The decision

AJH (Minority group-Swahili speakers) Somalia CG [2003] UKIAT 00094


Date heard: 13 June 2003
Date notified: 03/10/2003









1.The appellants, who claimed to be nationals of Somalia, have appealed with leave of the Tribunal against determinations of Adjudicators Mr W F Miles, Mr P G Farrell and Miss A Cheales respectively, dismissing their appeals against the decision refusing asylum. The first appellant was initially represented by Mr D Adams but the latter withdrew on instructions of solicitors Markandan & Co. Mr A Wilson of Counsel instructed by Norman H Barnet & Co represented the second appellant Mr G Shaw of Counsel instructed by Aaronsen & Co Solicitors appeared for the third appellant. Mr A Sheikh appeared for the respondent.

2. The Tribunal has decided to allow all three appeals.

3. The basic facts are these.

4. The first appellant, who gave evidence through a Swahili interpreter, claimed he was a Bajuni. At the hearing before the adjudicator the respondent did not dispute his claim to be a Somali national and the adjudicator accepted he was a member of the Bajuni tribe who had fled Somalia after members of the Isaaq tribe had raped his wife and beaten him. However, he did not allow the appeal because he considered that as the appellant and his wife had been victims of indiscriminate rather than targeted violence, he had not shown that he faced a real risk of serious harm. He placed reliance on paragraphs 5.25,5.26 and 5.28 of the CIPU Somali Assessment 2002 which stated that persecution solely on the basis of clan membership or ethnicity is now very unlikely in most areas of Somalia. Even if he were mistaken about risk in the appellant`s home area, the adjudicator considered there was a viable internal relocation alternative in Puntiland and Somaliland. In granting permission to appeal Vice President Mr Freeman noted that the appellant had given evidence through a Swahili interpreter and wrote: “It will be for him to satisfy the Tribunal he can speak Kibajuni or explain why not [see Laila Suleiman [2002] UKIAT 00416] if he wishes to have any reasonable prospect of success: a Kibajuni interpreter will be provided if possible”.

5. The basis of the second appellant`s claim was that he was a Barawan of Bravenese origin who claimed to speak Swahili and Arabic. His father spoke Chimbalasi (the traditional language of the Barawan). He claimed that he fled Somalia after his father and brother were killed in 1998 and went to Kenya. The adjudicator considered he showed poor knowledge of the Kismayo region of Somalia where he claimed to come from. He was not satisfied with the appellant`s explanation for why his mother and sister had remained in Kenya. The adjudicator concluded he was a Bajuni (we shall come back to this misdescription below) but that, because he did not speak Somali, he could not be a national of Somalia. He considered that the birth certificate produced by the appellant was forged. The grounds of appeal maintained that the adjudicator misjudged the appellant`s performance whilst giving testimony and accorded undue weight to the interview record which had been based on a very small number of questions. The appellant had shown a detailed and accurate knowledge of Somalia. The adjudicator was wrong to find that the birth certificate was inauthentic and wrong to make a negative finding of credibility based on the fact that the appellant had left his mother and sister behind in Kenya. Centrally, the adjudicator had erred, the grounds alleged, in assessing the appellant on the basis that he claimed to be a Bajuni. The appellant had in fact claimed from the outset to be a Barawan of Bravenese clan origins.

6. Apparently overlooking the point made in the grounds that the second appellant had said he was Bravanese not Bajuni, the Vice President Mr Freeman in granting leave stated that: “…it will be for the appellant, in the light of Laila Suleiman [2002] UKIAT 00416, either to show the Tribunal he can speak Kibajuni (for which purpose an interpreter in the language will be provided, if possible) or explain why not”.

7. The claim of the third appellant, Mr Hussein, was that he was a Bajuni from Kismayo in Somalia who had fled that country because of attacks by members of the Marehan. His family had been subjected to violence although nothing had happened to him personally. The Secretary of State did not accept he was Bajuni or that he was a national of Somalia. At the hearing before the adjudicator there was no appearance from the appellant. Noting that the appellant did not speak Somali and that he had produced a Somali birth certificate in poor condition, the adjudicator dismissed the appeal. The grounds of appeal contended that she had effectively endorsed the assessment made by the Secretary of State without giving any reasons of her own and she had been wrong to reject the birth certificate just because it was in poor condition. In granting leave the same Vice President, Mr J Freeman, stated in similar terms that this appellant “will need either to satisfy the Tribunal he speaks Kibajuni, or that there is some good reason why he does not”.

