Disasters and Refugee Protection: A Socio-legal Case Study from Yemen

Every year millions of people are forced to flee their homes in the context of climate change and disasters. Their needs and rights are unclear. This paper presents and discusses some findings from a socio-legal case study exploring the rights of disaster-affected Somalis and Ethiopians in Yemen. The first main findings relate to the challenges that Ethiopians faced in accessing, and succeeding with, the formal asylum process. This is discussed in light of legal aid theory and research as well as research on credibility assessments. Another category of findings relates to interactions of local, religious law and international law. This is discussed in light of legal pluralism, which helps in identifying an emancipatory potential. While complex, dynamic and dependant on regional politics and other factors, the way Islamic law was applied — and influenced other bodies of law — seemed to ensure better protection than the 1951 Refugee Convention alone. This potential should be further explored and possibly expanded in order to strengthen the rights of people displaced in the context of climate change and disasters more generally.


Introduction
Every year, millions of people are forced to flee their homes in the context of climate change and disaster. 1According to the Intergovernmental Panel on Climate Change, there has been an observed decrease in March-to-May and June-to-September rainfall in the Horn of Africa over the last few decades, and prolonged droughts may become even more frequent in the future. 2Furthermore, human vulnerability is as important as natural hazards -if not more -in creating the disasters. 3In the Horn of Africa there is widespread poverty, political tensions, violent conflicts and human rights violations.
In 2011, rainfall was late, which in combination with these other factors triggered a drought that eventually escalated into famine in certain areas. 4Hundreds and thousands of people were displaced, many across state borders.While the 2011 drought was particularly severe, there have been several droughts and other disasters since then, and affected people have been fleeing the Horn.
One of the main routes that people have been taking is going east, crossing the Gulf of Aden or the Red Sea to Yemen, where some remain and others move onwards to the Gulf states and elsewhere. 5Between 2006 and 2012, a conservative estimate indicates that 447,000 people set off to Yemen in boats from Djibouti or the Somali port city of Bossaso, almost all of them Somalis and Ethiopians. 6th regards to those displaced to another country in the context of climate change and disasters, lawyers and humanitarian agencies have identified a normative protection gap. 7They believe that people who are cross-border displaced will not be considered refugees according to existing law and thus be left without much protection.As opposed to doctrinal lawyers who mainly interpret formal legal texts and find the law sufficient or insufficient for certain scenarios, a socio-legal approach considers 'living law' and 'law in action' on the ground. 8his paper adopts a socio-legal approach in exploring protection challenges and possibilities for people who are cross-border displaced in the context of disasters.More specifically, the focus is on Somalis and Ethiopians in Yemen in 2013.It focuses on two main findings from a case study and explores and discusses these further in light of socio-legal theory and existing research.After a brief description of research methods, there is a description and discussion of social and economic challenges in getting formal refugee protection.Existing research and theory on legal aid and credibility assessments are particularly useful here.Then follows a description and discussion of how local, religious law interacts with national and international law.In this section, legal pluralism is a particularly helpful concept.The paper ends with some final remarks and recommendations.

Research Methods and Limitations
The case study that was carried out is based on a mixed-methods approach.A desk review of existing research, NGO reports, media reports, laws and policies available in English was carried out mainly during March and April 2013.The author, along with a research assistant, conducted field research in November 2013.At the time of the visit, the situation in Yemen was influenced by the consequences of the 2011 political upheavals in the context of the Arab Spring, continuing conflicts and the transition.We visited Sana'a, Aden and Al-Kharaz refugee camps.The Norwegian Refugee Council (NRC) facilitated the visit.Fifteen representatives from the government, the UN refugee agency UNHCR, the International Organisation for Migration (IOM) and other international and national organisations were interviewed.Focus group discussions with five to ten Somalis and Ethiopians were organised in all locations; seven discussions in total.The main selection criterion was that drought or another disaster played a role in their displacement (self-identified).Discussions were semi-structured and covered all phases of displacement.
