Skip to main content

Court of Appeal on credibility, standard of proof and appellate court's role

Written by
Asad Ali Khan
Date of Publication:
07 June 2023

MAH (Egypt) v Secretary of State for the Home Department [2023] EWCA Civ 216 (28 February 2023)

The central issue in this appeal was whether the Upper Tribunal was entitled to reach the conclusion that the appellant's claim for international protection failed because of his lack of credibility. Singh LJ said that at first sight such appeals are unlikely to succeed before the Court of Appeal; indeed it is unlikely that permission to appeal would be granted. Yet he said that on closer analysis, it was apparent that the case of MAH, an Egyptian national was not a typical case about credibility. The reason why the appellant was not believed by the UT was that it determined that there were further steps he could—and should—have taken to adduce evidence which would corroborate his account. The appellant's father was tried and convicted in Egypt in 2014, and sentenced to six years' imprisonment for being a member of the Muslim Brotherhood and was incarcerated in a prison complex, which is very notorious for torturing political prisoners, and specifically members of the Muslim Brotherhood. The appellant fled Egypt on 1 August 2015, when he was 14 years old, owing to fearing he would also be arrested and detained. From Egypt, he travelled to Italy and then France and Belgium, arriving in the UK on 25 October 2016 and claimed asylum on arrival. Meanwhile his father became ill in prison, was hospitalised and died on 26 April 2018. With a very complicated procedural history in the tribunal, the main decision was a re-determination and dismissal of the appellant's case by the UT on 17 September 2021.

Unusually, the appeal in the Court of Appeal was distinctive as the the UT was acting as the tribunal of fact rather than as an appellate tribunal. Further, at the hearing, which was conducted in person, the appellant gave evidence and was cross-examined. The court said that the absence of corroboration was not fatal in asylum cases and Singh LJ provided a wide-ranging exposition of the standard of proof in asylum cases, credibility and the role played by the appellate court in asylum appeals. Overall, the UT found that the appellant would not be at real risk of persecution and/or serious harm on return to Egypt. For the same reasons it was further satisfied that there would be no breach of articles 2 and 3 of the ECHR by his removal to Egypt; nor a real risk of serious harm for the purposes of article 15(b) of the Qualification Directive; nor paragraph 339C(iii) of the Immigration Rules (humanitarian protection). Finally, the UT dismissed the appeal under article 8 of the ECHR. The UT concluded he was not a credible witness because he never asked his mother or his father's lawyer what was on the indictment at his father's trial, although he knew that the accusation was that his father was involved in the Muslim Brotherhood. It was not unreasonable to expect him, now that he is an adult, to have enquired as to the identity of his father's lawyer with his mother/extended family and sought to get evidence confirming whether his father's imprisonment was in fact predicated on an accusation that he was a member of the Muslim Brotherhood.

For the UT, he failed to provide a reasonable explanation as to why he had not sought to make contact with the human rights organisation to which one of the raids on the family was reported and had not given a reasonable explanation for failing to provide supporting evidence from family members about his father's, or other family members', political involvement. The UT concluded that although the appellant had established that his father was imprisoned in Egypt for six years and that his father died of a brain haemorrhage in 2018, he did not credibly show his claim that the reasons for his father's imprisonment was because the Egyptian authorities considered him to have association with the Muslim Brotherhood.

Consequently, the UT arrived at the conclusion that the appellant was untruthful in respect of his claim of ongoing adverse interest from the Egyptian authorities against him either on the basis of the father's real or imputed political beliefs and connections or on the basis of the appellant's imputed political beliefs or connections.

The Court of Appeal

The Court of Appeal examined the relevant legal principles and addressed the standard of proof in asylum cases, credibility and the role of an appellate court. Singh LJ said that one of the striking features of the present case was that nowhere in its judgment did the UT set out what the "lower standard of proof" is, despite using that phrase many times. Setting it out expressly was a helpful discipline since it operates as a constant reminder of precisely what question the tribunal of fact has to determine.

