This appeal raises once again the issue, amongst others, of what constitutes a “particular social group” within the meaning of Article 1A(2) of the 1951 Convention Relating to the Status of Refugees (“the Refugee Convention”). That paragraph defines a refugee as a person who

The appellant is a female citizen of Ethiopia, born on 15 September 1986. She came to this country on 21 April 2002 and claimed asylum on arrival. She was at that time 15 years of age. She was granted limited leave to remain as a minor, a leave which would expire when she reached the age of 18, an event which has now happened. Nonetheless, the various appeals have as a result dealt with this matter solely in relation to the Refugee Convention and her status thereunder and not with her rights under the European Convention on Human Rights.

The appellant is of Oromo-Gille ethnicity, and part of her original asylum claim was based on that ethnicity. It was, however, rejected by the adjudicator and can now be put on one side. The other basis for her claim concerned the treatment of unmarried females in Ethiopia.

The primary facts about the appellant are not in dispute. The adjudicator accepted the “core elements” of the appellant’s account of events. These were that her older sister had been married at the age of 13 to a much older man, known as Amana. He ill-treated her and used her in black magic rituals. She became ill and eventually died while trying to escape from him.

He then sought to insist upon the appellant marrying him, according to local custom. She was then aged 14. Her mother refused to let this happen, but in December 2000 he abducted her from school, took her to his house, beat her and raped her. She became very ill because of this mistreatment. She too was used in his black magic rituals and was repeatedly raped by him. Eventually she escaped, and she and her mother fled to another town. He pursued her there, and so they fled to another town.

According to the adjudicator’s account of her evidence,

The adjudicator summarised in paragraphs 18 and 19 the background evidence about the treatment of women in Ethiopia as follows (the references are to the C.I.P.U. report from the Home Office, October 2003):

The adjudicator’s conclusion was effectively set out in paragraph 26 of her determination:

The Secretary of State sought permission to appeal to the IAT on four grounds. Since a point is raised in this appeal about whether those grounds spelt out a point of law, I summarise them briefly. First, it was said that the adjudicator failed to indicate which immutable characteristics were shared by members of this particular social group, and her approach to and finding on this issue were flawed. Secondly, it was argued that her approach to the objective evidence had been selective, though it was conceded that she had been entitled to draw the conclusion she did about insufficient state protection. Thirdly, it was said that the adjudicator had not adequately considered the credibility of the appellant. Finally, the fourth ground of appeal was that the adjudicator had failed to consider the internal flight alternative, when there was no evidence that the man Amana had followed the appellant to Addis Ababa or that she would not be safe there.

That summary of the grounds of appeal by itself is enough to indicate, in my judgment, that ground 3 raised no point of law. The concession in ground 2 to which I have referred might seem to rob that ground of any prospect of an error of law being established on that basis, but Ms. Chan who appears on behalf of the Secretary of State argues that it challenged the adequacy of the adjudicator’s reasoning. But ground 3 alleges no error of law at all. It does not assert that the adjudicator’s finding on credibility was perverse or that she had failed to take account of relevant matters or taken account of irrelevant ones. Sensibly Ms. Chan does not seek to argue that that ground raised any point of law.

When the appeal came before the IAT, that tribunal dealt with it very briefly. It referred to a passage from Lord Steyn’s judgment in Shah and Islam where he said

On behalf of Ms. RG, Ms. Plimmer raises first the issue to which I have referred already, namely whether the Secretary of State’s grounds of appeal to the IAT disclosed an error of law. If they did not, then she rightly submits that the IAT had no jurisdiction. She points out that the issue of whether a person belongs to a particular social group and is persecuted because of their membership of that group depends on the evidence; as Lord Hoffmann said in Shah and Islam, page 655, the matter has to be considered on a case by case basis. In Lord Steyn’s words in the same case at page 635:

Consequently it is submitted on behalf of the appellant that the Secretary of State’s first ground of appeal to the IAT raised only an issue of fact and not one of law.