The findings must be read and considered against a background of several limitations.The situation in Yemen has been particularly volatile and shifting in recent years; the findings described here are linked to the situation as it was in 2013.The field research period was relatively short (ten working days).The visit was postponed several times for security reasons.The security situation further meant that there were limitations on places that could be visited, access to people, as well as how much participatory observation was possible.We aimed for diversity in terms of sex, age, profession, regular and irregular migrant or refugee status, as well as people living in camps, rural and urban areas.However, most of the Ethiopians in Yemen are young men, and this is also reflected in our study.That so few Ethiopian women are included is a limitation.Generally, disasters, displacement and law impact differently depending on gender.Furthermore, while we managed to meet with some irregular migrants in Yemen, access to this group was particularly difficult compared to officially recognised refugees who were based in specific settlements and possibly less suspicious of strangers and less reticent when it came to relating their stories.
Reactivity occurs in all field research, but when informants are marginalised or in a dependent position, methodological problems may become ethical ones.9NGO participation in selection of interviewees and interpretation may have benefits and downsides. 10We clarified that NRC, or other partner NGOs we worked with, merely facilitated the meetings and that the main purpose was research and policy recommendations.Still, being associated with NGOs and the international community was unavoidable, and some interviewees may have highlighted certain aspects of their stories, such as the political, thinking that it might strengthen their possibility of getting formal refugee status, specific assistance, or even resettlement.Some initial findings from the case study were first briefly presented as part of a larger report. 11In the following, the focus is on two main findings that are further described, explored and discussed in light of existing research and theory.

The 1951 Convention and Formal Refugee Status
Yemen is party to several international human rights treaties as well as the 1951 Convention relating to the Status of Refugees and its 1967 Protocol. 12According to Article 1A of the 1951 Convention, as modified by the 1967 Protocol, a refugee is a person who: (…) owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group, or political opinion, is outside the country of his nationality, and is unable, or owing to such fear, is unwilling to avail himself of the protection of that country (…) Since climate change or disasters are not explicitly mentioned, some doctrinal lawyers believe that most -if not all -of those displaced in such contexts fall outside of existing refugee law protection. 13 socio-legal approach requires a closer look at law in action.As in several other developing countries, it is UNHCR that carries out refugee status determination and coordinates refugee assistance.Like other countries bordering on Somalia, Yemen automatically grants all Somalis refugee status due to the general situation of conflict and persecution in their country.They are so-called prima facie refugees.Getting formal refugee status is therefore not a major problem for disaster-affected Somalis.14Some Ethiopians, in particular young women, have a regular status as labour migrants, with many employed as domestic workers.Most of the Ethiopians in Yemen, however, are considered irregular.They can apply for asylum but would have to do so individually and show that their reason for flight is related to one of the recognised grounds in the refugee definition.

Legal Aid
One main group of findings were related to challenges in accessing, and succeeding with, the asylum process.Some of these can be better understood in light of existing legal aid theory and research.
A duty to provide legal aid, including rights information and legal literacy, is implicit in the state duty to implement human rights and the principle of effectiveness, ie that human rights should be effective and not merely theoretical.This duty has been reiterated and clarified in several cases, normative instruments and recommendations from treaty-governing bodies. 15The duty involves making rights information accessible and understandable.Legal aid measures should be proactive, include active outreach and enable people to assert their rights.The need for support -including material -to civil society organisations that provide information and legal aid has been stressed.
More specifically with regards to asylum seekers and refugees, Article 16 of the 1951 Refugee Convention clarifies that this group should have the same access to courts, including legal aid, as nationals of the country.Furthermore, UNHCR has stressed that asylum seekers have a right to legal counsel at all stages of the asylum procedure, including in UNHCR refugee status determination procedures. 16uch socio-legal research in the Nordic countries has focused on legal aid for marginalised groups through student and researcher-run projects. 17The Yemeni context is of course very different, but this may still provide interesting grounds for comparison.
