AA (Somalia) v SSHD

These two cases have been listed together because of an issue common to both. The issue arises in cases where there is overlapping evidence of fact in two separate cases. The issue is: “In X’s asylum/human rights appeal what weight, if any, should be given to a finding of fact made in Y’s favour which assists X and which was made by a tribunal when allowing an asylum/human rights appeal in respect of Y?”

In AA’s case it was and is submitted that the previous finding of fact in his sister’s appeal that she was of the Ashraf clan was binding on the AIT when considering AA’s separate appeal against the decision of the SSHD to refuse AA’s asylum and human rights claims. The AIT ([2006] UKAIT 00052) rejected that submission and the appellant appeals to this Court.

In AH’s case it was and is submitted that that the previous findings of fact in M’s appeal that AH was in a homosexual relationship with M, that M was the cousin of AH’s wife, that AH’s wife knew and strongly disapproved and that this made it more likely that AH would come to the attention of the Iranian authorities, were binding on the AIT when considering AH’s separate appeal against the decision of the SSHD to refuse AH’s asylum and human rights claims. The AIT (Appeal number HX/14896/2004) rejected that submission and the appellant appeals to this Court.

Mr Scannell submits on behalf of the appellants that the previous findings of fact are binding absent a very good reason why they should not be. Whilst accepting that fresh evidence might provide a very good reason, he was reticent about other possible good reasons. However, he submitted that if the material before the second tribunal was essentially no different to the material before the first tribunal, the previous findings of fact would be binding. If Mr Scannell is right then at the hearing before the second tribunal the appellant would be advised, in the absence of any new evidence from the SSHD, not to give or to call evidence and merely to rely on the earlier determination.

It is important to note that in both the two cases under appeal, it is now accepted that the decisions of the second tribunals were decisions which on the facts the tribunals were entitled to reach. In AA’s case the second tribunal was an adjudicator. In AH’s case the second tribunal was the AIT reconsidering AH’s appeal against the decision of the SSHD.

Mr Kovats submits, on behalf of the SSHD, that there is no such principle as that advanced by Mr Scannell. If two tribunals on the same material reach different (but rational) decisions, so be it. He is careful to stress that he does not submit that decisions adverse to the SSHD should be treated less favourably than decisions in her favour. He argues for what he describes as symmetry.

I start with Devaseelan [2004] UKIAT 00282. Devaseelan concerned second appeals made on human rights grounds by an asylum seeker whose asylum appeal had been dismissed earlier. Such appeals are no longer possible, although there remain the somewhat analogous cases involving fresh asylum and human rights claims.

The IAT gave guidance (paras. 37-42) as to the weight to be attached to the findings of the adjudicator who had rejected the asylum appeal. The IAT said, amongst other things, that the first adjudicator’s determination “should always be the starting point”. The AIT wrote:

37.

...The first Adjudicator’s determination stands (unchallenged, or not successfully challenged) as an assessment of the claim the Appellant was then making, at the time of that determination. It is not binding on the second Adjudicator; but, on the other hand, the second Adjudicator is not hearing an appeal against it. As an assessment of the matters that were before the first Adjudicator it should simply be regarded as unquestioned. It may be built upon, and, as a result, the outcome of the hearing before the second Adjudicator may be quite different from what might have been expected from a reading of the first determination only. But it is not the second Adjudicator’s role to consider arguments intended to undermine the first Adjudicator’s determination.

39.

The second Adjudicator must, however be careful to recognise that the issue before him is not the issue that was before the first Adjudicator. In particular, time has passed; and the situation at the time of the second Adjudicator’s determination may be shown to be different from that which obtained previously. Appellants may want to ask the second Adjudicator to consider arguments on issues that were not – or could not be – raised before the first Adjudicator; or evidence that was not – or could not have been – presented to the first Adjudicator.

In our view the second Adjudicator should treat such matters in the following way.

(1)

The first Adjudicator’s determination should always be the starting-point. It is the authoritative assessment of the Appellant’s status at the time it was made. In principle issues such as whether the Appellant was properly represented, or whether he gave evidence, are irrelevant to this.

(2)

Facts happening since the first Adjudicator’s determination can always be taken into account by the second Adjudicator. If those facts lead the second Adjudicator to the conclusion that, at the date of his determination and on the material before him, the appellant makes his case, so be it. The previous decision, on the material before the first Adjudicator and at that date, is not inconsistent.

(3)

Facts happening before the first Adjudicator’s determination but having no relevance to the issues before him can always be taken into account by the second Adjudicator. The first Adjudicator will not have been concerned with such facts, and his determination is not an assessment of them.

40.

We now pass to matters that could have been before the first Adjudicator but were not.

(4)

Facts personal to the Appellant that were not brought to the attention of the first Adjudicator, although they were relevant to the issues before him, should be treated by the second Adjudicator with the greatest circumspection. An Appellant who seeks, in a later appeal, to add to the available facts in an effort to obtain a more favourable outcome is properly regarded with suspicion from the point of view of credibility. (Although considerations of credibility will not be relevant in cases where the existence of the additional fact is beyond dispute.) It must also be borne in mind that the first Adjudicator’s determination was made at a time closer to the events alleged and in terms of both fact-finding and general credibility assessment would tend to have the advantage. For this reason, the adduction of such facts should not usually lead to any reconsideration of the conclusions reached by the first Adjudicator.

(5)

Evidence of other facts – for example country evidence – may not suffer from the same concerns as to credibility, but should be treated with caution. The reason is different from that in (4). Evidence dating from before the determination of the first Adjudicator might well have been relevant if it had been tendered to him: but it was not, and he made his determination without it. The situation in the Appellant’s own country at the time of that determination is very unlikely to be relevant in deciding whether the Appellant’s removal at the time of the second Adjudicator’s determination would breach his human rights. Those representing the Appellant would be better advised to assemble up-to-date evidence than to rely on material that is (ex hypothesi) now rather dated.

41.

The final major category of case is where the Appellant claims that his removal would breach Article 3 for the same reason that he claimed to be a refugee.

(6)

If before the second Adjudicator the Appellant relies on facts that are not materially different from those put to the first Adjudicator, and proposes to support the claim by what is in essence the same evidence as that available to the Appellant at that time, the second Adjudicator should regard the issues as settled by the first Adjudicator’s determination and make his findings in line with that determination rather than allowing the matter to be re-litigated. We draw attention to the phrase ‘the same evidence as that available to the Appellant’ at the time of the first determination. We have chosen this phrase not only in order to accommodate guidelines (4) and (5) above, but also because, in respect of evidence that was available to the Appellant, he must be taken to have made his choices about how it should be presented. An Appellant cannot be expected to present evidence of which he has no knowledge: but if (for example) he chooses not to give oral evidence in his first appeal, that does not mean that the issues or the available evidence in the second appeal are rendered any different by his proposal to give oral evidence (of the same facts) on this occasion.

42.

We offer two further comments, which are not less important than what precedes then.

(7)

The force of the reasoning underlying guidelines (4) and (6) is greatly reduced if there is some very good reason why the Appellant’s failure to adduce relevant evidence before the first Adjudicator should not be, as it were, held against him. We think such reasons will be rare. There is an increasing tendency to suggest that unfavourable decisions by Adjudicators are brought about by error or incompetence on the part of representatives. New representatives blame old representatives; sometimes representatives blame themselves for prolonging the litigation by their inadequacy (without, of course, offering the public any compensation for the wrong from which they have profited by fees). Immigration practitioners come within the supervision of the Immigration Services Commissioner under part V of the 1999 Act. He has power to register, investigate and cancel the registration of any practitioner, and solicitors and counsel are, in addition, subject to their own professional bodies. An Adjudicator should be very slow to conclude that an appeal before another Adjudicator has been materially affected by a representative’s error or incompetence; and such a finding should always be reported (through arrangements made by the Chief Adjudicator) to the Immigration Services Commissioner.

