NA (Sudan) v The Secretary of State for the Home Department

The general principles to be applied in considering whether the return of an asylum-seeker to Italy would constitute a breach of his or her rights under article 3 are not in doubt. They can be summarised as follows:

The last of those points does not of course mean that evidence about problems with the system for dealing with asylum-seekers and BIPs in the receiving state is irrelevant. Applicants will not be able to show that there is a real risk of their suffering inhuman or degrading treatment if returned merely by showing that there have been some particular cases of such treatment: they will in practice need to show that such cases are widespread. Proving that will often, or indeed generally, involve proving that there are systemic failings, or “major operational problems”, in the arrangements in the receiving country; and references to such failings or problems are found frequently both in the decision of the CJEU in NS (Afghanistan) and in numerous decisions of the ECtHR. The error of the Court of Appeal in EM (Eritrea) was only that it treated the proof of a “systemic” deficiency as a legal pre-condition for a breach of article 3 instead of a means by which a risk of such a breach might be established. That is clearly acknowledged in the judgment of Lord Kerr in the Supreme Court: see particularly para. 63.

Although the decision in MSS was concerned with an asylum-seeker, it is in my view clear that the relevant principles apply equally in the case of Dublin returnees who have already been granted international protection. In both Hussein and AME the applicants were BIPs, but the Court adopted the same approach in principle as in MSS: see, for example, para. 75 of the judgment in Hussein, finding no breach of article 3 in respect of the applicant’s treatment “either as an asylum seeker or as an alien having been accepted as a person in need of international protection”. That is consistent with Lord Kerr’s approach in EM (Eritrea) (see para. 85 above) and with what one would in any event expect. The special needs of refugees do not necessarily disappear at the moment that they are granted asylum, and those who were peculiarly vulnerable before the grant of asylum may remain peculiarly vulnerable thereafter. Access to integration facilities of the kind required by article 25 of the Qualification Directive, including a limited period of free accommodation, may – depending on the circumstances of the case – be essential if they are to avoid falling into circumstances sufficiently degrading to constitute a breach of article 3. I must emphasise that I am referring only to the question of principle – that is, whether it is open to a BIP to advance an “MSS-type” claim at all. I am not saying that the actual situations of asylum-seekers and BIPs are identical. On the contrary, it is clear that they are not, which is the point being made by the ECtHR at para. 179 of its judgment in Hassan (para. 80 above). On the whole, because of the more extensive rights enjoyed by BIPs it may be reasonable to regard them as being at lesser risk of suffering inhumane or degrading treatment than asylum-seekers; but that may not always be so, and it is necessary to look at the actual circumstances on the ground in each case.

So far so good, but the more difficult question is what light the authorities cast on how those principles apply in the case of returns to Italy. The Strasbourg cases are not formally authoritative on that question. It is for the domestic courts to decide on the application of the relevant principles to the facts of the particular case before them, and on the basis of the evidence adduced. However, how the ECtHR has applied the principles which it has itself enunciated to the particular cases before it gives valuable guidance as to the content and interpretation of those principles; and in a case where the facts and evidence are very similar the domestic decision-taker (whether that is a tribunal or the Secretary of State in forming a view as to what a tribunal would decide) is entitled to take its lead from the Strasbourg court. On the specific question of whether it is legitimate to return asylum-seekers to Italy it is indeed highly desirable that the ECtHR and the domestic tribunals adopt a consistent approach, provided always (a) that that approach is founded on good evidence and (b) that decision-makers recognise that the facts of a particular case, or evidence of significant changes in the situation in Italy, may require a departure from it. I do not believe that any of that is contentious. Both parties urged us to follow the Strasbourg decisions (or some of them): where they differed was about where those cases led.

The starting-point in considering the effect of the Strasbourg decisions must be the conclusion, reiterated in all the cases including Tarakhel (para. 114), that the situation in Italy is in no way comparable to that in Greece and that a general ban on returns to Italy cannot be justified. That conclusion was reached after consideration of all the reports considered at paras. 64-66 above.

However, Tarakhel decides that the problems about the availability of accommodation highlighted in the reports mean that at least in a case of the kind which was before the Court, namely that of a family with minor children, returns to Italy will not be lawful in the absence of specific assurances that suitable accommodation will be provided. The crucial question is whether that conclusion is limited to cases of that specific kind or whether it follows from the Court’s reasoning that the obligation to seek assurances arises in other cases also. That question was explicitly raised by the dissentients, who indeed raised the possibility that the reasoning of the majority entailed that (since it had been said in MSS that asylum-seekers as a class are particularly vulnerable) such assurances would have to be sought in every case of the return of an asylum-seeker to Italy. But a less extreme reading would be that the Court’s reasoning extends at least to the case of all other asylum-seekers who are peculiarly vulnerable, such as victims of rape or torture or those with mental health problems, so that returns of such asylum-seekers to Italy would be in breach of article 3 in the absence of specific assurances that they would be suitably accommodated.

Although in his original skeleton argument Mr Husain advanced both those possible readings, in the supplementary skeleton and his oral submissions he focused on the latter. He reminded us in particular that the Reception Directive required special provision to be made for “vulnerable persons such as minors, unaccompanied minors, disabled people, elderly people, pregnant women, single parents with minor children, victims of human trafficking, persons with serious illnesses, persons with mental disorders and persons who have been subjected to torture, rape or other serious forms of psychological, physical or sexual violence” (see para. 44 above). Since the basis of the Court’s conclusion was the risk of accommodation, or suitable accommodation, not being available, and that risk would impact on all peculiarly vulnerable persons, and not just children, the reasoning must extend to the entire class. The distinction which the Court drew between the situation in Greece and that in Italy was only that in the case of a return to Greece there would necessarily and in all cases be a real risk of inhuman and degrading treatment while in the case of Italy that risk could be removed by the giving of sufficiently specific assurances.

I do not accept Mr Husain’s submissions on this question. In my view it is wrong to treat the reasoning of the majority in Tarakhel as going any further than the particular kind of case with which the Court was concerned – that is to say, families with minor children. There is some indication to that effect in the judgment itself, but the position is confirmed by the subsequent authorities. I take them in turn.

So far as the judgment itself is concerned, para. 119 is a central part of the Court’s chain of reasoning. It refers to the “specific needs and extreme vulnerability” of children: that picks up on para. 99, which records the special protection accorded to asylum-seekers who are minors by reason of their specific needs. It says that it follows that they must have “reception conditions … adapted to their age”. The immediately following paragraphs, which address the circumstances of the particular case, are founded on that point. The conclusion in paras. 120 and 122 is that the Swiss authorities must obtain assurances that the applicants would be received in facilities “adapted to the age of the children, and that the family will be kept together”: the element of the family being kept together was evidently particularly important to the Court. The passage as a whole is very specifically directed to the position of children. The Court was explicitly referred to the 2013 decisions. Its decision would seem to represent a departure from some of them, since they included cases, one being Hussein, where the applicants included a child. But although the dissentients suggest that the departure was fundamental in character, there is no suggestion that the majority believed that that was the case, as opposed to applying the relevant principles (in connection with which it referred explicitly to Hussein) to a particular situation and on the particular evidence adduced. The Court’s analysis does not refer at any point to the provisions of Chapter IV of the Reception Directive or to any general concept of peculiar vulnerability.

As for the subsequent cases, the applicants in both AS and AM were peculiarly vulnerable inasmuch as they both suffered from PTSD and were at risk of suicide. If Tarakhel were authority for the proposition that specific assurances of suitable accommodation should be obtained in the case of all returnees who were peculiarly vulnerable both their claims should have succeeded.

I also regard it as of some significance that in the most recent cases referred to at para. 104 above the Court has departed from the position that specific assurances are required in the case of the return of a family with children and has held that a more general assurance about the availability of accommodation, as there recorded, was sufficient. That change of position cannot be read across directly to the present case, because the Italian authorities have not offered even that more generalised kind of assurance for vulnerable persons other than children. But the cases in question are a useful reminder of the nature of the relevant exercise. What is required in each case is an assessment of the risk that the Italian authorities will be unable comply with their article 3 obligations; and in a case where the evidence showed that the Italian authorities had adapted their system to reinforce the protection of particularly vulnerable returnees the Court was prepared to proceed on the basis that the system would work even in the absence of specific assurances.

I of course understand the argument that if, as the Court believed in Tarakhel, there is a real risk that there will be insufficient accommodation to meet the demand for suitable accommodation for families it must follow that there will also be insufficient accommodation for other classes of vulnerable returnees who need equivalent accommodation. But I think it is clear, for the reasons given, that the Court has not gone that far. That is not necessarily illogical. The problems of accommodating a family together (particularly a family of eight), in child-friendly accommodation, might reasonably be thought to be substantially greater than those of finding suitable accommodation for a single vulnerable person, and the risks of failing to do so might also be thought to be greater; so that the fact that specific assurances were required in the former case did not necessarily mean that they were required in the latter.

In short, therefore, reading the Strasbourg authorities as a whole, I believe that the message of the cases to date is that the evidence to which the ECtHR has been referred (including and in particular the evidence about the shortage of accommodation in CARAs and SPRARs) is not sufficient generally to undermine the presumption that Italy will comply with its obligations under article 3 and the relevant EU Directives, the only exception being for the case of families with minor children. It is important to emphasise that I am at this point concerned only with the question of how far the case-law has gone. I am not to be taken as saying that it is not open to us in this case, on the basis of the evidence before us, to reach an equivalent decision to that in Tarakhel in the Appellants’ cases: I am saying only that the reasoning in the cases to date, and in particular in Tarakhel, does not point towards such a decision.

It follows also that the 2013 decisions remain a legitimate source of guidance except to the extent that they concern families with minor children, and that, contrary to a submission made by Mr Husain, the post-Tarakhel decisions in AME, AS and AM are not inconsistent with Tarakhel itself. (That renders academic a point emphasised by him, which is not in itself controversial, that decisions of the Grand Chamber are inherently more authoritative than decisions of individual chambers.)

THE DECISION OF LEWIS J IN MS

At para. 99 of his judgment Lewis J summarised the claimants’ case that they were at risk of a breach of their article 3 rights if returned to Italy as follows:

“The Claimants presented their cases on the following basis. All the Claimants have significant vulnerabilities. All three have severe mental health problems. Two, NA and SG, are victims of repeated rapes (some of the rapes in each case occurring whilst they were in Italy). In the light of the Tarakhel judgment, the Claimants submit that, on one legitimate view, a First-tier Tribunal hearing an appeal against the Defendant's rejection of their human rights claim could conclude that (1) there was a glaring discrepancy between the number of asylum applications and the places available at relevant facilities in Italy (2) the three Claimants would be among a significant number of asylum seekers who may be left without accommodation or accommodated in overcrowded facilities or in unsalubrious conditions and (3) the three Claimants all have mental health problems and suicidal ideation which could not be properly managed.”

The case was accordingly essentially about the availability of accommodation. More particularly, it was about the availability of suitable accommodation for vulnerable Dublin returnees. Most such accommodation would be in SPRARs. (Footnote: ¹⁰) In the cases of MS and NA, who are BIPs, SPRARs are the only form of officially-provided accommodation potentially available; but, as we have seen, SPRARs are also used to accommodate peculiarly vulnerable asylum-seekers such as SG claimed to be.

However, as will appear, the claimants’ evidence also referred to practical problems facing returnees in obtaining access to the system at the point of return so as to be identified as peculiarly vulnerable and have the chance of being offered SPRAR accommodation, and specifically to the form of reception at the airports to which they were returned, which would in practice be either Rome (Fiumicino) or Milan (Malpensa). It is not disputed that the Italian government funds services run by NGOs at both airports to offer support and advice to asylum-seekers, including Dublin returnees; but it was the claimants’ evidence that the services of the airport NGOs were not available to BIPs, who would accordingly, after going through the usual formalities administered by the border police, simply be left to their own devices and have no reliable means of accessing SPRAR accommodation even if it were available.

Each of the claimants gave a witness statement describing his or her own experiences, both in their country of origin and in Italy prior to coming to England. I have given a summary of NA’s evidence at paras. 10-13 above. As for MS and SG, MS was homeless and destitute for a substantial period and was unable to access healthcare despite being ill, and SG lived partly on the streets and partly in a squat and was the victim of rape on three occasions. Both had, as already noted, severe mental health problems.