8. These cases were joined in view of the fact that each concerned a person who claimed to be from Kismayo in Somalia and each involved someone who either claimed to be of the Bajuni tribe or (in the case of the second appellant) was seen to be claiming to be Bajuni. The same Vice President who granted leave in each case saw the central issue as being whether the appellants could speak Kibajuni or, if not, whether there were good reasons explaining why they did not.

9. In listing these appeals before three legal chairmen there was an expectation that we would seek to clarify the general approach to be taken in cases involving persons claiming to be Bajuni from Somalia. Since the issues in these appeals turned largely on particular facts, we have only seen fit to give limited guidance on how adjudicators should approach such cases in future.

10. Plainly from the terms of Mr Freeman`s grant of permission to appeal it was envisaged that there would be a Kibajuni interpreter available at the hearing. In the event the appellants` representatives did not pursue a request for such an interpreter and rightly or wrongly none was instructed to attend. However, with permission from Mr Freeman in one of the For Mention hearings, a direction was made that oral testimony be given by Professor I M Lewis, Emeritus Professor of Anthropology at LSE (University of London) who was an Africanist specialising in the social institutions and culture of the Somali people of the Horn of Africa, which he had been studying since the early 1950s. His credentials bore out that he was generally regarded as one of the leading international academic authorities on Somali issues and was someone who was frequently consulted by governments and the media on Somali matters.

11. When giving evidence before us Professor Lewis spoke to two written reports he had completed on the second and third appellants.

12. Before summarising his evidence on the second and third appellants we should mention two observations of his on more general matters. One was that he considered that Kibajuni is just a dialect of Swahili with a few local words and a few slightly different constructions. It was “virtually the same language as Swahili”. He also mentioned, and this is relevant to the case of the second appellant, that another localised version of Swahili was Chimbalisi, which is spoken by Bravanese. The other concerned the relevance of knowledge of Somali. The Professor disputed the accuracy of the view expressed in the Fact-Finding Mission report that all or most Bajuni spoke Somali. He considered that the kind of Bajuni who would speak Somali would be those who had the most interaction with Somalis, in particular those in local political or business roles or elders or leaders of local communities.

13. Professor Lewis’s report on the second appellant described an interview conducted by telephone. The Professor was satisfied that this appellant was familiar with some words of Somali and that he demonstrated considerable familiarity with the region from which he claimed to originate. He did not agree with the Home Office assessment that this appellant showed scant or no knowledge of the Kismayo region or of the political structure and geography of Somalia. He knew sufficient words and expressions of Somali to suggest contact with Somalis. The Professor was adamant in rejecting the suggestion that the appellant could have prepared for the interview so as to deceive him. He did not find it odd that the appellant had only been able to name the Hawiya and Darood as major clan families in Somalia.

14. The Professor’s written report on the third appellant described an interview in which the latter was able, although he spoke only a few words of Somali, to give detailed esoteric information about local customs and fishing activities and techniques.

15. We gave the first appellant an opportunity to add to the account of his claim as already before us in the papers. This he did briefly. In their submissions Mr Wilson and Mr Shaw urged us to accept the Professor`s evidence in its entirety.

16. In his submission Mr Sheikh on behalf of the respondent asked us to find the Professor’s evidence as of limited value. The Professor had agreed, he reminded us, that he did not speak either Swahili or Kibajuni and therefore was not in a position to assess whether the appellants were Bajuni (or, in the case of the second appellant, Bravenese). Whilst he did not take issue with the interview methods used by the Professor he urged the Tribunal to prefer the evidence of the tribal elders as reported in the FFM report who said that most Bajuni would speak Kibajuni. He asked us to find that the two appellants concerned could well have “boned up” on matters to do with Somali society prior to their interviews with the Professor.

Our assessment
17. In reaching our conclusions on these appeals we have considered all the available evidence and taken careful account of what the first appellant told us and of the submissions made by the two representatives and Mr Sheikh. We have of course considered the FFM report as well as that by Mr Andrew Fleming who stated in November 2002 that Somalia is an extremely oral society and that no one will be able to get along in Somalia without some knowledge of Somali.