Standard of proof

The requirement that an applicant's fear of persecution should be well-founded means that there has to be demonstrated "a reasonable degree of likelihood" that the applicant will be persecuted for a Refugee Convention 1951 reason if returned to his own country, as decided by Lord Keith of Kinkel in R v SSHD, ex parte Sivakumaran [1988] AC 958.

Singh LJ said that strictly speaking it is not entirely accurate to refer to this as a standard of "proof" as the applicant does not in fact have to prove anything and is more accurately described as being an "assessment of risk". He also noted that various formulations of this standard were identifiable in the decisions of courts both in the English jurisdiction and elsewhere such as the USA. e.g. "reasonable possibility", Immigration and Naturalisation Service v Cardozo-Fonseca 480 US 421 (1987), at 440 (Stevens J, giving the Opinion of the US Supreme Court), cited by Lord Keith in Sivakumaran, at 994; "real chance": ibid.; or "real risk": Ravichandran v SSHD [1996] Imm AR 97, at 109 (Simon Brown LJ) and another formulation was that there must be a real as opposed to a fanciful risk that future events will happen: MH (Iraq) v SSHD [2007] EWCA Civ 852 (at paragraph 22, per Laws LJ).

The standard required is less than a 50% chance of persecution occurring and even a 10% chance that an applicant will face persecution upon return for a Convention reason may satisfy the relevant test: Lord Keith in Sivakumaran, at 994 citing Cardozo-Fonseca, at 440. Singh LJ observed that the High Court of Australia (Mason CJ) in Chan Yee Kin v Minister for Immigration and Ethnic Affairs (1989) 169 CLR 379 had found likewise and was cited with approval by Brooke LJ in Karanakaran v SSHD [2000] 2 All ER 449, at 464. The exercise required is not the familiar one used in ordinary civil litigation—when facts are taken to be true if they are proved on a balance of probabilities—this was made clear at page 459 by Brooke LJ. He said that when assessing future risk decision-makers may have to take into account the whole bundle of disparate pieces of evidence, namely (1) evidence they are certain about, (2) evidence they think is probably true, (3) evidence to which they are willing to attach some credence, even if they could not go so far as to say it is probably true, (4) evidence to which they are not willing to attach any credence at all. Brooke LJ said that the decision-maker is not bound to exclude category (3) evidence as they would be if deciding issues that arise in civil litigation. Nevertheless, Brooke LJ clarified that this approach does not entail the decision-maker purporting to find "proved" facts, past or present, about which it is not satisfied on the balance of probabilities. As he said:

What it does mean, on the other hand, is that it must not exclude any matters from its consideration when it is assessing the future unless it feels that it can safely discard them because it has no real doubt that they did not in fact occur (or, indeed, that they are not occurring at present). Similarly, if an applicant contends that relevant matters did not happen, the decision-maker should not exclude the possibility that they did not happen (although believing that they probably did) unless it has not real doubt that they did in fact happen.

when considering whether there is a serious possibility of persecution for a Convention reason if an asylum-seeker is returned, it would be quite wrong to exclude matters totally from consideration in the balancing process simply because the decision-maker believes, on what may sometimes be somewhat fragile evidence, that they probably did not occur.

Next, Singh LJ examined the crucial issue of credibility.


His lordship said that it was important to consider the issue of credibility in the context of the relatively low standard of proof. Indeed, as Lord Wilson JSC said in KV (Sri Lanka) v SSHD [2019] UKSC 10 (discussed here), "the conclusion about credibility always rests with the decision-maker following a critical survey of all the evidence" and "in an asylum case in which the question is only whether there is a real possibility that the account given is true, not even the decision-maker is required to arrive at an overall belief in its truth; the inquiry is into credibility only of a partial character". Furthermore, in SB (Sri Lanka) v SSHD [2019] EWCA Civ 160, Green LJ explained that appellate courts will accord due deference to the fact-finder who has assessed an applicant's credibility—however, the appellate court needs to be able to satisfy itself that the fact finder has at least identified the most relevant pieces of evidence and given sufficient reasons (which might be quite concise) for accepting or rejecting it.