A main finding in the existing Nordic research has been that there is a large unaddressed need for legal aid, and that this is unevenly distributed socially.The correlation with education and income is particularly strong.Papendorf also highlights the difficult position of groups that he describes as distant from law ['rettsfjerne'], including certain immigrant groups. 18(Women within these groups may face particular challenges. 19) The researchers tend to agree that active outreach is crucial to identify and help address the needs.They also emphasise the importance of organisations and material resources to mobilise the law successfully.(Similarly, on the basis of cross-country case studies, Epp concludes that a support structure and material resources to fund and organise legal mobilisation are crucial conditions in order for a rights revolution to happen. 20) There is also more specific existing research on legal aid for refugees in UNHCR refugee status determination procedures. 21Increasingly, NGOs offer legal aid in such contexts.
According to a survey carried out in Egypt, asylum seekers ranked legal aid as their first or second priority (after medical care) in terms of their humanitarian needs. 22Kagan finds that the provision of legal aid in Egypt corresponded with an increase in asylumseekers' chances of receiving protection from UNHCR. 23ucial to rights realisation are also factors that may not be considered legal in the strict sense.Studies have found that lawyers and others who apply the law spend most of their time establishing the relevant facts such as what has happened and what might happen, 24 and rightly so, since these are often the most difficult and crucial questions.Furthermore, it is a regular finding in socio-legal research that marginalised people lack not only knowledge about their rights, but also about which facts in their lives are legally relevant. 25During refugee status determination central questions are whether and how the asylum seeker was or might be persecuted according to one of the 1951 Convention grounds.Thus, the story and credibility of the asylum seeker are crucial.Studies show that a series of individual and contextual factors influence credibility assessments. 26As in the legal aid findings referred to above, education plays a role here as well.The level of education may affect the asylum seeker's ability to articulate the reasons for their application; their understanding of the context of certain events, including political conditions in their country; their ability to respond to questions; and, more generally, present a coherent and consistent account to the decision-maker.This needs to be taken into account by those providing legal aid as well as by the decision-makers.For example, Kagan found legal aid focusing on preparation of testimonies to be successful. 27cording to many socio-legal scholars and activists, legal aid projects should involve a politicisation of legal aid and an emphasis on the potential for countervailing power among the marginalised. 28In some cases, there may be needs but no corresponding, adequate formal rights and therefore not really any legal aid need in the strict sense.
(There has been a similar critique of the rights-based protection concept. 29) Much scholarship and activism relating to legal literacy recognises the importance of being critical of, and challenging, the law. 30An emancipatory strategy cannot limit itself to formal law, but must be part of a broader social and political mobilisation.This is in line with 'subaltern cosmopolitan legality' which entails inquiring into a combination of strategies through which transnational and local movements advance their causes. 31

Necessary Resources to Access the Asylum Process
Let us now return to the case of Yemen.Nomadic pastoralists and farmers, as many of the Ethiopians coming to Yemen are, are often directly and severely affected by droughts and other disasters.Most have no or very little formal education.Through focus group discussions, as well as interviews with national and international organisations, it became clear that many of them don't see the benefit of applying for asylum.Part of the reason may be that they want to move on to Saudi Arabia and the Gulf countries for work as soon as possible.However, according to interviewees in legal aid NGOs, many also lack necessary the language skills, networks and knowledge of the system.This is in line with findings in legal aid research more generally: unaddressed needs and lack of knowledge correlate with education and whether you belong to a group that is distant from the law, as certain immigrant groups are. 32me organisations do offer legal aid in Yemen, such as the Italian NGO Intersos.One major limitation, however, is that much of their work depends on people coming to their drop-in centres.(They do, however, also visit prisons, where they meet with those who have been detained and are waiting to be deported.)As we know from legal aid research 27 Kagan (n 21). 28 more generally, many people are unlikely to be reached unless there is much more active outreach. 