on a point on which all three members were agreed.

60. In this passage Buxton LJ said: “29. That suffices to dispose of this matter. However, we heard considerable argument as to the extent of the IAT’s powers had a point of law been properly identified before it. I will go on to discuss those issues, not only because they are of some importance in themselves, but also because they shed some further, albeit indirect, light on the issues of jurisdiction. The IAT’s powers in an appeal on a point of law

30. It will be recalled that in paragraph 16 of its determination the IAT concluded that, since it had identified an error of law on the part of the Adjudicator, it could therefore look again at the case in the light of the factual evidence as it existed at the time of the IAT hearing. I cannot agree. Since the IAT now has jurisdiction to determine only points of law, it cannot put itself in the position of the lower court and decide the whole of the case as it stood there. Unless the decision on the point of law determines the case on the basis of the facts already found below, the IAT has to remit. That the IAT cannot itself enter upon the facts has already been emphasised in two judgments of this court. In CA Mummery LJ at paragraphs 39-40 rejected in emphatic terms a contention in the opposite sense by the Secretary of State. And in paragraph 20 of B the Master of the Rolls, referring to that and other reported cases, said that ‘at the time of the IAT’s decision in this case members of the IAT were under the misapprehension that, once permission to appeal on a point of law had been given, it was open to the IAT to review the Adjudicator’s conclusions of fact.’ The same misapprehension appears to have obtained in the case before us.

31. The point is of some importance for the jurisdiction issue, because once it is assumed that the identification of any point of law opens the door to a general rehearing by the appellate body, the procedure becomes close to that in an ordinary civil appeal, where the appeal court does have all the powers of the lower court (CPR 52.10), and at least in the Court of Appeal has all the authority and jurisdiction of the court from which the appeal was brought ( Supreme Court Act 1981 , s 15(3) ). That is very significantly different from the position of a tribunal, like the IAT, whose jurisdiction is limited to appeals on points of law. I would venture respectfully to agree with the similar observations, albeit in a different statutory context, of Mummery LJ in paragraph 42 of his judgment in Bangs v Connex South Eastern [2005] EWCA Civ 14 .

32. Some difficulty has been perceived in this connection from the continuation, by section 102(2) of the 2002 Act , of the IAT’s power to receive evidence, including evidence of matters arising after the Adjudicator’s decision. But this power is expressly limited to the receipt of evidence ‘in reaching their decision on an appeal under section 101 ’: that is, an appeal against an adjudicator’s decision on a point of law. Therefore, as the President of the IAT stressed in paragraph 20 of the determination of the IAT in MA (fresh evidence) Sri Lanka [2004] UKIAT 00161, such evidence must be relevant to showing that there was an error of law committed by the Adjudicator. And it may be added as a footnote that the more expansive view taken by this court in paragraph 92 of its judgment in E v Home Secretary [2004] QB 1044 was, as the court itself emphasised, directed at the jurisdiction of the IAT before section 101 of the 2002 Act came into force.

33. In referring to that part of the President’s determination I must not be taken as approving the further observations as to the need to look at evidence of events subsequent to the adjudicator’s decision that are to be found in paragraphs 21-24 of that determination. The observations extracted from D (Croatia) [2004] UKIAT 00032 that the President sets out are, with great respect, full of commonsense; but they are very difficult to reconcile with the limited jurisdiction of the IAT that I have referred to in paragraphs 30-31 above. Since there is no error of law in this case the point does not arise for decision. If it did, further consideration of the approach of the IAT in the Sri Lanka case might well have been necessary.”

61. We can say at once that in CA and B , in the passages cited by Buxton LJ in para 30 of his judgment, Mummery LJ and Lord Phillips MR were concerned with a quite different idea, vigorously promoted by counsel for the Secretary of State in CA (see para 16 of Laws LJ’s judgment), to the effect that the IAT might look at the up-to-date merits of a case provided only that it had granted permission to appeal to itself on a point of law. Laws LJ gave very short shrift to this notion (in paras 16 and 18), Mummery LJ expressly agreed with him (at paras 39-40), and the Master of the Rolls repeated their conclusions in B (at para 20), in the passage quoted by Buxton LJ. What we are now concerned with is the quite different question, expressly left open by Laws LJ in CA (at paras 15 and 30: see paras 52 and 53 above), which relates to the courses of action that were open to the IAT under the new regime once it had detected an error of law.