Green LJ said that where the credibility of the appellant is in issue courts adopt a variety of different evaluative techniques to assess the evidence and consider. He had set out five points of which Singh LJ highlighted the fourth, i.e. the adequacy (or by contrast paucity) of evidence on relevant issues that, logically, the appellant should be able to adduce so as to support his or her case. Green LJ was clear that his five point list, consisting of things such as the overall plausibility of an appellant's account and the consistency of accounts given to investigators at different points in time and so forth, was not exhaustive and was not a "checklist", every part of which has to be satisfied in every case. Singh LJ stressed that "everything depends on all the circumstances of each individual case" and noted that in the present case the placed emphasis on Green LJ's fourth point. However what he said was not intended to be, nor should it be, read as if it were set out in a statute.

The UT referred to Y v SSHD [2006] EWCA Civ 1223 (at paragraphs 25-27). But Keene LJ had in Y observed that the tribunal of fact should be cautious before finding an account to be inherently incredible, as there is a considerable risk that it will be over influenced by its own views of what is plausible. Such views will have inevitably been influenced by its own background in this country and by the customs and ways of our own society. Hence, Singh LJ reiterated that it is therefore important that it should seek to view an appellant's account of events in the context of conditions in the country from which he or she comes.

This does not mean that the tribunal is required to take at face value an account of facts proffered by an appellant no matter how contrary to common sense and experience of human behaviour that account may be. The decision-maker is not expected to suspend its own judgment. In appropriate cases, it is entitled to find that an account of events is so far-fetched and contrary to reason as to be incapable of belief. Keene LJ supported that proposition by reference to the decision of Lord Brodie, sitting in the Outer House of the Court of Session, in Awala [2005] CSOH 73 where Lord Brodie cited a passage from a decision of the British Columbia Court of Appeal in Faryna v Chorny [1952] 2 DLR 354, at 357, where it was said by O'Halloran JA that:

in short, the real test of the truth of the story of a witness [where there is conflict of evidence] must be its harmony with the preponderance of the probabilities which the practical and informed person would readily recognise as reasonable in that place and in those conditions.

Singh LJ pointed out that Faryna itself was not an asylum case. It was a defamation case. The passage cited by Lord Brodie appeared in the context of consideration of the question of how the credibility of a witness should be gauged. The court said that credibility cannot be gauged solely by the test of whether the personal demeanour of the particular witness carries conviction of the truth. It was in that context that the story of the witness must be reasonably subjected to an examination of its consistency "with the probabilities".

Singh LJ said that in the context of the instant case it was important to keep in mind that the tribunal of fact is not concerned with establishing whether the facts have been proved on a balance of probabilities—as it would be in ordinary civil litigation—but is concerned with an assessment of risk. The required assessment, as held by Lord Bridge of Harwich in Bugdaycay v SSHD [1987] AC 514, must take place in the particularly sensitive context of a claim for asylum, in which there is the need for the "most anxious scrutiny". Overall, Singh LJ held that in this context, the present case was distinguishable from Y because the UT did not find that any part of the appellant's account was inherently implausible.

Role of an appellate court

The Court of Appeal noted that in the instant case the UT was acting as the tribunal of fact rather than as an appellate tribunal considering only matters of law from a decision of the FTT. The Court of Appeal will not, on an appeal, readily interfere with findings of fact by the tribunal of fact. The fact that the tribunal has considered all of the evidence, including hearing oral evidence; and is an expert tribunal in its field. However, the present case was not one in which anything turned on the appellant's oral evidence or the UT's assessment of it: the UT did not say, for e.g., that something he said in oral evidence was inconsistent with an earlier statement he had made to the Home Office. Green LJ explained in SB (Sri Lanka) thatif a judge makes material errors in the evaluation of evidence, for e.g. because the inference drawn from a fact found is logically not one that properly can be drawn, an appellate court will then interfere because "A material error in logic is an error of law".

The evaluation of the evidence must bear in mind that the relevant question that the court is dealing with is risk—not actuality. If the test were that of the balance of probability, a finding that a fact has been proved is a binary question, which can only be answered "yes" or "no". But where the question is whether there is a real risk, Green LJ said that does not squarely confront the relevant question.