Our evaluation of Professor Lewis’s evidence.
18. We found Professor Lewis an impressive witness. He did not seek in any way to exaggerate the extent of his knowledge and he demonstrated a comprehensive level of knowledge of southern Somalia. He was the author of numerous books, including A Modern History of the Somali, with an expanded u-dated fourth edition published in September 2002 and had written over a hundred articles in leading journals on various aspects of Somali society. He was a former Director of the International African Institute and was a Fellow of the British Academy. In preparation for his interview of the second and third appellants he had re-read the FFM report on minority groups in Somali which deals with a number of minority communities, including the Bajuni. He had also studied the ethnographic account of the Bajuni by the late Professor V L Grotanelli of Rome University entitled Pescatori dell Oceano Indiano, published in Rome in 1955 which is the standard study of the Bajuni.

19. We noted that the reports by the Professor on the second and third appellant were tendered at an earlier For Mention hearing and that the respondent did not take the opportunity of putting in his own report or offering a critical commentary on the Professor’s.

20. We do not think that the Professor`s evidence on the interrelationship between Kibajuni and Swahili (the main focus in the Tribunal decision in Laila Suleiman) took matters very far since the two cases he was asked to look at did not concern appellants who said they only spoke Swahili. What he had to say about the relevance of knowledge of Somali was more pertinent and we shall come back to its general implications later. However, in relation to the two appellants he interviewed, what is manifest in our view from the interviews he conducted with them was that he asked them nuanced questions which could not have been answered as they were unless the appellants had actually come from Kismayo and had the clan identity which they claimed. Very properly Mr Sheik cross-examined the Professor robustly. But in our view the principal effect was only to reinforce our view that the Professor’s assessment of the second and third appellants was extremely astute and established beyond peradventure that they were telling the truth.

21. In respect of the second appellant the Professor was able to satisfy us that this man was able to show close knowledge of other Somali groups of Brava and had given accurate details of the main Somali clan-family divisions and clan affiliations surrounding towns along the Somali coast between Kismayo and Mogadishu, the two main rivers, the local livestock and crops, the local pattern of religious life and the clan and political structure of the Somali world.

22. In respect of the third appellant, the Professor satisfied us that the appellant spoke fluent Kibajuni and that his very limited ability to speak Somali (he had only a few words in Somali) was entirely consistent with the account he had given of having attended an Islamic primary school where he studied Arabic and not Somali and not having ever attended a Somali school. The Professor noted that the appellant was able to correctly identify the lack of any unifying Bajuni clan structure, to pinpoint where his village was on a map and to name the resident chief, to name all the Bajuni islands along the coast to the south and north of Kismayo and various surrounding local towns and to describe local religious practices used by the Bajuni. In particular the appellant was able to describe highly esoteric fishing techniques used by fishermen in that region.

23. Mr Sheikh raised two main objections to the Professor’s evidence. One was that since he did not speak Swahili or any dialect thereof, he was not qualified to decide whether any of the appellants was a Bajuni or whether the second was a Bravanese. The other was that he was wrong to assess that the two appellants he interviewed could not have simply learnt up on the information they gave him when he interviewed them.

24. Dealing with the latter objection first, whilst answers to some of the questions put by the Professor, e.g. those about geography, were realistically capable of being prepared for, others, e.g. about fishing techniques, were plainly not. Furthermore, the Professor’s questions had far more range and depth than those put to these two appellants in their asylum interviews. Professor Lewis was quite right in our view to describe the series of questions asked by Secretary of State officials in both cases as amateurish. Placing his interview questions side by side with those asked by the Secretary of State representative during the asylum interviews we have no hesitation in finding the Professor’s better designed to establish the true facts. Whilst it can be said that neither the second or third appellants helped their own cases by failing at interview or in subsequent statements to furnish full particulars of their experiences, the evidence of the Professor is particularly strong in their favour.