The Claimants commissioned reports from four experts, or expert bodies, addressing reception conditions and living conditions for asylum-seekers and BIPs in Italy, with a focus on Dublin returnees. These were: Medici per I Diritti Umani (“MEDU”), a group working with asylum-seekers and BIPs in Florence; Loredana Leo, a lawyer specialising in asylum cases; Dr Chiara Marchetti, an academic at Milan University specialising in asylum law; and Amnesty International. Broadly, all attested to deficiencies in the provision of accommodation and support for Dublin returnees: I will give more detail when considering the grounds of appeal. In addition the claimants relied on a number of published reports from the international institutions and NGOs, most but not all of which had already featured in the case-law. I refer to all this material compendiously as “the reports”.

The claimants also provided a somewhat miscellaneous mass of press cuttings and short reports relating to conditions for migrants in Italy, accompanied by a 25-page “Schedule of Key Passages”. Lewis J was critical of the way in which this material was presented and noted that he had not been referred to any of it in oral submissions.

The Respondent relied on a witness statement dated 6 March 2015 from Mr Carl Dangerfield, a civil servant from the Home Office working with the Ministry of the Interior in Rome as liaison officer to the team dealing with Dublin returns. Mr Dangerfield had given evidence in the EM and Tabrizagh cases, and those statements were also relied on. The following points in his witness statements are particularly relevant for our purposes.

Mr Dangerfield gives figures, derived from Ministry of Interior statistics (which he exhibits), about the numbers of asylum-seekers in Italy and the numbers accommodated in government-provided accommodation. I discuss these at paras. 162-172 below.

In his witness statement for the Tabrizagh proceedings dated 13 May 2014 Mr Dangerfield explains that whenever an asylum-seeker is returned to Italy the UK authorities inform the Italian authorities of the travel arrangements. At that stage they will advise them if the returnee is particularly vulnerable or has particular needs (para. 18). In the case of returnees with health needs they will supply a copy of their medical records, provided they consent (para. 19). Mr Dangerfield says that if the information provided suggests that this is necessary medical staff will be present at the point when the ordinary verification procedures are performed by the border police. He says that following verification the police will refer the returnee to the airport NGO for assistance (para. 20). That evidence is of course consistent with what was said in the most recent decision letter in NA’s case: see para. 17 above.

In his witness statement for these proceedings Mr Dangerfield gives substantially the same evidence though in a slightly different form. He appears to say in the case of each category of returnee that they will be “medically assessed” at the airport (see paras. 20, 21, 23 and 24), but it may be that he means only that the NGOs will conduct a medical assessment where the UK authorities have provided prior notice of health problems or where they identify them for themselves: he deals with the latter category at para. 25 (making no distinction between BIPs and others). The point does not matter for our purposes because in the cases of both NA and MR the Respondent says that she will notify the Italian authorities of their health problems. In such a case he says that following the assessment the NGO will transfer the returnee either to a hospital or to “an appropriate accommodation centre”, where they will receive a further medical assessment.

As regards accommodation, Mr Dangerfield says in his witness statement for these proceedings that asylum-seekers will be entitled to CARA accommodation until their claim is decided, at which point (if it is granted) they will be entitled to SPRAR accommodation (para. 20). He says that BIPs will be entitled to SPRAR accommodation unless they have already been in such accommodation and that even in that case they will still be entitled to it if they are vulnerable (which must mean, in the terminology adopted above, peculiarly vulnerable). He says that in his experience “returnees with refugee status who are assessed to be vulnerable are always accommodated in some form, usually in a SPRAR” (para. 23). He says at para. 25 that certain accommodation centres have special provision for dealing with people with special needs, such as those suffering from PTSD, and the authorities will endeavour to arrange for returnees with such needs to be accommodated accordingly.

Mr Husain submitted that little weight should be given to Mr Dangerfield’s evidence because he does not state the sources of his information. I do not accept that. It is self-evident that as an official working in the Dublin unit in Rome he was very well-placed to acquire a good understanding of the workings of the relevant aspects of the Italian system.

The European Council on Refugees and Exiles (“ECRE”), an NGO partly funded by UNHCR and the EU, in partnership with three other NGOs, has established the Asylum Information Database (“AIDA”), which aims to provide authoritative and up-to-date information on asylum practice in a number of European countries, including Italy. AIDA’s first country report on Italy was published in May 2013; updated versions were published thereafter. The Respondent included the April 2014 version of the AIDA report on Italy in her bundle before Lewis J, but a version from March 2015 was submitted in the course of, or shortly after, the hearing, apparently by the claimants. The latter, to which I will refer as “the 2015 report”, stated the position as at the end of January. We were referred to passages concerning reception arrangements for Dublin returnees and SPRAR accommodation.

As regards reception arrangements, the report says that “once asylum-seekers arrive at the airport … they are assisted by a specific NGO and referred to the reception centre, on the basis of the individual situation (vulnerable or ‘ordinary’ categories)” (p. 31). The way the report is drafted suggests that the reference to “asylum-seekers” is meant to cover BIPs, but that cannot be said to be entirely clear.

As for availability of accommodation, Mr Husain asked us to note that the report says (at p. 53) that “the main negative aspect of the Italian reception system consists in the reduced number of places and structures that are not sufficient to provide to all asylum seekers in need an adequate reception”. It is necessary to look, however, at what it says specifically about the SPRAR accommodation. As to that, it notes the rapid expansion of SPRAR places, giving a figure of 19,900 as at the end of 2014 and noting a very substantial increase in the allocation of government funding (p. 53). But it makes clear that some of that extra capacity is taken up by asylum-seekers rather than BIPs. In the context of Dublin returnees it refers to the number of SPRAR places available as “limited” (p. 31), though the problem is to some extent mitigated by the ERF-funded accommodation, which it describes as available to “either all the Dublin returnees or to vulnerable categories among Dubliners” (loc. cit.): Mr Husain submits that there is a doubt whether the ERF accommodation is in fact available to BIPs as opposed to asylum-seekers, but it is not necessary to resolve the point. Beyond the reference to SPRAR accommodation being limited, AIDA gives no specific figures: in particular, nothing is said about waiting-times. It does refer to accommodation in SPRARs being dependent “on the availability of places and the urgency and vulnerability of the case” (p. 53), which suggests that an element of prioritisation occurs. It refers to the authorities sometimes filling centres beyond their official capacity and also to BIPs and asylum-seekers who are unable to access accommodation having to live in settlements such as the notorious “Selam Palace” in Rome where conditions are very poor.

I have referred to the AIDA report in some detail, even though the Judge did not (apparently because the claimants did not themselves rely on it), because both parties took us to it in their oral submissions, and Ms Giovannetti referred to it as being – like the UNHCR reports – of a more objective character than reports produced by campaigning bodies or for the purpose of litigation. But it does not significantly alter the picture appearing from the earlier evidence. The impression given is of a system which is operating under great pressure but which the authorities are taking active steps to improve.

Some further material (in addition to the AIDA January 2015 report) was submitted either during or shortly after the hearing. I need only refer to further witness statements from Mr Dangerfield and Ms Leo (with a colleague), and a letter from MEDU, about the applicable procedure where the residence permit of a returning BIP had expired or been lost, to which Ms Leo had referred in an earlier statement. The point appears to have arisen because MS’s permit had expired and NA had said (though not, so far as I can see, in either of her witness statements) that she had lost hers. It was not suggested before us that any difficulties which NA might encounter in this regard were important in considering the risk of an article 3 breach. (Footnote: ¹¹)

At paras. 1-70 Lewis J sets out the facts and the relevant background law. At paras. 71-91 he reviews the relevant case-law about article 3, focusing, so far as the Strasbourg cases are concerned, on MSS, Hussein, Daytbegova and Tarakhel and on EM (Eritrea). He states his view of the effect of those authorities at para. 91 as follows:

“… [I]n my judgment, the approach set out by Lord Kerr in EM (Eritrea) does not differ in substance from that adopted by the European Court of Human Rights in its case law, including the decision in Tarakhel v Switzerland. In both cases, the legal test is that identified in Soering v United Kingdom (1989) 11 EHRR 439. As is to be expected, different courts express themselves in different language, reflecting their differing judicial traditions. But in substance, the approach adopted by the Supreme Court involved recognising that Member States can be presumed to be complying with their obligations under EU law and international law. That can be rebutted in one of two ways. First, it may be rebutted in the case of all asylum seekers if there were sufficient evidence of substantial operational difficulties in the receiving state. Secondly, while the presumption would be the backcloth for considering individual cases, there may be situations where, on the individual facts of the case viewed against the overall situation (even if that situation did not rebut the presumption in all cases), there were still substantial grounds for believing that there would be a real risk on return. The European Court of Human Rights has taken a similar approach. In MSS v Greece, for example, it effectively considered that the circumstances were such that the assumption that an EU Member State was complying with its obligation was rebutted. In Tarakhel the overall situation did not lead to the conclusion that all removals to Italy needed to be stopped. Nevertheless, the situation of the individual applicants needed to be assessed against the overall situation to determine if, on the facts, there was such a breach.”

At para. 98 the Judge identifies the issues that he has to decide. The first and second issues are as follows:

“(1)

On one legitimate view, could a tribunal conclude that the evidential presumption that Italy will comply with its obligations under EU law and Article 3 ECHR had been rebutted in relation to:

(a)

all asylum-seekers and/ or all BIPs; or at least

(b)

in relation to vulnerable asylum-seekers and/or BIPs?

(2)

Considering the individual facts of each Claimant against the overall situation in Italy, could a tribunal, on one legitimate view, conclude that any of the three Claimants had shown that there were substantial grounds for believing that that individual faced a real risk of being exposed to treatment contrary to Article 3 ECHR if returned to Italy?”

I need not set out the third issue, which concerns the risk of suicide during the removal process itself, because there is no challenge to the Judge’s finding that standard Home Office practice governing the removal of returnees who pose a suicide risk would prevent the risk eventuating.

The distinction between the two issues identified by Lewis J – the first being concerned with whether the presumption of compliance had been displaced generally as regards Italy (or at least for particular classes of returnee), and the second with the risk in the individual cases – is unobjectionable as a working framework, particularly in the light of how the case-law has developed; and I dare say it reflected the way the case was put before him. But it is important to appreciate that there is no necessity for a two-stage approach of this kind. There is no sharp distinction between cases that depend on the general situation in the country of return, to be assessed by reference to whether the presumption of compliance has been rebutted, and cases that depend on some “individual facts” peculiar to the particular returnee. The ultimate issue is always whether the individual applicant has shown a real risk that if returned he or she will suffer a breach of their article 3 rights: the role of the presumption of compliance is to condition the assessment of that issue. In all cases the assessment will involve considering both the characteristics of the individual returnees and the general situation in the country of return so as to assess whether a returnee with those characteristics is at real risk of being subjected to inhumane and degrading treatment.

The first issue is considered at paras. 99-145 of the judgment. I have already set out how the Judge summarised the claimants’ case: see para. 121 above. He proceeds to address that case under three headings – the case-law, the statistics about accommodation for asylum-seekers and BIPs, and “the other reports and material relied upon”. I take those headings in turn.

As regards the case-law, the Judge says this:

“102.

First, any tribunal would need to have in mind the domestic case law and the case law of the European Court of Human Rights. Elisabeth Laing J. considered the statistical picture, and the evidence available, as at May 2014 in Tabrizagh. The judge concluded that, on a detailed consideration of the evidence available the evidential presumption that Italy would comply with its obligations if asylum seekers and BIPs were returned to Italy had not been rebutted.

103.

Secondly, two decisions of the European Court of Human Rights have expressly considered the position in relation to the return to Italy of vulnerable adults with severe mental health conditions, those judgments being delivered in 2013. The Court found that the claims in those cases were manifestly inadmissible. That must implicitly include a finding both that the situation in Italy was not such that, generally, the return of asylum-seekers to Italy would be incompatible with Article 3 ECHR and also that the return of the particular individuals in those cases, given their situations, would not be a breach: see Hussein v Netherlands and Italy application no. 27725/10, BAILII: [2013] ECHR 1341 and Daytbegova v Austria (2013) 57 EHRR SE12.

104.

Thirdly, the European Court of Human Rights in Tarakhel did not consider that the situation in Italy generally was comparable to Greece. At paragraph 115 of its judgment, the Court stated expressly that "the structure and overall situation of the reception arrangements in Italy cannot therefore in themselves act as a bar to all removals of asylum seekers". That is consistent with a finding that the evidential presumption has not, generally, been rebutted.