Having said that, we do accept that there will be occasional cases where the circumstances of the first appeal were such that it would be right for the second Adjudicator to look at the matter as if the first determination had never been made. (We think it unlikely that the second Adjudicator would, in such a case, be able to build very meaningfully on the first Adjudicator’s determination; but we emphasise that, even in such a case, the first determination stands as the determination of the first appeal.)

(8)

We do not suggest that, in the foregoing, we have covered every possibility. By covering the major categories into which second appeals fall, we intend to indicate the principles for dealing with such appeals. It will be for the second Adjudicator to decide which of them is or are appropriate in any given case.

It is of importance to note that at the outset of this passage the IAT wrote: “The first Adjudicator’s determination ... is not binding on the second Adjudicator”.

DB [2003] UKIAT 00053 was a case which involved an appeal from a second refusal of entry clearance, the first refusal having been upheld by an Adjudicator. The IAT said:

15.

There is nothing in Devaseelan which limits its principles to asylum and human rights appeals. There is no reason why they should be so limited. We are satisfied that the principles set out in Devaseelan apply to all categories of appeals coming before Adjudicators and the Tribunal.

16.

It follows that the Adjudicator was wrong to disregard the findings in the previous determination, and wrong to take care, as he described it, not to be affected in any way by the previous determination. It would be very unsatisfactory and not in the interests of justice if an Adjudicator paid no regard to a previous determination in such circumstances, particularly when the application the subject of the appeal before him was made only a matter of weeks after a refusal of leave to appeal by the Tribunal.

I turn to TK (Consideration of Prior Determinations) Georgia [2004] UKIAT00149, Mr C M G Ockelton, Deputy President, presiding. The decision is dated 3 June 2004. TK concerned an appeal by a family member, the success of which was entirely dependent upon her showing that the risks which she fears were based on the risks faced by another member of the family, in this case her father. In furtherance of the persecution of the father, it was being claimed, that his family members had also suffered persecution. At an earlier determination, the Tribunal had dismissed the father’s appeal. The IAT said:

19.

In these circumstances the Tribunal considers that not only was the Adjudicator entitled to read [the first] Determination, notwithstanding the arguments to the contrary which have been considered and dealt with above, but was also entitled to conclude that it would be wrong to revisit [the first] decision in relation to the Appellant's husband’s evidence. Were the Adjudicator not entitled to take this course, the following extraordinary circumstance could arise. The head of a family, call him X, claims asylum on the basis of his own account and loses on the grounds that his account is disbelieved. There follows thereafter a succession of separate members of X’s family who each makes his/her own asylum application and each expressly accepts that the risks which they fear are based on the risks to X as head of family. If Miss Record’s submissions were correct, then there could be a succession of hearings where a succession of Adjudicators, each deprived of all previous Adjudicator’s Determinations, could be asked to reappraise over and over again the same basic account from X, being an account on which all the successive family members were relying as showing that they were at risk because X was at risk. Unless some very good reason was advanced to the contrary, for example, compelling new evidence to show that X’s evidence (which originally had been disbelieved) was mistakenly appraised by the original Adjudicator, a future Adjudicator is, in the Tribunal's view, not merely entitled to read the Determination in X’s case but also to treat it as determinative as to X’s account.

I turn to LD (Algeria) [2004] EWCA Civ 804, also known as Djebbar, a decision dated 30 June 2004. This was another second appeal case. An unsuccessful attack was made on behalf of the appellant, LD, upon Devaseelan. The Court approved the guidelines saying:

30.

Perhaps the most important feature of the guidance is that the fundamental obligation of every special adjudicator independently to decide each new application on its own individual merits was preserved. The guidance was expressly subject to this overriding principle.

The Court then set out part of paragraph 37 of Devaseelan, starting with the words:

The first adjudicator’s determination … is not binding on the second adjudicator; but, on the other hand, the second adjudicator is not hearing an appeal against it … the outcome of the hearing before the second adjudicator may be quite different from what might have been expected from a reading of the first determination only.

The Court of Appeal in LD went on to emphasise that the Devaseelan guidelines were not to be read restrictively:

40.

... Having analysed the guidelines as a whole, in the light of the specific criticisms, it seems to us that it would be positively disadvantageous for this Court now to attempt to rewrite any part of the guidance by expressing the same ideas in different language. We have no reason to believe that adjudicators approach this guidance as if they were construing statute or regulation, or apply it as if it were, without regard to the true merit (or otherwise) of the fresh application. The great value of the guidance is that it invests the decision making process in each individual fresh application with the necessary degree of sensible flexibility and desirable consistency of approach, without imposing any unacceptable restrictions on the second adjudicator’s ability to make the findings which he conscientiously believes to be right. It therefore admirably fulfils its intended purpose. (Underlining added)

The relevance of Devaseelan to cases other than second appeals was considered in SK (Guidance on the application of Devaseelan) Serbia and Montenegro [2004] UKIAT 00149, Mrs Gleeson, Vice President, presiding. The decision is dated 5 October 2004. The decision is not easy to follow because it is so short and the facts of the appeal not set out. In para. 1 the Tribunal wrote:

Devaseelan does not purport to deal with decisions relating to the family member although increasingly it is misunderstood by Adjudicators as doing so. It deals only with the situation where a human rights claim is made by someone whose asylum appeal has already failed and a credibility and factual matrix been found by the first Adjudicator.

Both Mr Scannell and Mr Kovats agreed that the IAT was deciding that Devaseelan did not apply to cases not involving second appeals and that, by implication, the earlier decision should not be subject to the more restrictive approach laid down in that case.

In the instant case of AA, the AIT, Mr C M G Ockelton, Deputy President, presiding, examined the authorities including cases unconnected with asylum. Under the heading “Discussion” the AIT wrote:

60.

Generally speaking, parties to an action must regard the matter as finally settled between them by a subsisting order of a competent court. This is the rule of res judicata. ... It is sometimes said that there is no rule of res judicata or issue estoppel in immigration appeals. Technically speaking, that must be right. The fact that there has been a previous unsuccessful application and appeal does not prevent an individual making a new application for relief, whether by way of entry clearance, leave to enter, leave to remain, asylum, or non-removal. In in-country asylum and human rights cases, the possibility of a formal issue estoppel is effectively ruled out by the principle that matters have to be decided as they are at the date of the decision, whether administrative or judicial. That date will, ex hypothesei, be different from any consideration of a previous claim.

61.

On the other hand, a decision-maker considering a second application, or second claim, or second proceedings, to which a person involved in earlier proceedings was a party, should no doubt have regard to the previous judgment. There are two reasons. First, it may well summarise what was said on the appellant’s behalf on the earlier occasion. In a jurisdiction such as ours which has no hearsay rule, that material has evidential value of its own. Secondly, it is (so far) the authoritative decision on the matters that were raised at that time. If the parties did not take any opportunity available to them to challenge those findings then, the Tribunal should require a good reason for departing from them now. Considerations of that sort are behind the guidance in Devaseelan, which we set out earlier in this determination. The previous judgment is not binding, but it is not to be ignored. If there is no good reason for departing from it, it must, as between the parties to that litigation, be treated as settling the issues with which it was concerned and the facts on which the determination was based.

62.

When the parties to a second appeal are different from the parties to an earlier appeal, the latter consideration does not apply at all. An earlier determination, appearing to decide a fact which may be in issue in a later case, may, however, demand the attention of the court in the later case for a number of reasons.