25. Mr Sheikh contended that the Professor failed to adequately explain why the second and third appellants were not able to demonstrate the same level of knowledge at the interview stage. We would accept that the second and third appellants answered more questions when interviewed by the Professor than they had in their earlier interviews. However, we would observe that in the case of the second appellant the Reasons of Refusal letter was wrong to state he had failed to name any islands off the coast of Kismayo. And in any event we do not consider that questions about the name of the community of Asian families who own the best shops and a mosque in Kismayo or the name of the General in charge of the army defending Kismayo at the time were entirely apposite in the case of a person who had been sent by his father at the age of nine to an Islamic religious school and who had little to do with the surrounding Somali speaking population. The Reasons for Refusal letter appears to attach little or no weight to the number of questions the appellant did accurately answer. As regards the third appellant, although the Secretary of State considered he was unable to answer most of the common knowledge questions satisfactorily, we note he was able to identify the principal languages spoken in Somali, the local form of compensation, the faction that took power in Mogadishu in 1977, the two major clan families in Somalia, the clan of mainly Arab descent in Mogadishu and the name of the President’s palace. He rightly said there was no railway in Kismayo. The only specific question he was unable to answer related to the main crop export and, as Professor Lewis pointed out, persons of a fishing background would not necessarily know that. In short, we do not find the interviews undertaken by the Secretary of State`s representatives in either of these cases at all conclusive on the extent of the appellant`s knowledge of relevant matters. In our view, there is much to be learnt from Professor Lewis’s more wide-ranging and individual-specific series of questions.

26. Turning to Mr Sheikh`s first objection, we would agree that generally speaking an expert witness who does not speak language X (in this case Kibajuni) will not be as well-placed to assess whether someone speaks language X (Kibajuni) as someone who does. Since it is not in dispute that the main language spoken by Bajuni is Kibajuni, that is an important consideration. It was in the light of this type of consideration that Vice President Mr Freeman in granting leave saw the best way to resolve disputes about Bajuni identity for them to be heard with the use of a Kibajuni interpreter.

27. Professor Lewis`s credentials place him in an unusual position. He professed no linguistic knowledge of Kibajuni but did maintain that his lengthy first-hand experience of Bajuni and other clans of southern Somalia enabled him to recognise a Kibajuni speaker when he heard one. He was also adamant that, although he spoke neither languages/dialects he could tell apart a Swahili from a Kibajuni speaker. We do not find that this claim offends common sense. A layperson may not be able to tell apart a French-speaking Frenchman from a French-speaking Belgian, but someone who had studied both countries and their cultures in depth would. And in view of the great depth of experience of Professor Lewis as regards people from southern Somalia we are quite satisfied that he was in a position to identify someone as a Kibajuni speaker. His firm evidence that the third appellant was a Kibajuni speaker carried very considerable weight. For similar reasons we attach very considerable weight to his evidence that the second appellant was a Chimbalasi speaker.

28. Even if we were wrong to make this finding, however, we remain satisfied that being able to speak Kibajuni (or Chimbalasi in the case of the second appellant) is not the only means by which Bajuni identity can be assessed.

29. What these appeals illustrate in our view is that there is more than one way to establishing Bajuni identity. Apart from testing whether someone can speak Kibajuni, two other means are common. One is by testing a person as to his knowledge of Somali. The other is by testing whether a person has knowledge of matters familiar to persons who are Bajuni (or Bravanese) and who come from the region they say they do (in all three cases before us, this area was Kismayo).

30. As regards knowledge of Somali language we have already noted why this is pertinent. The Fact Finding Mission report and CIPU sources state that most Bajuni speak Somali. That is a very important indicator, but it must not be forgotten that it states “most”, not all. Hence it is not an indicator which can be determinative of an appeal in itself.

31. Professor Lewis`s evidence demonstrates the importance of assessing the level of knowledge of Somali by careful reference to the individual`s personal history. It was his assessment as both a fluent Somali speaker and a country expert that in respect of both the second and third appellants, albeit they only had a few words of Somalia, that was consistent with the evidence they had given of having had limited contact with the surrounding Somali-speaking population: the second appellant having been sent by his father to an Islamic religious school and the third appellant having attended an Islamic primary school where he studied Arabic and not Somali and never having attended a Somali school.