105.

Against that background, there would need to be some proper evidential basis before a tribunal could legitimately conclude that the situation in Italy had changed so that the evidential presumption had been rebutted. The Claimants rely upon the most recent statistical evidence and reports and evidential material.”

As to the alleged shortfall in accommodation, the Judge addressed the statistics at paras. 106-115. His conclusion was that the statistics relied on by the claimants were insufficient to displace the presumption that Italy would comply with its obligations under article 3. I consider the details of his reasoning at paras. 162-172 below.

The Judge considered the other material relied on by the claimants at paras. 116-135. Again, he concluded that none of it displaced the presumption of compliance. I return to this aspect at paras. 172-209 and 213-221 below.

Finally, at paras. 136-149 the Judge addresses the significance of the decision in Tarakhel. In short, he reaches the same conclusion as I have done at paras. 112-119 above, saying at para. 138 that the Court’s decision in the case of a family with children “does not suggest, or lead to the conclusion, that similar specific assurances must be sought in relation to other potentially vulnerable asylum seekers or BIPs such as those who have experienced trauma and have mental health difficulties”. As regards the claimants’ physical and mental health, he says:

“140.

… [T]he important consideration in Hussein and Daytbegova was that Italy provided access to health care, including psychological care. What was important was that Italy should provide a ‘well-planned reception’ … . The Court was aware that the Italian authorities would know in advance of the applicants' vulnerability and hoped that the Austrian authorities would provide the medical and psychological documentation available to them.

141.

That is the position here. All the reliable evidence is that the United Kingdom authorities will provide the medical information available to them to the Italian authorities for MS, SG and, if she consents, NA: see the decision letters. The evidence of how the Italians last proposed to effect the transfer of MS and SG confirms that the Italian authorities wanted 10 days advance notification of any health issue, including both physical or psychological issues. For completeness, that conclusion is reinforced by, but not dependent upon, the evidence of Mr Dangerfield, who has 7 years experience of working as the asylum and immigration liaison officer in Rome, and says that medical assessments are carried out on individuals where there are particular issues.”

The Judge summarises his decision on the first issue as follows:

“144.

The presumption that Italy will comply with its obligations under EU and international law is a significant evidential presumption. It is the backcloth against, or context in, which the question of a real risk of a breach of Article 3 ECHR falls to be addressed. In my judgment, on the material produced, there is no legitimate basis upon which a tribunal, properly directing itself, could reasonably conclude that that presumption was rebutted in relation either to asylum seekers or BIPs, or in relation to vulnerable asylum seekers or BIPs, that is, those such as MS who has mental health difficulties or NA and SG who have been the victim of rape and sexual violence, and have mental health difficulties.

145.

That has two consequences. First, this is not a situation such as that which prevailed in MSS v Greece where all removals of asylum seekers or BIPs, or all vulnerable asylum seekers or BIPs, should cease. Secondly, the presumption is that Italy will comply with its obligations, particularly in relation to those vulnerable asylum seekers such as SG, and vulnerable BIPs such as MS and NA. It is against the backdrop of that evidential presumption that consideration will need to be given to whether a tribunal could legitimately conclude that, given the particular circumstances of each of the Claimants viewed in the overall context, there are substantial grounds for believing that they would face a real risk of treatment which breached Article 3 if returned to Italy.”

The Judge states his approach to the second issue, at para. 146, as follows:

“The second issue involves consideration of the position in relation to each individual Claimant, viewed against the overall situation. As indicated, the backdrop against which that is assessed is that there is a significant evidential presumption that Italy will comply with its obligations. That presumption can, as explained by the Supreme Court in EM (Eritrea) be displaced in a particular case. That will involve an assessment of the situation in the receiving country and the foreseeable consequences of returning each Claimant there.”

He then proceeds to consider in turn the cases of MS, NA and SG. I need not summarise how he deals with MS’s case, but his consideration of SG remains relevant because she, like MR, was an asylum-seeker rather than a BIP.

As regards NA, the Judge says, at para. 150:

“NA … is a BIP. She will have access to health care on the same terms as Italian nationals. In terms of integration facilities, NA is eligible to be accommodated in a SPRAR for six months, and, if she is assessed as vulnerable, for a further six months. Given the overall situation in Italy, and the increase in the number of places at SPRARs, and the statistics considered above, there is no basis on which a tribunal could conclude that there is a glaring discrepancy such as would make it unlikely that NA would not be provided with a place in a SPRAR or left homeless or without support. The evidence of difficulties associated with this by reason of the need to replace her residence permit is not capable of constituting reliable evidence and, even if such difficulties did exist, they could not, on any legitimate view, be regarded as establishing substantial grounds for believing that there was a real risk of treatment contrary to Article 3 ECHR if NA were returned for the reasons given above.”

At para. 151 he considers the case based on her mental ill-health and risk of suicide. He finds that when and if arrangements were put in place to return her to Italy the UK government would, unless she continued to refuse her consent, provide the Italian authorities in advance with copies of the medical information in their possession in order to enable them to make any appropriate special arrangements. He refers to the judgment in Daytbegova, where the Italian government confirmed that where it was informed of any particular vulnerability of a Dublin returnee, appropriate medical measures would be taken – see para. 68, quoted at para. 76 above. He says that “there is no evidential basis upon which a tribunal could conclude that the Italian authorities would be unable to make effective arrangements to deal with [NA’s health issues]”. Specifically in relation to the suicide risk he concludes that “no tribunal could legitimately conclude that there were substantial grounds for believing that there was a real risk of a breach of Article 3 because of the risk of suicide on return”. Para. 152 reads:

“In terms of the risk of re-victimisation, and exposure to rape or further sexual violence, Mr Husain accepted in argument that if accommodated on return, rather than being homeless, that risk would be removed. For the reasons given, no tribunal could legitimately consider that NA would be left homeless on return.”

(It is convenient to note at this stage that Mr Husain did not seek to resile from that acceptance before us.) The Judge concludes, at para. 153:

“For all those reasons, in my judgment, there is no legitimate basis upon which a tribunal, properly directing itself, could conclude that, if NA were returned to Italy, there are substantial grounds for believing that there is a real risk that the conditions in Italy would be such as to amount to a breach of Article 3 ECHR.”

The Judge’s reasoning and conclusions as regards SG, at paras. 153-158, are essentially similar. He notes her peculiar vulnerability, as a result of her mental ill-health and history of rape, but he says that full information about that vulnerability will be given to the Italian authorities on her return. He refers to their obligations under the Reception Directive to provide her with appropriate treatment and support, and he finds that there is no basis for a conclusion that they will not do so. He refers, as in NA’s case, to Hussein and Daytbegova. He summarises his conclusion in her case in the same terms as para. 153 in NA’s case.

THE GROUNDS OF APPEAL IN NA’s CASE

As noted above, initially all three claimants appealed against Lewis J’s decision. Five grounds of appeal are pleaded, but they fall into three groups and are indeed so treated in Mr Husain’s original skeleton argument. They can be summarised as follows:

Essentially the same grounds are relied on by MR. But it will be more convenient if I start by considering NA’s appeal on its own and then consider MR’s appeal in the light of my conclusions in her case.

GROUND 1: THE EFFECT OF TARAKHEL

Mr Husain in his skeleton arguments refers to four errors in Lewis J’s approach to the decision in Tarakhel, but on analysis there is an element of overlap between them, and at least one other point is made which does not appear to correspond to any of the four errors. I address the substance of the points made as follows.

Mr Husain submits that the Judge was wrong to treat the reasoning in Tarakhel as confined to the case of families with children and as not extending to the case of vulnerable persons generally. For the reasons given at paras. 112-119 above I do not accept that submission.

Mr Husain’s other principal point is that the Judge was wrong to “focus on the question whether an evidential presumption had been rebutted rather than on whether an arguable article 3 case had been shown”. In his submission the message of paras. 103-105 of the judgment in Tarakhel is that “the presumption of compliance is rebutted without more where there is evidence of a real risk of harm” and that accordingly the only question is whether such a risk had in fact been demonstrated.

I have no difficulty with the premise of Mr Husain’s submission, namely that the ultimate question in any case is whether there is a real risk, having regard to the practical realities, that if the asylum-seeker is returned his or her article 3 rights will be breached; and that asking whether or to what extent “the presumption has been rebutted” as regards a particular country should not be allowed to distract from that question. I can see that the way that the Judge framed the issues could have led to his losing focus on the essential question in that way: see my observations at para. 140 above. But I do not believe that it in fact did so. He framed the second issue squarely in terms of the Soering test and answers it in those terms in the cases of the individual claimants (see, as regards NA, para. 153, quoted at para. 148 above).

In so far as Mr Husain’s submission is that the reasoning of the ECtHR meant that the presumption of compliance had no application in the case of Italy, I do not accept that. In answering the correct, Soering, question the Judge was entitled – indeed positively obliged – to take as his “backdrop”, or starting-point, the expectation that Italy would meet its obligations under the Convention: Lord Kerr says so in terms at paras. 64 and 68 of his judgment in EM (Eritrea) – see para. 62 above. That is a “significant evidential presumption” not for some reason of theory but because it is indeed reasonable as a matter of practical reality to believe that a member state of the Council of Europe will comply with its obligations under the Convention unless the evidence demonstrates that there is a real risk that it will not. I see no sign that the ECtHR in Tarakhel intended to depart from that approach. On the contrary, in the very paragraphs on which Mr Husain relies it uses the language of a rebuttable presumption and refers specifically to its own previous decision in MSS acknowledging the existence of such a presumption and to the decision of the CJEU in NS (Afghanistan), which it discusses at length (see para. 94).

Mr Husain also submits that Lewis J adopted a “binary approach, whereby Italy is totally and irremediably unsafe or presumed to be compliant in all respects”, ignoring the possibility of any more nuanced position. I can see no sign in the judgment that that was his approach.

GROUNDS 2-4: LEWIS J’s ASSESSMENT OF THE EVIDENCE

The essence of the factual case advanced by NA was (a) that her vulnerability, as a single woman and a rape victim, with serious mental health problems, is such that her rights under article 3 would be infringed if on her return to Italy she was unable to access suitable accommodation and support (in practice that meant accommodation in a SPRAR or equivalent (Footnote: ¹²)); and (b) that on the evidence if she were returned to Italy there was a serious risk – to put it no higher – that she would not be provided with such accommodation, primarily because there was an inadequate supply but also because of practical or procedural difficulties that she was likely to encounter in accessing it.

Since this is a certification case I do not think that element (a) in that case can be challenged, and I did not understand Ms Giovannetti to be seeking to do so. The essential issue for the Judge was thus simply whether there was a serious risk that if she were returned NA would not be accommodated in a SPRAR: this subsumes the question whether she would receive proper healthcare, and specifically psychiatric care, since it is not argued that even if she were in a SPRAR her needs would not be attended to. The only evidence that was material, at least in her case, was evidence that bore, directly or indirectly, on that question.

Ms Giovannetti reminded us that as a matter of principle the only right of a BIP or an asylum-seeker under the Convention is not to be subjected to inhumane and degrading treatment. Although the terms of the Reception Directive were relevant in the way identified in MSS (see para. 55 above), article 3 does not import an obligation to comply with its terms as such, and still less with those of article 25 of the Qualification Directive (integration facilities). Long-term homelessness in conditions of the kind described in MSS would no doubt constitute such inhumane and degrading treatment but short-term homelessness need not, nor would a situation where she was obliged to live in accommodation which was merely sub-standard. I accept all that, but I do not think that it is relevant in NA’s case, for the reason already given, namely that she is (at least arguably) sufficiently vulnerable to require not only accommodation but SPRAR accommodation. Ms Giovannetti’s point is, however, a useful reminder of the “minimum level of severity” qualification. It would not in my view even arguably be a breach of article 3 if there were, for example, a short delay in SPRAR accommodation being found for her or if the quality of such accommodation and the associated support were merely sub-optimal.

The evidence relating to the prospects of NA accessing SPRAR accommodation was of various kinds and the Judge is said to have erred in his treatment of each. I take them in turn.

THE STATISTICS

It was an important part of the claimants’ case that the published figures showed a drastic shortfall between the numbers of migrants (Footnote: ¹³) requiring accommodation and the places that were in fact available, with the result that there was inevitably a grave risk that they would not be found suitable accommodation, or indeed any accommodation, if returned: they calculated the shortfall in overall accommodation at between 60,000 and 90,000 places. The Respondent did not accept that the figures showed any such shortfall.