63.

The first, which we mention only for the purpose of excluding it from further discussion, is where the Tribunal has issued country guidance. ...

...

66.

Returning to the reasons which might be given for citing a decision or determination made in an application or appeal by a related claimant, there is surely no reason, in principle or authority, to give the previous determination evidential value to the case now under consideration. The previous determination is not the result of the application of the rigorous requirements of the criminal law; and the fact that a previous court or other decision-maker has reached a view on facts which are in issue in the present appeal is not of itself any evidence as to those facts. On the other hand, in the general interests of good administration, it is probably true to say that decisions should not be unnecessarily divergent. It is that principle of good administration which, so far as we can see, provides the sole basis in logic or on authority for saying that the result of the previous litigation may be relevant in the present appeal.

67.

What then is its relevance? It can surely only be this: that the previous decision can be taken as establishing the issue in question unless there is any reason not to take it as establishing that issue in question. It has no evidential effect. It does not even give rise to a presumption. It is simply a starting point. That is, indeed, what was decided in TK, as we have seen. ... [T]he old decision remains, but only as long as there is no reason for displacing it.

68.

We can see no possible basis for the assertion that a determination in one appellant’s case has any binding effect on any other individual. All the authorities, as well as principle, are against that. Still less can we find any reason for saying that favourable decisions are binding but unfavourable decisions have no lasting effect at all. That latter submission, if we may say so, is only too obviously a demonstration of the way in which the appellate process may be the subject of cynical manipulation.

69.

In fact, comparison with the considerations relating to the admission of criminal convictions in the proof of civil claims suggests that the reverse may be nearer the correct position. In asylum and human rights appeals, the standard of proof is very low. It suffices, it is usually said, to establish that there is a real risk of the apprehended harm. On the other hand, in refusing a claim, or dismissing an appeal, the Tribunal decides that there is no real risk that the claimant is or will be in the danger he claims. Thus the dismissal of an asylum and human rights appeal indicates a level of certainty about the effect of the evidence that is not apparent in allowing an appeal. We would not press this issue further, but it seems to us to be worthy of mention.

70.

The previous determination stands as a determination of the issues between the claimant in that case and the Secretary of State. If the claimant in that case brings another claim, there will need to be a good explanation for why it is said (if it is said) that the second decision, or the basis for it, should be different from the conclusion reached in the earlier case. But, by contrast, the earlier decision can be nothing more than the background or the starting point for the determination of a claim made by a different person altogether.

71.

It would be dangerous for us to attempt to set out what the factors would be that would entitle or require the Tribunal to depart from the earlier finding in a later case. Obviously one factor, relevant in some cases although not in others, will be the passage of time. Another factor, which is likely to be of importance in a very great many cases, is simply whether the evidence is different. Evidence available to the earlier decision-maker may be reinforced, supplemented, contradicted, effectively withdrawn or simply be not available to the second decision-maker. It is the very fact that he is required to make his independent decision on the evidence before him that makes it so clear that it would not be right for him simply to rely slavishly on a decision made by someone else. We can see no good reason in general why, if a beneficial earlier decision is said to be relevant, the beneficiary of that decision should not give evidence before the second Tribunal and be subject to cross-examination if the Secretary of State chooses to challenge the evidence before the second Tribunal. It is for the claimant in the second case to prove his case on the evidence, and if it is clear that available evidence is not being adduced, the Tribunal is entitled to draw appropriate conclusions.

72.

There is one final point we would make. As we have said above, the contents of a previous determination or decision may be of value as evidence of what was said before that decision was reached. The decision itself, however, is only a starting point for the second Tribunal. It is the point from which a departure may be made. Crucially, the conclusion of the previous decision-maker is not in itself any evidence of the facts upon which the conclusions appears to have been based.

73.

The result may be, as occurred in the cases before us and in TK and AC [a case very much like the present case of AA], that the second decision is on its facts inconsistent with the first. That is, in our view, no reason for thinking that the first decision rather than the second is correct. It is for the Respondent to decide, in such circumstances, whether he can or should seek to change the first decision or its effect, in the light of the further or different information or evidence that supported the second decision. We are not called upon to make any further comment in this determination about how that is to be done. ...

The AIT is making it clear in these passages that, in cases like the present involving different parties, the earlier decision is not binding and that “in the general interests of good administration” the earlier decision should be taken as no more than a starting point. Mr Kovats even questions the use of the words “starting point”. However he recognises that treating the previous decision as a “starting point” would not impose a significant restriction on the second tribunal.

The passages highlight the differences between Mr Scannell and Mr Kovats. For Mr Scannell there can only be a very limited category of “good reasons” for departing from a previous finding of fact in a case even though the parties are different. One reason would be, he accepts, fresh evidence. The AIT in AA’s case is making it clear that there are no such limits on the kind of good reasons for departing from the earlier decision. If, on the evidence, the second tribunal (rationally) makes different findings of fact to those reached by the earlier tribunal and has taken that decision into account, then the second tribunal’s decision cannot be impeached on this ground alone.

Absent any further authority, I find the reasoning of the AIT in AA’s case very persuasive. Restrictions of the kind suggested by Mr Scannell are not necessary or desirable. If Mr Scannell were right, the second tribunal would not be doing that which it has to do: examine the evidence before it and reach the appropriate conclusions of fact. Mr Scannell’s approach would lead to numerous appeals in which the reviewing or appellate court would be asking itself: “Was there a sufficient good reason here to justify departure?” Another layer of complexity would be added to an already far too complicated area of law. There would be numerous appeals in which the issue to be litigated would be: Was the second tribunal “bound by” the decision of the first tribunal? There is a tension between the need for consistency in public law and the need to ensure that the right to stay in the country and not be deported is only granted to applicants who show (to a low standard of proof) that they are entitled to do so. In my view in this field the need for consistency should take second place. I bear in mind the variable quality of advocates before immigration tribunals (indeed in a significant number of cases the Secretary of State is not represented) and the inevitable effect this has on decision making. I therefore agree with the position adopted by the respondent: there are no principles here. The second tribunal should have regard to the earlier decision but only as a starting point.

I now turn to Ocampo v. SSHD [2006] EWCA 1276. In Ocampo the daughter of the appellant had earlier appealed the decision of the SSHD refusing her asylum. Both the daughter and her father gave evidence in the course of that appeal. The adjudicator believed their evidence and found the account given by them of the threats by FARC to his life giving rise to similar threats to her after his flight, to be credible and allowed the appeal. The SSHD granted her refugee status, with its attendant international rights. In 2003 she was granted British citizenship.

The father then appealed the refusal of the SSHD to grant him asylum. At the appeal both he and his daughter gave evidence. In the words of Auld LJ:

... the accounts of the appellant and his daughter in their respective applications for asylum and in the evidence each gave in support of the other’s appeal were mutually supportive, both depending essentially upon the truth of the appellant’s account of having had to flee Colombia for fear of what FARC would do to him if he remained there.

At the hearing of the appellant’s appeal the SSHD specifically sought to undermine his credibility by adducing a copy of a record of an interview of him on 10^th May 1997 in the course of his own application for asylum, a record demonstrating material inconsistencies with the account he had later given in support of his daughter’s successful appeal in 2000. The Secretary of State maintained that it was “new and compelling” evidence that could not have been placed before the Adjudicator in her appeal.

The appellant, on the other hand, sought to rely upon the grant of refugee status to his daughter and the findings of the Adjudicator as to her credibility in her successful appeal. He maintained that those findings bound the AIT in his appeal unless the Secretary of State could show that they were fraudulently obtained. However, he did not suggest that the Adjudicator’s findings as to his credibility in his daughter’s appeal were similarly binding on the AIT.

