In re B (FC) (Appellant) (2002)
Regina v. Special Adjudicator (Respondent) ex parte Hoxha (FC) (Appellant)
60. True it is that 1C (5), no less than 1A (2), appears in the Convention under the heading "Definition of the Term 'Refugee'". True it is, too, as para 28 of the Handbook neatly points out, that someone recognised to be a refugee must by definition have been one before his refugee status has been determined. But it by no means follows that, because someone has been a refugee before his status comes to be determined, any change in circumstances in his home country falls to be considered under 1C (5) rather than under 1A (2). Quite the contrary. As has been seen, the Handbook is replete with references to the "determination" of a person's refugee status and his "recognition" as such. Article 9 of the Convention itself, indeed, allows certain provisional measures to be taken "pending a determination by the Contracting State that that person is in fact a refugee". The whole scheme of the Convention points irresistibly towards a two-stage rather than composite approach to 1A (2) and 1C (5). Stage 1, the formal determination of an asylum-seeker's refugee status, dictates whether a 1A (2) applicant (who may, indeed, be someone previously held not to qualify as a statutory refugee by the International Refugee Organisationsee the second paragraph of 1A (1)), is to be recognised as a refugee. 1C (5), a cessation clause, simply has no application at that stage, indeed no application at any stage unless and until it is invoked by the State against the refugee in order to deprive him of the refugee status previously accorded to him.
61. Para 112 of the Handbook makes all this perfectly plain. So too, more recently, did the UNHCR Lisbon Roundtable Meeting of Experts held in May 2001 in their Summary Conclusions:
62. Many other of the documents and writings put before your Lordships point the same way. And so, of course, does the language of 1C (5) itself. The words "the circumstances in connection with which he has been recognised as a refugee" could hardly be clearer. They expressly postulate that the person concerned "has been recognised as a refugee", not that he "became" or "was" a refugee.
63. This provision, it shall be borne in mind, is one calculated, if invoked, to redound to the refugee's disadvantage, not his benefit. Small wonder, therefore, that all the emphasis in paras 112 and 135 of the Handbook is upon the importance of ensuring that his recognised refugee status will not be taken from him save upon a fundamental change of circumstances in his home country. As the Lisbon Conference put it in para 27 of their conclusions: " the asylum authorities should bear the burden of proof that such changes are indeed fundamental and durable".
64. Many other UNHCR publications are to similar effect. A single further instance will suffice, taken from the April 1999 Guidelines on the application of the cessation clauses:
65. The reason for applying a "strict" and "restrictive" approach to the cessation clauses in general and 1C (5) in particular is surely plain. Once an asylum application has been formally determined and refugee status officially granted, with all the benefits both under the Convention and under national law which that carries with it, the refugee has the assurance of a secure future in the host country and a legitimate expectation that he will not henceforth be stripped of this save for demonstrably good and sufficient reason. That assurance and expectation simply does not arise in the earlier period whilst the refugee's claim for asylum is under consideration and before it is granted. Logically, therefore, the approach to the grant of refugee status under 1A (2) does not precisely mirror the approach to its prospective subsequent withdrawal under 1C (5).
66. That said, however, it would seem to me appropriate that in the initial determination of an asylum claim under 1A (2) the decision-maker, in a case where plainly the applicant fled his home country as a genuine refugee from Convention persecution, should not too readily reach the view that he could now safely be returned to it. Not only, as both Lord Slynn and Lord Lloyd observed in Adan  1 AC 293, may historic fear constitute important evidence tending to establish a current fear; so too it justifies some scepticism on the part of the decision-maker as to whether in truth the change in home circumstances is sufficiently clear and firm as to warrant the refusal of refugee status. That essentially is the point I was trying to make in the Court of Appeal in Mohammed Arif v Secretary of State for the Home Department  Imm AR 271 where, at p 276, I suggested that, depending always on the particular facts of the case, there might well be "an evidential burden on the Secretary of State to establish that [the asylum seeker] could safely be returned home." Although "some reservations as to the utility of the language of burden of proof" were expressed in the later Court of Appeal decision in S v Secretary of State for the Home Department  INLR 416,431, I remain unrepentant. It seems to me only right that in a case where the Secretary of State is contending that a country once plainly unsafe (like, say, Sri Lanka or Kosovo) has now become safe, he should place before the appellate authority sufficient material to satisfy them of that critical fact. There can, of course, be no doubt that the Arif approach was satisfied here. As Jackson J observed in Mr Hoxha's case: "In cases arising from Kosovo the Secretary of State can discharge that evidential burden by pointing to the presence of UNMIK and KFOR since June 1999". (June 1999, your Lordships may note, was before either appellant in fact arrived in this country.)
