LORD GOFF OF CHIEVELEY
I have had the advantage of reading in draft the speech prepared by my noble and learned friend, Lord Lloyd of Berwick. For the reasons he gives I would allow the appeal.
LORD SLYNN OF HADLEY
I have had the advantage of reading in draft the speech prepared by my noble and learned friend Lord Lloyd of Berwick.
As to the first issue raised in the case, there is, it seems to me, force in Mr. Blake Q.C.'s argument that on humanitarian grounds a person who leaves his own country because of a well-founded fear of being persecuted for a Convention reason and later is unable, or, owing to that fear is unwilling, to avail himself of that country's protection even when the grounds for his fear have gone, should be able to claim the status of a refugee.
I am satisfied, however, that the Geneva Convention in Article 1A(2), does not confer that status. The first matter to be established under the Article is that the claimant is outside the country of his nationality owing to a well-founded fear of persecution. That well-founded fear must, as I read it, exist at the time his claim for refugee status is to be determined; it is not sufficient as a matter of the ordinary meaning of the words of the Article that he had such fear when he left his country but no longer has it. Since the second matter to be established, namely that the person "is unable or, owing to such fear, is unwilling to avail himself of the protection of that country" (emphasis added) clearly refers to an inability or unwillingness at the time his claim for refugee status is to be determined, it seems to me that the coherence of the scheme requires that the well-founded fear, the first matter to be established, is also a current fear. The existence of what has been called an historic fear is not sufficient in itself, though it may constitute important evidence to justify a claim of a current well-founded fear.
Like Lord Lloyd of Berwick I also attach importance to the passages in Professor James Hathaway's book set out in his speech and to the state practice recommended in the Joint Position dated 4 March 1996 of the Council of the European Union although the latter is not conclusive.
Reference has been made in argument to Article 1C(5) of the Convention. That Article is, however, dealing only with the situation where a person has qualified as a refugee but (a) the circumstances have changed so that he has no longer a well-founded fear of persecution for a Convention reason, and (b) the protection of the country of his nationality is available. If (a) is satisfied then he cannot say that he is unwilling because of the previous fear to accept the protection of his country of nationality. I do not think that the special circumstances in which the Convention is said to cease to apply to someone who was within Article 1A assist in determining the general question as to whether for the purposes of Article 1A(2) it is a current or an historic fear which has to be proved.
As to the second issue there is on the face of it more difficulty once it is accepted, as on the authorities and in principle it must be accepted, that there can be persecution of a group and that the individual in the group does not have to show that he has a fear of persecution distinct from, or over and above, that of his group. Thus if in a state two groups exist, A and B, and members of group A threaten to or do persecute members of group B the latter should, other necessary matters being established, be able to claim refugee status. If at the same time members of group B are persecuting or threatening to persecute members of group A the claim should be the same. The position is even stronger if the persecution is not exactly simultaneous but those in power change from time to time so that the persecutors become the persecuted.
Looking, however, at the language of the Convention and its object and purpose I do not consider that it applies to those caught up in a civil war when law and order have broken down and where, as in the present case, every group seems to be fighting some other group or groups in an endeavour to gain power. In such a situation what the members of each group may have is a well-founded fear not so much of persecution by other groups as of death or injury or loss of freedom due to the fighting between the groups. In such a situation the individual or group has to show a well-founded fear of persecution over and above the risk to life and liberty inherent in the civil war. The line may be a fine one to draw in some situations but I agree with my noble and learned friend that the Immigration Appeal Tribunal was entitled in the present case to find that such persecution over and above the risk of the civil war was not established, though I share his satisfaction that on the facts of this case exceptional leave to remain in the United Kingdom has been given to the Respondent, his wife and children.
Accordingly I too would allow the appeal of the Secretary of State.
LORD LLOYD OF BERWICK
Hassan Hussein Adan is a Somali national who fled from Somalia in June 1988 owing to a well-founded fear of persecution at the hands of the then government. On 15 October 1990 he arrived in the United Kingdom with his wife and two children. He was refused asylum on arrival, but he and his family were granted exceptional leave to remain. There is no question of Mr. Adan being returned to Somalia as things stand.
But there are certain benefits in being accorded refugee status, which are not available to those who have exceptional leave to remain. These are well described in the judgment of Simon Brown L.J. I need not repeat them. Mr. Adan wishes to take advantage of those benefits. He claims that he is entitled to refugee status under article 1A(2) of the Geneva Convention. Mr. Pannick Q.C. for the Secretary of State submits that Mr. Adan is not entitled to refugee status because he no longer has any fear of persecution. There has been a change of government in Somalia. President Barre has fallen from power.