The appellant’s appeal was dismissed.

Auld LJ, having reviewed the authorities, said:

24.

In my view, it is at the very least doubtful whether the principles of res judicata or issue estoppel have any application, certainly in their full rigour, to appeals before immigration tribunals, any more than they do to successive claims for judicial review; ... And the recent approach of the Court of Appeal in E & R v Secretary of State for the Home Department [2004] EWCA Civ 49; [2004] INLR 268, was to treat Ladd v Marshall principles as starting points, with a discretion to depart from them in exceptional circumstances; see per Carnwath LJ, giving the judgment of the Court at para 82. The Devaseelan guidelines, in their application to fact finding by successive immigration tribunals, represent much the same approach, as Judge LJ, as he then was, giving the judgment of the Court in Djebbar [LD(Algeria], indicated in approving the Devaseelan guidelines:

“28.

… The second application is a fresh application, requiring proper consideration on such merits as it may enjoy, approaching the issues contemporaneously. Although it is indeed a ‘fresh’ application, a second or subsequent application is not and is not deemed to be a first application, and it is not properly to be treated as if it were. Re-litigation of issues which have already been resolved is contrary to the public interest, and nothing in the process suggests that the first application should or must automatically be treated as irrelevant to second applications arising in cases like those with which we are presently concerned. If the first application may be relevant, then the extent of its possible relevance and the proper approach to it should be addressed as a matter of principle. That is what the [Devaseelan] guidance purported to provide.”.

25.

In my view, the Devaseelan guidelines are as relevant to cases like the present where the parties involved are not the same but there is a material overlap of evidence, as the Immigration Appeal Tribunal observed in TK Georgia, at paragraph 21 of their determination. Clearly, the guidance may need adaptation according to the nature of the new evidence, the circumstances in which it was given or not given in the earlier proceeding and its materiality to securing a just outcome in the second appeal along with consistency in the maintenance of firm immigration control. It should also be borne in mind, as Hooper LJ pointed out in the course of counsel’s submissions, that admission of new evidence may, as a matter of fairness, operate for, as well as against, a claimant for asylum. In immigration matters, as in other areas of public law affecting individuals, public policy interests of firmness, consistency and due process may have to be tempered with considerations of fairness in particular circumstances.

26.

Accordingly, in my view, the AIT rightly rejected any application in the circumstances of this case of the strict principles of res judicata or issue estoppel and, with them, the contention that they could only take account of, and rely upon, the new evidence if the dishonesty of the appellant had not been previously establishable. Equally untenable, in my view, was Mr O’Callaghan’s recourse, in his submissions to this Court to the suggestions: 1) that the AIT, in acting as it did, in some way interfered with the daughter’s status as a refugee acquired as a result of the Adjudicator’s finding in her favour; and/or 2) that that finding was a judgment in rem so as to render by implication his acceptance of her evidence, and effectively that of the appellant in the same proceeding, as conclusive of its effect against all persons including the Secretary of State in the appellant’s appeal. The daughter’s status as a refugee is not affected by any finding in reliance on new and cogent evidence that the appellant lyingly supported her in her successful appeal against refusal of asylum. To the extent that it may reflect on the credibility of her evidence in her appeal and in the appellant’s appeal, it has no effect at all on her status as a refugee. And, as I have said and Mr O’Callaghan has acknowledged, any finding of the Adjudicator in the daughter’s appeal in reliance upon the appellant’s evidence, cannot, as a matter of res judicata or issue estoppel or, under the principles of Ladd v Marshall, be binding against him.

Auld LJ went on to hold that the AIT did not have to resort to the principles in Ladd v Marshall before “receiving” “new” evidence.

Mr Scannell submits that the Court in Ocampo, by approving Devaseelan,was adopting the restrictive approach to previous decisions for which Mr Scannell argues. He submits that Ocampo decided that if none of the circumstances identified in the Guidelines applied, then the second tribunal was bound by the findings of fact made by the first tribunal. If that is what Ocampo decided it would be quite contrary to the decision in LD, paras. 14-15 above, specifically approved by Auld LJ in Ocampo. In Ocampo there was fresh evidence in the form of the appellant’s interview, evidence which had not been relied upon by the SSHD at the hearing of the daughter’s appeal. If Auld LJ was going as far as Mr Scannell submits (and I do not believe he was), it would have been obiter.

In my judgment it is time for the Court of Appeal to adopt the submissions made by Mr Kovats. In cases where the parties are different, the second tribunal should have regard to the factual conclusions of the first tribunal but must evaluate the evidence and submissions as it would in any other case. If, having considered the factual conclusions of the first tribunal, the second tribunal rationally reaches different factual conclusions, then it is those conclusions which it must apply and not those of the first tribunal. In my view Ocampo and LD do not stand in the way of this simple approach. Both cases make it clear the first decision is not binding and that it is the fundamental obligation of the judge independently to decide the second case on its own individual merits. All that I am doing is simplifying and clarifying the law. Simplification and clarification have the advantages of making it easier for immigration judges for whom the law is already far more complicated than it should be and of making it less likely that there will be appeals on whether the second tribunal was, or was not, bound by the decision of the first. It also has the advantage that the same rule applies whether the previous decision was in favour or against the Secretary of State.

I take the facts from the decision of the AIT (presided over by Mr CMG Ockleton) on appeal from the decision of Mr Hulme, Adjudicator.

21.

AA came to the United Kingdom on 28 August 2004 by train. The passport he produced was one he was entitled to use. He was interviewed under caution at Ashford Police Station. He said that he did not know where he had boarded the train and was unable to explain his apparent possession of certain papers. He claimed asylum the following day. He was interviewed then and on at least one subsequent occasion, by which time he was assisted by solicitors. He said that he was an Ashraf, and gave details of events which had caused him and other family members to leave Somalia or Ethiopia in 1992. He said that those who left with him were his mother, his father, his sister, his brother-in-law and himself: “five of us” altogether. There is no record of his having been asked any more detailed questions about his family.

22.

The Secretary of State considered his claim to asylum and decided to refuse it. He considered that the Appellant knew too little about the Ashraf and its sub-clans to be credibly regarded as a member of the sub-clan he claimed. He further considered that any difficulties which the Appellant had had in Somalia before 1992 were incidents of the civil war and not of persecution. Other factors caused the Secretary of State to disbelieve the Appellant’s account of his history. ...

23.

The Appellant appealed. The grounds of appeal challenged the Secretary of State’s conclusions on the Appellant’s credibility and his knowledge of the Ashraf clan structure. They add the following:

“The Secretary of State has failed to take into account the fact that the applicant’s sister has already been accepted as being a refugee. She was granted asylum on the 14/3/2003. In her application SEF form she gives details of the applicant as being her brother (copy relevant documents enclosed).”

24.

The documents in question include the grant of asylum status to the person in question (whose name we will abbreviate for the purposes of this determination as Ouma) and her children, and her own SEF form, completed by the claimant with the assistance of her solicitors and dated 27 September 2002. ... In Ouma’s SEF form she names as one of her brothers, then living at an unknown address in Ethiopia, a person with the same year of birth as the Appellant (month and day not stated) ... [I]t appears to have been accepted, and we accept for the purposes of this determination, that Ouma is the Appellant’s sister.

25.

The Appellant’s appeal came before an Adjudicator, Mr S C D Hulme, on 25 January 2005. He heard oral evidence from the Appellant and from Ouma. He became aware or was told that the grant of refugee status to Ouma followed a successful appeal to another Adjudicator. He was not, however, shown the determination in Ouma’s case. We do not know specifically what submissions were made to the Adjudicator about the relevance of Ouma’s status. There was a full skeleton argument signed by Mr Barcello, counsel for the Appellant before the Adjudicator. The only reference to this issue is at paragraph 11:

“In support of [AA], [Ouma] his sister and a recognised refugee has provided a statement in support of their relationship and will be in attendance at court to give evidence. Her letter granting indefinite leave to remain is at page 96 of the bundle.”