33en Ethiopians who know their rights may still choose not to apply for asylum.Some simply don't have the necessary economic resources.Saleh, a 28-year-old man from a farming background in Ethiopia, was living irregularly in Sana'a during our visit.He explained that he had been to the UNHCR office to apply for asylum and was told to come back later on several occasions.There is a massive backlog of cases at this office.Eventually, Saleh stopped coming.He had a job in the informal sector at a car wash and simply couldn't afford the time off from work and risk being fired.This indicates that one must have a minimum of resources and rights to even be able to apply for asylum.Many can't or won't spend the necessary amount of time and wait for what may be years.This is also in accordance with legal aid research more generally: there are several thresholds, including material, for why people don't approach legal institutions and realise their rights. 34

Necessary Resources to Succeed with the Asylum Process
In addition to challenges in accessing the asylum process, disaster-affected people may also have particular challenges in succeeding with the process.While it is likely that many of the Ethiopians would not qualify as Convention refugees, focus group discussions indicated that ethnic and politically based discrimination and persecution in several cases interacted with droughts or other disasters in Ethiopia.This is in line with findings in previous studies. 35Many focus group participants of Oromo and Ogadeni ethnicity mentioned how they couldn't access assistance before, during or after disasters because they were suspected of supporting the Oromo Liberation Front or the Ogaden National Liberation Front.Ogadenis, including an ex-soldier of the Ethiopian army, even claimed that Ethiopian soldiers killed their livestock during drought, thereby contributing to the creation of disaster -which is not unprecedented in Ethiopian history. 36With a proper understanding of contextual vulnerability and multi-causality of disasters and displacement, the 1951 Convention would remain relevant and applicable: People flee persecution based on political or ethnic grounds -which is recognised in the Convention -in the context of drought. 37Yet, according to UNHCR interviewees, only 20 percent of Ethiopian asylum seekers were accepted.According to the same interviewees, a main reason for rejection was lack of credibility.
33 ibid. 34ibid; Epp (n 20). 35 As mentioned, many of the Ethiopians are nomadic pastoralists and farmers with no or very little formal education.Interviewees from the legal aid NGO Intersos as well as other organisations claimed that some of them have 'a problem of vocabulary'.This is in line with research which shows that an asylum seeker's level of education may affect their ability to articulate the reasons for their application; their understanding of the context of certain events, including political conditions in their country; their ability to respond to questions; and, more generally, present a coherent and consistent account to the decision-maker. 38Formulating the challenge as 'a problem of vocabulary' clearly indicates the need for legal literacy.This applies even more so to the substantial number of unaccompanied minors who come to Yemen.Mohammed, who was 13 at the time of our visit, initially told us he came to Yemen seeking education and a better livelihood.
Only later, did he and others in the focus group explain that his family had been affected by the 2011 drought and denied assistance from the Ethiopian government.If this was related to their ethnicity or perceived political affiliation, Mohammed should arguably be considered a Convention refugee.
The lack of networks, suspicion towards other Ethiopians and asylum seekers, and insufficient legal aid may contribute to many Ethiopians lacking knowledge of the system and how to present their story.In addition, culture and language influences credibility assessment, with misinterpretation of statements and behaviour being a particular risk in the context of cross-cultural communication. 39en we recounted some of the stories, UNHCR staff were clear that these Ethiopians would qualify as refugees.In line with Kagan's findings, this indicates that provision of legal aid could increase asylum-seekers' chances of receiving protection. 40Furthermore, Kagan also recommends that UNHCR should inform all applicants of the right to counsel and encourage the development of legal aid, and legal aid providers should focus on helping asylum seekers develop a coherent and detailed testimony.
In conclusion, it is clear that some of the disaster-affected Ethiopians may in fact qualify as Convention refugees, and the major challenges concern lacking legal aid and literacy and the process of credibility assessments.

Legal Pluralism
Another important category of findings concerns interactions of local, religious law and international law.In sociology of law this situation can be discussed in light of the concept legal pluralism.
