Judgment - Islam (A.P.) v. Secretary of State for the Home Department Regina v. Immigration Appeal Tribunal and Another Ex Parte Shah (A.P.) (Conjoined Appeals)  continued

(back to preceding text)

Counsel for the Secretary of State also tried to rely on dicta by two members of the majority in the High Court of Australia in A. v. Minister of Immigration and Ethnic Affairs (1997) 142 A.L.R. 331. This case involved a claim to refugee status by a husband and wife who had come from China to Australia. They said that they feared sterilization under the "one child policy" of China if they were returned. The majority (Dawson, McHugh and Gummow J.J.) rejected the argument that the appellants were members of a particular social group. Brennan C.J. and Kirby J. dissented. Contrary to counsel's submission I consider it clear that Dawson J. did not accept the Sanchez-Trujillo theory. On the contrary, he said that Sanchez-Trujillo is unpersuasive so far as it suggested that "the uniting particular must be voluntary": at p. 341. Gummow J. may have adopted the Sanchez-Trujillo principle. And McHugh J. and the dissenting judges took a broader view. In any event, in a case such as A. v. Minister of Immigration and Ethnic Affairs a significant difficulty in the way of claimants to refugee status is the fact that the one child policy is apparently applied uniformly in China. There is no obvious element of discrimination. That may be the true basis of the decision of the Australian High Court. Far from assisting the argument of the Secretary of State the trend of the dicta in A.'s case (except for the observations of Gummow J.) is against a requirement of cohesiveness. Moreover, in Canada the Supreme Court has adopted a broader approach which depends on the reasoning in Acosta and is inconsistent with Sanchez-Trujillo: see Attorney-General of Canada v. Ward (1993) 103 D.L.R. (4th) 1; see also Chan v. Minister of Employment and Immigration 128 D.L.R. (4th) 213, 247. This is made explicit by La Forest J. in Ward. He said, at p. 34 that "social group" could include individuals fearing persecution on "such bases as gender, linguistic background and sexual orientation."

Apart from the judgment of Staughton L.J. in the present case, there is no English authority for the view that cohesiveness is an indispensable requirement for the existence of a "particular social group." Counsel for the Secretary of State cited the decision of the Court of Appeal in Savchenko v. Secretary of State for the Home Department [1996] Imm. A.R. 28. The judgments in that case contain references to Sanchez-Trujillo but no adoption of its reasoning on the element of cohesiveness. The ratio of Savchenko is that the alleged group (Russian security guards at a hotel who feared victimisation by the mafia) did not exist independently of the persecution. But MacCowan L.J. recorded, at p. 34:

     "The Secretary of State submits, we were told by Mr. Pannick, that the concept of membership of a particular social group covers persecution in three types of case: (1) membership of a group defined by some innate or unchangeable characteristic of its members analogous to race, religion, nationality or political opinion, for example, their sex, linguistic background, tribe, family or class; (2) membership of a cohesive, homogeneous group whose members are in a close voluntary association for reasons which are fundamental to their rights, for example, a trade union activist; (3) former membership of a group covered by (2)."

Paragraph (1) is in line with Acosta's case (1985) 19 I. & N. 211 and inconsistent with Sanchez-Trujillo's case (1986) 801 F.2d. 1571. It was not explicitly adopted by the Court of Appeal but it was also not rejected. In these circumstances Savchenko's case [1996] Imm.A.R. 28 cannot assist the argument of the Secretary of State. Counsel for the Secretary of State informed the House that the Secretary of State no longer supports his submission in paragraph (1). That is understandable: his earlier submission weakens his case on the present appeals.

The support in the case law for the Sanchez-Trujillo view is slender. In the literature on the subject there is no support: see the criticism in Hathaway, The Law of Refugee Status, (1991), at p. 161 (note 182.) Considering that view on its merits I am satisfied that for the reasons given in Acosta's case the restrictive interpretation of "particular social group" by reference to an element of cohesiveness is not justified. In 1951 the draftsmen of article 1A(2) of the Convention explicitly listed the most apparent forms of discrimination then known, namely the large groups covered by race, religion, and political opinion. It would have been remarkable if the draftsmen had overlooked other forms of discrimination. After all, in 1948 the Universal Declaration had condemned discrimination on the grounds of colour and sex. Accordingly, the draftsmen of the Convention provided that membership of a particular social group would be a further category. It is not "an all-encompassing residual category": Hathaway, The Law of Refugee Status, 1991, at p. 159. Loyalty to the text requires that one should take into account that there is a limitation involved in the words "particular social group." What is not justified is to introduce into that formulation an additional restriction of cohesiveness. To do so would be contrary to the ejusdem generis approach so cogently stated in Acosta. The potential reach of the Acosta reasoning may be illustrated by the case of homosexuals in countries where they are persecuted. In some countries homosexuals are subjected to severe punishments including the death sentence. In Re G.J. [1998] 1 N.L.R. 387 the New Zealand Refugee Status Authority faced this question. Drawing on the case law and practice in Germany, The Netherlands, Sweden, Denmark, Canada, Australia and the U.S.A., the Refugee Status Authority concluded in an impressive judgment that depending on the evidence homosexuals are capable of constituting a particular social group with the meaning of article 1A(2): see pp. 412-422. This view is consistent with the language and purpose of article 1A(2). Subject to the qualification that everything depends on the state of the evidence in regard to the position of homosexuals in a particular country I would in principle accept the reasoning in Re G.J. as correct. But homosexuals are, of course, not a cohesive group. This is a telling point against the restrictive view in Sanchez-Trujillo's case. Finally, the restrictive interpretation is at variance with the principle that a treaty ought to be construed in a purposive sense: see article 31 of the Vienna Convention on the Law of Treaties.