26.

Nothing in the skeleton argument suggests that the Adjudicator was bound by the findings made in respect of Ouma or refers to any authority on that issue.

The AIT then set out the conclusions of Mr Hulme:

It is the underlined passage which Mr Scannell must show contains a material error of law if AA is to succeed in this appeal. Mr Kovats submits (para. 36 of his skeleton argument) that the clan membership of the appellant’s sister was not a live issue before Mr Hulme. I do not follow this submission. According to the AIT (in para. 4), it was accepted that, for the purposes of the appeal, if the clan membership was made out, the appeal would succeed. If Ouma is an Ashraf so also is her brother, the appellant. I believe that Mr Hulme would have realised that the previous adjudicator would probably have found that the sister was Ashraf.

The AIT said:

75.

In AA’s case, there had been a previous judicial determination. We reject the suggestion that it was for the Home Office to produce it, or for the Adjudicator to enquire for it. The case before the Adjudicator was that of the Appellant and the Adjudicator was to determine it on the material before him. The position is simply that a determination in Ouma’s case was not before him and in those circumstances he did not err in law by failing to take account of its contents. Whether he had the determination or not, we also reject the submission that it was binding on him in the sense that it regulated, or ought to have regulated, his determination of the appeal of the Appellant before him. He was bound to determine that Appellant’s appeal on the whole of the evidence before him, as an independent judge of fact and law. He did exactly that.

I agree.

The AIT then went on to consider whether the position would have been materially different if the appellant had decided to place before Mr Hulme the decision in the sister’s case.

76.

For the avoidance of doubt, we do not consider that the position would have been materially different if he had had the determination in Ouma’s case before him. He would have seen from it that the Adjudicator had decided that Ouma was a member of the clan she claimed, but that that conclusion had been reached from a starting point that, in her case, the Secretary of State chose not to question it. Although in those circumstances he ought to have treated the determination in Ouma’s case as the starting point, there was so much extra evidence before him that he would have been bound to move away from the starting point. When he had done that, the mere fact that on other evidence (or the lack of it) another Adjudicator had found that Ouma was of the clan she claimed could have no conceivable impact on his own task.

The AIT concluded that Mr Hulme did not make made a material error of law and so dismissed the appeal.

Again I agree with the conclusions of the AIT and would dismiss the appeal.

The appellant arrived in the UK clandestinely on 1 November 2000, claiming asylum on arrival. The appellant is a citizen of Iran. The thrust of his claim was that, being a homosexual, a return to Iran would put him at risk of persecution or treatment contrary to the Human Rights Convention.

The SSHD refused his asylum and human rights claims in June 2004. The refusal letter did not challenge the appellant’s homosexuality but did challenge the claim about a party. The thrust of the refusal letter was that notwithstanding that the punishment for sodomy is death, AH would not be targeted for being a homosexual given the difficulties of proving sodomy to the required standard and given the practice of the police not to pursue actively “homosexual activity of any kind that is performed behind the ‘veil of decency’ of closed doors”.

Mr Aitken allowed his appeal in October 2004. The SSHD appealed Mr Aitken’s decision to the AIT. On 26 April 2004 a panel of the AIT held that Mr Aitken had made errors of law and ordered reconsideration. The panel did not identify the reasons and gave no reasons for its conclusion.

On 16 August 2005 another panel of the AIT dismissed the appellant’s appeal from the decision of the SSHD in June 2004. According to paragraph 18 of the decision, Mr O’Ryan, during the hearing, had “observed that he had not been informed why the [26 April] Tribunal had decided to allow the Respondent’s appeal against Mr Aitken’s determination.” The AIT then set out the errors which it said had been found by the April Tribunal. The alleged error of law with which this appeal is concerned was that said to have been made by Mr Aitken in his approach to an earlier decision of another Adjudicator, Mr Cope.

Mr Cope had allowed an appeal brought by a HRM against a refusal of the SSHD of his asylum and human rights applications. HRM was, the appellant said, his homosexual partner. The appellant had given evidence at HRM’s hearing and had been both disbelieved in part and believed in part. The appellant and HRM were disbelieved when they gave contradictory evidence about a party during which, on their account, HRM was arrested for homosexual activities and others escaped arrest. Mr Cope did not accept their evidence about the party, the arrest and the consequences that were said to have followed and would follow from the police raid.

However, in contrast, Mr Cope had accepted their evidence, unchallenged by the respondent, that HRM and AH were in a homosexual relationship in England. He also accepted the evidence of HRM, corroborated by AH, that HRM had had homosexual relationships in Iran. Mr Cope made the further following findings. HRM’s cousin, AKN, was the wife of the appellant AH. She knew about the homosexual relationship between HRM and AH. The fact that she knew and her emotional reaction to the relationship made it more likely that HRM would come to the attention of the authorities as a homosexual and for this reason would be persecuted.

I turn to the decision of Mr Aitken. By the time of the hearing before Mr Aitken the appellant’s relationship with HRM had come to an end and he had formed another homosexual relationship. Malayeri did not give evidence. Mr Aitken wrote (paras. 12 and 16-20):

12.

It was submitted on the Appellant’s behalf that since the facts of this case had already been litigated when his former partner (his wife’s cousin) Mr HRM’s claim was assessed in a determination promulgated in April 2002 and this Appellant gave evidence in that hearing this should be considered the staring point in accordance with Devaseelan and TK [2004] 00149, and further since there was no new evidence to speak of this gave the facts of the case, which were found to be that although there was no gay party, the appellant did have an estranged and vengeful wife in Iran and the relationship with her cousin placed him at risk of being reported and persecuted.

...

16.

Having considered all of the matters raised, I consider that I am bound by the previous adjudicator’s findings as to fact. That is that the appellant is homosexual, that he had a relationship with his wife’s cousin, that his wife is aware of this and is likely to seek revenge by informing on him if he returns to Iran. That was the basis upon which the Adjudicator allowed Mr HRM’s appeal.

17.

I have come to that conclusion following the case of TK [2004] 00149 in particular. At paragraph 19 the Deputy president postulated the situation where a family could individually carry on bringing claims which had already been rejected in respect of the first member of a family, he rejected this and said “Unless some very good reason was advanced to the contrary, for example, compelling new evidence to show that X’s evidence (which originally had been disbelieved) was mistakenly appraised by the original Adjudicator, a future Adjudicator is, in the Tribunal’s view, not merely entitled to read the Determination in X’s case but also to treat it as determinative as to X’s account.” I find that the reverse must also be true and that if an account is accepted it cannot be re-litigated without compelling new evidence.

18.

There was said to be lines of cross examination which could have been taken but which were not, clearly they are not compelling new evidence. There were also said to be the refusals of the Appellant’s sister and brother in law. Whatever the reason for the refusals they could not directly impact on the facts as found by the original adjudicator.

19.

As to the question of the risk faced, it was submitted that this was also a fact found by the previous adjudicator, and there was no evidence that it had abated, even if this were not the case having read the experts report and I accept that it is detailed and persuasive of the judicial arrangements in Iran, and therefore I come independently to the conclusion that if the historic facts are as found and that the appellant’s wife will report him for his homosexual behaviour he would then be at risk of persecution. His actions would be perceived as a threat to the established religion and he is at serious risk of being flogged.

20.