Lewis J did not accept the claimants’ case on this point. His reasoning can be summarised as follows:

His conclusion, at para. 115, was as follows:

“Ultimately, … in my judgment, the court is not able to resolve precisely the amount of demand against provision for the reasons given. An attempt at absolute precision is likely to be futile. The following conclusions are, however, clear and any tribunal would be bound to accept them. First, the amount of provision made by the Italian authorities has increased over the period up to February 2015. Secondly, when the Italian authorities faced a crisis in 2011 with a large influx of migrants, they ‘responded to the extraordinary strains placed upon the system by creating extra accommodation places’ (per Elisabeth Laing J. in Tabrizagh at paragraph 72). Thirdly, the figures of possible shortfalls presented by the Claimants, and which are said to show a glaring discrepancy between demand and the provision of places are unrealistic and could not reasonably be taken by any tribunal as representing the likely position. Fourthly, on any reasonable analysis, no tribunal could consider that the current official figures demonstrate a shortfall due to the absence of key figures (the number of persons coming to Italy who claim asylum and accommodation and the number of BIPs who ceased to occupy a place at a SPRAR during 2014 or by the end of February 2015). Consequently, having regard to all those matters, no tribunal could reasonably conclude on the statistical evidence that the Italian authorities were leaving persons who claimed asylum, or who were BIPs entitled to some form of integration programme, in a situation of homelessness. A tribunal could not, in my judgment, reasonably conclude on the evidence available that there was a glaring discrepancy between demand for facilities and provision and it could not conclude, on the figures, that there were omissions to provide facilities on a widespread or substantial scale, amounting to substantial operational problems, sufficient to displace the presumption of compliance.”

Before considering Mr Husain’s challenge to that reasoning I should note that a comparison between overall supply and demand of the kind which the Judge was asked to undertake is of very limited value in resolving the specific issue in NA’s case, which depends on the availability of SPRAR accommodation. A shortage of places in CDAs or CARAs would be immaterial if there were sufficient places in SPRARs for BIPs and for the minority of peculiarly vulnerable asylum-seekers who might also be accommodated there. I am not to be taken to be criticising the Judge for entering into the exercise: he was asked to do so, and it may well have been more relevant to the case of SG. But I am sceptical of its relevance to NA’s case. I will nevertheless address the challenges to it.

By ground 2 of the Appellant’s Notice Mr Husain submitted that the Judge’s reasoning was flawed on what, on analysis, comes down to two grounds. I take them in turn.

First, he submitted that the ECtHR had in Tarakhel “placed the onus on the member states, rather than on the individual who asserted a risk, to displace the doubts raised by controversial and incomplete statistics”, and that since the Judge himself found that the statistics were disputed and incomplete he was bound to find that there was a risk that SPRAR accommodation would not be available for NA, as the Court had done in Tarakhel itself.

I do not accept that submission. I do not believe that Mr Husain’s characterisation of the reasoning in Tarakhel is correct. In finding that there was a risk that the applicants would not be properly accommodated the Court did not rely on any question of onus but on the facts (a) that it was common ground that the applicants would have to be accommodated in a SPRAR and (b) that there was in mid-2013 a “glaring discrepancy” between the number of asylum-seekers and the number of SPRAR places available: see para. 110 (Footnote: ¹⁴). The question for Lewis J was whether the claimants had adduced evidence about the shortage of accommodation sufficient to show the same risk in the case before him.

Mr Husain sought to reinforce his submission by reference to the decision of this court in Gashi v Secretary of State for the Home Department [1999] Imm AR 415. In that case a Kosovar asylum-seeker was resisting being removed to Germany under the Dublin Convention on the basis that if returned he was at risk of refoulement to Yugoslavia in breach of his rights under article 3. Part of his case depended on a remarkable discrepancy between the rate of grants of asylum to Kosovars in the UK and in Germany which it was said should have put the Secretary of State on enquiry as to whether unlawful refoulement was a real risk. The Court accepted that case. Buxton LJ, giving the leading judgment, said at p. 429 that the statistics called for an explanation and that it was for the Secretary of State to investigate the reasons for the discrepancy before deciding whether the claim could be certificated. I do not regard that case as analogous to the present. This is not a case where the available statistics raised a prima facie case that Italy was unable or unwilling to provide accommodation where required by its obligations under article 3.

Mr Husain’s other submission was that the Judge was wrong to take into account the 37,000 “temporary structures” when measuring the accommodation against the demand. He said that such evidence as there was about these structures showed that they were set up to provide accommodation for migrants arriving by boat from North Africa immediately following their arrival, many of whom would not even claim asylum: they were not intended for the longer-term accommodation of asylum-seekers, still less BIPs. They had, rightly, been ignored by the Court in Tarakhel in its discussion of the figures.

I do not believe that this submission undermines the Judge’s essential reasoning. I am content to assume that Mr Husain is right that many (Footnote: ¹⁵) of those accommodated in temporary structures at any given date will not have claimed asylum and thus that the spaces that they occupy should not be treated as available for asylum-seekers – and still less for BIPs such as NA. But the Judge was not seeking to demonstrate that the figures for accommodation fully matched the figures for demand: rather, his point was that the available figures did not allow such an exercise to be usefully performed. If anything, Mr Husain’s submission reinforces that point.

I can accordingly see nothing wrong in the Judge’s conclusion that the statistics did not demonstrate a risk that NA could not be accommodated in a SPRAR.

THE REPORTS

Both the expert reports commissioned by the claimants in MS and the various published reports on which they relied contain statements about the availability and accessibility of SPRAR accommodation. As we have seen, the Judge addressed those reports at paras. 116-131 of his judgment and held that none of them were capable of displacing the presumption that Italy would comply with its obligations under article 3. Mr Husain submits, as ground 3, that in relation to at least some of those reports that conclusion was unjustifiable. He also submits, as part of ground 4, that the Judge ignored important evidence contained in them.

In order to assess those submissions I will have to summarise the contents of the reports that were before the Judge, though I do so as shortly as I may, focusing on the questions of the availability of SPRAR accommodation and of support for returning BIPs. Some of the reports contain a good deal of more general material about the availability and standard of CARA accommodation; but it would not conduce to clarity of analysis if I allowed myself to be diverted from the issues before us. I will also have to set out and examine the Judge’s observations on some of the reports, an exercise which at some points, regrettably but unavoidably, involves descending to minutiae. I take the reports in the order that they are dealt with in the judgment.

I have summarised the criticisms made in these reports at para. 64 above. Lewis J dealt with them at para. 117 of his judgment as follows:

“I deal first with the two UNCHR reports. The Supreme Court in EM (Eritrea) has recognised that these reports need careful consideration and will provide useful factual information and that respect should be accorded to the views of the UNHCR. The 2012 and 2013 report considers the position in Italy in detail. They should be read in their entirety. Taking the 2013 report, that notes a series of problems in relation to matters such as limited capacity, overcrowding, and lack of opportunities for integration. It made a number of specific recommendations for the improvement of facilities within Italy for both asylum-seekers and BIPs. In my respectful judgment, Elisabeth Laing J. correctly summarised the contents of these two reports at paragraph 167 of the judgment in Tabrizagh as follows: ‘in its two most recent reports on Italy, the UNCHR, while making robust and objective criticisms, has not painted a picture which begins to meet the relevant test’, that is, the test for determining whether, generally, the presumption of compliance has been rebutted. As noted at paragraph 87 of the judgment in Tabrizagh, ‘the reports reveal a picture of general compliance by Italy with its EU and international obligations, while disclosing some operational difficulties’. It would not be open to a tribunal to conclude on the basis of the UNHCR reports that there were ‘omissions on a widespread or substantial scale, or operational problems’. I agree.”

There is no challenge to that passage in the grounds of appeal or Mr Husain’s skeleton arguments.

I have explained the origin and general conclusions of this report at para. 66 (3) above. Chapter 5 deals with the reception of BIPs. The following points are particularly relevant for our purposes.

Section 5.1 addresses the procedures for returning BIPs at the airport. It reports conflicting evidence about their access to the airport NGOs. UNHCR told the authors that the arrival of returning BIPs was reported to the NGOs and the border police said such returnees they could go to the NGOs if necessary, but the representatives of the airport NGOs at Fiumicino and Malpensa said that they only saw asylum-seekers.

Section 5.2.1 deals with the availability of SPRAR accommodation. It refers to the fact that there were only 3,000 SPRAR places in 2012 and to Mr Muižnieks’ description of that as “woefully inadequate”. It says that “5,000 people were on the waiting list for a SPRAR place in 2012” and goes on to describe the consequent difficulties in accessing a place, which it says is particularly acute for people with mental illness. It says that vulnerable BIPs who are not able to find accommodation in SPRAR centres “end up staying in [ERF] housing for Dublin returnees for a very long time”. It does not suggest that they are rendered homeless. The conclusion reads:

“The SPRAR system offers good support to those who get in. However, it does not provide enough places by far. The waiting list is long and obtaining a place is a matter of luck. Plans exist to significantly increase the number of places from 3,000 to 16,000 starting in 2014. It remains to be seen whether implementation will be successful and the extent to which this will defuse the accommodation problem in Italy. SPRAR places are always temporary, and the length of stay in the project is not long enough to ensure lasting independence afterwards. Dublin returnees make up only a small percentage of SPRAR participants. It is striking how few of those returned from Switzerland are accommodated in SPRAR. For this reason, the chances of finding a place in a SPRAR programme appears to be very small for returnees with protection status.”

Sections 5.2.3-6 deal with alternatives to SPRAR accommodation in Milan and Rome. They conclude that the various forms of municipal and charitable provision are inadequate and that a large number of BIPs live in squats or slums.

Lewis J considered the SRC report at para. 120 of his judgment. He said:

“In general terms, in my judgment, the report does not begin to establish that the conditions [which the authors] found in Italy in May to June 2013 demonstrate the existence of substantial operational problems such as would rebut the evidential presumption. Furthermore, that was the conclusion of Elisabeth Laing J. in Tabrizagh at paragraphs 91 to 99 who had subjected the report to what McCloskey J. has described as a ‘more intense, more penetrating’ analysis than the European Court of Human Rights, in the Upper Tribunal decision of Weldegaber v Secretary of State for the Home Department. I agree that the material in the report discloses some failings in May and June 2013 but not on the sort of scale which would entitle a tribunal, or other objective observer, to conclude that Italy was breaching its obligations on a widespread scale. Furthermore, the reports were based on visits which took place 2 years ago. There is now further information, discussed above, which indicates the current position in relation to the supply and demand for accommodation. Nor, in my judgment, are the Claimants correct to say that the … report was approved by the European Court of Human Rights in Tarakhel v Switzerland so that it is no longer appropriate to rely upon Elisabeth Laing J.'s assessment. The applicants in Tarakhel relied upon the report and the Court referred to it. However, the Court's conclusions are based on the statistical and other information provided by the Italian authorities (see, in particular, paragraph 110 of its judgment). For all those reasons, no tribunal, properly directing itself could, at this time, regard the … report as establishing that there are substantial operational problems in Italy.”

That may seem a rather short analysis, but it is fair to say that in her judgment in Tabrizagh, which Lewis J effectively adopts, Elisabeth Laing J had gone into a good deal more detail. Her essential reason for finding that the SRC report did not justify the conclusion that there was a risk that the Italian authorities would not discharge their obligations under article 3 was that it did not have the same authority as the UNHCR reports. But she also said, at para. 91, that there were some reasons to doubt the objectivity and accuracy of the report. She characterised the SRC report overall as disclosing “some failings, but nothing on the sort of scale which would entitle an objective observer to declare that Italy was breaching its obligations on a widespread scale”.

Mr Husain said that the only reason given by Lewis J for “rejecting” the SRC report was that it was based on research that was two years old. He submitted that that was illegitimate in the context of a certification case, since the extent to which the report’s findings were out of date clearly required careful factual examination. He also said that that reasoning was inconsistent with what he said was the ECtHR’s reliance on the report in Tarakhel.

It is not correct to say that Lewis J relied only on the date of the SRC report. He relied also on the criticisms by Elisabeth Laing J in Tabrizagh, which Mr Husain did not address. But I accept that the age of the data was an important part of the Judge’s reasoning. I do not believe that it was illegitimate for him to attach weight to that factor in the particular circumstances of this case. The crucial question for his purposes was the availability of SPRAR accommodation. The SRC report had drawn attention to the plans for a massive expansion of the SPRAR network and had said that it remained to be seen whether implementation would be successful: see para. 179 above. The evidence before the Judge was that by 28 February 2015 over 20,000 SPRAR places were being occupied, as against 3,000 in the period covered by the SRC report. In those circumstances the Judge was entitled to place little weight on evidence about shortages in 2012/2013.

