AA (Zimbabwe) v Secretary of State for the Home Department

This is the judgment of the Court.

The Secretary of State acknowledges that some people returning from the United Kingdom to Zimbabwe may face a real risk of inhuman or degrading treatment or punishment. Anyone who does face that risk should not be required to return involuntarily, since that would constitute a violation by the United Kingdom of his fundamental rights under Article 3 of the European Convention on Human Rights.

In January 2002, after a newspaper report of ill-treatment of an asylum seeker returning after he had been expelled from the United Kingdom, the Secretary of State suspended removals to Zimbabwe. There were no such expulsions until November 2004 when removals were resumed.

On 15 February 2005, the Immigration Appeal Tribunal heard and subsequently determined three appeals which raised a number of common issues relating to the then current situation in Zimbabwe. The IAT’s determination, notified on 11 May 2005, is SM and Others (MDC – Internal Flight – Risk Categories) Zimbabwe CG [2005] UKIAT 00100. The Tribunal concluded that all failed asylum seekers forcibly returned to Zimbabwe would be regarded with contempt, suspicion and hostility in the belief that at least some of them would be British-trained agents or saboteurs, so all would be subject to close scrutiny. If the Zimbabwe authorities then discovered any reason to believe the returnee to be politically active, the initial scrutiny would be followed up. At that stage, “there is a reasonable degree of likelihood that this [follow up] will include treatment sufficiently serious to amount to persecution” (paragraph 42). Those potentially at risk included people whose political activity was identified and described in paragraph 43 of the determination. The existence of records alone did not materially add to the assessment of the risk of persecution in an individual case, which depended on the individual’s profile and background. In reaching these and other conclusions, the Tribunal considered and accepted the evidence of Professor Ranger, whose expertise arises from more than 45 years’ familiarity with Zimbabwe. MDC (in the title to this decision) is an acronym for the Movement for Democratic Change, an opposition political party in Zimbabwe formed in 1999.

On 14 July 2005, the Secretary of State again suspended removals to Zimbabwe pending decisions on judicial review applications on behalf of failed asylum seekers. There was one further involuntary removal only to Zimbabwe in August 2006, but the Secretary of State now wants to resume removals in some cases.

This appeal comes before this court as a test case in which the Secretary of State wishes to obtain and confirm a finding on the facts that those who are returned involuntarily to Zimbabwe from the United Kingdom do not by that fact alone face a real risk of torture or of inhuman or degrading treatment. For reasons which we shall indicate, we are sceptical whether a concluded determination in this case is capable of being applied uncritically to other cases in 2007.

The background facts concerning AA, the present Appellant, may be found in paragraphs 5 to 10 of the determination of the Asylum and Immigration Tribunal of July 2006 [2006] UKAIT 00061. The Chairman of this Tribunal was the President of the AIT, Hodge J. In brief summary, AA is a citizen of Zimbabwe, born on 8 April 1975. He arrived in the United Kingdom on 6 November 2002. He was granted temporary admission, but absconded. Upon his detention two-and-a-half years later, he claimed asylum. The Secretary of State refused his claim on 27 July 2005. He appealed to an immigration judge, who rejected as untrue his claim to be at risk on return on account of anything he had done in Zimbabwe. But the immigration judge allowed the appeal on human rights grounds, holding that the Appellant would face real risk of persecutory ill-treatment by the authorities in Zimbabwe simply because he would be returning to Zimbabwe as a person who had unsuccessfully claimed asylum in the United Kingdom.

Reconsideration of this decision was ordered, and a first Tribunal reconsidered the matter in October 2005. This first Tribunal found that there had been a material error of law, because the immigration judge had relied on evidence which he had found on the internet after the hearing. The first Tribunal made a fresh decision, which concluded, in agreement with the immigration judge, that the Appellant’s asylum claim was, in all its substantive parts, fraudulent. AA is therefore a failed asylum seeker, who claimed to have been involved in some low-level political activity in Zimbabwe in opposition to the ruling party, but whose claim has been held to have been fraudulent. The case has proceeded to this court on this appeal on the basis that AA has had no material political involvement in or concerning Zimbabwe and that his only relevant attribute, if he returns to Zimbabwe, would be that his claim for asylum had failed. His personal case is, of course, important. But the wider significance of it is said to be that he may be taken as typical of Zimbabweans with no material political background whose claim for asylum failed because the facts on which it was based were found to be untrue.