Both Article 4(5) of the Qualification Directive and paragraph 339L of the Immigration Rules provide that, where certain criteria are met, corroborative evidence is not required. But it does not follow that, where one or more of those criteria are not met, corroborative evidence is required. Indeed, the correct legal position is accurately summarised in the Home Office guidance entitled, Assessing credibility and refugee status in asylum claims lodged before 28 June 2022. In those circumstances the decision-maker (here the tribunal of fact) must still consider whether—on the precise facts of the case—it is appropriate to give the appellant the benefit of the doubt, bearing in mind the relatively low threshold of "reasonable degree of likelihood". The benefit of the doubt is to be given in the guidance which states that Paragraph 339L sets out that where a claimant's account is not supported by documentary or other objective evidence there will be no need for further confirmation when the following conditions are satisfied (i) the claimant has made a genuine effort to substantiate their claim, (ii) all material factors at their disposal have been submitted, and a satisfactory explanation regarding any lack of other relevant material has been given, (iii) their statements are coherent and plausible and do not run counter to the available specific and general information relevant to their case, (iv) they have lodged an asylum or human rights claim at the earliest possible opportunity, unless they can demonstrate good reason for failing to do so, and (v) their general credibility has been established.

The guidance instructs caseworkers give the claimant the benefit of the doubt if they can show that they satisfy all five criteria. If the claimant only meets one or more criteria, the caseworker must still consider whether, on the facts of the case, it is appropriate to give them the benefit of the doubt, bearing in mind the relatively low threshold of "reasonable degree of likelihood" applicable and all of the credibility indicators must be considered in the round.

Grounds of appeal

There were seven grounds of appeal, namely (i) misdirection in law on standard of proof, (ii) misdirection in law on approach to assessing risk of return, (iii) misdirection in law on corroborating evidence, (iv) misdirection in law as to approach to the appellant's age, (v) failure to take into account the material facts, (vi) taking an irrelevant consideration into account, and (vii) irrationality.

The Court of Appeal

It was the appellant's view that despite citing the lower standard of proof, the UT failed to apply it in concluding that his account of events in Egypt were incredible. The UT instead essentially sought certainty (or something very near to it) when no more than a reasonable degree of likelihood was required. Singh LJ accepted the first ground in substance and he held:

81. … It is not necessary to say that the UT sought certainty, or something close to it. It suffices to observe that, although the UT used the phrase "lower standard of proof" many times in its judgment, it never defined what that standard was. If that were the only criticism that could be made, that would be unlikely to be sufficient to allow the appeal. More fundamentally, in my view, as a matter of substance, the UT applied a much higher standard of proof than the one which is appropriate in this context.

He however rejected that the second ground had been made out and pointed out that the UT expressly directed itself that it needed to assess the overall evidence, citing Karanakaran and Ravichandran for that proposition.

And as to the third ground, the appellant submitted that there is no legal duty to provide corroborating evidence in a protection claim, which the UT acknowledged but did not apply. Singh LJ accepted the third ground and said that while UT directed itself that there is no legal duty on the appellant to corroborate his claim, that was in substance the basis on which it proceeded. Singh LJ held that:

87. … In the circumstances of this case, bearing in mind both the relatively low standard of proof and the fact that the appellant had adduced positive evidence which supported his claim (as the UT recognised), evidence both of what he had himself witnessed and evidence of experts which was consistent with his claim, I have reached the conclusion that the UT required more of him than was necessary. It then fell into error by concluding that the failure to adduce corroborative evidence undermined his credibility with the result that his evidence was found not to be "truthful", at para. 87.

88. The Appellant had not said anything that was in fact untruthful. He still does not know whether his father was in fact a member of the Muslim Brotherhood or whether that was the reason why the Egyptian authorities arrested and detained him. Nor does he know whether they will take a similar interest in him if he is returned to Egypt. His evidence was to the effect that that is his fear and that fear is well-founded. When the UT concluded that he had not been truthful, what it meant in substance was that it did not accept that his fear was indeed well-founded. But that has nothing to do with the Appellant's credibility; it is simply an expression of the conclusion that his claim for asylum is not made out. In this context it is of interest to note that … the UT said that it rejected "the reliability of the Appellant's claim as to the reasons for his father's arrest." (Emphasis added) I would observe that "reliability" is not the same thing as "credibility".