As regards the references to the SRC report in Tarakhel, Lewis J was right to say that its evidence had not been explicitly adopted in the crucial part of the ECtHR’s reasoning. It may nevertheless be fair to say that the Court regarded the report as contributing to its overall conclusion that there was a risk that SPRAR accommodation would not be available for the applicants. However that does not meet the point that the evidence in question was, by the time of the hearing before Lewis J, two years old and that the number of SPRAR places had doubled in the meantime.

Mr Husain also relies on SRC’s evidence about reception arrangements at Fiumicino and Malpensa, but I deal with those separately below (see paras. 213-216).

The MEDU report is dated 16 December 2014. The authors work in a project helping migrants living in squats in Florence. The number about whom they have data is 105, of whom 64 are Dublin returnees who are BIPs (referred to conveniently as “Dublin BIPs”). They make clear that they have no data about Dublin BIPs in Rome or Milan (though they do hear from some of “their” BIPs about their experiences elsewhere). They explain that since almost all the people with whom they work are BIPs they feel unable to comment on reception and living conditions for Dublin returnees who are asylum-seekers.

The substance of the report is to be found in section C, which addresses reception and living conditions for Dublin BIPs. In bare outline, and confining myself to the points relevant to NA’s case, the authors’ evidence, based on what they are told by “their” BIPs, is that no assistance of any kind is made available to Dublin BIPs at the airport. When they leave the airport they may get in touch with an NGO, who may then be able to arrange for accommodation in a SPRAR or ERF facility (if they have not been so accommodated in the past), but the waiting-lists are very long, sometimes as long as a year: the authors make it clear that they do not have access to authoritative figures for waiting times but are dependent on what the BIPs tell them. Unless and until they can enter a SPRAR they are dependent on the ordinary services available to the homeless, which are limited, and many accordingly have to live in squats or on the street. They give a number of individual examples of Dublin BIPs who are destitute.

At para. 12 of the report the authors address the availability of facilities for peculiarly vulnerable BIPs. Their evidence is that only a very small number of such facilities exist. They refer to official national figures for the years 2010/11, 2011/12 and 2012/13, which say that places for vulnerable persons represent about 16% of the total SPRAR places, though they say that the true figure in Tuscany is much lower. They refer to thirteen cases of people encountered between September 2013 and March 2014 who were seriously ill but had been living in a squat with no access to medical help or other support: several did not receive immediate accommodation even when MEDU referred their cases to the authorities. It is not clear whether all or indeed any of these were Dublin BIPs or BIPs at all. (Mr Husain argues that they were, but that is not stated in the report itself.)

The Judge deals with the MEDU report at para. 121 of his judgment, as follows:

“The Claimants rely upon a report by a group, Doctors for Human Rights ("MEDU"). The members of MEDU, which is a not-for-profit organisation, frequently come into contact with beneficiaries of international protection who have ceased to be eligible for places in SPRARs. The MEDU report confirms that it is not in a position to provide information on the reception and living conditions for asylum seekers. The report could not, therefore, assist in relation to that group of persons (who include SG). In relation to the report, it is relevant to note its inherent limitations. It is based on BIPs living in Florence and the report expressly states that its expertise is limited to Florence. It refers to serious shortcomings on the part of the authorities in identifying BIPs and vulnerable people. It refers to 13 cases in a squat in Florence, occupied by asylum seekers and BIPs, during the period from September 2013 to May 2014 when such persons had no contact with health care services. The report does not indicate whether these were asylum seekers or BIPs. It does not indicate whether they were returned under the Dublin Regulation. No tribunal could properly rely upon this report to determine whether the arrangements and conditions in Italy were such that Italy would not comply with its obligations in relation to persons returned to Italy.”

The statement at the start of that passage that MEDU’s clients are BIPs “who have ceased to be eligible for places in SPRARs” is, with respect to the Judge, wrong. The authors say in terms that “based on our experience, we can confirm the presence of large numbers of Dublin BIPs on Italian territory who do not have access to accommodation or support although they are entitled to it [my emphasis]” (section C, para. 8). The thrust of the entire section is that the Dublin BIPs on whom they have data are living in squats because they have not known how to get into a SPRAR and/or because the waiting lists are too long; and all the individual cases which they describe are of that kind.

It does not appear, however, that that error is relevant to the Judge’s decision that no tribunal could rely on the MEDU report. It seems that his reasons for that conclusion are based on the “inherent limitations” which he specifies. With respect to him, I do not accept that those limitations are such as to justify discounting the report entirely. Specifically:

Mr Tom Southerden of Amnesty International UK’s Refugee Programme wrote a report for the claimants in MS dated 13 February 2015 reviewing conditions for asylum-seekers and refugees in Italy. Although it covers a number of other issues, at a level of some generality, the relevant parts for our purposes are those relating to the availability of SPRAR accommodation, and more particularly for Dublin returnees. As to that, the report gives a figure of 13,020 SPRAR places as at January 2014. It says that there are plans to expand numbers to 19,000 but says that it is not known how close to achievement those plans are. It says that the quality of provision is variable. In relation to Dublin returnees it makes a particular point, irrelevant for our purposes, about the position of asylum-seekers whose claim has been rejected. Otherwise it says only (at para. 49):

“In relation to accommodation provision, to Amnesty’s knowledge, absent some special provision for an individual return a Dublin returnee will face the same risks as any other asylum arrival in relation to the overcrowding and related chronic under supply of appropriate accommodation. As a result, in Amnesty’s view, they would face a risk of homelessness on return to Italy.”

The Judge addressed the Amnesty report at para. 122 of his judgment, as follows:

“Next, the Claimants rely upon a report by Amnesty International Limited. That body is, of course, a well-known organisation. The report, read as a whole and fairly, could not on any justifiable view be capable of supporting a conclusion that the evidential presumption that Italy would comply with its obligations is rebutted. Nor could any tribunal legitimately regard the report as capable of supporting the conclusion that those returned to Italy would be exposed to treatment in breach of Article 3. Part of the report is a commentary on the judgment in Tarakhel. Part is an acceptance that the two UNCHR reports are broadly in line with Amnesty International's understanding of the position in Italy. The report does not, itself, include, and is not based upon, any further analysis or work done by Amnesty International. It relies upon earlier data, such as Eurostat data, discussed above, or information about the numbers accommodated within SPRARs which are now out of date. It states that, in its view, ‘absent some specific special provision for an individual returnee’ such a person will face the same risk on return as any other asylum arrival in relation to overcrowding and undersupply. However, it does not seek to address the question of the arrangements in place for those being returned to Italy and could not reasonably be relied upon as demonstrating any difficulties in this regard.”

Mr Husain describes that reasoning as “opaque”. I do not agree. It gives reasonable reasons for finding the Amnesty report of little assistance on the essential issue before him, certainly as regards the availability of SPRAR accommodation. I note in particular the Judge’s points (a) that Amnesty’s data about the availability of SPRAR places is out-of-date and (b) that the report does not address the position of Dublin returnees returned under arrangements of the kind that will apply in NA’s case.

Mr Husain is also critical of the Judge for treating the essential question as being whether the report was “capable of supporting a conclusion that the evidential presumption that Italy would comply with its obligations is rebutted”, saying that the real question was whether it shed light on whether, as a matter of (in Lord Kerr’s phrase) “practical reality”, there was a risk that the claimants would suffer serious harm if returned to Italy. But the Judge’s language simply reflects the way that he had formulated the issues. I have already said that I do not believe that his references to the presumption of compliance distracted him from addressing the essential question: see para. 155 above.

Following the judgment Mr Southerden wrote to Wilsons setting out a number of concerns about the Judge’s reasoning. Mr Husain referred to that letter in his skeleton arguments, though not in his oral submissions, and it was included in the bundle before us. This is an inappropriate procedure. Amnesty was fully entitled to draw the claimants’ advisers’ attention to what it believed to be errors in the Judge’s approach, but if those advisers believed the points to be good the right course was to adopt them in counsel’s submissions. (Footnote: ¹⁶) It appears that Amnesty understood Lewis J’s observation to the effect that the report was not based on specific data-gathering or studies of the kinds embodied in some of the other reports as meaning that no value could be accorded to its experience based on its presence in Italy and its ongoing monitoring of the position of migrants there. I do not think that that is what the Judge meant; but he was entitled to point out that the report is more in the nature of a general overview than, say, those of MEDU or SRC. I am quite sure that the English courts and tribunals will approach any report emanating from Amnesty International with careful attention, but of course what weight should be given to it in any particular case will depend on the extent to which its contents add to the knowledge available from other evidence.

Ms Leo is a lawyer at the Rome bar specialising in immigration and asylum law. She is active in the Association for Legal Studies on Immigration (“ASGI”) and has participated in various studies under its auspices. Three reports from her were submitted to the Court. They were written in Italian. Only the English translations were before the Court.

Ms Leo’s first report is dated 15 September 2014. It is 30 pages long and divided into six sections. Sections I and II contain introductory matter and section III gives an overview of the Italian asylum system. Sections IV and V address the position of Dublin returnees. Like other witnesses and reports, Ms Leo refers to the number of BIPs who are homeless or living in and in circumstances of destitution: she refers to the Selam Palace. The specific points that she makes that bear directly on the risk that NA might not receive SPRAR accommodation if returned are twofold.

First, she gives evidence about the availability of SPRAR accommodation. In section IV (d) she says:

“With regard to the possible access to SPRAR facilities, it should be noted that although there has been a recent increase in the number of places available, the centres are now largely filled, as reported by the Ministry of the Interior in the above-cited circular dated 19.03.2014 where is written: “all the centres of government and those backed by some local authorities within the SPRAR system were saturated” [a footnote gives a link is given to the report]. For this reason, the waiting list for a SPRAR place is very long: according to reports by some operators, it is now around 2-3 months.”

Secondly, in section V (b) she deals with the assistance available to Dublin BIPs at Fiumicino and Malpensa. In her first paragraph she makes the point that they have no rights qua asylum-seekers. In the next two paragraphs she acknowledges that although they will be entitled to SPRAR accommodation, that is only so if they have not previously had such accommodation and that there are in any event waiting lists of up to three or four months. She then continues:

“As has already been stated, [ASGI] is, during three months, doing a thorough research on the situation of asylum seekers and the beneficiaries of protection when they arrive at Fiumicino airport. Although, as has been said, the results of the research have not yet been made public, it is possible to predict what will be revealed by the research from the data gathered to date. The beneficiaries of protection on whom data was collected as part of the research and who had been returned to Italy did not receive any kind of reception on their arrival in the country. Indeed, after the initial formalities at the relevant border police station, the beneficiaries of protection were invited to leave Fiumicino airport and were not offered any form of reception or social support. It is important to note here that Fiumicino airport is the arrival point of most of the people sent back to Italy under the Dublin Regulations. Many of the people sent back to Italy are, in fact, beneficiaries of protection.

It is reasonable to assume that the same thing occurs when beneficiaries of protection arrive at other airports because, as has been shown, in Italy, a beneficiary of protection does not have an enforceable right to reception and social support.”

She goes on to quote from a report commissioned by the operators of Malpensa airport about reception procedures there.

A second report from Ms Leo was submitted dated 18 March 2015 and addresses various points made by Mr Dangerfield and the procedure for obtaining a replacement for a lost residence permit.

Lewis J characterised Ms Leo’s first two reports, at para. 124 of his judgment, as follows:

“There are, in my judgment, many features of these two reports which mean that this court, and indeed any tribunal, would have to regard the reports as so lacking in objectivity as to be incapable of constituting reliable evidence. Viewed objectively, the reports do not provide a fair, balanced objective picture capable of assisting in the resolution of the disputes in this case. Rather they are partisan and advocate a particular view.”

In support of that conclusion he makes at paras. 125-126 four particular criticisms of Ms Leo’s reports, though he says that they are not comprehensive. He concludes:

“Other parts of her reports give rise to similar criticisms of lack of objectivity and analysis. In my judgment, no tribunal could possibly treat the two reports as constituting reliable evidence of the position in Italy generally or the position in relation to what any individual, including these Claimants, would face if returned to Italy.”