67. From all this it follows that, even were the proviso to 1C (5) capable of availing the appellants, their appeals must fail since clearly 1C (5) has no application. Recognising, however, that were the proviso indeed to encompass them as well as statutory refugees this would involve a substantially greater (and in this event rather disturbing) mismatch between the approach respectively to recognition under 1A (2) and cessation under 1C (5), it is right for your Lordships to address the point.
68. The appellants advance two principal arguments with regard to the proviso, pointing first to its drafting history and secondly to current state practice. As to the drafting history there can be no doubt whatever that, when first the Convention took effect, the proviso was intended to apply, as indeed it expressly states, to statutory refugees only. As Grahl-Madsen said in The Status of Refugees in International Law, vol 1 (1966) p410:
69. It is the appellants' submission, however, that the 1967 Protocol, recognising as it did the need thenceforth to deal also with new refugee situations and not just those which had arisen before 1 January 1951, must be recognised as having changed the scope of the proviso notwithstanding the Protocol's failure actually to amend it. The appellants point to the Preamble to the Protocol"it is desirable that equal status should be enjoyed by all refugees covered by the definition in the Convention irrespective of the dateline, 1 January 1951"and urge too the great importance accorded generally in international law to the principle of non-discrimination. To my mind, however, the argument is an impossible one. The language of the proviso is clear and unambiguous. The failure in 1967 to amend it, as it could so easily have been amended simply by deleting the "(1)" from the reference to "section A (1)", seems to me eloquent of the continuing intention of the contracting parties to confine the benefit of the proviso to the diminishing number of statutory refugees who would otherwise have remained vulnerable to the loss of their Convention rights under the cessation clause. (It may be noted for good measure that the Protocol, by article 1 (3), expressly preserved certain geographical limitations on the Convention's scope, namely any declarations previously made by Contracting States under article 1 B limiting their Convention obligations to those claiming refugee status under 1A (2) from "events occurring in Europe".)
70. Professor Guy Goodwin-Gill, having pointed out that the UNHCR statute on the Convention itself dealt differently with the position, describes the Convention's limitation of the right to invoke compelling reasons for non-return to statutory refugees as "perverse" (The Refugee in International Law, 2nd ed (1996), p 87). To my mind, however, that puts it too high: one can surely understand why some Contracting States at least would wish to limit the proviso's scope. There is, indeed, a striking modern example of precisely this evident reluctance to extend the range of Convention protection. Despite the EU Commission's 2001 proposal in connection with the then proposed Council Directive on the Minimum Standards for Refugee Protection that any "Member State invoking this cessation clause should ensure that an appropriate status, preserving previously acquired rights, is granted to persons who are unwilling to leave the country for compelling reasons arising out of previous persecution or experiences of serious and unjustified harm", the Directive as enactedCouncil Directive 2004/83/ECconspicuously omitted any such provision. Instead, all that article 11.2 requires with regard to the relevant cessation clauses is that "Member States shall have regard to whether the change of circumstances is of such a significant and non-temporary nature that the refugee's fear of persecution can no longer be regarded as well-founded."
71. I conclude, therefore, that the appellants' argument derives no support whatever from the drafting history of the proviso. The 1967 Protocol, indeed, so far from advancing their case, constitutes yet another obstacle in their path.