Notwithstanding the finding to which we have referred, on 18 October 2005 the first Tribunal upheld AA’s claim on Refugee Convention and human rights grounds. On 12 April 2006, this court allowed the Secretary of State’s appeal against that decision and remitted the case for further reconsideration by the AIT. The matter now before this court is a further appeal, this time by AA, from the decision of the second Tribunal of 31 July 2006 dismissing AA’s appeal on both Refugee Convention and human rights grounds.

Buxton LJ gave AA limited permission to appeal, suggesting that those advising him might consider reformulating somewhat grounds of appeal which Buxton LJ considered to be reasonably arguable. Mr Andrew Nicol QC and Mr Mark Henderson, counsel for AA, have helpfully done this and there are now five reformulated grounds of appeal. Of these, in substance Buxton LJ gave permission for ground 1. AA renews his application for permission on the other four grounds. The substance of ground 1 is that the Tribunal failed to consider significant parts of the oral evidence of two former members of the Zimbabwean Military Intelligence, referred to anonymously as witnesses 5 and 6 (W5 and W6); and failed properly to address evidence about attitudes and practices of the Central Intelligence Organisation (CIO).

As Buxton LJ ruled, the extensive written submissions on behalf of the appellant contain manifold criticisms of the Tribunal’s handling of the facts. An appeal to this court is only on a point of law. We note in this context Mr Kovats’ overarching submission that, when the very careful decision is read as a whole, the appeal is no more than a disagreement with the Tribunal’s findings of fact. The five reformulated grounds of appeal sensibly reduce the ambit of AA’s challenge. As will appear, we regard the challenge to the Tribunal’s assessment of the evidence of W5 and W6 as pivotal.

The issue which the second Tribunal had to address was whether the evidence established a real risk of serious ill-treatment for a person who had been found to have no objectively well-founded fear of being so treated for any reason other than that he was being forced to return to Zimbabwe, and who would be regarded as a failed asylum seeker when he arrived at the airport.

The Tribunal heard oral evidence from a number of witnesses, including Professor Ranger and two witnesses from the Home Office, Mr Walker and Mr Walsh. The Tribunal gave an overview of the evidence (paragraphs 14 to 22). In the period between 16 November 2004 and 7 July 2005, there had been 210 people removed from the United Kingdom to Zimbabwe. It was possible to glean evidence or information about what had happened to some of these, but these were a relatively small proportion. Direct particular evidence relevant to what might happen to a failed asylum seeker returned to Zimbabwe in or after July 2006 was therefore not great. There was evidence from and about Non-Government Organisations (NGO’s) in Zimbabwe, some of which those representing AA said was of no value.

The Tribunal considered the test which they should apply. They held in paragraph 31 that:

“The issue is whether the evidence establishes a real risk. The Appellant does not need to show a certainty or a probability that all failed asylum seekers returned involuntarily will face serious ill-treatment upon return. He needs to show only that there is a consistent pattern of such mistreatment such that anyone returning in those circumstances faces a real risk of coming to harm even though not everyone does. So is there evidence pointing to a substantial number of cases in the context of general evidence showing that involuntarily returned failed asylum seekers are at real risk of being subjected to serious ill-treatment on that account alone?”

The Tribunal considered the existing country guidance in SM and Others (paragraphs 32 to 33). They considered objective country guidance (paragraphs 34 to 35) and the evidence of Professor Ranger (paragraphs 36 to 56). They then considered procedures at Harare airport with extended reference to the written and oral evidence of W5 and W6 (paragraphs 57 to 72). There was then consideration of the activities of the International Organisation for Migration (IOM), an organisation which had to date provided reintegration assistance to 117 Zimbabweans (paragraphs 73 to 86); and of information obtained by the Secretary of State from NGO’s (paragraphs 87 to 93) which the Tribunal found did not take the position very far either way.