I can accept that the topic of membership of a particular social group is fact sensitive. But it does not follow that there cannot be an error of law in an adjudicator’s consideration of it. The issues involved are not pure questions of fact, but involve the application of a legal test to the facts as found. That is illustrated by Lord Steyn’s reference at page 637 of Shah and Islam to “the question of law whether appellants could claim to be members of the particular social group under Article 1A(2)”. Both he and the other members of the House of Lords in that case were seeking to identify the correct interpretation of the words “membership of a particular social group” in that provision of the Convention. That is an exercise in legal definition.

In the present case, the first ground of appeal to the IAT was not very clearly formulated, but in my judgment it did raise a point of law, namely whether the appellant was a member of a particular social group within the meaning of Article 1A(2).

It is also clear that the fourth ground of appeal, the alleged failure by the adjudicator to consider the alternative flight option, raised a point of law. Ms Plimmer seeks to argue that the adjudicator did consider that option, but that argument, even if sound, would not prevent the point raised from being one of law. It would merely show that the point should not succeed, but that is quite different from whether or not it is one of law. On normal principles, if an adjudicator fails to consider a relevant matter, that amounts to an error of law: see Railtrack plc v. Guinness Limited [2003] EWCA Civ 188; [2003] RVR 280; and E v. Secretary of State for the Home Department[2004] EWCA Civ 49; [2004] QB 1044, paragraph 42. The IAT thus had jurisdiction to consider the point.

Ground 2 amongst the Secretary of State’s grounds of appeal to the IAT is said by the appellant to raise no point of law, because it concedes in terms that the adjudicator had been “entitled to draw the conclusion she does at paragraph 26”. That paragraph (set out earlier in this judgment at paragraph 8) found that there was insufficient state protection for women and girls in Ethiopia by reason of the cultural and legislative discrimination. Consequently, submits Miss Plimmer, criticisms of the adjudicator’s analysis of the objective evidence on this issue cannot amount to an allegation of an error of law, once it was conceded that the conclusion was one which the adjudicator was entitled to arrive at.

I see the force of this submission. Miss Chan for the Secretary of State argues that ground 2 challenged the adequacy of the adjudicator’s reasoning on the sufficiency of state protection. If it did, it was poorly drafted, because that challenge is nowhere spelt out in ground 2. However, the adequacy or otherwise of state protection is, as I explain later, relevant to the concept of a particular social group, and since the adjudicator’s approach to that concept was alleged to be wrong in law, it is of little significance whether or not ground 2 in its wording did actually raise a point of law.

I conclude on this first issue raised by the appellant that the IAT did have jurisdiction in this matter.

It is then submitted by Miss Plimmer that the IAT erred in law in the way in which it dealt with the issue of membership of a particular social group because it gave wholly inadequate reasons for finding that women and girls in Ethiopia are not such a group. That is a contention which the Secretary of State accepts. Miss Chan concedes that the decision lacks adequate reasoning for that finding.

That concession seems to me to be rightly made. The finding of the IAT on this appears to derive solely from the fact that the situation in Ethiopia is different from that found by the House of Lords to exist in Pakistan in Shah and Islam. That, put bluntly, is not enough. The situation in every country is going to be different to some degree from that in Pakistan, either by being better or by being worse, just as the situation in Pakistan itself will probably vary over time. The mere existence of such a difference does not answer the critical question about membership of a particular social group, because the House of Lords in Shah and Islam was not asserting that a case had to be in every respect identical to that found in Shah and Islam in order for the test under Article 1A(2) to be met.

The only other reasoning to be found for the IAT’s conclusion on this issue is the reference to “objective evidence that women do have recourse to the police and the courts” in Ethiopia. That is a reference to a passage from the adjudicator’s summary of the background evidence set out earlier in this judgment. It comes word for word from the October 2003 CIPU report, and in full it actually states:

This analysis of the IAT’s decision demonstrates that, while brevity is often a desirable characteristic of judgments, in the present case it was carried to a degree which is unacceptable. The IAT’s reasoning is patently deficient, and as a result it erred in law.