Article 1 of the Convention provides:
Mr. Blake, for the respondent, submits that it is unnecessary for Mr. Adan to show a present fear of persecution. It is enough that he had a fear of persecution when he left Somalia ("historic fear"), so long as the historic fear is the cause of his being outside his country today, and so long as he is unable to avail himself of his country's protection.
There is a second issue. Mr. Blake draws attention to the political situation in Northern Somalia where the local clans are engaged in civil war. If Mr. Adan were to return to Somalia he would be in danger of his life owing to his membership of one of the warring clans. Mr. Blake argues that this amounts to persecution for a Convention reason, of which he has a current well-founded fear. If so then he would be entitled to succeed on the second issue, even if he fails on the first.
Mr. Pannick on the other hand submits that in a state of civil war between clans, where everybody is subject to the ordinary risks of war, then a person who is at no greater risk than anybody else, whether members of his own clan or any other clan, cannot claim the protection of the Convention. Such a person may indeed be in fear for his life. But it cannot be said that he is in fear of persecution.
Simon Brown L.J. and Hutchison L.J. decided the first issue in favour of Mr. Adan, with Thorpe L.J. dissenting. The court was unanimous in deciding the second issue in favour of Mr. Adan. The Secretary of State now appeals.
It is common ground that the words in square brackets in article 1A(2), which were repealed by the 1967 Protocol, can be ignored. They throw no light on the true construction of the article.
It was also common ground that article 1A(2) covers four categories of refugee:
(1) nationals who are outside their country owing to a well-founded fear of persecution for a Convention reason, and are unable to avail themselves of the protection of their country;
(2) nationals who are outside their country owing to a well-founded fear of persecution for a Convention reason, and, owing to such fear, are unwilling to avail themselves of the protection of their country;
(3) non-nationals who are outside the country of their former habitual residence owing to a well-founded fear of persecution for a Convention reason and are unable to return to their country, and
(4) non-nationals who are outside the country of their former habitual residence owing to a well-founded fear of persecution for a Convention reason, and, owing to such fear, are unwilling to return to their country.
It will be noticed that in each of categories (1) and (2) the asylum- seeker must satisfy two separate tests: what may, for short, be called "the fear test" and "the protection test." In categories (3) and (4) the protection test, for obvious reasons, is couched in different language.
Mr. Blake's case is that Mr. Adan falls within category (1). He left Somalia because of a well-founded fear of persecution. He is outside Somalia now because of that well-founded fear in the past, and he has been unable to avail himself of the protection of his country at any time since he left. He submits that the words "any person who . . . owing to well-founded fear . . . is outside the country of his nationality" is capable, linguistically, of including those who have had a fear of persecution in the past, as well as those who have a present fear. If the words are confined to those with a present fear of persecution, then the words "and is unable . . . to avail himself of the protection of that country" are otiose. They serve no purpose. This latter argument is the one which weighed most heavily with the majority of the Court of Appeal, and which in the end tipped the balance in what they clearly regarded as a difficult case.
In addition, Mr. Blake relies on the travaux préparatoires. He took us on a voyage of discovery through the early drafts of the Convention. He pointed out, for example, that in a draft prepared and circulated by the Economic and Social Council on 16 August 1950 there appeared the words "who has had, or has, well-founded fear of being the victim of persecution." The words "has had, or has," dropped out in the draft circulated by the Third Committee on 12 December 1950, and did not reappear in the final draft. While these changes are no doubt of interest to historians, from a lawyer's point of view they are inconclusive. For we do not know why the changes were made. All we know is that successive drafts (as one would expect) were subject to continual changes in the light of comments by governments and specialist agencies. It may be therefore that the changes in language were intended to reflect a change in substance. Or it may be that they were intended to reflect the same meaning in different words. We do not know. In these unfavourable circumstances your Lordships did not feel it necessary to call on Mr. Pannick in reply on the travaux préparatoires.
I return to the argument on construction. Mr. Pannick points out that we are here concerned with the meaning of an international Convention. Inevitably the final text will have been the product of a long period of negotiation and compromise. One cannot expect to find the same precision of language as one does in an Act of Parliament drafted by parliamentary counsel. I agree. It follows that one is more likely to arrive at the true construction of article 1A(2) by seeking a meaning which makes sense in the light of the Convention as a whole, and the purposes which the framers of the Convention were seeking to achieve, rather than by concentrating exclusively on the language. A broad approach is what is needed, rather than a narrow linguistic approach.