A large section of the Tribunal’s determination was then devoted to evidence and information concerning the treatment of individual involuntary returnees. The evidence and information was from various sources with varying detail and reliability. It had largely been gathered together under the direction of Ms Sarah Harland of the Zimbabwe Association in London. The Tribunal considered this material, which concerned 39 individual returnees, in great detail. Their conclusions about this evidence, which we refer to later in this judgment, are in paragraphs 229 to 239. In short, they concluded that a very small minority of the 210 failed asylum seekers returned involuntarily may have been subjected to ill-treatment; and that these accounts cannot be relied upon to demonstrate that returnees face a real risk of being subjected to serious ill-treatment on account only of being involuntarily returned failed asylum seekers. Mr Kovats, for the Secretary of State, presented us with an analysis to the effect that only three of the 39 returnees considered were or may have been failed asylum seekers only and were or may have been significantly ill-treated on their return to Zimbabwe. The three were returnee 5 (paragraph 124); returnee 20 (paragraphs 160 and 162); and returnee 22 (paragraphs 167 and 169). There were, he said, approximately 14 others who were or may have been ill-treated, but they had political or other connections and were not merely failed asylum seekers. Mr Nicol and Mr Henderson challenged this analysis both generally and in particular in relation to returnees 4 and 31. Returnee 31 was not a failed asylum seeker, but she claimed to have been struck across the mouth during interrogation and to have heard shouts and groans from two other deportees. The tribunal held of her that the treatment to which she claimed to have been subjected did not amount to serious ill-treatment such as to engage Article 3. This finding is challenged in this court as part of the reformulated fifth proposed ground of appeal, as is the Tribunal’s finding in relation to returnee 4. We refer to these later in this judgment.

The Tribunal summarised its conclusions in paragraph 244 as follows:

“A person who is returned involuntarily to Zimbabwe having made an unsuccessful asylum claim in the United Kingdom does not face on return a real risk of being subjected to persecution or serious ill-treatment on that account alone. That is so whether or not the removal is escorted. Each case must be considered on its own facts. We reaffirm the country guidance in SM and Others (MDC – Internal Flight – Risk Categories) CG [2005] UKIAT 00100. The evidence before us demonstrates that those at risk upon return to Zimbabwe continue to fall into the risk categories identified and set out in SM. This is subject to what we say about those whose military history discloses issues that will lead to further investigation by the security services upon return to Harare airport and those in respect of whom there are outstanding and unresolved criminal issues.”

The Tribunal noted that there continue to be three flights a week from the United Kingdom to Harare airport. These are generally fully booked with ordinary travellers who pass freely and without difficulty in and out of Zimbabwe. An unsuccessful asylum seeker returning voluntarily would be indistinguishable from the ordinary traveller. As to those returning involuntarily, the Tribunal’s conclusion in the light of all the evidence was that at Harare airport all deportees would be diverted for questioning to establish whether the deportee was of any interest to the Central Intelligence Organisation (CIO) or the other military and police security services. He would be of interest if, for example, it emerges that he has a political profile considered adverse to the regime. In that event he would be taken from the airport for interrogation and that might involve serious ill-treatment. But the Tribunal found that the CIO was not shown to have any interest in manufacturing or fabricating evidence to create suspicion that was otherwise absent. So in the case of AA, who was found by an immigration judge to have no involvement with the MDC, there was, said the Tribunal,

“…no reasonable likelihood that the Appellant would be prevented from passing through the airport after the initial screening interview. He would then be able to return unhindered to live at one of the two rural homes available to him and his family”.

We take these first. The Appellant needs permission for these grounds. They seem to us to be devoid of substance. They read as follows:

“3.

If and in so far as the Tribunal required the Appellant to show a substantial number of cases where returned asylum-seekers had been subjected to ill-treatment notwithstanding the inferences to be drawn from the other evidence, the Tribunal misdirected itself in law and departed from the correct test of whether there was a real risk (taking all the evidence into account) of such ill-treatment.

4.

If and in so far as the Tribunal drew an inference that returnees had not encountered ill-treatment from the absence of information about such returnees, it acted in a way which was not reasonably open to it.”

Nothing in the determination justifies the conclusion that the Tribunal made the requirement postulated in ground 3, and we see no reason to conclude that the Tribunal drew the general inference postulated in ground 4, although it did of course have regard to those who indicated that if there had been ill-treatment they would expect to have heard about it.