62. In at least three of the appeals with which we are concerned counsel sought to re-argue this question because these observations, if correct in law, have already been perceived to raise very great difficulties in practice. As it happened, it was in T (Eritrea) , the final appeal in our list, that we heard the most sustained challenge by counsel for the Secretary of State to the correctness of this part of Buxton LJ’s judgment in Miftari , but we did not then take steps to reopen the other appeals because we were able to decide them without recourse to our decision on this point.

63. It might be thought that the matter is of academic interest only now that the IAT has ceased to exist and the powers of the AIT on a reconsideration of one of their own decisions are phrased in different terms. But quite apart from the presence in the Civil Appeals Office of a number of outstanding appeals from the IAT which date from the period in which their powers to interfere were limited, this passage in Miftari , if correct, calls into question the legal validity of all the CG determinations of the IAT during that period in cases where the IAT admitted new evidence and gave country guidance on all the up-to-date evidence that had been adduced. In NM and Others ( Lone Women – Ashraf) Somalia CG [2005] UKIAT 00076, for instance, the IAT heard three appeals in February 2005 in which the adjudicators’ decisions had been made in June 2003 and August and September 2004 respectively. It was only on the first of these appeals that the IAT detected an error of law, but the voluminous material about Somalia which they listed in the appendices to their decision included many IAT decisions and reports of experts and international bodies which post-dated 23rd June 2003. If Buxton LJ is correct, then as he also recognised (see paras 32-33 of his judgment in Miftari ) the IAT had no jurisdiction to enter upon this very valuable exercise at all in the way in which they undertook it.

64. We would be very reluctant to differ from the considered view of three members of this court so recently expressed – albeit, not binding on us – unless we felt compelled to reach a different conclusion. This is particularly the case because another division of this court in HC v SSHD [2005] EWCA Civ 893 at [25} has already applied para 30 of the judgment in Miftari in the belief that it was binding on them without any debate as to its correctness (although this aspect of the matter did not affect the result of that appeal). But the powerful arguments addressed to us by the Secretary of State in these appeals deserve and demand a careful reappraisal of the meaning of ss 101 and 102 of the 2002 Act . If we are to obtain a proper understanding of Parliament’s intentions in enacting these alterations to the previous statutory scheme, it is incumbent on us to understand clearly the whole of the new scheme, and not merely parts of it.

65. Under the new regime introduced by the 2002 Act , the adjudicator was the fact-finder and the IAT the reviewing body: the IAT had no power to re-hear the appeal (see, by contrast, the express powers of the Court of Appeal in CPR 52.11 to hold a re-hearing, to receive oral evidence on the re-hearing, and to draw inferences of fact). It was the duty of the adjudicator, as the fact-finder, to determine whether the removal of the appellant would breach the UK’s obligations under the Refugee Convention or would be unlawful under section 6 of the Human Rights Act. It was the duty of the IAT as the review body to determine whether the adjudicator’s decision was vitiated by an error of law.

66. We have set out the relevant elements of s 102 in para 5 above. The language of s 102(2) (“In reaching their decision on an appeal under section 101 the Tribunal may consider evidence about any matter which they think relevant to the adjudicator’s decision, including evidence which concerns a matter arising after the adjudicator’s decision”) appears to have been borrowed from s 77(3) of the 1999 Act : “(3) In considering – (a) any ground mentioned in s 69, or (b) any question relating to the appellant’s rights under Article 3 of the Human Rights Convention, the appellate authority may take into account any evidence which it considers to be relevant to the appeal (including evidence about matters arising after the date on which the decision appealed against was taken).” This statutory provision was considered by the IAT in their starred decision in S & K (Croatia) [2002] UKIAT 05613 at paras 19-22.

67. When this language was transposed to s 102(2) of the 2002 Act , the draftsman decided not to limit the power to consider new evidence to the time when the IAT was considering any ground of appeal or any question under ECHR Article 3 (as in s 77(3) (a)). Instead this power was to be available to the IAT when they were reaching their decision on the appeal , and the words “their decision” would ordinarily be understood to refer to a decision to take one or other of the six courses open to them under s 102(1) once they had identified an error of law in the adjudicator’s decision.

68. Indeed, it is not at all clear to us how it could be thought legitimate or relevant for the IAT to consider evidence which concerned a matter arising after the adjudicator’s decision at the stage when they were considering whether the adjudicator had made an error of law. The examples of cases where the appellate courts have been willing to take into account evidence of matters arising after a trial at first instance (see paras 34-35 above) show that these powers were not invoked to show that the lower court had made an error of law - indeed, the idea that a first instance judge had erred by failing to take into account matters which by definition he could not possibly have known about unless he was a soothsayer is one worthy of Lewis Carroll - but to enable the appellate court to arrive at a just result through using the powers open to it to correct injustice (for which in the context of the IAT see para 5 above).