I find that the appellant has discharged the burden of proof of having a well-founded fear of persecution for a Convention reason. I come to the conclusion that the appellant’s removal would case the United Kingdom to be in breach of its obligations under the 1951 Convention.

He also allowed the appeal on human rights grounds.

In paragraph 18 in the instant case the AIT, in setting out what it said were the errors of law found by the April Tribunal, said that Mr Aitken had erred in law in applying Devaseelan “thereby binding himself by facts found by an Adjudicator, Mr Cope” and that he had erred in applying TK Georgia and not SK (Guidance on the application of Devaseelan) Serbia and Montenegro [2004] UKIAT 00149.

The AIT was right, in my view, to find that Mr Aitken had made an error of law (albeit not for the brief reasons given). At the most the earlier findings could only provide a “starting point”. They were not binding. Mr Aitken did not carry out any evaluation of the evidence or arguments presented to him. It is as simple as that.

Before examining the decision of the AIT, I shall deal with a now abandoned ground of appeal which relates to the failure of the 26 April Tribunal to identify the errors of law which had been made by Mr Aitken. Mr Kovats accepts that it should have done so (see Wani v. SSHD [2005] EWHC 2815 (Admin)). However, Mr Scannell can point to no prejudice to the appellant’s case caused by this failure and, in these circumstances as now conceded, the failure affords no ground for allowing the appeal.

Before the AIT, the appellant gave evidence as did Malayeri. The AIT concluded, as had Mr Cope, that the evidence about the party, the arrests and the consequences was fabricated. The AIT concluded in paragraph 24:

For these reasons we have concluded that the appellant’s claims are totally fictitious. We are not satisfied that the Iranian authorities suspect him of homosexuality, nor are we satisfied that arrest warrants exist and that he is on a blacklist. We are satisfied that the police have never called at his home or his parents’ home. In view of this lack of credibility, we are not satisfied that the appellant is homosexual or bisexual. We are not satisfied by HRM’s evidence that he and the appellant had a homosexual relationship. Mr HRM succeeded in his asylum appeal before Mr Cope only because he persuaded Mr Cope that he and the appellant were in a homosexual relationship. We are satisfied that HRM, who is related to the appellant through marriage, has a motive to tailor his evidence to help the appellant and to stick to the story that gained him indefinite leave to remain here. In a statement of 1 October 2004 the appellant said that his current boyfriend, Scott Hutchinson, would give evidence at the hearing of their homosexual relationship. Mr Hutchinson failed to do so.

There is no dispute that the AIT was entitled to reach these conclusions of fact on the evidence before the AIT.

For these reasons I would also dismiss this appeal.

I gratefully adopt Hooper LJ’s explanation of the facts and the issues in these two appeals. The central question in both is how far, and with what effect, are the so-called Devaseelan guidelines to be applied in cases involving different applicants, but closely related factual circumstances. I will consider first Devaseelan itself; secondly the controversy within the AIT over the breadth of the principle; thirdly the principle of consistency in administrative law generally; and finally the application of those principles to the present cases.

In Devaseelan itself it was the Secretary of State who was seeking to rely on the previous decision. The applicant’s claim for asylum, on the basis of feared persecution in Sri Lanka, had been rejected; but, following the coming into effect of the Human Rights Act, he made a human rights claim based on substantially the same facts. The guidelines were the tribunal’s attempt to provide a consistent approach to such cases. Hooper LJ has set out the relevant passage in full. I extract what seem to me the most relevant points for present purposes (including the AIT’s emphasis):

(1)

The first Adjudicator’s determination should always be the starting-point. ….

(4)

Facts personal to the Appellant that were not brought to the attention of the first Adjudicator, although they were relevant to the issues before him, should be treated by the second Adjudicator with the greatest circumspection. …

(6)

If before the second Adjudicator the Appellant relies on facts that are not materially different from those put to the first Adjudicator, and proposes to support the claim by what is in essence the same evidence as that available to the Appellant at that time, the second Adjudicator should regard the issues as settled by the first Adjudicator’s determination and make his findings in line with that determination rather than allowing the matter to be relitigated…

(7)

The force of the reasoning underlying guidelines (4) and (6) is greatly reduced if there is some very good reason why the Appellant’s failure to adduce relevant evidence before the first Adjudicator should not be, as it were, held against him… “

As Hooper LJ has noted, this passage is prefaced by a statement that the first determination is not “binding” on the second Adjudicator. However, I understand this to be saying to more than that it is not binding in the technical sense of issue estoppel or res judicata. The whole purpose of the guidelines is to indicate the circumstances in which it is appropriate to follow the first decision rather than allow the issue to be relitigated. This is most explicit in the above extract from guideline (6) (directed specifically at an Article 3 claim based on the same reasons as a refugee claim). The same point is underlined by the remainder of guideline (6), which explains the limits to the new evidence which might justify reopening the first decision:

“We draw attention to the phrase ‘the same evidence as that available to the Appellant’ at the time of the first determination. We have chosen this phrase not only in order to accommodate guidelines (4) and (5) above, but also because, in respect of evidence that was available to the Appellant, he must be taken to have made his choices about how it should be presented. An Appellant cannot be expected to present evidence of which he has no knowledge: but if (for example) he chooses not to give oral evidence in his first appeal, that does not mean that the issues or the available evidence in the second appeal are rendered any different by his proposal to give oral evidence (of the same facts) on this occasion.”

The legal considerations underpinning the guidelines can be seen in the tribunal’s treatment of the respective arguments. Mr Lewis for the applicant is recorded as submitting that -

“…the previous determination is merely' a relevant matter to be taken into account' in the human rights appeal, but that neither the findings nor the conclusions of the first Adjudicator are binding upon the second Adjudicator…” (para 31)

Miss Giovannetti, for the Secretary of State, accepted that the first Adjudicator's determination “cannot be regarded as binding on the second Adjudicator”, but submitted that it was proper for a second Adjudicator to have regard to the first Adjudicator's findings, and that “the second Adjudicator should only differ from those findings where there is good reason to do so”:

“If the human rights claim was based on a different factual matrix, it would generally be necessary to make new findings, probably on additional evidence. The different factual matrix would itself be a good reason for not following and applying the first Adjudicator's determination. Otherwise, however, legal and policy considerations demanded that the Appellant's second appeal be determined in line with his first.” (emphasis added)

Miss Giovannetti identified four “legal and policy considerations” leading to that proposition:

The first is fairness: it would be unfair to an Appellant, who had satisfied the first Adjudicator that his account of events was credible, to deprive him of the benefit of that finding. If that is right, it must follow that an Appellant who has failed to satisfy an Adjudicator of his credibility is not entitled to have the same evidence re-assessed by a second Adjudicator. It is not fair to the public for there to be a system in which favourable findings stand but unfavourable findings are always questionable. Secondly, general principles of consistency and finality in litigation are important even in the absence of a rule of res judicata. Thirdly, the general approach to findings of fact in immigration cases both on appeal to the Tribunal and outside the IAA (e.g. ex parte Danaie [1998] 1mm AR 84) is that findings of fact stand unless there is good reason to displace them. Fourthly, it would, in Miss Giovannetti's submission, be contrary to good administration to have a system which allowed for the continuing existence of two undisturbed determinations of the IAA containing inconsistent findings of fact in relation to the same individual.” (para 33-4)

The tribunal thought the answer lay somewhere between the two submissions, but “considerably nearer” that of the Secretary of State.