We do not consider that the definition of the test which the Tribunal articulated is open to criticism. We have referred to it in paragraph 14 above. They reached this definition after considering the approach of the first Tribunal, which was not criticised in this court; the decision of this court in Hariri v Secretary of State [2003] EWCA Civ 807, and in particular the judgment of Laws LJ at paragraph 8; and the decision of this court in Batayav v Secretary of State [2003] EWCA Civ 1489, and the judgment of Sedley LJ at paragraphs 37 to 39. The Tribunal considered there to be no real tension between each of these. The test which they formulated was a neat and, we think, correct synthesis of the ways in which Laws LJ and Sedley LJ each addressed the question. Mr Nicol suggested that the Tribunal’s application of the test, in particular in reaching in paragraph 229 their conclusions derived from the individual accounts of failed asylum seekers returned involuntarily, was arguably flawed. The Appellant, he submitted, did not have to establish or demonstrate a consistent pattern that such people were subjected to ill-treatment, when it was clear from other evidence that there was a real risk of ill-treatment. He submits that the Tribunal’s view of these cases was influenced by the view they had formed about procedures at Harare airport derived from an erroneous or incomplete assessment of the evidence of W5 and W6. He submits that the evidence about individuals, properly assessed, is consistent with the case advanced by the Appellant based on the general evidence – the more so, if the Tribunal’s errors with the evidence of W5 and W6 are corrected. The evidence about individuals does not provide a basis for a rational decision that it was safe to return failed asylum seekers to Zimbabwe.

We are entirely unpersuaded that the Tribunal, having articulated a correct test in paragraph 31 of its determination, proceeded to apply it incorrectly in paragraph 229 and succeeding paragraphs, supposing for this purpose that their evaluation of the various strands of factual evidence was sustainable. There is much perhaps to be said about their evaluation of the evidence about individuals (see below). But given that evaluation, we think that paragraph 229 correctly applied a correct principle. Evidence of what has happened to individuals is capable of contributing to an evaluation of this kind. The Tribunal’s reference in paragraph 229 to “a consistent pattern of such returnees being subjected to ill-treatment” is a clear reference back to the test which they had correctly propounded in paragraph 31.

We are not, therefore, persuaded that the Tribunal misdirected itself by applying a wrong test, and we refuse the renewed application for permission on the reformulated grounds 3 and 4. In doing so, however, we do not preclude a reconsideration of the Tribunal’s evaluation of the material relating to the 39 involuntary returnees, not least in the light of a reconsideration of the evidence W5 and W6, to which we now turn.

As we have said, Buxton LJ gave permission to appeal on ground 1, which relates to the Tribunal’s assessment of the evidence of W5 and W6 and their handling of more general evidence about the attitude and practices of the CIO. It is said in substance that the Tribunal erred in law because when dealing with the evidence of W5 and W6 and coming to the conclusion that all deportees would be faced at Harare airport with a screening interview, they failed to give any or sufficient weight to the evidence of those witnesses and others who indicated that at the screening interview itself there was and is a real risk of the interviewee being subjected to serious violence. The Tribunal did not refer to significant parts of the oral evidence of these two witnesses. It is said that their conclusion was wholly incompatible with the evidence about the attitudes and practices of the CIO, and was not one which was open to the Tribunal. By ground 2, it is said in the alternative that, if the Tribunal did have regard to the full evidence of these witnesses, they must have applied an excessively demanding standard when considering whether ill-treatment would amount to inhuman or degrading treatment for the purposes of Article 3. In the light of the view that we have reached on ground 1, it will be unnecessary to proceed to the alternative ground 2.

We must first enlarge somewhat on our account of the evidence which the Tribunal considered other than that of W5 and W6, to set that evidence in its context.