32. As regards knowledge of matters that would generally be known to Bajuni or Bravanese from Kismayo, it is important to emphasise why this test of local knowledge is central in many cases brought by persons claiming to be from Somalia. It is simply not possible to make any simple equation of ability to speak Kibajuni and being a Bajuni from Somalia. That is because we know from the objective materials that the Bajuni are a highly mobile population and that a small but significant number reside in Kenya. Insofar as knowledge of matters to do with Bajuni in Somalia, we have already seen that the Professor found both the second appellant (in respect of being a Bravanese) and the third appellants to show extensive knowledge of their respective tribal geographies.

33. What is needed therefore in cases in which claims to be Somali nationals of Bajuni clan identity are made is first of all: (1) an assessment which examines at least three different factors:

(a) knowledge of Kibajuni,
(b) knowledge of Somali varying depending on the person`s personal history; and
(c) knowledge of matters to do with life in Somalia for Bajuni (geography, customs, occupations etc).

But what is also needed is (2) an assessment which does not treat any one of these three factors as decisive: as the Tribunal noted in Mohamed Ali Omar [2002] UKIAT 06844, it is even possible albeit unusual that a person who does not speak Kibajuni or Somali could still be a Bajuni.

34. Our principal conclusions can now be set out as follows.

35. It is not clear to us on what basis the adjudicator in the case of the second appellant thought he was claiming to be Bajuni, since he had stated he was Barawan. But in any event we are satisfied his claim should not have been considered as one from a person claiming to be a Bajuni and that the adjudicator erred in this respect. We find that he is a Barawan of Bravenese tribal origin whose home area is Kismayo.

36. We are satisfied that the third appellant is a Bajuni, also from Kismayo.

37. As for the first appellant, whilst we note that the Vice President who granted leave appeared to consider the burden was on him to prove he was Bajuni, the fact of the matter was that the respondent had conceded at the hearing that he was from Somalia and the adjudicator had then given sound reasons for concluding that he had given a credible account as someone who was a Bajuni. Mr Sheikh has advanced no reasons before us as to why we should re-open the findings of the adjudicator in this case. And in our view it cannot be right to treat demonstrated ability to speak Kibajuni through a Kibajuni interpreter as the only or even necessarily the best means of establishing whether someone is a Bajuni from Somalia.

38. Having established tribal identities, it remains to consider whether the appellants in these three cases have demonstrated that they would face a real risk of serous harm upon return to Somalia.

39. In relation to the first appellant, his account of adverse experiences was accepted as credible, but the adjudicator refused his asylum and Art 3 grounds of appeal on the basis that the latest objective country materials did not demonstrate that persecution on account of clan membership or ethnicity was likely. He also though that the first appellant would have a viable internal relocation alternative in either Puntiland or Somaliland.

40. We are not persuaded that the adjudicator was correct to assess that members of the Bajuni who come from Kismayo are no longer at risk on account of their clan membership or ethnicity. It is true that the CIPU Assessment of October 2002 at 5.25 (6. 85 of the April 2003 Assessment) states that persecution on such a basis was now unlikely. But that statement was made in the context of a generalised description of Somalia as a whole. It is also true that the CIPU Assessment in its latest April 2003 version no longer states (as did the October 2001 Assessment at 5.26) that the future of the Bajuni is uncertain and that the Bajuni continue to face very serious difficulties. At paragraph 6.88 it states: “With the Bajuni, their position is more one of denial of economic access by Somali clans than outright abuse”. However, there has been no withdrawal or amendment of the October 2002 Operational Guidance Note concerning Bravanese, Benadiri and Bajuni which continues to describe them as “persecuted minorities” in need of international protection. Bearing in mind that other background materials have continued to express concerns about the vulnerable position of the Bajuni, we think that it would be premature to describe them as being no longer a persecuted minority in the absence of clear-cut evidence saying so.

41. Nor were we persuaded that the adjudicator’s reasons for concluding that the first appellant would have a viable internal relocation alternative in Somaliland or Puntland were sustainable. He correctly noted that the International Organisation for Migration (IOM) has stated that political, economic and social conditions in Puntland are suitable for return and reintegration and that UNHCR has facilitated the return of one tribe from southern Somalia. However the October 2002 IND Operational Guidance Note states that:

“The authorities controlling Somaliland, Puntiland and the Bay and Bakol regions have each made it clear that they would only admit to the territory they control those who are of the same clan and who were previously resident in that particular area. Internal flight for other Somali groups to these relatively safe areas is not therefore a viable option” (emphasis added).