How we define law has always been a normative and political question as well as a descriptive and analytical one. 41In the modern Western world there has been a dominant understanding of law as a coherent system of rules that is linked to a specific state.However this has also been challenged since the early days of sociology of law, most prominently by Ehrlich. 42Much later, in the 1970s, it was further challenged by research from colonies and previous colonies where Western colonial powers had introduced elements of their own legal systems and at the same time often recognised some version of the local systems. 43Legal anthropologists used the term 'legal pluralism' to describe this coexistence of several legal orders within the same official legal system.
In the 1980s, John Griffiths, a sociologist of law, was more radical in rejecting what he called 'the ideology of legal centralism', the positivist notion that law is necessarily the law of the state, uniform, exclusive and administered by state institutions. 44He linked law to society instead, defining law as 'the self-regulation of a "semiautonomous social field"'. 45For Griffiths, legal pluralism entailed the coexistence of an official legal order and other unofficial legal orders.He called this a strong version of legal pluralism and criticised the weaker version of the earlier legal anthropologists.According to Griffiths, describing and conceptualising different legal orders within the same official legal system, as they had done, involved a tendency towards legal centralism and served the interests of the state.
In the 1990s, research and theoretical development on postmodernism and globalisation also fed into the conceptualisation of legal pluralism.Boaventura de Sousa Santos called for a relative uncoupling of law from the state, and brought in more time and space scales. 46While Günter Teubner wrote about 'global law without a state', 47 de Sousa Santos claimed the nation-state remains central in the capitalist system, but 'state and non-state, local and global social relations interact, merge and conflict in dynamic and even volatile combinations'. 48He coined the terms 'legal porosity' and 'interlegality' to describe how our lives are influenced by interactions of several, different legal orders. 49In contrast to, for example Griffiths, de Sousa Santos claimed, 'there is nothing inherently good, progressive, or emancipatory about "legal pluralism"'. 50According to de Sousa Santos the benefit of the expanded concept of law and consequently politics is its usefulness in uncovering social relations of power and unsuspected sources of oppression as well as emancipation.(Feminist scholars have further emphasised the importance of considering who in the local context has the power to interpret and represent different norms, such as the Islamic. 51) The emancipatory potential needs to be tested in each concrete case.

The 1951 Convention and Assistance through International Organisations
Yemen has a mixed legal system consisting of Islamic law, Napoleonic law, English common law, international law and customary law.Against this background, one can expect to find interactions of state and non-state actors as well as different local, national and international instruments in line with de Sousa Santos' conceptualisation of legal pluralism.
One could perhaps say that the 1951 Refugee Convention is what de Sousa Santos calls a 'globalized localism', that is, 'a given local phenomenon is successfully globalized'. 52It was created mainly by European men for the specific problems of European men after the Second World War.Today, however, a great majority of the world's states are parties.Importantly, it is applied differently in different contexts; it is a 'localised' as well as globalised phenomenon.
The categorisation and formal status of Ethiopians and Somalis in Yemen was important to interviewees from UNHCR and other international organisations.In addition to UNHCR having the responsibility for formal refugee status determination, the agency as well as other international organisations, are constrained by their own mandates as well as funding from donors when providing assistance. 53With regard to mandates, UNHCR focuses on asylum seekers and refugees while IOM focuses on migrants.IOM explained that they had had difficulties accessing humanitarian funding since migrants are not normally seen as a humanitarian issue.There were some instances of flexibility such as the World Food Programme -according to IOM interviewees -providing assistance to Ethiopian migrants under their refugee programme.In sum, however, Somalis displaced in the context of drought or other disasters automatically get refugee status and thereby qualify for some assistance while Ethiopians have to apply for asylum, risk being rejected, and only the most vulnerable groups of those considered irregular migrants or rejected asylum seekers get basic life-saving assistance.