The decision in Shah and Islam establishes that women in a given society, or a sub-set of women, may in appropriate circumstances constitute a particular social group, membership of which results in a risk of persecution. The House of Lords in that case emphasised that the group in question had to exist independently of the persecution and could not be defined by the fact of persecution. Nonetheless, discrimination against women, coupled with an involvement of the state in such discrimination, could and did result in them constituting such a group. There seems little doubt that in Shah and Islam some degree of state involvement of that kind was regarded as important for the terms of the Article 1A(2) to be met in this way. It is of course necessary in any event for there to be an absence of adequate state protection when the persecution is alleged to emanate from non-state agents. But it does seem that the presence of a degree of institutionalised discrimination by organs of the state in Pakistan was regarded as important in the decision in Shah and Islam. Lord Hoffmann emphasised that at page 655, and it is reflected also in Lord Steyn’s statement at page 635 that

There have been a number of decisions by this court since that case, seeking to apply those principles. Miss. Plimmer relies upon P and M [2004] EWCA Civ 1640, where it was held that it had been open to the adjudicator to conclude that women in Kenya were a particular social group. The court there endorsed, at paragraph 37, the decision of the Australian High Court in Applicant S v. MIMA [2004] 8 CA 25, where that court had at paragraph 36 summarised the law on determining whether a group amounts to a particular social group under the Refugee Convention as follows:

The decision in P and M may be seen as one which reduces the necessity for the discrimination against women to be part of the law of the country in question. The emphasis in the judgment of the court, delivered by Lord Woolf CJ, was on the existence in Kenya of societal discrimination against women and a lack of adequate police protection, not on whether such discrimination was enshrined in the law.

Consequently it is submitted by Miss Plimmer that the adjudicator in the present case was entitled to find that in Ethiopia women and girls faced serious discrimination attributable to cultural traditions, and indeed reflected in certain legal provisions, and that there was a lack of state protection for them amounting to discrimination. Those findings of fact were central to any determination of whether women and girls constituted a particular social group. The appellant therefore contends that on the approach spelt out in Shah and Islam and in P and M the adjudicator cannot be said to have erred in law.

Our attention has been drawn to the background material before the adjudicator. The Secretary of State points out that the adjudicator did not refer to the fact that a new family law in Ethiopia raised the legal age of marriage for girls to 18 as from July 2000. However, the appellant relies upon another passage from the 2003 CIPU report, which states that, despite the new law, “early childhood marriage particularly in rural areas is common” (paragraph 6.129) and that

Another document, which we are told was before the adjudicator but not referred to in her decision, is a paper dated March 2002 entitled “Poverty and Illiteracy aggravate Abduction.” Its status is unclear, but it says that marriage by abduction is common in Oromiya, where one in five women marry by abduction, and that in almost all cases rape follows abduction because it guarantees to the abductor success in keeping the woman or girl. It is submitted on behalf of the appellant that this demonstrates the inadequacy of state protection.

Miss Chan for the Secretary of State points to various passages in the October 2003 CIPU report not referred to by the adjudicator. At paragraph 6.117 it is stated that in 1997 the Ethiopian government adopted a plan of action to enhance the status of women and that certain harmful practices, such as early marriage by abduction, appeared to be declining. In September 2001, it was reported that over 500 charges “were filed against rape cases” in Addis Ababa in the preceding 12 months (paragraph 6.122). According to the report, Human Rights Watch report a move towards enforcement of laws protecting women and children, especially the latter, though women’s groups claimed that police often did not investigate reports of adult rape. (paragraph 6.123).

It is also submitted on behalf of the Secretary of State that the adjudicator failed to give adequate reasons for her decision. Some of the material before her pointed towards an increasing attempt by the state to protect women, yet this is not discussed in the adjudicator’s reasoning. The same contention is advanced in respect of the finding of inadequacy of protection.

I have already set out earlier the approach adopted to this issue of membership of a particular social group adopted in Shah and Islam and in P and M. The Secretary of State acknowledges in argument that the latter case lessens the need for the discrimination against women to be part of the law of the land before they can be regarded as a particular social group. Miss Chan concedes that the discrimination need not be embodied in law if there is in practice a systematic lack of protection. Some more recent guidance may also be obtained from a passage in the judgment of Baroness Hale of Richmond in R (Hoxha) v. Special Adjudicator [2005] UKHL 19; [2005] 1 WLR 1063, where at paragraph 37 she summarised the law on this:

In the present case, it does seem that the Ethiopian government had been taking some steps to achieve greater protection of women, but the passages from the CIPU report relied on by the Secretary of State also reveal an estimate of over 1,000 rapes a year in Addis Ababa alone, but only 168 rape convictions across the whole country in the year ending September 2000. Thus, while charges may be increasing, there is no evidence of increasing convictions, and the evidence indicates a persistence of abduction and marriage by abduction, as well as under-age marriage, despite the raising of the marriage age.