42. Thus we conclude that the first appellant has established a real risk of serious harm upon return to Somalia. His appeal is accordingly allowed.

43. In relation to the second appellant because the adjudicator did not believe his account, he did not go on to consider whether he would be at risk in view of his clan or tribal identify. Having found that the second appellant was a Barawan of Bravenese, we, however, are satisfied that he would be at real risk of serious harm if returned to his home area. Whether the account he gave of his past experiences at the hands of a neighbouring tribe is accepted or not, his clan identity would put him at risk. We base this view on the latest objective country materials.

44. However, as with the first appellant, we do not consider that this appellant could succeed in his asylum or Art 3 claim unless he could demonstrate that he would not have a viable internal relocation alternative in Puntliand or Somaliland. For reasons already given we think the second appellant has demonstrated this: he is not from clans making up the authorities who control Somaliland, Puntland (or from the Bay and Bakool areas). Accordingly we allow his appeal also.

45. That brings us to the third appellant. Whether or not his account of adverse experiences at the hands of Marehan is accepted as credible or not, we find for reasons already given, that he will be at risk in Kismayo on account of his tribal identity as a Bajuni. However, once again, he can only succeed in his asylum and Art 3 claims if he can establish he would not have a viable internal relocation alternative. We find that he too would not have a viable internal relocation alternative in either Puntliland or Somaliland. Accordingly his appeal too is allowed.

The relevance of Home Office policy regarding Bajuni and Bravenese
46. It is salient to say something about the relevance in Somali cases of issues to do with Home Office policy. Mid-way through the hearing we asked Mr Sheikh to take instructions as to whether there was any current Home Office policy regarding persons who were accepted as Bajuni and Bravenese. Due to a CIPU Policy Unit Officer being on leave, he was only able to offer limited information. We do not blame him for that, although we are somewhat surprised that nothing in written form was immediately accessible or available to the court. On his description there was no established policy but nevertheless, even though such cases were looked at in the light of their particular facts, persons accepted as Bajuni or Bravenese would be “likely” to be given some sort of leave. All we had in written form was the IND Operational Guidance Note – Somalia dated 24 November 2002. It could be said that this supports his view that there is no policy as such since it states that the Home Secretary has announced an end to country specific ELR policies and thus that the country specific policy for Somalia in place since the early 1990s and last modified in July 2001 “has now ended”. But this statement only applies to cases where the starting point is that refugee status is refused.

47. Mr Wilson and Mr Shaw urged us to consider that what Mr Sheikh described was a de facto policy under which those who could establish they were Bajuni or Bravanese or Benadiri would qualify as refugees or at least for limited leave to remain. We have some sympathy with their submissions. In recent years the Tribunal has been given to understand that subject only to proof of Bajuni (or Bravanese) or Benadiri identity, the Secretary of State would accept such persons met the criteria for refugee status. This is confirmed by the wording found in the November 2002 Operational Guidance Note as follows:

“Members of the minority Bravanese, Benadiri and Bajuni groups are likely to be able to establish a need for international protection. The UN has assessed them as persecuted minorities and their situation in Somalia remains, at best, uncertain”.

48. Albeit the operative word in this passage is “likely”, we think it does amount to recognition that being a member of the minority Bravanese, Benadiri and Bajuni groups creates a strong presumption that one will be considered as a member of a persecuted minority and so in need of protection under the Refugee Convention.

49. However, even if we are right and there is therefore a continuing de facto policy, what has to be decided in appeals brought under s. 69 is solely whether the decision is contrary to the Refugee Convention. We could only allow these appeals if satisfied that the appellants had established key elements of the refugee definition. The existence of a Home Office policy is at best one source of evidence as to the existence of a real risk. (We consider that under Art 3 the test is similarly limited to whether a claimant can establish a real risk of serious harm on the evidence). How important a source that is depends on the state of the overall evidence in relation to any particular country: in these appeals we have attached considerable weight to the IND Operational Guidance as a source, since it is based on UN assessments and is not contradicted by any conclusive CIPU entry or other evidence.

50. (It may be that under Art 8 a concessionary policy can play an additional role in establishing whether a decision was or was not proportionate since the existence of such a policy qualifies the extent to which there can be said to be a state interest in enforcing effective immigration control in respect of categories covered by such a policy: see Soleiman [2002] UKIAT 07140. But that is not an issue in any of these three appeals).