Islamic Refugee Law
While international organisations were concerned with the formal categorisation of Somalis and Ethiopians on the basis of the 1951 Refugee Convention, the situation was somewhat different among Yemeni interviewees.Many mentioned longstanding ties between the countries and a sense of neighbourliness.Several also mentioned a duty of hospitality that they linked to Islam. 51See eg Hellum and Taj (n 15). 52De Sousa Santos, Toward a New Legal Common Sense (n 31) 179. 53See also Roberts (n 35).
While there are different schools of law within Islam -and several of these are represented in Yemen -different scholars largely agree in their descriptions of Islamic law relating to migration and displacement, so-called 'hijrah' law. 54Perhaps this is because hijrah law is so clearly based on the primary source of all Islamic law, the Quran, considered divine revelation.Abd al-Rahim particularly highlights elements of Surah Al-Tawba, 9:6, which he translates as follows: '(…) If any [one, even] of the idolaters seeks thy protection, grant him protection [forthwith] and then convey him to a place where he can feel safe (….)' 55 The Arabic terms for seeking and granting protection, 'istijara' and 'ijara', refer to an ancient Arabic custom and are literally metaphors for seeking to become someone's neighbour and honouring and protecting a neighbour.The Arabic term used for 'a place where he can feel safe' is 'aman', which Abd al-Rahim also translates as 'sanctuary'.It corresponds to the modern refugee law terms 'asylum' or 'protection'.According to a widely respected and referenced, twentieth century Muslim scholar, Professor Muhammed Hamidullah, the meaning of the Surah can be summarised as follows: '[I]f any human being asks for asylum, it can on no account be refused'. 56cording to Abd al-Rahim and Elmadmad, 57 Islamic law gives better protection than the 1951 Convention in several respects: First, rights are extended to any foreign person, Muslim or non-Muslim, who seeks protection without requiring them to prove that they have been, or risk being, persecuted.Second, it gives them a right to asylum and the hosts a corresponding duty to provide asylum.Third, it is not only the state that is entitled and obliged to grant asylum, but any individual or group of people.Finally, asylum includes a range of rights, notably the right to work, to education, to freedom of movement, the freedom to practice one's own religion, and particular needs of women and children are also addressed.
According to Elmadmad, however, this law is rarely referred to or invoked in today's Muslim world, and displaced people are instead often considered illegal migrants. 58She describes how European states are putting pressure on surrounding Muslim states to enact harsh refugee and migration laws, policies and practices.On the other hand, she describes how for example the Sudanese have a modern legislation based on the refugee conventions while the practice of the state and individuals is what she calls 'semimodern' and greatly influenced by Islamic law.Similar trends, both restrictive and generous, were found in Yemen.
Local and religious law plays a role at the level of ordinary citizens in Yemen.Many Ethiopians and Somalis, including Saleh who was living irregularly in Sana'a, recounted how they had received assistance such as food, shelter and help to access medical facilities from ordinary Yemeni citizens.In Bassateen in Aden, one of the major settlements of Ethiopians and Somalis, a local imam and mosque provide much of the assistance.In Haradh, where thousands of destitute irregular migrants were stranded for months, many survived partly due to the generosity of the local population.Among many Yemeni citizens then, Islamic law seemed to be more crucial than the 1951 Convention in terms of how they related to the Ethiopians and Somalis.
Local and religious law may also interact more intimately with the 1951 Convention.
After having built a certain rapport with them, two Yemeni women working with refugee status determination and protection for UNHCR distinguished between their 'professional' and 'personal' answers.Professionally, they said, the 1951 Convention was not so easily applied to drought situations.Personally, however, they believed that the drought-affected should be allowed to stay.Both highlighted the Islamic duty of hospitality.At the same time, they felt somewhat constrained by the Convention text.
There was a tension between the personal and professional.One way that they mentioned that this could be negotiated this was through using the flexibility and discretion that the Convention allows for, such as generously applying the benefit of the doubt to asylum seekers.In this example, it is clear how the local context shapes the application of international law, but also how international law affects -in this case constrains -the local law and context.