Having considered SM and Others, the Tribunal looked at objective country guidance, which indicated that the situation in Zimbabwe was poor and was deteriorating. That theme was developed by Professor Ranger, who provided three reports. By July 2006, he was of the opinion that failed asylum seekers did indeed face a real risk of being subjected to serious ill-treatment. That opinion was formed as a result of what he learnt of the fate of returnees. But, as he conceded, he had not in his report drawn attention to some political rhetoric to the opposite effect, that is to say welcoming returnees; nor had he drawn attention to regular and well-patronised passenger air traffic between Zimbabwe and the United Kingdom. This did suggest that in reality not everyone known to be connected with the United Kingdom is suspected of treachery and criminalised. The Tribunal therefore decided to approach with some caution Professor Ranger’s view that a failed asylum seeker about whom nothing else was known was at risk of serious ill-treatment simply because he or she was a failed asylum seeker.

We have referred to the International Organisation for Migration. Since 2002, this organisation has been assisting those returning voluntarily to Zimbabwe. It has worked out a draft Memorandum of Understanding with the Secretary of State to provide a pilot project to assist any involuntary returnee who might want help, and to monitor those returned. The Tribunal paid tribute to the work of the IOM, but found that:

“The proposed monitoring of returnees under the pilot project described in the draft Memorandum of Understanding is of no direct relevance to the assessment of risk on return to any particular returnee as the IOM is not in a position to intervene to prevent such abuse should it occur.”

Before us neither party sought to challenge that conclusion. We have also referred to evidence of and concerning NGO’s and the Tribunal’s conclusion relating to it.

As to the 39 individual involuntary returnees, some were not failed asylum seekers, but it is relevant to bear in mind that the Zimbabwean authorities would not initially be able to distinguish between a failed asylum seeker and any other category of involuntary returnee. Also, many of those who were failed asylum seekers had, when claiming asylum, asserted that they were linked to the main opposition party in Zimbabwe, the MDC. That claim was often rejected, and their credibility impugned, so that the Tribunal was not always prepared to accept what they said about what happened on their return to Harare. As to this category of evidence as a whole the Tribunal’s detailed conclusions were:

“229.

We find that the individual accounts of those who have been involuntarily returned to Zimbabwe, considered together and evaluated with care in the context of the evidence overall, do not establish or demonstrate a consistent pattern of such returnees being subjected to ill-treatment upon being involuntarily returned simply on account of being regarded as someone who has made an unsuccessful asylum claim in the United Kingdom. At its highest this evidence can only demonstrate that a very small minority of the 210 failed asylum seekers returned involuntarily may have been subjected to ill-treatment. Put another way, this does not point to a substantial number of cases in the context of the available evidence being subjected to ill-treatment simply on account of a person being identified as an involuntarily returned failed asylum seeker.

230.

An examination of those accounts that survive scrutiny in any form at all reveals that there is only a very small handful of cases in which it is said that there was no reason other than mere fact of an involuntary return and the perception on the part of the authorities of being a failed asylum seeker that gave rise to these difficulties. Of those, some were bare assertions of that being the case with no real detail of the nature or severity of the ill-treatment or the circumstances in which it was inflicted. In our judgment, and for the reasons we have set out, little weight can be given to such accounts. We have explained why we approach the accounts with caution and why these accounts cannot be relied upon to demonstrate that returnees face a real risk of being subjected to serious ill-treatment on account only of being involuntarily returned failed asylum seekers.

231.

This is in accordance with the evidence of procedures at the airport which suggest that while all deportees will be questioned, often in a hostile fashion, it is only in those cases where some further suspicion arises, above and beyond the asylum claim in the United Kingdom, that the deportee is moved on to the next stage of the process which involves interrogation which carries with it a real risk of serious ill-treatment.”

As to what would happen to an involuntarily returned failed asylum seeker, the Tribunal heard evidence from W5 and W6, who were formerly involved with airport security, and both of whom still had contacts serving there. There was a third witness who worked in air traffic control, but whose evidential contribution was not in the end of great value. Both W5 and W6 have been granted asylum in the United Kingdom, and now live and work here. But from about 1993 to May 1998, W5 was the officer in charge of the military intelligence unit of the airport. For two-and-a-half years up to 1996, W6 was a senior non-commissioned officer there. W5 and W6 gave written statements and testified orally, not only as to their own experience, but also as to what they knew about current conditions having spoken to their contacts who were still serving at the airport. We do not need to set out in detail what the witnesses said. It is summarised by the Tribunal between paragraphs 58 and 69 of its determination. For the purposes of this appeal, the Tribunal’s summary of this evidence is important. It is as follows:

“70.