One notes especially the reference to penal law in Ethiopia legitimising the marriage of abducted and raped girls to their violators, which marriage then exempts the latter from punishment. Though that reference in the adjudicator’s determination comes from a report probably produced in about 2001, counsel for the Secretary of State accepts that this provision of Ethiopian penal law still operates. Indeed, it is endorsed in the April 2004 CIPU report (not before the adjudicator) at paragraph 6.141, which quotes a US 2003 Human Rights Report as follows:

This institutionalised legal discrimination must be regarded as of considerable importance. Along with the evidence of a lack of protection of women against sexual abuse and serious discrimination, it shows a degree of complicity by the state in this treatment of women in Ethiopia, sufficient to entitle the adjudicator to conclude that women constituted a particular social group. She did not err in law in so concluding. She was likewise entitled to find that there was generally insufficient state protection for women from serious abuse.

Did she give adequate reasons for those findings? The principle to be applied is well-established: the reasons must be intelligible and deal with the substantial points raised: In re Poyser and Mills Arbitration [1964] 2 QB 467, 478. Reasons can be briefly stated: per Lord Scarman in Westminster City Council v. Great Portland Estates plc [1985] I AC 661. That proposition was endorsed by the House of Lords again in South Bucks District Council v. Porter [2004] UKHL 33, [2004] 1 WLR 1953, at paragraph 36, where Lord Brown of Eaton - under-Heywood emphasised that

It follows from these and other authorities that, while a decision must show to the losing party why he has lost, it most certainly need not deal with all the evidence placed before the decision-maker. That must especially hold good when one is dealing with background material dealing with conditions in the country from which an asylum seeker has come. Such material is often voluminous and it would place an intolerable burden on adjudicators to expect them to refer expressly to all the relevant factual material. It is, of course, a long-established principle of administrative law that it is not to be assumed that a decision-maker has left a piece of evidence out of account merely because he does not refer to it in his decision. One also needs to bear in mind that the decision of an adjudicator is not binding on other adjudicators: it is a decision simply reflecting the evidence before him, evidence dealing necessarily only with the situation in the country in question at a particular moment in time. Indeed, even the IAT’s decisions in cases not categorised as “Country Guidance” cases are not to be cited as evidence of the background situation in a country: Eshete[2002] UKIAT 01963.

Applying these principles and considerations to the adjudicator’s decision in the present case, it seems to me that her reasons for concluding as she did are sufficiently clear. In paragraph 26, quoted earlier, she explicitly refers to the “objective evidence regarding the serious discrimination faced by women and young girls in Ethiopia by reason of their gender.” She refers back to the “cultural traditions and the Civil and Penal Code in Ethiopia as described above” (my emphasis). That is clearly a reference back to the material taken almost verbatim from the Home Office CIPU report and set out at paragraphs 18 and 19 of the decision. Those passages provide an intelligible basis for the adjudicator’s finding in paragraph 26 that the appellant was a member of a particular social group. They also explain her finding about the lack of state protection: whatever attempts were being made by the government to improve the situation, those passages indicate that wife beating and marital rape are “pervasive social problems”; that the availability of protection through the police and the courts is reduced by societal pressures and limited court facilities, especially in rural areas; that abduction still appears to be practised widely; and that the law still allows rapists to escape punishment by marrying their victims. These features of Ethiopian society were, in my judgment, quite sufficient to enable the adjudicator to conclude properly that women were inadequately protected by the state. If there was such protection such widespread serious sexual abuse would not exist. The facts speak for themselves.

I conclude, therefore, that the adjudicator gave adequate reasons for her decision on these issues.