Finally, local and religious law may play a role among higher-level decision-makers.One government official highlighted the historically good connections and relations between the countries and said that in general the government is not very strict with Ethiopian migrants but is guided by Islamic principles of good neighbourliness and humanitarianism.The challenge is that they themselves have very little to offer, and he hoped UNHCR, IOM and other organisations could assist the refugees and migrants more.UNHCR and IOM interviewees confirmed the generally good will from, and cooperation with, the government.

Good and Bad Neighbours
Along with the duty of hospitality, several interviewees mentioned that Yemenis, Somalis and Ethiopians are neighbours.Being neighbours was, as we have seen in Islamic law, a metaphor for the idea of extending and receiving protection.Being neighbours may also be a concrete, geographical and historical fact.
There is a long history of seeking and granting exile, trade, proselytising, inter-marriages and other connections between the Horn of Africa and Yemen. 59Social psychologists claim that the reciprocity norm or principle is rather universal but the degree to which 59  we feel obliged to reciprocate varies, and people are more likely to give help in return for help given in the past or anticipated in the future. 60'Drought means death,' said one of the Yemeni UNHCR staff and described how Yemenis themselves had fled to the Horn of Africa during droughts and received help there in the past.Decision-makers approach their tasks from the perspective of their own background and experiences -asking themselves what would I, or someone I know, do in this situation. 61It is reasonable to believe that shared experiences and connections between the countries in the region may influence the application of Islamic law and the 1951 Convention.
However, while government officials interviewed mentioned the good relations with the neighbouring countries, such relations at an official level don't necessarily result in protection for displaced people.Until 2010 Ethiopians as a group were in fact officially considered illegitimate asylum seekers, detained and deported. 62One possible reason for this treatment was the tight relationship between the Yemeni and the Ethiopian governments and Ethiopia considering asylum seeking a serious crime.According to Human Rights Watch, the Ethiopian embassy in Yemen may have been heavily involved in the detentions and deportations. 63The situation improved as of March 2010 after pressure from UNHCR and other international organisations.While the relations between the countries and peoples may positively affect some of those migrating or fleeing, people who apply for formal asylum may also be negatively affected, at least in terms of state action.This recent history of maltreatment may also help explain why many Ethiopians still don't apply for asylum.
The legal position of Somalis in Yemen cannot be taken for granted either.While they have received prima facie refugee status since the early 1990s, in late 2010 the government signalled that this would change.They believed many Somalis were militants seeking to join al-Qaeda groups in Yemen and destabilise the country. 64During the political upheavals in 2011, opposition media, on the other hand, reported that the government was recruiting Somali refugees as mercenaries to help put down the protest movement. 65During the field visit, some Somali focus group participants reported abuse from police, soldiers and even instances involving ordinary Yemeni citizens.The 'public opinion perspective' in sociology of law 66 can help us understand how the application of formal as well as Islamic refugee law may change: as members of society, we have collective, subjective experiences and perceptions of the world that affect law.These may develop quickly on the basis of a collective sense of a great and threatening change in society such as the political turmoil and ongoing conflicts in Yemen, sometimes resulting in 'moral panic'. 67Quite often, certain marginalised groups of the population such as asylum seekers and refugees are used as scapegoats.Good neighbours may turn bad.
Finally, Yemen is under strong pressure from another neighbour, Saudi Arabia, and other rich countries in the region to halt the flow of people transiting through Yemen into their territory.While Yemen and Saudi Arabia have had a rather turbulent relationship historically, many Yemenis are also working in the neighbouring country and it is clearly a regional super-power that Yemeni officials are likely to consider.
In this final section, we have explored some of the complex interactions of state and nonstate actors, and local, religious law and international law.We can conclude with de Sousa Santos, 'state and non-state, local and global social relations interact, merge and conflict in dynamic and even volatile combinations.' 68The expanded concept of law and politics that we have followed helped to identify an emancipatory potential in Islamic law.The application of hijrah law may mitigate some of the obstacles and negative effects related to accessing, and succeeding with, the formal asylum process.