Mr Husain submits that the Judge’s particular criticisms are unfair and unfounded, and that the rejection of Ms Leo’s evidence in support of which they are made is unjustified. He also relies on the fact that the Respondent did not in her evidence or written submissions prior to the hearing advance the criticisms relied on by the Judge, so that neither she nor he as counsel had the opportunity to answer them. In support of his submissions Mr Husain seeks to adduce a further witness statement from Ms Leo, dated 31 July 2015: I will refer to this as her “appeal statement”. He submits that the statement should be admitted because its purpose is to give factual evidence relevant to procedural and other mistakes made by the Judge. To the extent that that is indeed what it purports to do I can see no objection to it being admitted. In fact much of what it contains is argumentative and could have been made by way of submission; but the convenient course is to admit it in its entirety.

In order to address that submission I need to consider the Judge’s four particular criticisms in a degree of detail which risks obscuring the overall pattern of this judgment. I accordingly relegate the exercise to an appendix. For the reasons there given, I accept substantial parts of Mr Husain’s submissions. I do not believe that the Judge’s particular criticisms, to the extent that they were valid, justified his conclusion that “no tribunal could possibly treat [Ms Leo’s] reports as constituting reliable evidence of the position in Italy generally or the position in relation to what … these Claimants … would face if returned to Italy”.

I would add that I also see force in Mr Husain’s point that there is a real risk of unfairness in a Court reaching conclusions about the reliability of an expert witness on the basis of particular criticisms which the witness has had no opportunity to rebut. Where a criticism is flagged up in advance (for example in the other party’s skeleton argument), there is usually an opportunity to answer it by way of submission or by provision of a further statement or other clarificatory material, though there will occasionally be cases – even in judicial review proceedings – where oral evidence and cross-examination may be appropriate; but the problem in the present case is that the criticisms in question were not foreshadowed in submissions of any kind. Having said that, I would not wish to be understood as saying that a judge in judicial review proceedings must always ensure that a witness has the opportunity specifically to meet every point of criticism that may contribute to their evidence being rejected, in whole or in part.   In many, perhaps most, cases the nature of the issue will be sufficiently clear to enable the judge to make a fair assessment without such an opportunity being given. What justice requires will inevitably depend on the circumstances of the particular case.

However, my conclusion about the Judge’s particular criticisms of Ms Leo’s evidence is not the end of the matter. Those criticisms were put forward in support of his conclusion that the reports did not give a “fair, balanced [and] objective picture” and were “partisan”. I have carefully read both reports in full, and I believe that he was entitled to form the impression that they were written more from the standpoint of a campaigner than as a dispassionate observer. Although I do not in any way question Ms Leo’s integrity or her knowledge of her subject, her reports do at some points have the flavour of advocacy.  It is because of the importance of obtaining, so far as possible in this difficult field, dispassionate evidence that the courts tend to put particular weight on the reports of established institutions such as UNHCR or the Commissioner for Human Rights of the Council of Europe or AIDA.  I do not propose to lengthen this already lengthy judgment by giving examples of where Ms Leo may have erred in this direction, since, even though I think the judge’s overall assessment was legitimate, that does not affect my view that he was wrong to exclude her report from consideration altogether. It is necessary to distinguish between the weight to be given to her broader conclusions about the risks facing NA if returned and the factual evidence that she gives on specific points. Even if the former are discounted the latter still require to be considered.

In his original skeleton argument Mr Husain confined his challenge to the Judge’s reasoning to the way in which he dealt with the four reports discussed above – that is, the reports from the SRC, MEDU, Amnesty and Ms Leo. In his supplementary skeleton argument he added some criticisms of how the Judge dealt with other reports in the evidence – such, for example, as the Brunswick report (see para. 66 (1) above) – though these were not developed in oral submissions. In the circumstances I can deal with those criticisms compendiously. Not all the reports in question were concerned with SPRAR accommodation, or the position of Dublin BIPs, at all; but to the extent that they mentioned those aspects none of them addressed the situation as it stood in late 2014.

It follows from the foregoing discussion that I believe that the Judge was over-dismissive of the evidence of MEDU and Ms Leo. But it does not follow that the appeal must be allowed. It remains to decide whether, giving their reports such weight as they merit, the evidence overall showed, or arguably showed, that NA was at real risk of not being found SPRAR accommodation if returned to Italy. I consider that question at paras. 213-221 below.

THE EVIDENCE OF THE CLAIMANTS’ EXPERIENCES IN ITALY

Mr Husain submits, as part of ground 4, that Lewis J failed to consider, save very indirectly, the relevance of the experiences of the individual claimants while they were in Italy first time round. I have given a sufficient outline of NA’s experience at paras. 11-13 above and of MS’s and SG’s at para. 125. Mr Husain says that those experiences were “central to the enquiry” for a number of reasons. They were evidence of the practical realities of the Italian system, as opposed to how it might be supposed to work, showing that there was, to put it no higher, a real risk that the claimants would be exposed to homelessness and destitution if returned. He submits, on the basis of such cases as Demirkaya v Secretary of State for the Home Department [1999] EWCA Civ 1654, [1999] INLR 441, that “past ill-treatment is a good indicator of future risk, absent a material change of circumstances”. Finally, he submits that their experiences tended to show that they were the kinds of person who would not be able to cope with the system if returned.

I do not accept that the Judge erred in this respect.  Generalisations about past ill-treatment being a good indicator of future risk are well enough in the context of the risk of persecution, but they should not be unthinkingly applied in a different context.  On her account NA had to live on the streets, where she had the appalling experiences that she did, because she was compelled to leave Cassa Calenga, and she was never offered, or given help to apply for, SPRAR accommodation or other support that would have prevented her undergoing those experiences.  But the question is whether she will now receive that support if she is returned in the different circumstances of a Dublin BIP.   That was the question on which the Judge had to focus, and her past experience sheds no real light on that.  The Judge was well aware that what is supposed to happen in theory does not always happen in practice; and he did not need to refer to the claimants’ particular experiences to make that point.

I should make it clear that I accept that NA’s experiences in Italy are relevant to the extent that they establish that she is peculiarly vulnerable, not least, though not only, because they will have contributed to the serious depression from which she is suffering. But that only brings us back to the question whether she will in fact receive proper accommodation, healthcare and support if returned.

The first question is whether there is a real risk that if she is returned NA will not be given the support and advice necessary to enable her to access SPRAR accommodation. The starting-point is that she is peculiarly vulnerable. There is no reason to reject Mr Dangerfield’s evidence that accordingly, in accordance with established procedures, her return would be pre-notified to the Italian authorities, who would also (unless she maintains her refusal to consent) be provided in advance with her medical records.

I accept that there is at first sight a conflict in the evidence about what the consequences of such pre-notification would be. Mr Dangerfield says that it means that NA would on arrival be referred by the border police to the airport NGO, which would give her the necessary support and advice. That is consistent with what UNHCR told SRC (see para. 178 above). It is also consistent with the passage from the AIDA March 2015 report quoted at para. 134, though I would not attach decisive importance to it because of the ambiguity there referred to. Against that is the evidence of MEDU that the airport NGOs do not assist Dublin BIPs (para. 188); of SRC that what they were told by the airport NGOs was different from what they were told by UNHCR (para. 178); and of Ms Leo that Dublin BIPs “did not receive any kind of reception on their arrival in the country” (para. 201).

I do not believe that the existence of that conflict meant that Lewis J was obliged to conclude that there was a risk that NA might receive no assistance or support at the airport to which she was returned, or in any event that there was an area of factual dispute that meant that the claim could not properly be certified. It may be – though I express no view either way – that the evidence relied on by NA raises a real question as to whether BIPs whose arrival has not been pre-notified in the way described by Mr Dangerfield would be referred to the airport NGOs. I take the point that the formal position is that a BIP has the full rights of an Italian citizen and does not, other things being equal, require any support: that would explain the response which SRC and Ms Leo apparently received on this question. But the evidence relied on by NA does not relate to the particular case of a “pre-notified” peculiarly vulnerable Dublin BIP. I do not think it could reasonably be regarded as outweighing the specific evidence of Mr Dangerfield about what happens in such cases. That evidence is consistent with the presumption of compliance: to put it less technically, it conforms to what one would expect of a country like Italy which, for all the stresses which it is encountering, has been recognised by UNHCR as having a functioning system for the treatment of asylum-seekers and BIPs. It is in truth most implausible that a peculiarly vulnerable person, returned in accordance with the procedures described by Mr Dangerfield, and in all probability accompanied by a medical escort, would be offered no assistance by the airport NGOs.

I should also mention in this connection a particular point made by Mr Husain. It is said that whereas when the Italian authorities accepted responsibility for MS and SG under the Dublin procedures their letter requested at least ten days’ notice “about any particular health situation, both from the physical and from the psychical point of view, as well as about any disability or delicate situation which can entail considerable reception problems” – which would in practice require the provision of her medical records – no such request appears to have been made when it was first sought to remove NA in May 2013. Mr Husain suggests that this reflects the fact that the Ministry did not believe that they had any responsibility for her as a BIP and thus also that it would take no steps to assess her condition on arrival. It is not clear from the papers before us what communications there were between the UK and the Italian authorities on that earlier occasion. I accept that it is not implausible that the standard forms employed would be different in the case of a returning BIP – though in fact MS was also a BIP – and thus there might not have been an explicit request from the Italian authorities for ten days’ notice of any medical condition. But even if that were so it could not possibly be the basis for rejecting the explicit evidence of Mr Dangerfield that NA’s medical records would be sent (if she consented) and that the Italian authorities would conduct an assessment accordingly.

Mr Husain also refers in his skeleton argument to two passages in reports that were before the Court. The former says that “Italy … do[es] not have standard practices in place to conduct a vulnerability assessment”; and the latter that assessments are not performed “systematically” and are performed only in the accommodation to which the asylum-seeker is allocated, where the quality will depend on the expertise available. He complains that the Judge did not deal with this evidence. It is far from clear to me that the evidence in question was ever relied on by the claimants: I see no reference to it in their skeleton argument below, and the documents referred to were in the Respondent’s bundle rather than the claimants’. But even if the Judge was referred to these passages I can understand why he felt it unnecessary to address them specifically. Once it is accepted, as he found, that NA’s return would be pre-notified to the Italian authorities, with notice of her vulnerability and (subject to her consent) a copy of her health records, there is no reason to believe that there is a real risk that that vulnerability will simply be ignored – still less, again, if she is accompanied by a medical escort. It may be that the procedures for assessment are variable and that some may be sub-optimal, but that fact alone cannot come close to the attaining the minimum threshold of severity required to give rise to a breach of article 3.

The question then is whether there is a risk that even if NA is offered support and assistance by the airport NGO – and by any other organisation to which they may refer her – they will be unable to offer her SPRAR accommodation (in which term I include other forms of accommodation suitable to her condition). I do not believe that such a risk has even arguably been established. There was no doubt an acute shortage of SPRAR accommodation in 2012, when there were only some 3,000 places. But the official figures, relied on by both parties, show that by 28 February 2015 over 20,000 people were being accommodated in SPRARs: see para. 163 (2) above. I accept that SRC, MEDU and Ms Leo all refer to the existence of waiting lists, and SRC and MEDU refer to the waiting time sometimes being as long as a year. But SRC’s data comes from the period before the recent increase in provision (as it explicitly acknowledges), and MEDU’s is of uncertain date and admittedly anecdotal. Ms Leo’s is the most recent. It refers to a circular from the Ministry of the Interior from March 2014 referring to the SPRARs being “saturated”, and to evidence of waiting-lists experienced by “some operators” of two to three months: see para. 200 above. The lack of particularity about the “operators” in question is unsatisfactory, but I would not for that reason discount it altogether. However, the date of the Ministry’s circular is almost a year before the date with which the Judge was concerned. In other circumstances such a time-lapse might not be significant, and I appreciate that it will never be possible to present the Court with evidence that is completely up-to-date. But the crucial point about the present case is that it was accepted by all concerned that the SPRAR system was being expanded as a matter of urgency, and that the figures showed that its capacity had grown very rapidly over the previous year: in those circumstances the passage of time was likely to have made a real difference. I note also that the possible waiting period of two to three months referred to by Ms Leo’s evidence is already much less than that reported in the earlier reports. Taken as a whole, this evidence does not in my view justify the conclusion – contrary to the evidence of Mr Dangerfield – that there is a real risk that NA would not if returned be found a SPRAR place if she were judged sufficiently vulnerable to require one.