But having said that, the starting-point must be the language itself. The most striking feature is that it is expressed throughout in the present tense: "is outside . . . ," "is unable . . . ," "is unwilling . . . ." Thus in order to bring himself within category (1) Mr. Adan must show that he is (not was) unable to avail himself of the protection of his country. If one asks "protection against what?" the answer must surely be, or at least include, protection against persecution. Since "is unable" can only refer to current inability, one would expect that the persecution against which he needs protection is also current (or future) persecution. If he has no current fear of persecution it is not easy to see why he should need current protection against persecution, or why, indeed, protection is relevant at all.
But the point becomes even clearer when one looks at category (2), which includes a person who is (a) outside the country of his nationality owing to a well-founded fear of persecution and (b) is unwilling, owing to such fear, to avail himself of the protection of that country. "Owing to such fear" in (b) means owing to well-founded fear of being persecuted for a Convention reason. But "fear" in (b) can only refer to current fear, since the fear must be the cause of the asylum-seeker being unwilling now to avail himself of the protection of his country. If fear in (b) is confined to current fear, it would be odd if "owing to well-founded fear" in (a) were not also confined to current fear. The word must surely bear the same meaning in both halves of the sentence.
I turn from these linguistic points to the more general point which concerned the majority of the Court of Appeal. They considered that if "owing to well-founded fear . . . is outside" is confined to current fear, then "is unable . . . to avail himself" would serve no purpose. If category (1) were confined to refugees who are subject to state persecution, then I can well see that such persons would, ex hypothesi, be unable to avail themselves of state protection. On that view the words would indeed serve no purpose. But category (1) is not so confined. It also includes the important class of those who are sometimes called "third party refugees," i.e. those who are subject to persecution by factions within the state. If the state in question can make protection available to such persons, there is no reason why they should qualify for refugee status. They would have satisfied the fear test, but not the protection test. Why should another country offer asylum to such persons when they can avail themselves of the protection of their own country? But if, for whatever reason, the state in question is unable to afford protection against factions within the state, then the qualifications for refugee status are complete. Both tests would be satisfied.
So I would not regard the inclusion of a protection test as standing in the way of Mr. Pannick's construction. On the contrary, I consider that the existence of the protection test, couched as it is in the present tense, adds positive support for the view that "owing to well-founded fear" is also confined to current fear. In this way the two halves of the sentence are linked together.
That brings me to article 1C which provides:
I had at first thought that article 1C(5) provided a complete answer to Mr. Blake's argument. If a present fear of persecution is an essential condition of remaining a refugee, it must also be an essential condition for becoming a refugee. But it was pointed out in the course of argument that article 1C(5) only applies to refugees in category (2). It does not help directly as to refugees in category (1). This is true. But the proviso does shed at least some light on the intended contrast between article 1A(1) and 1A(2). Article 1A(1) is concerned with historic persecutions. It covers those who qualified as refugees under previous Conventions. They are not affected by article 1C(5) if they can show compelling reasons arising out of previous persecution for refusing to avail themselves of the protection of their country. It would point the contrast with article 1A(1), and make good sense, to hold that article 1A(2) is concerned, not with previous persecution at all, but with current persecution, in which case article 1C(5) would take effect naturally when, owing to a change of circumstance, the refugee ceases to have a fear of current persecution.
Mr. Pannick also founded an argument on article 33. But for my part I found the argument unconvincing. As Simon Brown L.J. said in the Court of Appeal, it approaches the question from the wrong end. It throws no light on the definition of refugee in article 1A(2).
I now turn to the authorities on the first issue. There is no judicial authority which is directly in point, other than In re C, 22 September 1997, a very recent unreported decision of the New Zealand Refugee Status Appeals Authority. In a lengthy, and carefully reasoned judgment, the Authority concluded that the majority decision of the Court of Appeal in the present case was wrong in law, and was not to be followed in New Zealand.
Of equal and perhaps of greater importance are the views of academic writers, since it is academic writers who provide the best hope of reaching international consensus on the meaning of the Convention. One of the leading figures in the academic field is Professor James Hathaway. In The Law of Refugee Status published in Canada in 1991 he says, at pp. 68-69:
In a document headed Joint Position dated 4 March 1996 the Council of the European Union adopted certain guidelines for the application of article 1 of the Convention. Paragraph 3 provides:
Paragraph 9.1 reads:
These and other passages indicate that the essential criterion for determining refugee status (other than refugees covered by article 1A(1)) is a current well-founded fear of persecution for a Convention reason.