As to the fourth ground, the appellant submitted that the UT failed to properly take into account his age when assessing his credibility, despite noting that it was mindful of this. The submission was rejected as the UT had the fact that the appellant was a child at all material times well in mind.

The Court of appeal accepted the argument on the fifth ground that the UT failed to take into account and make findings as to whether the Egyptian authorities visited the family home of the appellant on a number of occasions before his father's arrest, looking for his father "to do with the Muslim Brotherhood issue". In Singh LJ's view, it was important, in a case where the appellant's evidence was ultimately found not to be truthful, to set out, at least in summary, what his own evidence was. Notably an important part of the positive evidence, which supported the appeal, was not mentioned in the UT's judgment. This was a material omission, in particular at para. 19, where it said that the appellant had not been present during his father's arrest. Although that was true so far as it went, what the UT did not mention was that the appellant's evidence in his witness statement was that he had been present "sometimes" when the authorities had come to the family's house some five or six times, every one or two weeks in the period before his father was arrested. He said that his father was not there on those occasions and the authorities would search the house and destroy things. There were normally about six or seven men, sometimes eight. They did not usually come in uniform but in civilian clothing, although there might have been one wearing a uniform and they had told the appellant's mother to persuade his father to hand himself in. Singh LJ added that:

94. I also accept Mr Jones's submission that the evidence in relation to human rights organisations and lawyers in Egypt (in particular whether it was unreasonable not to approach them to provide corroborative evidence) had to be considered in the context of the conditions in Egypt at the material time, in 2017 and 2018 in particular. That evidence included the evidence of Mr Armbrust, that there was a crackdown at that time, in particular the mass arrests of human rights lawyers and activists, which had been reported by the international NGO Human Rights Watch. There was similar evidence before the UT from Ms Pargeter, to the effect that, from April 2017, the President of Egypt had maintained a nationwide state of emergency that gave security forces unchecked powers: see para. 2.22 of her report dated 14 September 2020.

As to the sixth ground (taking into account an irrelevant consideration), Singh LJ said that he would not have been inclined to accept it. However, in the end, this was not material as the Court of Appeal reached the conclusion that the UT fell into more fundamental errors, in particular in relation to the standard of proof and the failure to adduce corroborative evidence (the first and third grounds).

As to the seventh and final ground (irrationality), Singh LJ held that the argument was on point and held that:

99. … When one applies the correct standard of proof, the positive evidence which supported the Appellant's claim, including his own evidence of what he had observed before his father was arrested; the circumstances of that arrest and his father's imprisonment; and the expert evidence, which was consistent with – although not directly probative of – his case, could reasonably lead only to one conclusion: that the Appellant does have a well-founded fear of persecution if he is returned to Egypt. I stress again that the Appellant does not have to show that this will happen or even that it is likely to happen on a balance of probabilities. It suffices that there is a reasonable degree of likelihood. He therefore qualifies as a refugee.


The judgment of the Court of Appeal neatly sets out the principles in asylum appeals as to credibility, standard of proof and the role of an appellate court. As Singh LJ explained it is important, in a case where the appellant's evidence is found not to be truthful, to elaborate at least in summary what his own evidence was. This way omissions can be avoided.

As to credibility, he stressed that the appellate court needs to be able to satisfy itself that the fact finder has at the very least identified the most relevant pieces of evidence and has given sufficient reasons for accepting or rejecting it. He also highlighted the adequacy (or by contrast paucity) of evidence on relevant issues that, logically, the appellant should be able to adduce so as to support his or her case. Credibility cannot be gauged solely by the test of whether the personal demeanour of the particular witness carries conviction of the truth. Moreover, material errors in the evaluation of evidence warrants interference from an appellate court because a material error in logic is an error of law.