The guidelines were approved by this court, also in the context of a second appeal by the same claimant, in LD (Algeria) [2004] EWCA Civ 804 (decision dated 30 June 2004). The appellant’s first claim, based on alleged risks to his life from Arab extremists, had been rejected by the first adjudicator. On the second appeal he had redefined the nature of the alleged threats, and had relied on new expert evidence. The second adjudicator admitted the new evidence and allowed the appeal, but the decision was reversed by the Appeal Tribunal. It held that there had been no “very good reason” under the Devaseelan guidelines for reopening the decision, and also that the adjudicator’s assessment of the evidence had been flawed. In this court the IAT’s decision was upheld.

The guidelines had been challenged (by Mr Rabinder Singh QC, for the appellant) as deriving from principles of estoppel, which were incompatible with the statutory scheme (para 27). The challenge was rejected. The guidelines were held to be a proper exercise of the IAT’s role as a specialist body, in order to secure consistency, while respecting the “fundamental obligation” of each adjudicator to decide each case on its own merits (para 29-30). The court thought it would be “positively disadvantageous” for it to rewrite the guidance to express “the same ideas in different language”; but emphasised the need for them to be applied flexibly:

“The great value of the guidance is that it invests the decision making process in each individual fresh application with the necessary degree of sensible flexibility and desirable consistency of approach, without imposing any unacceptable restrictions on the second adjudicator’s ability to make the findings which he conscientiously believes to be right.” (para 40)

At AIT level there was controversy as to whether the guidelines had any wider application than to a second appeal by the same applicant. The contrasting positions were exemplified by TK (decision dated 3rd June 2004, Mr Ockelton presiding) and SK (decision dated 5^th October 2004, Mrs Gleeson presiding).

The former, TK (Consideration of Prior Determinations) Georgia [2004] UKIAT00149), was an appeal by a wife. Her claim for asylum was based on the fear of risks identical to those relied on by her husband in his own appeal which had been rejected. It was held that the tribunal was entitled not only to read the first determination, but also to conclude that “it would be wrong to revisit the first decision in relation to the Appellant's husband’s evidence”. The tribunal noted the odd results which would otherwise follow:

“Were the Adjudicator not entitled to take this course, the following extraordinary circumstance could arise. The head of a family, call him X, claims asylum on the basis of his own account and loses on the grounds that his account is disbelieved. There follows thereafter a succession of separate members of X’s family who each makes his/her own asylum application and each expressly accepts that the risks which they fear are based on the risks to X as head of family.”

The tribunal did not accept that it was necessary for “a succession of Adjudicators… to reappraise over and over again the same basic account…” They concluded:

“Unless some very good reason was advanced to the contrary, for example, compelling new evidence to show that X’s evidence (which originally had been disbelieved) was mistakenly appraised by the original Adjudicator, a future Adjudicator is, in the Tribunal's view, not merely entitled to read the Determination in X’s case but also to treat it as determinative as to X’s account.” (para 21)

There was no such “compelling new evidence”; the only extra material being supporting evidence from the wife as “a small portion of her husband’s account”. Although the case was distinguishable from Devaseelan because a different party was involved, the same “general approach … as to the extent to which matters can properly be relitigated” was relevant. Applying guideline (6) (see para above), they held that the Adjudicator had acted “consistently with the Devaseelan principles” in treating the previous determination as –

“authoritative of the credibility of the husband’s evidence as given to the (first adjudicator) and offered again to the Adjudicator” (para 21)

A different view was taken four months later in SK (Guidance on the application of Devaseelan) Serbia and Montenegro [2004] UKIAT 00149. The factual context is not fully apparent from the decision. Since the title of the case (like that of TK) implies that it was designed to give general guidance on the issue, it seems surprising that there was no reference to the earlier decision. The tribunal held that the adjudicator had erred in applying the guidelines to a case of another family member:

“Devaseelan does not purport to deal with decisions relating to the family member although increasingly it is misunderstood by Adjudicators as doing so. It deals only with the situation where a human rights claim is made by someone whose asylum appeal has already failed and a credibility and factual matrix been found by the first Adjudicator.”

As will be seen, the continuing controversy is reflected in the two cases before us. However, in October 2006, it was overtaken by the decision of this court in Ocampo which, as I read it, resolved the controversy in favour of the wider view, holding, in line with TK, that the guidelines were not limited to cases between the same parties. It also confirmed (as had perhaps been implicit in Devaseelan) that the guidelines could work both ways: either for or against the Secretary of State.

In that case, the claimant had sought to rely on the findings of a previous tribunal in respect of his daughter’s claim to asylum, which was based largely on the same facts. Auld LJ (with the agreement of the rest of the court) said:

“In my view, the Devaseelan guidelines are as relevant to cases like the present where the parties involved are not the same but there is a material overlap of evidence, as the Immigration Appeal Tribunal observed in TK Georgia, at paragraph 21 of their determination. Clearly, the guidance may need adaptation according to the nature of the new evidence, the circumstances in which it was given or not given in the earlier proceeding and its materiality to securing a just outcome in the second appeal along with consistency in the maintenance of firm immigration control. It should also be borne in mind, as Hooper LJ pointed out in the course of counsel’s submissions, that admission of new evidence may, as a matter of fairness, operate for, as well as against, a claimant for asylum. In immigration matters, as in other areas of public law affecting individuals, public policy interests of firmness, consistency and due process may have to be tempered with considerations of fairness in particular circumstances.” (para 25)

On the facts of the case, the court held that the reception of new evidence was justified, because –

“as a matter of common-sense and fairness, the AIT rightly took into account what (counsel for the claimant) has acknowledged to be material inconsistencies in the appellant's two accounts, the second of which was not before the Adjudicator in his daughter's appeal and which logically, legally and fairly affects the final outcome of his appeal….” (para 29)

Thus Ocampo demonstrates both the width of the Devaseelan guidelines, and also the flexible approach required in their application. In TK there had been no significant new evidence; in Ocampo there was a second statement which cast serious doubt on the accuracy of the statement relied on at the first hearing.

As I understand his submissions, Mr Kovats for the Secretary of State does not accept that a previous decision should be given any particular weight, at least in a case involving a different appellant. The tribunal may have regard to it, but it is not obliged to follow it, whether or not there is new evidence; its duty is simply to decide the case on its own merits on the evidence before it.

I could understand this submission on the basis of the law as it stood before Ocampo. The normal principle is that previous tribunal decisions do not establish a precedent (see Mukarkar v Home Secretary [2006] EWCA Civ 1045). “Country guidance” cases are a well-recognised exception (see S v Home Secretary [2002] EWCA Civ 539). In Otshudi v Home Secretary [2004] EWCA Civ 893, a case involving inconsistent decisions arising out of appeals by two brothers. Sedley LJ noted that no legal submission had been based on the discrepancy as such, and commented:

“This is not the class of case which involves what Laws LJ has called a "factual precedent" - for example a finding about the political situation in a given country at a given moment. It is an illustration, if an alarming one, of the fact that two conscientious decision-makers can come to opposite or divergent conclusions on the same evidence. But it is no more material to the legal soundness of the present adjudicator's decision than hers would be to the soundness of the second adjudicator's decision….” (para 11)

As he made clear later in the judgment, he regretted that position:

“The discrepancy between the two decisions, while giving rise to no legal challenge, must be a matter of concern. If the second adjudicator is right, this appellant's life too is at risk. If he is wrong, of course, neither brother may be at risk; but asylum law - for example by demanding something less than proof positive - deliberately errs on the side of caution….” (para 23)

He noted that normally arrangements would have been made for such linked cases to be heard together. He invited the Home Office to reconsider the case on humanitarian grounds.

That, however, was before the decision in TK (Georgia),that the Devaseelan principles could be extended to such a situation, and before that extension had been approved by this court in Ocampo. In the light of that decision I do not see how we can accept Mr Kovats’ argument. I note that Hooper LJ, who was himself a party to Ocampo, takes a different view of its significance. Respectfully, however, the reasoning of Auld LJ’s judgment seems to me carefully considered and entirely clear. Whether or not it is technically binding, I would not think it right to depart from it unless I thought it clearly wrong, which I do not.