But even more significant than the positive support for Mr. Pannick's construction of article 1A(2) among the academic writers is the complete absence of any support for Mr. Blake's construction, whether in Professor Hathaway's book, or in other academic writings, or in the United Nations Handbook, or elsewhere. So far as I am aware the suggestion that anything other than a current fear of persecution will suffice has never even been mooted.
So with great respect to the majority of the Court of Appeal, I would hold that on the first issue the views of Thorpe L.J. are to be preferred. I am glad to have reached that conclusion. For a test which required one to look at historic fear, and then ask whether that historic fear which, ex hypothesi, no longer exists is nevertheless the cause of the asylum-seeker being presently outside his country is a test which would not be easy to apply in practice. This is not to say that historic fear may not be relevant. It may well provide evidence to establish present fear. But it is the existence, or otherwise, of present fear which is determinative. I would therefore answer the first issue in favour of the Secretary of State.
The second issue raises more difficult questions, at least in theory, since, if Mr. Pannick is right, it involves drawing a line between the persecution of individuals and groups, including very large groups, on the one hand, and the existence of a state of civil war on the other. Mr. Pannick accepts that protection under the Convention is not confined to individuals. He accepts further that the persecution of individuals and groups, however large, because of their membership of a particular clan is very likely to be persecution for a Convention reason. But he says that where there is a state of civil war between clans, the picture changes. Otherwise the participants on both sides of the civil war would be entitled to protection under the Convention. Indeed, as Simon Brown L.J. pointed out, the only persons who would not be entitled to protection, on that view, would be those who were not the active participants on either side but were, as Simon Brown L.J. put it, "lucklessly endangered on the sidelines." Simon Brown L.J. found this unappealing. So do I. It drives me to the conclusion that fighting between clans engaged in civil war is not what the framers of the Convention had in mind by the word persecution.
What then is the critical factor which distinguishes persecution from the ordinary incidents of civil war? Mr. Blake sought to draw a distinction between the armed forces on either side, who would, he said, be governed by the rules of war, and the targeting of individual civilians or groups of civilians. I doubt, however, whether, in the context of clan warfare in Somalia, it is realistic to think in terms of rules of war, or the conventional distinction between civilians and members of the armed forces. Mr. Adan's own evidence was that most of the population is armed.
Mr. Blake is nearer the mark when he refers to the targeting of civilians or groups of civilians. If an asylum-seeker can show that he is being targeted for Convention reasons, other than his membership of one of the warring clans, then he might qualify for refugee status. This indeed comes near to Mr. Pannick's submission. In a state of civil war between clans an asylum seeker must be able to show that he is at greater risk of ill-treatment than other members of his clan. There must, he said, be a differential impact.
One can find a good deal of authority to support Mr. Pannick's submission. In Salibian v. Canada (Minister of Employment and Immigration) (1990) 3 F.C. 250 the Federal Court of Appeal in Canada was concerned with a claim for refugee status by a citizen of the Lebanon. The Refugee Division had decided that the plaintiff was not entitled to refugee status because there was no evidence that he personally had been singled out for persecution. He was a victim of the disruption in the Lebanon, like all other Lebanese citizens. The Court of Appeal held that the Refugee Division had fallen into error, both in fact and law: in law, because it is unnecessary for a victim of persecution to show that the persecution has been directed against him in particular; in fact, because the evidence was that he had suffered persecution, not as a Lebanese citizen, but as an Armenian and a Christian. The court stated the following proposition, among others, as having been established by previous authority:
In support of that proposition, the court cited two passages from Professor Hathaway's book. The second at p. 97 was as follows:
This passage, with its rejection of a differential test ("the issue is not whether the claimant is more at risk than anyone else in her country") might at first sight appear to be inconsistent with Mr. Pannick's argument. He hinted that Professor Hathaway might be wrong. But the passage is not dealing with a country in a state of civil war. It is disposing of an argument that had prevailed in early decisions both in Canada and the United States that to qualify for refugee status the asylum-seeker had to have been "singled out." It was for this reason, no doubt, that it was cited with approval in Salibian. It is now accepted that generalised oppression may indeed give rise to refugee status, as Professor Hathaway makes clear. It is not necessary for a claimant to show that he is more at risk than anyone else in his group, if the group as a whole is subject to oppression. This is clearly right. But it does not touch on the more difficult questions which arise when a country is in a state of civil war. Professor Hathaway deals with these problems in a later chapter.