The Secretary of State contends that the adjudicator did not deal at all with the possibility of internal relocation or internal flight, as it is sometimes described. It was an issue which had been expressly raised by the Home Office Presenting Officer at the hearing, as paragraph 15 of the decision indicates. Yet nowhere does the adjudicator give her findings on this.

Miss Plimmer seeks to resist this argument by reference to a sentence in the decision where the adjudicator said (paragraph 24):

I would allow this appeal because of the inadequacy of the reasoning in the IAT’s decision, which amounts to an error of law. For the reasons given above, I would remit this case for the internal flight option to be considered and determined. I would, however, limit the scope of any reconsideration to that issue.

- In the areas in which my Lords agree, I also agree. In the area in which they disagree, I find myself, after initial hesitation, in agreement with Keene L.J.

The charge against the adjudicator is not that there was an absence of material which might, following proper survey, have entitled her to conclude that women and girls in Ethiopia constituted a particular social group. It is that her conclusion is inadequately reasoned in that in particular there was insufficient analysis of the objective evidence placed before her. In truth the objective evidence placed before her was very limited; no doubt the assistance given to her in analysing it was far less sophisticated than that given to us by two excellent counsel; and, in our appraisal of her determination, we must search not for some breathtaking exegesis but for the bare minimum required by law.

Miss Chan identifies for us the passages in the CIPU report dated October 2003 which, in her submission, the adjudicator should at least have expressly considered. Most of the passages are equivocal in the sense that, while they refer to attempts by the government in Ethiopia to provide greater protection for women and girls, they note little improvement on the ground. The high water-mark of Miss Chan’s complaints is that the adjudicator failed to refer to the reform in 2000 under which the minimum legal age for marriage in Ethiopia was raised, in the case of a female, to 18. But, in that the paper presented to the CSW, New York, in March 2002 purported to address a continuing significant problem in Ethiopia of the kidnapping of girls “mostly less than 15 years old”, followed by rape and forced marriage “at childhood age”, the evidence gives rise to real doubt as to whether, as elsewhere, well-intentioned legal reform in Ethiopia has yet been matched by delivery on the ground.

It was in my view not unlawful for the adjudicator to fail to weigh such material expressly. On the evidence submitted to her upon the appeal her identification of women and girls in Ethiopia as a particular social group was in my view lawful and so may not now be reopened.

I gratefully adopt Keene LJ’s statement of the facts and issues. I too would remit the case to the IAT but on broader grounds than those proposed by Keene LJ.

As Keene LJ points out, the case raises the issue of what constitute “a particular social group” within the meaning of Article 1A(2) of the 1951 Convention Relating to the Status of Refugees. Membership of such a group, if established, may form the basis for a claim to asylum under the Convention. As summarised by Baroness Hale of Richmond in R (Hoxha) v Special Adjudicator [2005] UKHL 19, at paragraph 37, women may constitute a particular social group for the purpose of the Convention because they share the common immutable characteristic of gender, and where they are “discriminated against as a group in matters of fundamental human rights, from which the state [gives] them no adequate protection”.

Complex issues are involved and these have been given considerable attention in the courts (Islam v Secretary of State for the Home Department, R v IAT ex parte Shah [1999] 2 AC 629, P & M [2004] EWCA Civ 1640). We have also been referred to the Tribunal cases of HM (Somalia) [2005] UK IAT 00040, NS (Afghanistan CG) [2004] UK IAT 00328 and TB (PSG – Women) Iran [2005] UK IAT 00065. In those cases, the Tribunal conducted a detailed and thorough investigation and analysis of the evidence and the issues which arise in the present context before reaching a conclusion.

In my judgment, the determinations both of the Adjudicator and of the IAT in the present case are defective for want of such investigation. In relation to the IAT, I am in agreement with Keene LJ, who described the Tribunal’s reasoning as “patently deficient”. I agree with Keene LJ that the IAT had jurisdiction, save on ground 3, and notwithstanding the poor drafting of ground 2. Points of law did arise but the IAT was not entitled to set aside the Adjudicator’s decision in the summary way it did.