On the contrary the reasoning is in line with the principles relied on by the Secretary of State himself, through Miss Giovannetti, in Devaseelan (see above). They in turn reflect the well-established principle of administrative law, that “persons should be uniformly treated unless there is some valid reason to treat them differently”. As was said in Matadeen v Pointu [1998]1 AC 98 PC (per Lord Hoffmann):

“Their Lordships do not doubt that such a principle is one of the building blocks of democracy and necessarily permeates any democratic constitution. Indeed, their Lordships would go further and say that treating like cases alike and unlike cases differently is a general axiom of rational behaviour. It is, for example, frequently invoked by the courts in proceedings for judicial review as a ground for holding some administrative act to have been irrational: see Professor Jeffrey Jowell QC, Is Equality a Constitutional Principle? [1994] Current Legal Problems 1, 12-14 and De Smith, Woolf and Jowell, Judicial Review of Administrative Action, paras. 13-036 to 13-045.” (p 109C-D)

The same principle was relied on by this court in the asylum context R(Iran) v Home Secretary [2005] EWCA Civ 982 para 22, in the context of country guidance cases. Looking at the matter in 2004, I might have shared Sedley LJ’s doubts as to the application of such principles outside the context of country guidance cases. However, I would have also shared his concerns at the potential unfairness which that limitation can cause. Now that the jump has been made, I see no reason to question it, or to regret it.

In one of the present cases, Mr Ockelton appears to have had second thoughts about the width of the approach taken in TK. In AA he was party to a lengthy discussion of the question why a decision of fact in one appeal should be considered of any relevance in an appeal by a related claimant. The discussion included reference to decisions in administrative law (including one of my own, R v Cardiff County Council ex parte Sears [1998] 3 PLR 55), and to cases under the general law of evidence relating to the admissibility of a previous court decision (such as Hollington v Hewthorn [1943] KB 587). The latter line of authority seems to have led the tribunal, while not in terns departing from TK, to express rather more doubt as to the relevance of previous decisions in cases between different parties, suggesting that it is no more than “background” (para 66-71).

Unfortunately, the decision in that case was given in July 2006, a few months before Ocampo. Had the tribunal had the benefit of that decision much of the discussion might have been rendered unnecessary. Furthermore, I think the doubts were misplaced. As I have said, the basis of the Devaseelan approach, and of its extension (if correct), must be found, not in the civil or criminal law of evidence, but in general principles of administrative law.

While I do not think it is open to us to depart from Ocampo I would suggest two qualifications, which seem to me consistent with it. First, Auld LJ said that the guidelines are relevant to “cases like the present” where the parties are not the same but “there is a material overlap of evidence”. The term “material” in my view requires some elaboration. It recognises I think that exceptions to the ordinary principle that factual decisions do not set precedents (see above) should be closely defined. To extend the principle to cases where there is no more than an “overlap of evidence” would be too wide, and could introduce undesirable uncertainty. In all the cases in which the principle has been applied so far, including Ocampo, the claims have not merely involved overlapping evidence, but have arisen out of the same factual matrix, such as the same relationship or the same event or series of events. I would respectfully read Auld LJ’s reference to “cases such as the present” as limiting the principle to such cases.

Secondly, in applying the guidelines to cases involving different claimants, there may be a valid distinction depending on whether the previous decision was in favour of or against the Secretary of State. The difference is that the Secretary of State was a direct party to the first decision, whereas the claimant was not. It is one thing to restrict a party from relitigating the same issue, but another to impose the same restriction on someone who, although involved in the previous case, perhaps as a witness, was not formally a party. This is particularly relevant to the tribunal’s comments, in Devaseelan, on what might be “good reasons” for reopening the first decision. It suggested that such cases would be rare. It referred, for example, to the “increasing tendency” to blame representatives for unfavourable decisions by Adjudicators, commenting:

“An Adjudicator should be very slow to conclude that an appeal before another Adjudicator has been materially affected by a representative's error or incompetence…”

I understand the force of those comments where the second appeal is by the same claimant, but less so where it is by a different party, even if closely connected. Although I would not exclude the Devaseelan principles in such cases (for example, the hypothetical series of cases involving the same family, cited in TK), the second tribunal may be more readily persuaded that there is “good reason” to revisit the earlier decision.

In AA it is said that the adjudicator should have regarded himself as bound, on the issue of ethnicity, by the earlier favourable decision on the sister’s claim. The adjudicator had been aware of the previous decision, but he was not shown the actual determination. Although the claimant was represented by counsel, there is no indication that he sought to rely on the previous decision in respect of the issue of ethnicity. The simple answer to this point, in my view, is the first answer given by the AIT:

“We reject the suggestion that it was for the Home Office to produce it, or for the Adjudicator to enquire for it. The case before the Adjudicator was that of the Appellant and the Adjudicator was to determine it on the material before him. The position is simply that a determination in Ouma’s case was not before him and in those circumstances he did not err in law by failing to take account of its contents.”

In view of the way in which the case was presented by a legally represented claimant, there was no error of law in the adjudicator’s approach.

For these reasons I would dismiss the appeal in AA,but I would allow the appeal in AH and direct that the adjudicator’s decision be restored.

This appeal raises an interesting and, as far as I am concerned, a troublesome problem. Take the hypothetical example I put in argument. There is a meeting of political dissidents in Ruristan at which A and B are present. The security police raid the meeting. A and B dive out of a window and flee to England where they claim asylum. The other dissidents are rounded up and summarily executed by the police. Immigration judge X hears A and B give evidence, believes them, and grants A asylum. Immigration judge Y, who has X’s decision before him, hears A and B give exactly the same evidence, but disbelieves them and refuses B asylum. Can that be right?

I have no doubt that the man in the street would answer, “Of course it cannot be right. Either X or Y has got it wrong. Both cannot be right. Justice has not been done.” The logic is unassailable. On the other hand, the lawyer would reply, “There is no estoppel and the important principle of judicial independence demands that each judge try every case on the evidence before him or her.”

It seems to me, after careful reflection, that the apparent conflict between those two positions is met within the Devaseelan principles to which my Lords have referred and that they should be held to apply in a case such as this.

The guidelines, as such, were approved by this Court in LD (Algeria) [2004] EWCA Civ 804, with the emphasis on the flexibility of the approach. They were then extended by this Court in Ocampo v SSHD [2006] EWCA 1276 to apply to a case where there is “a material overlap of evidence”, the guidelines to be adapted as might be needed “according to the nature of the new evidence, the circumstances in which it was given or not given in the earlier proceedings and its materiality to securing a just outcome in the second appeal along with the consistency in the maintenance of firm immigration control”.

Like Carnwath L.J. I can see no reason for this Court now to depart from Ocampo and to rewrite the guidelines for cases where different parties are involved. Much as I sympathise with Hooper L.J.’s desire to simplify and clarify the law, I have concluded that we should not depart from Devaseelan as explained by this Court in those two cases. I am, however, happy to agree to the two qualifications suggested by Carnwath L.J. although I confess that I had thought that there would not be a material overlap of evidence unless the second case arose out of the same factual matrix.

I also agree with his second qualification. It does seem to me also that there is a valid distinction depending on whether the previous decision was in favour or against the Secretary of State.

I am, therefore, persuaded by Carnwath L.J.’s judgment and agree with him and with the reasons he gives for dismissing the appeal in AA but allowing the appeal in AH and directing that the adjudicator’s decision be restored.