Given the unequivocal acceptance by Staughton L.J. of the restrictive theory of the interpretation of article 1A(2) I have thought it right to explain at some length why in my view his conclusion was not justified. In oral argument counsel for the Secretary of State expressly conceded that Staughton L.J. erred in ruling that cohesiveness is an indispensable requirement. Instead counsel submitted that "particular social group" normally requires cohesiveness. What the practical implications of this qualification are I do not know. For my part the position is as follows. Cohesiveness may prove the existence of a particular social group. But the meaning of "particular social group" should not be so limited: the phrase extends to what is fairly and contextually inherent in that phrase.

The second issue: The different theories of "particular social group."

In oral argument different foundations for treating the two appellants as members of a particular social group were explored and tested. First, counsel for the appellants argued that three characteristics set the appellants apart from the rest of society viz gender, the suspicion of adultery, and their unprotected status in Pakistan. He submitted that this combination of characteristic exists independently of persecution. Secondly, while counsel UNHCR made no submissions on the merits of the cases of the appellants he placed before the House general submissions as to the meaning of "particular social group." He submitted that individuals who believe in or are perceived to believe in values and standards which are at odds with the social mores of the society in which they live may, in principle, constitute "a particular social group" within the meaning of article 1A(2). Women who reject those mores--or are perceived to reject them--are capable of constituting "a particular social group." The third way of approaching the matter was suggested in argument by my noble and learned friend Lord Hoffmann. It involves the proposition that women in Pakistan are a particular social group. ˇKˇˇKCounsel for the Secretary of State pointed out that this way of approaching the case is a new development. But it was thoroughly explored and tested in oral argument in the House. In these circumstances it must be considered on its merits. Indeed as the wider theory it seems right and convenient to examine it first.

Women in Pakistan as a Group

The idea so incisively put forward by Lord Hoffmann is neither novel nor heterodox. It is simply a logical application of the seminal reasoning in Acosta's case 19 I. & N. 211. Relying on an ejusdem generis interpretation the Board interpreted the words 'persecution on account of membership in a particular social group' to mean persecution "that is directed toward an individual who is a member of a group of persons all of whom share a common immutable characteristic." The Board went on to say that the shared characteristic might be an innate one "such as sex, color, or kinship ties." This reasoning covers Pakistani women because they are discriminated against and as a group they are unprotected by the state. Indeed the state tolerates and sanctions the discrimination. The analogy of discrimination against homosexuals who may in some countries be a "particular social group" supports this reasoning. What is the answer to this reasoning? It avoids any objection based on the principle that the group must exist dehors the persecution. The objection based on a requirement of cohesiveness foists an impermissible restrictive requirement on the words of article 1A(2). What then is left by way of counter-argument? Counsel for the Secretary of State said that there is a clear answer to this line of reasoning. That turned out to be the fact that some Pakistani women are able to avoid the impact of persecution, e.g. because their circumstances enable them to receive protection. In such cases there will be no well founded fear of persecution and the claim to refugee status must fail. But this is no answer to treating women in Pakistan as a relevant social group. After all, following the New Zealand judgment in Re G.J. [1998] 1 N.L.R. 387 I regard it as established that depending on the evidence homosexuals may in some countries qualify as members of a particular social group. Yet some homosexuals may be able to escape persecution because of their relatively privileged circumstances. By itself that circumstance does not mean that the social group of homosexuals cannot exist. Historically, under even the most brutal and repressive regimes some individuals in targeted groups have been able to avoid persecution. Nazi Germany, Stalinist Russia and other examples spring to mind. To treat this factor as negativing a Convention ground under article 1A(2) would drive a juggernaut through the Convention. My Lords, on careful reflection there is no satisfactory answer to the argument that the social group is women in Pakistan.

The narrower group

If I had not accepted that women in Pakistan are a "particular social group," I would have held that the appellants are members of a more narrowly circumscribed group as defined by counsel for the appellants. I will explain the basis of this reasoning briefly. It depends on the coincidence of three factors: the gender of the appellants, the suspicion of adultery, and their unprotected position in Pakistan. The Court of Appeal held (and counsel for the Secretary of State argued) that this argument falls foul of the principle that the group must exist independently of the persecution. In my view this reasoning is not valid. The unifying characteristics of gender, suspicion of adultery, and lack of protection, do not involve an assertion of persecution. The cases under consideration can be compared with a more narrowly defined group of homosexuals, namely practising homosexuals who are unprotected by a state. Conceptually such a group does not in a relevant sense depend for its existence on persecution. The principle that the group must exist independently of the persecution has an important role to play. But counsel for the Secretary of State is giving it a reach which neither logic nor good sense demands. In A. v. Minister for Immigration and Ethnic Affairs 142 A.L.R. 331, 359 McHugh J. explained the limits of the principle. He said:

     "Nevertheless, while persecutory conduct cannot define the social group, the actions of the persecutors may serve to identify or even cause the creation of a particular social group in society. Left-handed men are not a particular social group. But, if they were persecuted because they were left-handed, they would no doubt quickly become recognisable in their society as a particular social group. Their persecution for being left-handed would create a public perception that they were a particular social group. But it would be the attribute of being left-handed and not the persecutory acts that would identify them as a particular social group."

The same view is articulated by Goodwin-Gill, The Refugee in International Law, 2nd ed., (1996) at p. 362. I am in respectful agreement with this qualification of the general principle. I would hold that the general principle does not defeat the argument of counsel for the appellants.

My Lords, it is unchallenged that the women in Pakistan are unprotected by state and public authorities if a suspicion of adultery falls on them. The reasoning in Acosta, which has been followed in Canada and Australia, is applicable. There are unifying characteristics which justify the conclusion that women such as the appellants are members of a relevant social group. On this additional ground I would hold that the women fall within the scope of the words "particular social group."