51. Mr Wilson and Mr Shaw also sought to argue that we should allow the appeals on the basis that there had been an express or implied agreement between the parties at the hearing before the adjudicator that they would succeed so long as the appellants could establish they were Bajuni or Bravanese. However, such agreements, if made, were not on issues of fact. They concerned legal interpretation. And even if we thought that the adjudicator in these appeals had accepted such agreements, the matter is now before us. We do not consider our assessment of these cases can be predetermined by any type of agreement between the parties as to the outstanding legal issues. Persons are not entitled to protection under either the Refugee Convention or the ECHR unless they can demonstrate substantively, not by procedural means, that they face a real risk of serious harm in their country of origin.

52. We would make three other general observations raised by the cases we have considered.

53. Firstly, we do not consider that any evidence we have heard in this case displaces the conclusions reached in Laila Suleiman [2002] UKIAT 00416 regarding the interrelationship between Kibajuni and Swahili. That decision stated:

“…we think adjudicators should approach any [case] involving someone who claims to be a Bajuni resident of Somalia, but who cannot speak Kibajuni (let alone Somali) with great caution.”

54. Professor Lewis plainly did not think the differences between Swahili and Kibajuni were as great as claimed by the FFM elders, but he did not develop this at length and it was not a critical issue in the appeals before us.

55. Secondly, whilst we continue to view Laila Suleiman as correct in regard to the ability to speak Kibajuni and the ability to speak Somali as two very important indicators of whether or not someone is a Bajuni from Somalia, they are not necessarily determinative in every case. As Laila Suleiman itself pointed out, if there is strong evidence based on the appellant`s ability to show knowledge of such matters as clan, geography, politics, that can also be a very important indicator and, in some case, a decisive one.

56. Thirdly, whilst Professor Lewis rightly identified methodological limitations to the findings set out in the FFM Report, we do not consider that his evidence or any other evidence we have been made aware of undermines the accuracy of the conclusion reached by this report that most Bajuni speak Kibajuni as well as Somali. However, his evidence does indicate that this test must always be applied in an individual-specific way. As already noted, in cases where an appellant is not able to demonstrate any or much knowledge of Somali, it will be necessary to consider whether there is a satisfactory explanation for this in terms, for example, of limited contact with the surrounding community. Furthermore, particular note should be taken of Professor Lewis`s finding that: “Most Bajuni do not speak Somali…The kind of Bajuni who would [speak Kibajuni] are those who have most interaction with Somalis, minor local political or business role or elders, leaders of local communities.”

57. To summarise, therefore, we would reiterate what we said earlier about the proper approach to be conducted in such cases.

“What is needed therefore in cases in which claims to be Somali nationals of Bajuni clan identity are made is first of all: (1) an assessment which examines at least three different factors:

a) knowledge of Kibajuni,
b) knowledge of Somali varying depending on the person`s personal history; and
(c) knowledge of matters to do with life in Somalia for Bajuni (geography, customs, occupations etc).

But what is also needed is (2) an assessment which does not treat any one of these three factors as decisive: as the Tribunal noted in Mohamed Ali Omar [2002] UKIAT 06844, it is even possible albeit unusual that a person who does not speak Kibajuni or Somali could still be a Bajuni.”

58. Finally, we would observe that our findings in relation to internal relocation in respect of the three appellants of concern to us in these appeals are to be taken as demonstrative that Puntland and Somaliland cannot be viable relocation options for appellants of Bajuni (or Bravanese) clan origin. Even if we were wrong about the basis on which we reached this conclusion (the insistence by those who control these areas that they would not admit persons from clans other than their own) and so internal relocation was not subject to any such general restriction, it would remains, in our view, that much would depend on the particular circumstances, including the extent to which an appellant is a member of a minority clan or sub-clan who continues to face hostility from majority clans. At the very least this factor may create real problems of access to safe areas. Gender may also be a relevant factor in view of the evidence that violence against women and girls, including rape, is reportedly not uncommon in Somalia, particularly in camps for displaced persons and against women and girls of rival clans and those of minority groups: see [2003] UKIAT 00011 G [Somalia].

59. For the above reasons these appeals are allowed.