In reaching that conclusion I do not exclude the possibility that all may not go perfectly for NA if returned. It is not realistic to suppose that the recent efforts by the authorities to improve support and provision for Dublin BIPs, among others, have removed all the problems previously encountered; and it is possible that there may be some difficulties, delays and uncertainties before she is fully assessed and securely settled in SPRAR accommodation. But temporary inadequacies in provision would not meet the minimum level of severity on which the ECtHR case-law insists. What, on the evidence, I am sure of is that as a Dublin returnee NA will not be simply abandoned by the system in the way that occurred when she was required to leave Cassa Calenga and be left homeless and destitute. I should also say that Mr Husain expressly accepted that the evidence showed that if NA is adjudged to be in need of immediate medical attention she will receive it.

I should also say that my conclusion does not ignore the evidence that there are many BIPs, including some Dublin BIPs, in Italy living in deplorable conditions, and also that some at least of them are peculiarly vulnerable. But the individual histories that led them to that condition will be very variable. The evidence does not support the proposition that they include BIPs who have been returned from this country, in the circumstances that NA would be returned, and have nevertheless ended up on the streets because no provision has been made for them.

Although I have, as I believe was necessary, engaged with the particulars of the evidence before Lewis J, and the points taken on it, my conclusion is consistent with that reached by the ECtHR in similar cases. In Hussein the applicant was, like NA, a vulnerable BIP. The evidence about the problems in the Italian system for both asylum-seekers and BIPs was similar to that in the present case: indeed in some respects it showed a worse position, particularly as regards the availability of SPRAR accommodation. But the Court found that there was no risk of a breach of her article 3 rights, having regard to the fact that her return would have been pre-notified and that she was entitled to special consideration as a vulnerable person: see paras. 76-78 of its judgment, quoted at para. 71 above. Daytbegova is not on all fours because the applicants were not BIPs, but it is nevertheless significant that the Court was prepared to proceed – without requiring specific assurances – on the basis that hand-over arrangements could and would be made between the Austrian and Italian authorities which would meet the daughter’s particular health needs.

I believe that the Judge was correct to conclude on the evidence before him that the conditions for BIPs returned to Italy under the Dublin Regulation were not such that there was a real risk that NA would suffer inhuman or degrading treatment if so returned.

GROUND 5: MENTAL HEALTH AND SUICIDE RISK

NA’s pleaded case under this head is that the Judge treated his conclusion that NA would receive appropriate accommodation and support if returned to Italy as a complete answer to her case that her removal would arguably amount to a violation of her article 3 rights by reason of her mental ill-health; and that that involved him failing to take into account other factors which were relevant to the assessment of whether she would in fact commit suicide on return.

That case was not developed by Mr Husain in his oral submissions. He did, however, put in a note summarising the way the case was put, which in turn cross-referred to his supplementary skeleton argument, and I can best address this ground by reference to that.

There is now a good deal of law about the application of article 3 in cases where it is said that a returnee is liable to commit suicide if returned. The position was most recently summarised in my judgment in R (Wasif) v Secretary of State for the Home Department [2016] EWCA Civ 82, [2016] 1 WLR 2793: see para. 44 (p. 2807). There would certainly normally be no arguable breach of article 3 in returning a person who is at risk of suicide as a result of mental illness to a country such as Italy where appropriate treatment is readily available. That is illustrated by the decisions of the ECtHR in both Daytbegova and AS (Switzerland) – see paras. 75-78 and 102 above.

However Mr Husain contends that NA’s case falls within the exceptional category recognised by this Court in Y and Z (Sri Lanka) v Secretary of State for the Home Department [2009] EWCA Civ 362, [2010] INLR 178. In that case the appellants were Sri Lankan Tamils, brother and sister, who had been the victims of torture and rape by the security forces and suffered from PTSD and depression in consequence. Their claims for asylum in the UK failed on the basis that there was, objectively, no risk of further ill-treatment if they were returned; but they resisted removal to Sri Lanka on the basis that if they were returned they were likely to be driven to suicide. Their case was they had a genuine and overwhelming fear, whether or not well-founded, that on return they would again be tortured and abused by the security forces, and also that they had no-one to turn to there for support because their entire family had been killed in the tsunami. That case was accepted by this Court. Its essential reasoning appears from para. 61 of the judgment of Sedley LJ, where he said:

“… [A]lthough some psychiatric care is available in Sri Lanka, these two appellants are so traumatised by their experiences, and so subjectively terrified at the prospect of return to the scene of their torment, that they will not be capable of seeking the treatment they need. Assuming (what cannot be certain) that they come unscathed through interrogation at the airport, with no known family left in Sri Lanka and no home to travel to, the chances of their finding a secure base from which to seek the palliative and therapeutic care that will keep them from taking their own lives are on any admissible view of the evidence remote.”

Mr Husain submitted that NA’s case was substantially comparable to that of the appellants in Y and Z. She too had suffered rape and ill-treatment in the country to which it was proposed to return her and was traumatised at the prospect of return to the scene of her suffering. He conceded that that treatment was not at the hands of agents of the state, but he said that it was nevertheless the consequence of the authorities’ culpable neglect in ejecting her from Cassa Calenga without any means of support. She too had no family to support her.

I do not accept that submission. The circumstances of the present case are materially different from those in Y and Z. The relevance of the appellants’ past experiences in that case, coupled with the total absence of any home or family support, was that those circumstances rendered it practically impossible for them to seek the support and treatment from the authorities that they needed. But it follows from Lewis J’s conclusion that on NA’s return to Italy she will be found SPRAR accommodation, and that she will receive the support necessary to enable her to access the treatment which she needs. If that is correct, the issue of whether or to what extent the Italian authorities were responsible for her previous suffering does not arise.

My conclusion on this point is consistent with the decision of the ECtHR in AM v Switzerland, where the applicant’s mental ill-health appears to have been caused at least in part by the treatment which he had received from the police in Italy: see para. 103 above.

Mr Husain submitted that any argument based on the availability of support and treatment in Italy foundered on the fact that NA’s medical records would not be available to the authorities there. He makes three points, which I take in turn.

First, it is said that on a previous occasion when it was sought to remove her no attempt was made to transfer her records. I will assume that that is the case, but the fact that the Respondent failed to comply with her policy on that occasion, if she did, is no basis for concluding that she will not do so if and when NA is removed in the future; and the importance attached to the issue in this case in fact makes it all the more certain that she will (subject to the question of NA’s consent).

Secondly, he relies on the point with which I have already dealt at para. 216 above.

Third, it is said that NA’s refusal of her consent to the transfer of her records means that if she is returned the Italian authorities will have no details of her medical history (Footnote: ¹⁷). Mr Husain seeks to equate this with the unwillingness of the appellants in Y and Z to seek treatment. The Judge noted at para. 31 of his judgment that there was no suggestion that she lacked capacity to make that decision, and no such suggestion was made before us. However, Mr Husain did rely on a paragraph in Dr Obuaya’s second report which reads as follows:

“I note that [NA] has refused to give consent for her medical records to be shared with the Italian authorities. Irrespective of whether or not this is a wise decision, from a clinical perspective, this is an understandable reaction to her previous experience of being left unsupported and vulnerable in Italy; she has clearly developed a strong sense of distrust of the Italian authorities due to the trauma she has described there.”

Given the nature of these proceedings, that opinion should be accepted: that is, it would not be right to treat NA’s refusal as tactical. But I do not think that it follows that a decision to return her to Italy would constitute a breach of her rights under article 3. In the first place, there is no certainty that she will persist in her refusal: on the contrary, it must be more likely that, when she appreciates that she will indeed be returned and has the benefit of further advice, she will change her mind. But, even if she does not, the result of her refusal will not be to deprive her of access to treatment. It will still be the Respondent’s responsibility to notify the Italian authorities of her vulnerability and of Dr Obuaya’s diagnosis, including the risk of suicide, and explain why it has not been possible to supply her records. The lack of the records may be a hindrance to treatment but it will not mean that she cannot be treated at all or that appropriate steps cannot be taken to ensure against the risk of suicide.

That conclusion means that I need not consider a further question raised in Mr Husain’s skeleton argument which may not be entirely straightforward. There is plainly a powerful argument that even if the non-transfer of the records did prevent steps being taken to treat NA and mitigate the risk of suicide, that is the result of her own deliberate decision and not of the act of the UK government in returning her to Italy: it would be different if she were not competent to take that decision, but that has not been argued. However, Mr Husain seeks to draw analogies with the situation considered by the ECtHR in Saadi v Italy (2009) 49 EHRR 30 and/or with that considered by this Court in Danian v Secretary of State for the Home Department [1999] EWCA Civ 3000, [2000] Imm AR 96, which both in (very) different ways involve the proposition that in an article 3 context it may be necessary to disregard risks created by the applicant’s own conduct. I doubt if those are useful analogies. If I had a concern in this area it would be more about whether the question whether NA was legally competent to take the decision to refuse to consent to the disclosure of her notes is the only touchstone: perhaps it might be enough for her to show that, as Dr Obuaya says, her decision was significantly affected by her mental ill-health. But I am not willing to express any definitive view on these points where it is unnecessary to do so and where we did not have the benefit of oral argument.

I have reached my conclusion on this aspect of the case without reference to the Strasbourg authorities. I do, however, note that a very similar situation arose in one of the post-Tarakhel cases referred to at para. 104 above – JA v Netherlands – where the mother of the family in question, who was said to be suffering mental ill-health and to be at risk of suicide, refused to give her consent to the transfer of her medical data to the Italian authorities. The ECtHR held that this did not mean that her return to Italy would constitute a breach of her article 3 rights. It said, at para. 32:

“The first applicant having refused to give consent to communicate her medical data to the Italian authorities, it cannot be said that the latter have not been duly notified of her mental health condition, or that she runs a real risk of not receiving the required medical care in Italy.”

The reasoning is very condensed, but the decision is nevertheless consistent with that which I have reached.

MR’s APPEAL

It will be recalled that Whipple J’s reason for refusing MR’s claim was essentially that she should follow the decision of Lewis J in MS. Unsurprisingly, therefore, MR’s grounds of appeal adopt those in MS, in the form of the summary appearing in the original skeleton argument. No separate skeleton argument was lodged in MR’s case, but it was understood that the submissions directed to the case of SG, as an asylum-seeker rather than a BIP, were applicable to MR, and any further points peculiar to him could be identified in the supplementary skeleton argument (though in the event none were).

As regards ground 1 – that is, the effect of the judgment in Tarakhel – my reasoning and conclusion in NA’s case apply equally to MR. As regards ground 5, MR’s case differs from NA’s only by being weaker: he can rely neither on a history of previous ill-treatment in Italy nor on any refusal to consent to the transfer of his medical records.

However, the position as regards grounds 2-4 requires slightly fuller consideration. The primary difference between MR’s case and NA’s is that he is an asylum-seeker rather than a BIP. As explained above, the legal positions of asylum-seekers and BIPs in Italy are different, and that difference has practical consequences as regards accommodation and support. The essential question is thus whether Lewis J was entitled to find that there was no risk that SG – in whose shoes MR stands for these purposes – would be subjected to inhuman and degrading treatment if returned to Italy. Again, the focus of the case is squarely on the availability of suitable accommodation: see para. 99 of the judgment, quoted at para. 121 above. I consider in the following paragraphs whether that question is answered by the conclusions that I have reached in NA’s case.

As we have seen, there is an issue in NA’s case as to whether she would have access to the airport NGOs because she was a BIP. That issue does not arise in MR’s case. It is common ground that as a returning asylum-seeker he would be referred to the NGO for appropriate advice and support. The question is whether, even with that assistance, he would be found appropriate accommodation pending the determination of his claim.

As to that, it is MR’s case that, like SG, he has serious mental health problems – see paras. 22-23 above – and accordingly that he is peculiarly vulnerable. Asylum-seekers who are peculiarly vulnerable are typically accommodated in SPRARs. It follows from my conclusions in NA’s case that appropriate accommodation will be available for MR if required.