Having regard to the importance and complexity of the issues involved, the approach of the Adjudicator was also, in my judgment, defective. The decision of the Adjudicator depends on paragraphs 18, 19 and 26 of her determination, set out in paragraphs 7 and 8 of the judgment of Keene LJ, the finding at paragraph 26 being based on what is set out in paragraphs 18 and 19.

If there is to be a condemnation of the way a state treats its women, or allows them to be treated, thorough analysis of the available material is required. The Adjudicator had before her the CIPU Report of October 2003, an OMCT [World Organisation against Torture] Report, probably of 2001, wrongly stated in the determination to be 1997, and, a paper entitled ‘Poverty and Illiteracy Aggravate Abduction’ presented to a session of the CSW in New York in March 2002, which I single out because of the reliance placed on it by counsel for the appellant and by Keene LJ in his analysis. That document, as presented to the court, does not identify authorship. It is not mentioned by the Adjudicator. Its objective is stated to be “to propose conceptual interventions towards the reduction of poverty and illiteracy in order to stop early forced marriages in the rural areas”. Other material was also before the Adjudicator.

The analysis of the material conducted by Keene LJ, in his judgment in upholding the conclusion of the Adjudicator, is very much fuller than that conducted by the Adjudicator. It includes, at paragraph 34, reliance on a document, the CIPU report of April 2004, which was not before the Adjudicator and, indeed, post-dates her decision. The analysis is, as one would expect and with respect, persuasively written. In my judgment it is, however, for the adjudicator to conduct the analysis of current background material and, particularly having regard to the likely general consequences of a conclusion on this subject, a thorough and systematic analysis is required. In my judgment, the absence of such an analysis in this case amounts to an error of law. It is for the fact-finding tribunal, and not for this court, to consider the material, and decide what the facts are and whether they speak for themselves.

The error I find to be present is not primarily in a lack of reasoning, as to which Keene LJ has referred to appropriate authority, but in a somewhat different sector, that of insufficient consideration of the evidence. As Miss Chan for the Secretary of State submits, no mention is made in the determination of paragraphs in the CIPU’s report of October 2003 which refer to the progress made towards enforcing laws protecting women and children, including a new family law which took effect from 4 July 2000. Keene LJ refers, though the Adjudicator did not, to evidence of the small proportion of rapes which result in convictions but, judged by that criteria, statistics produced for England and Wales are hardly reassuring. Be that as it may, the task of weighing the evidence is that of the fact-finding tribunal. The central factual issues must be addressed by that tribunal.

A finding in this case that a state permits the fundamental human rights of women to be ignored has implications in other claims for asylum, the more so if it appears to have the approval of this court. It is of course for Adjudicators, and now Immigration Judges, to decide cases before them on the evidence presented and I accept that a different Adjudicator might come to a different conclusion on the same in-country evidence. However, Keene LJ’s reminder, at paragraph 37, of that principle, does not, in my view, help to justify upholding a less than thorough fact-finding exercise. Indeed the possibility of conflicting decisions, with consequent perception of unfairness as between appellants, reinforces the need for a thorough analysis to be conducted. The new single tier appellate system with its greater flexibility in determining the constitution of the Tribunal hearing a case may permit and encourage a more effective process.

I have expressed agreement with the question of internal relocation being remitted to the Tribunal. That issue includes consideration of the safety of the appellant, in Convention terms, in parts of Ethiopia other than where she lived. Safety must be considered upon a remission in any event and I would remit to allow the issue to be considered generally.

I would allow the appeal and remit to the Tribunal on the basis stated. It will be clear that I am expressing no view as to what the result of a remission should be. Reconsideration of the Adjudicator’s acceptance of the ‘core elements’ of the appellant’s ‘accounts’ should not be permitted.

Finding myself in the minority, I do venture to add that, of the majority approaches, I prefer the narrower conclusion of Wilson LJ stated in the last paragraph of his judgment, namely that the conclusion of the Adjudicator was lawful and may not be reopened. While I have expressed my disagreement, it is in my respectful view a conclusion preferable to one based on a much fuller review, in this court, of the evidence than that conducted in the fact-finding tribunal, and which includes reliance on material not before that tribunal. It is for the judges of the IAT to find the facts.