At p. 185 he states the first of two "essential points" as follows:
At pp. 186-187 he states the following general proposition:
To the same effect is paragraph 164 of the United Nations Handbook which provides:
Finally one can refer again to the Council document dated 4 March 1996. Para. 6 provides:
With regard to the passage last quoted, I would not agree that fear of persecution must be individual in nature. As the decision of Taylor J. in Reg. v. Secretary of State for the Home Department, Ex parte Jeyakumaran  Imm.A.R. 45 shows, and the Salibian case confirms, the Convention encompasses the persecution of groups as well as individuals. But otherwise I agree.
I conclude from these authorities, and from my understanding of what the framers of the Convention had in mind, that where a state of civil war exists, it is not enough for an asylum-seeker to show that he would be at risk if he were returned to his country. He must be able to show what Mr. Pannick calls a differential impact. In other words, he must be able to show fear of persecution for Convention reasons over and above the ordinary risks of clan warfare.
What I have said so far applies only so long as the state of civil war continues. Once the civil war is over, and the victors have restored order, then the picture changes back again. There is no longer any question of both sides claiming refugee status. If the vanquished are oppressed or ill-treated by the victors, they may well be able to establish a present fear of persecution for a Convention reason, and in most cases they would be unable to avail themselves of their country's protection.
Obviously it may prove difficult, in the case of warring clans, to establish precisely when one side or the other has won. By way of example, Professor Goodwin-Gill in his book The Refugee in International Law, 2nd ed. (1996), p. 76 cites a number of French and German decisions in which a distinction is drawn between the civil war in Somalia and the civil war in Liberia, on the ground that in the former country none of the competing clans has yet emerged "as an authority in fact, controlling territory and possessing a minimum of organisation." I agree with the Court of Appeal that refugee status ought not to depend "on casting around for the current underdog." But the difficulty of establishing the facts does not undermine the principle that those engaged in civil war are not, as such, entitled to the protection of the Convention so long as the civil war continues, even if the civil war is being fought on religious or racial grounds. In so far as the second issue is capable of a generalised answer, I feel bound to disagree with the Court of Appeal, and answer it in favour of the Secretary of State.
I turn to the facts. The political situation in northern Somalia is well described in the decision of the Special Adjudicator. It is unnecessary for me to go into any detail. The Special Adjudicator found that Mr. Adan had a well-founded fear of persecution by the government when he left Somalia in June 1988. But she also found that he no longer had a well-founded fear from that source at the time of her decision, owing to the change of regime. Instead he had a well-founded fear of persecution from a different source, namely, the opposing forces in the civil war. "The agents of persecution in the case of this appellant" she said "are not the authorities of the country but the members of the armed groups or militias of other clans or alliances." Accordingly she held that Mr. Adan was entitled to refugee status.
The Immigration Appeal Tribunal ("I.A.T.") disagreed. There could be no doubt as to the existence of a state of civil war in northern Somalia. But that would not, in the I.A.T.'s view, be enough by itself to give Mr. Adan a well-founded fear of persecution. I quote the critical paragraph in the I.A.T.'s Decision and Reasons:
Mr. Adan's evidence was that members of his own sub-clan were particularly at risk because they had attacked a militia stronghold of the main opposing sub-clan. But I do not consider that this throws doubt on the I.A.T.'s conclusion that all sections of society in northern Somalia are equally at risk so long as the civil war continues. There is no ground for differentiating between Mr. Adan and the members of his own or any other clan.
If I am right in the answer I have given to the two issues of principle, it follows that the I.A.T. were justified in differing from the Special Adjudicator. Mr. Adan is not entitled to refugee status and the Court of Appeal were wrong to hold otherwise.
Less there be any misunderstanding, I repeat what I said at the outset: there is no question of Mr. Adan being returned to Somalia as things stand. He and his wife and children have been given exceptional leave to remain in the United Kingdom on humanitarian grounds. The only effect of a decision to refuse refugee status is that they will be denied the additional benefits which refugee status attracts.
I have had the advantage of reading in draft the speech prepared by my noble and learned friend, Lord Lloyd of Berwick. For the reasons he gives I would allow the appeal.
LORD HOPE OF CRAIGHEAD
I have had the advantage of reading in draft the speech which has been prepared by my noble and learned friend, Lord Lloyd of Berwick. I agree with what he has said on both issues, and for the reasons which he has given I also would allow the appeal.