72. I turn therefore to the second limb of the appellant's argument, Mr Manjit Gill QC's reliance on state practice. The argument arises under article 31 of the Vienna Convention on the Law of Treaties, 1969:
73. Sir Ian Sinclair QC, in The Vienna Convention on the Law of Treaties, 2nd ed, (1984) says at p138:
74. Anthony Aust in Modern Treaty Law and Practice (2000) states, at p 195:
75. That, submit the appellants, with some support from Professor Goodwin Gill, puts it too high: universal agreement by all the treaty parties is not necessary; a state can be bound by the general practice of other states even against its wishes.
76. For the purposes of the present appeals it seems to me unnecessary to resolve the apparent differences between the several commentators upon the degree of uniformity of approach necessary in the case of multi-lateral treaties to support an interpretation based on state practice. Having regard to the clarity of the "ordinary meaning" born by the proviso to 1C (5), only the most compelling case founded on "subsequent practice" could properly give rise to a different and apparently contradictory interpretation from that obviously first intended. Until very recently, as we shall see, none of the many UNHCR pronouncements down the years appeared to support such a case. So far from suggesting that, as a result of state practice, contracting states had become obliged to treat the proviso as extending to all refugees, one repeatedly finds instead the language of aspiration and exhortation. Paragraph 26 of the Handbook expresses "the hope" that nations will grant refugee protection beyond the Convention's "contractual scope". Paragraph 27 characterises this as a "recommendation". Paragraph 136 points out encouragingly that the proviso, reflecting as it does a general humanitarian principle, "could" also be applied to A1 (2) refugees. Executive Committee Conclusion 69 of 1992:
77. Paragraph 31 of the UNHCR guidelines of 1999 states:
78. A change of view, however, appears to have overtaken the UNHCR at around the time of the Lisbon Conference in 2001. Para 18 of the Roundtable's conclusions states:
This statement was repeated verbatim in guidelines issued by the UNHCR in February 2003see para 21. This was supported solely by a footnoted reference to a forthcoming UNHCR publication: Refugee Protection in International Law edited by Feller, Türk and Nicholson. At p 32 of that publication (which came out later in 2003) appears this:
79. Those 2003 publications, it may be noted, post-date the Court of Appeal's judgment in this case. In turn they were followed by a research paper, "The Applicability of the 'Compelling Reasons' exception to cessation for refugees and asylum-seekers" published under the auspices of UNHCR as recently as November 2004. The findings of that paper were summarised as follows:
80. In the International Journal of Refugee Law published in January 2004, Vol 16, No 1, p 92 in an article "Exemption from Cessation of Refugee Status in the Second Sentence of article 1C(5)/(6) of the 1951 Refugee Convention", David Milner had expressed the view that "whilst it is clearly the case that extension of the exemption to Convention refugees is indeed well established, it is harder to argue that it has 'generally been accepted', and the practice is certainly far from being universal." The November 2004 subsequent research document refers to Mr Milner's article but suggests that "despite [its] recent date of publication some of the information regarding national legislation and case law is outdated."
81. It is upon these recent UNHCR publications and above all the 2004 research findings that Mr Manjit Gill founds his main argument on this limb of the case. They invalidate, he submits, an important finding in the judgment below,  1 WLR 241, 254, para 47:
The research is also relied on by Mr Tim Eicke in his written case on behalf of the UNHCR as Intervener. It shows, he submits (at para 10.12):
82. For my part I would reject these submissions. No doubt the Court of Appeal's judgment can now be seen to have put the appellants' case too low. But the Intervener to my mind puts it too high. I do not regard 15 out of 20 as a "vast majority" and nor do I think it capable of overriding the obvious disagreement of countries as important to the asylum system as Australia, New Zealand and the UK. Nor, indeed, is the appellants' difficulty in establishing the necessary consistency of state practice solely one of numbers. It is noteworthy too that both Feller, Türk and Nicholson and the subsequent summary of research findings describe the grant or continuation of refugee status merely as "best state practice", acknowledging in terms that a number of other states (even amongst the majority who give general effect to the "compelling reasons" exception) do so only by according "subsidiary statuses" to those they let stay. These grants of permission do not necessarily provide a secure legal status or preserve acquired rights. This "subsidiary form of protection," as the research paper calls it, manifestly falls short of continuing refugee status and is inconsistent therefore with a strict Convention obligation to treat A1 (2) refugees in the same way as A1 (1) refugees for the purposes of the 1C (5) proviso. On this ground too, therefore, the state practice argument is in my opinion unsustainable.