Drawing together this part of the evidence, considered in the context of the evidence overall including what we say about the evidence discussed below, we reach the following conclusions. Those being returned involuntarily to Zimbabwe will face a two-stage process. Upon arrival all deportees will be separated from ordinary travellers and will be interviewed. This is a screening or filtering process. It is assisted by having recourse to any intelligence that might be available. It is designed to distinguish between deportees about whom there is nothing else known or suspected to give rise to any interest and those who may be of interest because of a relevant military history, outstanding criminal issues or who may have some form of political profile, at whatever level. A relevant military history will be one in respect of which enquiries reveal aspects to be followed up such as being absent without leave or being involved in military activities outside Zimbabwe.

71.

Those conducting this initial interview at the airport are likely to prepare a report upon each deportee. If there is nothing to suggest anything in a person’s military history that requires further investigation or there are no outstanding criminal matters to be followed up and if there is no reason to suspect any involvement with political activity adverse to the regime the deportee will be allowed to pass through the airport. This process may involve the deportee being detained at the airport for several hours but carries with it no real risk of serious ill-treatment. The report is likely to be available to the CIO officers in the deportee’s home area to inform the process of monitoring that is likely to take place thereafter.

72.

Where the screening interview does give rise to any suspicion that the deportee has any form of adverse political profile or where it is established that there is a military history requiring further investigation or outstanding criminal matters it is likely that the deportee will be taken from the airport by the CIO, military intelligence or the police, depending upon the nature of the suspicion that has arisen, for the purpose of a rigorous interrogation. In view of what is known from the country evidence about the CIO that does given rise to a real risk that the deportee will be detained for a period of time and will be subjected to serious ill-treatment.”

The case advanced on behalf of AA in this appeal is both general and particular. The general case is that it was not rationally open to the Tribunal on the evidence as a whole to find that there was a two-stage screening interview process at the airport, and that the first stage was to be regarded as risk-free for those without an adverse political profile, a questionable military history or outstanding criminal matters. The direct experience of W5 and W6 was out of date. It ended in 1998 and 1996 respectively. It was at a time when there were few returning asylum seekers from the United Kingdom. It was at a different and less spacious airport. The Tribunal’s findings were incompatible with what is known about Zimbabwe generally, with what is known about how the CIO operates, and the experiences of a significant number of the individual returnees properly considered. It is said that, in contemporary circumstances, the division of the procedure into two stages was illusory. It was extrapolated from evidence which did not include suspicion of the person returning to Zimbabwe. Now a failed asylum seeker returning from the United Kingdom would be regarded with suspicion and hostility and would probably be revealed to be a failed asylum seeker. In these circumstances, interrogation by intelligence services, whom W6 regarded as no longer professional, for a period of several hours must constitute a real risk of serious ill-treatment in the light of the evidence as a whole.

We are not persuaded that this general case alone predicates an error of law sufficient to sustain this ground of appeal. We have carefully considered the written and oral evidence of W5 and W6. Their direct experience was not contemporary, but they both had contacts in Zimbabwe. Their evidence did sustain a finding of a two-stage process. Apart from particular points about their evidence, which we consider below, and subject to possible further consideration of the evidence and information about individual returnees in the light of the particular points and generally, we consider that it was open to the Tribunal to make the factual evaluative judgment in this respect which they did.

The particular part of this ground of appeal is, however, more persuasive. Those advising AA considered that the Tribunal’s written determination had failed to take account of parts of the evidence of W5 and W6 which supported the case that involuntarily returning failed asylum seekers faced a real risk of serious ill-treatment even at a first stage screening interview. There was no transcript of the evidence of these witnesses, but the notes of evidence taken by members of the Tribunal have been provided to us. In summary the relevant parts of that evidence are as follows.

W5 said in his statement that his current contact at the airport told him that all returned asylum seekers were handed over to the CIO who carried out thorough questioning and then decided what should be done. In re-examination, as noted by the Chairman and another member of the Tribunal, he explained that the thorough questioning, as he understood it, involved the use of crude techniques, which he referred to as coercion.