69. The power to consider matters that arose after the adjudicator’s decision would of course be highly relevant when the IAT was considering what course they should take after they had detected an error of law. They were, indeed, given an express power to make any decision the adjudicator could have made. If they decided that the adjudicator had committed an error of law when dismissing an appeal, they were empowered to consider for themselves afresh whether the appellant’s removal from this country would breach this country’s obligations under the Geneva Convention or would be incompatible with the appellant’s ECHR rights. To perform this role they must inevitably consider the up-to-date position, for the reasons identified by this court in Ravichandran (see paras 41-44 above).

70. If Parliament had not intended them to perform that role, it would hardly have conferred on them in primary legislation the power to admit evidence of changed circumstances. Their power to remit the case to the adjudicator under s 102(2) (c), whether to make fresh findings him/herself ( s 102(4) (a)) or to make findings and then remit the appeal back to them for decision ( s 102(4) (b)), would have been quite sufficient, and the ordinary power to admit further evidence on an appeal could have been conferred by procedural rules. (Compare Rule 27(3)(a) of the 1996 Rules, for which see para 40 above. The equivalent provision in the Immigration and Asylum Appeals (Procedure) Rules 2003, Rule 21 is strikingly different from Rule 27(3)(a) of the 1996 Rules - and indeed Rule 22(2) of the equivalent rules made in 1999 - because it confers no express power on the IAT to admit further evidence: this power is provided for exclusively by primary legislation, and the procedural rule is purely regulatory).

71. For these reasons, we are satisfied that the words “in reaching a decision on an appeal under s 101 ” embraced the entirety of the IAT’s decision on the appeal and are not limited to the first stage of a successful appeal, at which the IAT was considering whether the adjudicator made an error of law. Maurice Kay LJ, who was a member of the court in Miftari , is convinced that this conclusion is right, having heard much fuller argument on this occasion than was available to the court in Miftari.

72. The interpretation of s 102(2) which we prefer would leave unimpaired the IAT’s power to give CG decisions on up-to-date evidence, as they did in NM (Somalia) CG (see para 63 above), provided only that they had first detected an error of law in the adjudicator’s decision.

73. The mischief which s 101(2) was enacted to correct was surely the endless streams of appeals to the IAT by appellants seeking, very often on spurious grounds, to dislodge the findings of fact made by an adjudicator about an appellant’s personal circumstances, after the adjudicator had usually heard and seen the appellant giving evidence, an advantage generally denied to the IAT. We can identify no good reason why Parliament should have wished at the same time to deprive the IAT of their beneficent and valuable role in giving country guidance decisions, even though the occasions on which they would now be able to perform that role would be limited to those occasions when the adjudicator had made an error of law. This conclusion of ours follows the grain of what Laws LJ said in his judgment in CA (see paras 52-53 above) at paras 15 and 30: “[O]nce a material error of law is shown, I for my part would accept that the Tribunal must then decide what if any relief to grant in the light of the facts arising at the time it is considering the case.” (para 15) “[T]he jurisdiction under s 101 of the 2002 Act forbids in effect the Tribunal deciding the merits itself unless at least it first concludes that the adjudicator’s decision cannot stand because it is marred by error of law.” (para 30)

74. Since the hearings ended we have been told by counsel for the Secretary of State that there appears to be no relevant White Paper, Parliamentary Statement or other document which shows that Parliament intended to retain the IAT’s power to reopen factual matters once the error of law gateway had been passed. This, we fear, may be the reason why difficulties have now arisen in practice, because points of this complexity may not have been clearly identified and talked through in the public and Parliamentary discussions which preceded the enactment of the 2002 Act . However that may be, in the light of the very great importance of the IAT’s role in achieving consistency in decision-making through their CG decisions, we are satisfied that it would require very clear Parliamentary language before we could discern an intention that this role should be abandoned, and of this there is none.

75. In these circumstances it is clear to us that what Buxton LJ said in the obiter passage of his judgment in Miftari (see para 60 above) was stated without the benefit of all the argument we have now heard, and that it should not be followed. Part 12 The application of these principles to the five appeals

76. The question whether the IAT could properly consider matters that came to light by way of country background after the date of the adjudicators’ decisions arose in different guises in M and T (Afghanistan), and in T (Eritrea). However, the three appellants in these cases faced significant difficulties in surmounting the difficulty that the IAT now had to identify an error of law before they could exercise any of their powers in s 101(2) , or in taking advantage of the decision in E and R.