On that basis it is unnecessary for me to consider what the position would be if MR were not peculiarly vulnerable and accordingly did not qualify for accommodation in a SPRAR and fell to be accommodated in a CARA. Although some of the reports relied on by the claimants in MS dealt with the availability of CARA places and the conditions of accommodation there, Lewis J did not consider that evidence, and it was not the focus of the submissions before us. I would only observe that the clear message of the post-Tarakhel decisions of the ECtHR discussed at paras. 100-104 above – and indeed of Tarakhel itself – appears to be that the acknowledged problems about reception conditions for asylum-seekers in Italy are not such as to act as a bar to the return of those without any peculiar vulnerability: see in particular the passages from the judgment in AME v Netherlands quoted at para. 101 above. That is consistent with the fact that UNHCR has not made any general recommendation against returns to Italy equivalent to that made in the case of Greece.

DISPOSAL

I would dismiss both appeals.

There is one observation that I wish to make by way of coda. I think it is a pity that the issues raised in these cases have had to be decided in the context of a judicial review challenge to a certification decision. In a certification challenge the Court has to focus not on what it believes is the right decision but on what a tribunal – which is in principle the primary forum for determination of the underlying issues – might decide: that is an inherently awkward exercise, and it carries the risk that if the Court believes that the answers are less than clear-cut the litigation will have to start all over again in the tribunal (as nearly happened in EM (Eritrea)). Also, the judicial review procedure is less well-adapted to deciding disputed issues of primary fact or expert evidence: indeed the types of issue raised by a case of this kind would be peculiarly suitable for the employment of a version of the “country guidance” procedure of the Upper Tribunal. To the extent that the Respondent enjoys a discretion whether to certify a claim of the present kind, I would urge her to consider very carefully whether that is always the best course, at least if a challenge is practically inevitable by one route or another. However, I appreciate that the scope for the exercise of any discretion is constrained by the way in which the applicable provisions are drafted: by paragraph 5 (4) of Schedule 3 to the 2004 Act she is positively obliged to certify the claim unless “satisfied that [it] is not clearly unfounded”.

On 4 March 2014 Mr Muižnieks (who was, it will be recalled, the Council of Europe’s Commissioner for Human Rights (Footnote: ¹⁸)) gave a lecture in London entitled “Refugee Protection, Migration and Human Rights in Europe: Notes from the Field”. It was wide-ranging in character but it included a paragraph on returns under the Dublin regime which reads as follows:

“Another challenge faced by those who reach Europe is the risk of ‘Dublin returns’ to countries whose asylum systems are dysfunctional. At the end of my thematic visit concerning Syrian refugees last December I urged EU member states to refrain from using the “Dublin Regulation” for returning Syrian refugees to other European countries whose asylum systems are already overstretched or highly dysfunctional, in particular Bulgaria, Greece, Italy and Malta. Bulgaria was caught unprepared for the influx, and its staffing levels and reception conditions are woefully inadequate. Shortly after my visit, UNHCR came to a similar conclusion about ‘Dublin returns’ to Bulgaria. I was glad to learn that on 5 February the Danish authorities announced their decision to suspend the return of asylum seekers to Bulgaria.”

In section III (a) of her first report Ms Leo summarised that passage by saying that in his lecture Mr Muižnieks had “urged European Union member states to not return refugees to EU countries whose asylum systems are not functioning, particularly, Bulgaria, Greece, Italy and Malta”. The Judge pointed out that Mr Muižnieks had in fact referred to the countries in question as countries where the asylum systems are “already overstretched or [my underlining] highly dysfunctional” and that there was a clear distinction between the two: he said that the Italian system was indeed plainly overstretched in March 2014 but that it did not follow, and Mr Muižnieks could not fairly be taken as saying, that it was “highly dysfunctional”.

Mr Husain denied that there was any inaccuracy in Ms Leo’s summary. The phrase which she had used was “not functioning” – not in fact, as the Judge’s criticism suggested, “highly dysfunctional” – and that was a fair summary of Mr Muižnieks’ language.

In my view the Judge’s point is formally correct. As he said, Mr Muižnieks deliberately used two phrases, and I think it likely that he included the reference to systems which were “overstretched” in order to recognise the distinction between Italy on the one hand and Greece and Bulgaria on the other, whose systems were generally recognised to be dysfunctional in a way that could not fairly be said of Italy’s: that is indeed essentially the distinction that was made in Tarakhel. Ms Leo’s summary ignores that distinction, and it makes no difference that she used the phrase “not functioning” rather than “highly dysfunctional”.

Mr Husain submits that even if there was a trivial inaccuracy of this kind it could not rationally be regarded as undermining Ms Leo’s credibility. I agree with that as far as it goes, but it must be recalled the Judge was relying on it as an example of a generally tendentious approach on her part, and it is necessary to look at all his examples, and indeed all three of the reports, together.

On 5 February 2015 the Southern Europe office of UNHCR published a Note containing recommendations on how Italy should transpose the recast Reception and Qualification Directives into its national law. The Note runs to some 28 pages. The version in the bundle is in English. Ms Leo says in her appeal statement that it represents a translation commissioned by the Secretary of State of an Italian original.

The introductory section of the Note contains the statement that “despite major improvements, problems remain in the Italian system in terms of both the law and its implementation, particularly as regards the reception and integration of recipients of international protection”, and reference is made to the July 2013 UNHCR report. A host of detailed recommendations follow. For present purposes I need only refer to a passage in the section dealing with the Reception Directive (p. 14). A formal recommendation is made that “the transposition of the reception conditions directive should be seen as an opportunity for creating uniform and consistent regulation of the reception system starting at the initial reception of new arrivals within mixed flows”. The supporting text includes the following passage:

“Experience in recent years has demonstrated the importance of moving beyond CARAs as reception centres. The problems created by the size and management of these centres have led to a worrying drop in reception standards. Originally designed to host asylum seekers for at most 25/30 days, the lack of places across Italy and the slowdown in asylum procedures mean that seekers are held in them for extremely long periods (up to a year) at high cost to the state given the low quality of services they offer and the possibility of abuse and inefficiency within them and particularly their inability to provide the services that might assist the seekers through possible subsequent integration processes.”

In the section of her second report addressing Mr Dangerfield’s evidence as to the number of places available in CARAs, SPRARs and CSAs, Ms Leo deals inter alia with whether Dublin returnees could be accommodated in CARAs. She says that the standards in such centres are “very very low” and quotes a Parliamentary report about a particular centre. She then refers to the UNHCR Note, describing it as “highlighting the absolute need to do away with the CARA system” and proceeds to quote from a different translation of the passage which I have set out above.

As regards that evidence at para. 124 of his judgment Lewis J says:

“Reading Ms Leo's 2015 report, the reader would assume from the language that a fair summary of the UNCHR view is that the UNCHR considers that there is ‘an absolute need’ to do away with the CARA system. The full text, not exhibited to Ms Leo's statement, gives a very different view. In context, the document sets out recommendations for implementing the recast Directives. It notes that, despite major improvements, problems remain in relation to reception and integration. The first sentence of the paragraph containing the quotations relied upon by Ms Leo, but not cited by her says: ‘Experience in recent years has demonstrated the importance of moving beyond CARAs as reception centres’. That is far removed from the impression that Ms Leo seeks to give of the document that there is ‘an absolute need’ to do away with the CARA system.”

Mr Husain submits that that is an unfair criticism. Ms Leo says in her appeal statement that the difference between the phrases “do away with” and “move beyond” simply reflect two different translators’ (i.e. those employed by the Secretary of State to translate the Note and those employed by Wilsons to translate her statement) translation of the same Italian word (superamento). She was not intending to use any different language from UNHCR. As to the addition of the word “absolute”, she says that that was done in her own text rather than in the passage which she quoted, and it will accordingly have been quite apparent that it represented her own emphasis. She says that it is a legitimate emphasis. She points out that in any event in a later passage the English version of the Note recommends “abandoning the CARA model”, which seems pretty “absolute”.

I do not agree with most of the Judge’s criticism of this part of Ms Leo’s report. Even without the benefit of her explanation about the different translations I do not believe that there is, in this context, a substantial difference between “do away with” and “move on from”: the substance is that UNHCR was recommending that the Italian authorities abandon the CARA model because of the deficiencies to which it refers. I do not believe that that point required to be qualified by reference to other parts of the Note, which deal with other matters; and if, as seems to be implied, the Judge was intending to criticise Ms Leo for not exhibiting the full Note, which was a public document fully identified by her in a footnote, I cannot agree. I am not clear whether the Judge meant to say that Ms Leo did not quote the sentence beginning “Experience in recent years”; but if he did he was simply wrong, perhaps being misled by the different translations. The one point on which I can agree to some extent is that the use of the word “absolute”, even if it can be justified as accurate, may tend to suggest a degree of advocacy rather than strict impartiality; but it is no more than a small straw in the wind.

This criticism concerns the passage which I have set out at para. 201 above. At para. 126 of his judgment the Judge said this:

“Ms Leo's reports give the impression of being based upon research but, on analysis, they are not. In relation to arrivals at Rome airport, for example, Ms Leo notes in her 2014 report that a particular association is undertaking thorough research. The results of that research, if it is available, has not been presented to this court. Ms Leo then goes on to say that ‘is possible to predict what will be revealed by the research from the data gathered to date’. No details of that data are provided but Ms Leo goes on to conclude that no persons are offered any form of reception or social support at the airport. She then infers the same would be true of Milan, based on (unspecified) research done ‘a few years ago’. None of that could legitimately be viewed as properly sourced, objectively presented material.”

I agree with the Judge that this part of Ms Leo’s evidence was unsatisfactory but I believe that the way that he expresses his criticism is unfortunate. He was quite right to point out that it is unsatisfactory that the data behind her statement that the BIPs who were the subject of the study “were not offered any form of reception or social support” (Footnote: ¹⁹) was unavailable. It is not enough to say that the results of the research were not published: if the data was good enough for Ms Leo to rely on it in her evidence it could and should have been produced. Without it it is impossible to evaluate the reliability of her statement. But that is not the same as saying, as the Judge appears to do in his introductory sentence (though it may not be quite what he meant), that Ms Leo’s evidence was not based on research at all. On the contrary, there is no reason to doubt her statement that it was based on the three-month ASGI study: the problem is that she gives us no details. The Judge was also, with respect, wrong to describe the Milan research as “unspecified” and (which is the clear implication of his putting “a few years ago” in quotes) undated. The published report is identified in a footnote, with its date (2010). It is true that no details are given of the methodology, but that could presumably be ascertained by consulting the report.

In her appeal statement Ms Leo says that the Fiumicino research was in fact published on 20 March 2015, i.e. a few days before the hearing, under the title Il Sistema Dublino e Italia: un Rapporto in Bilico (“The Dublin System and Italy: a Relationship on the Edge”), to which I will refer as “the March 2015 report”. She says that it “confirmed the interim conclusions which I reflected in my [first] report”. Ms Leo’s third report, at the end of a short section headed “Assistance at the airport”, contains the single sentence “In this regard we also refer to what is stated in [the March 2015 report] pages 39-40” and gives the link. But the third report, as I have said, post-dated the hearing and there was nothing in that brief reference to alert the Judge to the fact that the March 2015 report constituted the final version of the research referred to in the first report; and if he had nevertheless followed it up the document was only available in Italian. It was not part of Mr Husain’s submissions that the Judge erred in not referring to the March 2015 report. (Footnote: ²⁰)

This criticism relates to the passage that I have set out at para. 200 above. At para. 126 of his judgment, as a further example of Ms Leo’s reports giving the impression of being based upon research when in fact they are not, Lewis J says about that passage:

“Similarly, Ms Leo refers … to a Ministry circular dated 19 March 2014 and gives a quote saying that all the centres of government and those backed by some local authorities are saturated. Ms Leo then says there is a waiting list and, based on ‘reports by some operators’, there is a waiting list of around 2-3 months. The identities, roles and numbers of such operators are unspecified. No properly sourced data giving an up to date picture of the position or demonstrating that there is such a waiting list is provided.”

My view about that criticism is similar to my view about the third. The Judge is right to point out that the reference to the length of the waiting list – presumably waiting lists – is unsatisfactory both because it is not specifically sourced and because it is very general. But a criticism that Ms Leo’s report is inadequately particularised is not the same as a criticism that it is not based on research at all – which would be a much more serious matter.

In her appeal statement Ms Leo says that if she had been asked to particularise the sources for her statement about waiting lists she would have explained that she had interviewed twelve SPRAR operators as part of her research for the ASGI project (and forty by the time that the research was finalised) and that what she said about waiting-lists was based on what they told her. But of course that information was not before the Judge. Nor does she say how many of the twelve (or the forty) operators reported waiting lists of two or three months: her report referred only to “some”.