83. As Lord Bingham of Cornhill observed in Sepet v Secretary of State for the Home Department  1 WLR 856, 864, para 11, with regard to a claimed right of conscientious objection to compulsory military service as the basis for the grant of asylum in the light of the many expressions of international opinion supporting that view:
84. From paragraph 72 onwards I have been discussing state practice with regard to the proviso for all the world as if the UK's practice is to invoke 1C (5) irrespective of any "compelling reasons" not to do so. In fact, as Miss Carss-Frisk QC told us, there is only one recorded case of the Secretary of State ever invoking 1C (5) to remove a refugee and the scope of the proviso was not there in point. The reality appears to be that, once the UK grants refugee status, it does not subsequently withdraw it however safe the home country may become. In truth, therefore, for the appellants to succeed in these appeals, they would have to establish that modern state practice requires not merely that refugee status be not withdrawn in "compelling reasons" cases but that it be granted in such cases, a proposition which, as already explained, flies directly in the face of Adan and is in any event irreconcilable with the UNHCR's repeated emphasis on treating the initial determination of refugee status and its subsequent cessation as "separate and distinct processes".
85. It is one thing to invite this House to construe the Convention as a living instrument generously and in the light of its underlying humanitarian purposes; quite another to urge your Lordships effectively to rewrite it so as to create a fresh entitlement to refugee status based upon no more than historic fear and present compelling reasons for non-return, with no need at all for any current fear of persecution. That would be to distort entirely the language and structure of the text and in my judgment do a serious disservice to the cause of human rights generally. As Lord Bingham of Cornhill said, first in Brown v Stott  1 AC 681, 703, and then again in R (European Roma Rights) v Prague Immigration Officer  2 WLR 1, para 18:
Lord Steyn too recognised in the European Roma Rights case (at para 43), the limitations upon the "very important principles of interpretation" set out in the Vienna Convention: " they are not capable of filling gaps which were designedly left in the protective scope of the Refugee Convention."
86. The Convention does not meet all humanitarian needs. It only avails those unable to return to their home country who have a present fear of persecution (Adan  1 AC 293). It does not avail conscientious objectors (Sepet  1 WLR 856). It cannot be invoked by those who have yet to leave their home country (European Roma Rights). It does not protect those whose life is threatened by famine, or civil war, or by persecution for non-Convention reasons. As Lord Hope of Craighead observed in Horvath v Secretary of State for the Home Department  1 AC 489, 498:
87. However compelling therefore may be the reasons why these appellants or others like them should not be returned to their home country, there is no basis in the Convention, absent a continuing well-founded fear, for granting them refugee status (or even, indeed, for allowing them to retain such status once granted). That, however, is not to deny such refugees all possibility of relief. Rather, as paragraph 72 of the respondent's written case makes plain:
88. A refusal of discretionary leave on compassionate grounds might itself, in an extreme case, be judicially reviewable. No such challenge is, however, mounted here. These appellants' claims stand or fall on their entitlement to refugee status. For the reasons given, reasons substantially the same as those appearing in Keene LJ's admirable judgment below, I would hold that their claims fail. I would accordingly dismiss these appeals.
Lords Parliament Commons Search Contact Us Index
|© Parliamentary copyright 2005||Prepared 17 March 2005|