Final Remarks and Recommendations
In this paper, some findings from a case study concerning disaster-affected Somalis and Ethiopians in Yemen have been presented and discussed.The 1951 Refugee Convention has been applied in a way that ensures Somalis automatic refugee status and certain assistance while Ethiopians have to apply for asylum.The first main findings presented concerned the challenges that these Ethiopians faced in accessing and succeeding with the asylum process.These were discussed in light of legal aid theory and research as well as research on credibility assessments.Unaddressed legal aid needs, lack of knowledge and poor resources to mobilise the law correlate with education and belonging to a group that is distant from the law.Most of the disaster-affected and displaced Ethiopians were nomadic pastoralists and farmers who had no or very little formal education.There needs to be much more active outreach by legal aid NGOs and others.UNHCR should take the situation and characteristics of Ethiopians into account when approaching and evaluating their cases, and encourage the development of legal aid.still managed to prevail, can also be very resilient and resourceful.In a different context, Hardy shows how Ghanaians in Canada worked through the system and managed to increase asylum acceptance rates. 69They provided relevant country information to the government; helped individual Ghanaians secure refugee status through education, information, organisation, and support; contributed to media coverage of Ghanaian refugees trying to create a favourable public image (which links to the public opinion perspective on law); and worked with other, more established NGOs.Ethiopians displaced to Yemen could perhaps also strengthen the organisation and mobilisation for their own rights in cooperation with existing legal aid and other NGOs.

Much
Another important category of findings presented here concerned interactions of local, religious law and international law.This was discussed in light of legal pluralism.The two main groups of findings were mostly discussed separately, but relationships between the two have been hinted at: legal aid projects should be part of broader social and political mobilisation.Legal pluralism allows for an expanded concept of law and politics, moving beyond merely formal state laws and protection through these.In our case, this helped in identifying an emancipatory potential in Islamic law.The application of Islamic norms may have mitigated some of the obstacles and negative effects related to accessing, and succeeding with, the formal asylum process.Even with increased legal aid, many of the disaster-affected Ethiopians would probably not qualify as refugees according to the 1951 Refugee Convention.While complex, dynamic and depending on regional politics and other factors, the way Islamic law has been applied seems to provide better protection than the 1951 Refugee Convention alone.This potential should be further explored and taken into account in order to strengthen the rights of people displaced in the context of climate change and disasters more generally.
This paper has also shown that a socio-legal approach may complement and correct a doctrinal approach to law.Protection gaps may be different on the ground than they are according to a traditional lawyer sitting in an office in Geneva or Oslo analysing international law.While there is clearly a challenge that many disaster-affected people are unlikely to be considered refugees, this paper has clarified that quite a number may in fact qualify as refugees according to the 1951 Convention, that a major challenge relates to social, economic and other obstacles in accessing and succeeding with the asylum process, and that we could draw upon law in a broader sense, such as Islamic law, to influence and complement the application of the 1951 Convention.
See eg Bradford G Martin, 'Arab Migrations to East Africa in Medieval Times' (1974) 7(3) The International Journal of African Historical Studies 367; Mariane de Regt, 'Migration To and Through Yemen: The Case of Migrant Domestic Workers' (Discussion paper prepared for the The Forced Migration & Refugee Studies Program (Migration and Refugee Movements in the Middle East and North Africa), The American University in Cairo, 23-25 October 2007).
of this paper has focused on vulnerability and what is lacking for displaced people, but clearly many of them, such as Saleh who had experienced much hardship and yet 67 Stanley Cohen, Folk Devils and Moral Panics: The Creation of the Mods and Rockers (MacGibbon & Kee 1972). 68De Sousa Santos, Toward a New Legal Common Sense (n 31) 94.