W6 also explained in oral evidence what happened at the airport in the screening interview. As noted by the Chairman of the Tribunal, he said “there was abuse at the airport; kicking, beating, not torture”. The note made by another member of the Tribunal was to the same effect, but noticeably that was when W6 worked at the airport. He left, he said, because of corruption in government and things going the wrong way. These days, he said, the Zimbabwe Intelligence Services were no longer professional.

This evidence of W5 and W6 as to significant violence at the airport did not stand alone. It was reflected in some of the complaints made by or on behalf of some of the individual returnees, and in our judgment it should have been addressed. Mr Kovats accepts that the Tribunal’s decision says nothing about hitting and kicking at the airport. The reference to thorough questioning is quoted, but there is no reference to the explanation by the witness of what he understood those words to mean. Whether or not there was violence at the airport was, in the context of this case and in the context of the Tribunal’s own conclusions as outlined above, an important issue. Not having heard the evidence, we are unable to say with any confidence how, if this had been addressed, it may have affected the evaluation as a whole. It might thus be seen, as we indicated earlier, as pivotal. It could have been determinative of the appeal, as is apparent from the structure of the Tribunal’s judgment.

We shall therefore allow this appeal on the first ground. In doing so, we bear fully in mind the specialist experience of the Tribunal, and its advantage in seeing and hearing the witnesses. But this is not a matter which appears to be affected by specialist experience or specialist assessment of oral testimony. Mr Kovats also pointed out that the Tribunal rightly looked at the evidence as a whole, including that from individual returnees, when deciding what happened at the airport. But that does not, we think, sufficiently deal with what appear to be potentially important parts of the evidence of W5 and W6 which the Tribunal did not sufficiently address. We are not persuaded by Mr Kovats’ submission that the omitted evidence was indirect and insignificant and that, in the context of the evidence as a whole, it could not have tipped the balance for a category of failed asylum seeker whose claim has been rejected.

Having to this extent allowed the appeal on ground 1, we shall remit the matter to the Tribunal for yet further reconsideration. We are concerned that, in a matter of such importance, this court is unable to achieve a degree of finality. But we have already questioned whether a decision in this case on increasingly out-of-date evidence, including information about a small sample only of those returned during a few months between November 2004 and July 2005, can be uncritically applied for the future. We shall remit the matter to the same Tribunal, subject to any directions to the contrary by the President of the Tribunal for administrative reasons only. We understand concerns of the Appellant about remitting the matter to a tribunal whose constitution has already dismissed his appeal. But it would be disproportionate to do otherwise, when this experienced and careful Tribunal has already done so much work on the appeal. It will be for the Tribunal to decide what additional evidence it may hear. But we anticipate that this will be relatively confined and would not include a purely factual reconsideration of the evidence about the individual returnees, for which Buxton LJ, in our view correctly, refused permission to appeal.

However, the question whether failed asylum seekers with no adverse political profile or relevant military or criminal attributes returning involuntarily to Zimbabwe face a real risk of inhuman or degrading treatment is obviously a finely balanced one. We have indicated that, in our view, a reconsideration of the evidence of W5 and W6 might tip the balance. Since we regard the evaluation of this evidence about procedures at the airport as pivotal, and since it is intrinsically bound up with the general evidence about the attitude and practice of the CIO, we shall not embark on an analysis of the Tribunal’s handling of that evidence in isolation. We note in particular, however, the submissions in paragraphs 97 to 102 of AA’s skeleton argument to the effect that the Tribunal failed to take explicit account of the evidence of W1 and W2 as to physical ill-treatment of those questioned by the CIO.

Reconsideration of the evidence of W5 and W6 may also require reconsideration, in the light of all the evidence, of the impact which the evidence and information about the 39 individual returnees, taken as a whole and with the other evidence, may have on the appeal. We say this for two reasons. First, ground 5 of the present appeal seeks to challenge the Tribunal’s conclusions about three of the individual returnees.

There were two inconsistent accounts of R4’s treatment when he was removed in January 2005. According to the first, he was intensively questioned, then released; according to the second, he was beaten by the CIO during intensive questioning at the airport. The Tribunal regarded the second account as unreliable for evaluative reasons which are by themselves sustainable, if their earlier conclusions about procedures at the airport are also sustainable, but which otherwise may require reconsideration. One of their reasons was that the first account given by R4 was consistent with the evidence they had received concerning procedures at the airport. If, as we think, the Tribunal’s conclusion about procedures at the airport requires reconsideration, so too may their conclusion about R4.