77. In M and T (Afghanistan) the appellants were seeking to rely on a decision by the IAT in RS (Hibi-e-Islami – expert evidence) Afghanistan [2004] UKIAT 00278 , which was promulgated after both the adjudicators’ first instance decisions. In that case the IAT recited with approval the expert evidence of Dr Lau, an academic from London University, who was concerned to identify the risks currently faced by those suspected of involvement with Hisb-e-Islami, a radical Islamic group, on their return to Afghanistan. Both M and T had fought as members of that group ten years previously.

78. We would have been willing to hold that insofar as this evidence referred to country background matters prior to the adjudicator’s decision, these appellants should not have been barred by the first rule in Ladd v Marshall from adducing this evidence in support of an appeal to the IAT based on errors of fact. But it would be stretching Carnwath LJ’s decision in E and R impossibly far to say that the adjudicators had made errors of law on the grounds that they had made mistakes of fact that resulted in unfairness through their ignorance of Dr Lau’s views and the evidential material he adduced.

79. Carnwath LJ was concerned in E and R with a fact or evidence which must have been “established” in the sense that it was uncontentious and objectively verifiable. We have set out the practical examples he gave in para 29 of this judgment. The main thrust of Dr Lau’s report was drawn from what a number of people had told him. Other people said different things to Danish fact-finders, and if we were to accept Mr Gill’s submission we would be reintroducing an appeal based on errors of fact through the back door, and this cannot have been Parliament’s intention.

80. In T (Eritrea) the appellant’s problems were even more pronounced. In that case the appellant sought to rely on a UNHCR report that was published ten days before the hearing before the adjudicator, and the IAT’s CG decision in MA (Female draft evader) CG [2004] UKIAT 00098, which was published four months later. On the adjudicator’s findings the case before him related to the risks faced by an Eritrean lady who, he found, was not a conscientious objector, but had made this assertion shortly before the hearing in order to bolster her asylum claim. The UNHCR report to which we have referred highlighted for the first time the potential risks that faced those perceived to be draft evaders, and this was the topic addressed by the IAT in their CG decision.

81. Again, we would not have been disposed to hold that the UNHCR evidence should have been excluded through a rigid application of the first rule in Ladd v Marshall if appeal lay to the IAT on issues of fact. As it was, we were unable to detect any error of law in the way the adjudicator handled the evidence before him, and we dismissed the appeal on this ground.

82. Mr Gill argued in M and T that if we were to adopt this approach there was a danger that like cases would not be determined in the same way: for this important principle see para 22 above. When we observed that it would be open to his clients to make representations to the Secretary of State based on changed circumstances pursuant to rule 353 of the Immigration Rules, he was unenthused by this prospect. This was, we believe, partly because he was used to a post- Ravichandran appellate regime in which all these matters could be resolved by the IAT in what was intended to be a one-stop jurisdiction. It was also partly because of the length of time, in his and his instructing solicitors’ experience, that the Secretary of State habitually takes to decide such applications (coupled with the prospect of judicial review proceedings if he then makes a decision that can be challenged on a point of law). During this waiting period, he told us, his clients will be disadvantaged in relation to the benefits available to them as compared with clients who are still within the appellate process.

83. The views of experienced counsel and solicitors deserve respect, but however well based and sincerely held they may be, they cannot draw us away from the inevitable consequences that had to follow when Parliament decided to restrict the scope of appeal to the IAT to errors of law. Under this new scheme, the decision of the adjudicator was to be final, unless it was found to be vitiated by an error of law, and arguments based on changed circumstances now had to be addressed to the Secretary of State, not the IAT, if no error of law was found.

84. R (Iran) is a case where the IAT detected an error of law, in that the adjudicator had failed to make findings on a relevant issue. They then attempted to make findings themselves on a wholly artificial basis, namely that everything said by the appellant in his interview with an immigration officer should be presumed to be true, and any additional evidence he gave in his witness statement should be presumed to be false. There is an unhappy dispute between the IAT and counsel who then appeared for the appellant as to whether this was a course to which he had assented. However that may be, the IAT did not in practice give credence to everything said in the interview, and we were satisfied that this was a case that should be remitted to the AIT for reconsideration on this limited issue, on which the appellant should give factual evidence untrammelled by any artificial presumptions.

85. This case affords a good example of the way in which under the appellate regime in the 2003-5 period remittal to an adjudicator would normally be the appropriate course when factual evidence relevant to the appellant had to be investigated further once an error of law had been detected in the original decision. Factual evidence relating to country background matters was quite another matter.