Second and generally, the Tribunal’s conclusions about the individual accounts taken as a whole (see paragraphs 229 ff to which we have already referred) drew (in paragraph 231) on their earlier conclusions about the evidence of procedures at the airport. Since, as we think, the balance is a fine one, reconsideration of the evidence of W5 and W6 will require reconsideration also of the relevance of evidence about the risk of violence to voluntary and involuntary returnees, who were not merely failed asylum seekers, to those of whom AA is taken to be representative. We note, for instance, submissions on behalf of AA that the evidence of R25 (W7) and R26 (W8) was inconsistent with a conclusion that a real risk of serious ill treatment only arises when a returnee is taken away from the airport. R25 and R26 were MDC activists (not merely failed asylum seekers) whose evidence complained of serious ill treatment and showed that there are sufficient facilities to enable beatings to be inflicted during questioning at the present international airport.

Ground 5 also seeks to criticise the Tribunal’s findings in relation to R19, but the criticism is insubstantial. The information about R19 was extremely vague and there was no indication at all of the nature of the problem he was said to have encountered. That remained so with the addition of Ms Harland’s evidence that the problems, whatever they were, occurred at the airport.

Criticism of the Tribunal’s decision in relation to R31 seems to us to have more substance. She was not an asylum seeker and the information about her came from what appears to be an internet news report. She was a student who had been refused an extension of her leave to remain in the United Kingdom. At the airport, she was, according to a report, subjected to a hostile interview during which she was struck across the mouth when she asked why the interviewers would not believe she was just a student. After about three hours of interview, she said that she had an uncle in the Zimbabwean national army. He was contacted and she was released. As she left, she could hear the shouts and groans of two other deportees. In paragraph 205 of its determination, the Tribunal said that the treatment to which this witness claimed to have been subjected did not amount to serious ill-treatment such as to engage Article 3. We have difficulty understanding why not. We agree that trivial violence to an interviewee might not engage Article 3. But Mr Nicol pointed to what was said by the European Court of Human Rights in Ribbitsch v Austria (1995) 21 EHRR 573 at paragraph 38 about injuries deliberately inflicted on a person in police custody, as follows:

“The court emphasises that, in respect of a person deprived of his liberty, any recourse to physical force which has not been made strictly necessary by his own conduct diminishes human dignity and is in principle an infringement of the rights set forth in Article 3 (art 3) of the Convention.”

The brief account in relation to R31 does not give us the impression that, properly considered and in context, the violence was trivial. In saying this, we take account of Mr Kovats’ submission that the physical violence in Ribbitsch was gross and that the issue was whether the injuries had been sustained accidentally; and that other cases in which the court has found physical mistreatment of those in detention by state officials went well beyond that complained of in the case of R31. Mr Kovats referred in Ireland v United Kingdom (1978) 2 EHRR 25, paragraphs 92-130; Tomasi v France (1992) 15 EHRR 1, paragraphs 105-115; Selmouni v France (2000) 29 EHRR 403, paragraphs 82-89; and Balogh v Hungary, [2004] ECHR 361, application 47940/99 (20 July 2004), paragraphs 10 and 45-46. The Tribunal went on to point out that the news report was clearly intended to fulfil a journalistic point, and that R31 could not be identified, so the Secretary of State was unable to test the account in any way. That certainly affects the weight to be given to the evidence. But, unless the account was to be wholly disregarded, the point in relation to the Tribunal’s perception of the threshold for Article 3 violence for this returnee did need to be addressed.

For these reasons we allow the appeal on ground 1 and order reconsideration by the same Tribunal as and to the extent that we have indicated. It is not necessary to proceed to ground 2, nor, we think, to make a specific order on ground 5. Permission to appeal is refused on grounds 3 and 4.

Finally, we note that Mr Kovats deliberately did not argue that, because AA could return voluntarily to Zimbabwe without risk and is therefore in no position to invoke the Refugee Convention, the Secretary of State is entitled to enforce his return even if that would put him at risk. He reserved the right to make that submission elsewhere.