86. The appeal in A (Afghanistan) cast light on the circumstances in which the IAT may interfere with a decision by an adjudicator under Article 8 of the ECHR to the effect that it would be disproportionate to interfere with an appellant’s family life in this country by sending him back to his country of origin.

87. A is a 30-year old Afghan man who joined the rest of his family in this country three months before the adjudicator’s decision. His father was dead. His mother and two of his siblings had succeeded in their claims for asylum, while two other brothers had been given exceptional leave to remain for a further two years.

88. The first question we had to decide in A was whether the adjudicator did correctly direct himself as to the nature of his task. If he did, it would be difficult for the IAT to interfere with his decision as a matter of law, as they did, merely because he gave more weight to certain matters and less weight to others. In the event we concluded that he did not, because he decided the appeal before the decision of the House of Lords in Razgar (see para 18 above), and concentrated on the difficulties facing the appellant (and his family) and not at all on the consideration to be accorded to immigration control. He also committed an error of law by paying no regard to the existing jurisprudence on family life in relation to adult siblings.

89. The IAT were therefore entitled to set his decision aside and determine the question afresh as if they were in the adjudicator’s shoes. We were able to detect no error of law in their approach, and we therefore dismissed A ’s appeal. Part 13 A summary of the main points in this judgment

90. It may now be convenient to draw together the main threads of this long judgment in this way. During the period before its demise when the IAT’s powers were restricted to appeals on points of law:

1. Before the IAT could set aside a decision of an adjudicator on the grounds of error of law, it had to be satisfied that the correction of the error would have made a material difference to the outcome, or to the fairness of the proceedings. This principle applied equally to decisions of adjudicators on proportionality in connection with human rights issues;

2. A finding might only be set aside for error of law on the grounds of perversity if it was irrational or unreasonable in the Wednesbury sense, or one that was wholly unsupported by the evidence.

3. A decision should not be set aside for inadequacy of reasons unless the adjudicator failed to identify and record the matters that were critical to his decision on material issues, in such a way that the IAT was unable to understand why he reached that decision.

4. A failure without good reason to apply a relevant country guidance decision might constitute an error of law.

5. At the hearing of an appeal the IAT had to identify an error of law in relation to one or more of the issues raised on the notice of appeal before it could lawfully exercise any of its powers set out in s 102(1) of the 2002 Act (other than affirming the adjudicator’s decision).

6. Once it had identified an error of law, such that the adjudicator’s decision could not stand, the IAT might, if it saw fit, exercise its power to admit up-to-date evidence or it might remit the appeal to the adjudicator with such directions as it thought fit.

7. If the IAT failed to consider an obvious point of Convention jurisprudence which would have availed an applicant, the Court of Appeal might intervene to set aside the IAT’s decision on the grounds of error of law even though the point was not raised in the grounds of appeal to the IAT.

91. Although we have phrased this guidance in relation to the appellate regime that came to an end in April 2005, with the demise of the IAT and the birth of the AIT many of the principles we have set out in this judgment will be equally applicable when issues arise as to the identity of an error of law under the new statutory regime. Part 14 Appeals to the Court of Appeal

92. Finally, so far as access to this court is concerned, the Court of Appeal has traditionally shown itself willing to take an appropriately modest view of its supervisory role when invited to grant permission to appeal from decisions of specialist tribunals: (see Cooke v Secretary of State for Social Services [2001] EWCA Civ 734 ; [2002] 3 All ER 279 ; Napp Pharmaceutical Holdings Ltd v Director-General of Fair Trading [2002] EWCA Civ 796 , [2002] 4 All ER 376 : and compare now Hinchy v Secretary of State for Work and Pensions [2005] UKHL 16 at [30]. In Koller v SSHD [2001] EWCA Civ 1267 , however, this court distinguished its approach in relation to appeals from decisions of the IAT on the grounds that such appeals often raise complex issues of fact and law; the law was still developing; and it had been the experience of this court that determinations of some panels of the IAT were of uncertain quality.

93. Much has changed since then. In particular, the influence of High Court judges as President of the Tribunal has led to a noticeable improvement in the quality of decision-making; reconsideration of a decision of the AIT is restricted to issues of law; and the law is, temporarily at any rate, in a far more settled state than it was when Koller was decided. In future this court will be slower to grant permission to appeal in these cases. For a fuller exposition of the relevant principles, see the judgment of Carnwath LJ in Akaeke v SSHD [2005] EWCA Civ 947 at paras 26-30.