Journal of Ethnic and Migration Studies Vol. 27, No. 4: 615-634 October 2001 Refugees and transnationalism: the experience of Bosnians and Eritreans in Europe Nadje Al-Ali, Richard Black and Khalid Koser Abstract The study of transnationalism has largely bypassed refugees, or in rare cases has focused specifically on their political activities. Proceeding from recent perspectives in international migration studies which suggest that there may be at best only a blurred conceptual distinction between refugees and other migrants, this article subjects two refugee groups - Eritreans and Bosnians in various European countries - to the type of transnational analysis more commonly found among labour migrants. It extends the focus from political activities to show how refugees can become involved in a range of economic, social and cultural transnational activities. At the same time, the paper identifies a range of obstacles which differentially influence the desire and capacity of the study populations to participate in these activities. On the basis of this empirical evidence, we make the case for a fuller incorporation of refugees in the contemporary study of transnationalism. At a more conceptual level, the paper charts the evolution of transnational characteristics among the study populations. The implication, which extends beyond the refugee context alone, is that transnationalism is not a 'state of being, as is sometimes implied by the existing literature, but rather that transnationalism is a dynamic process. KEYWORDS: REFUGEES; TRANSNATIONALISM; BOSNIA; ERITREA Within the burgeoning literature on transnationalism, refugees and other exile groups remain relatively understudied. Two exceptions are provided by research on the activities of Haitian and Salvadorean exiles in the USA (Basch et al. 1994; Landolt et al. 1999). Still, even this literature has focused primarily on the political activities of these two groups: in the case of Haitians targeted initially against Duvalier during the late 1950s, and in the 1990s against the coup that deposed Aristide; and in the case of Salvadoreans, mobilised by the anti-government forces of the FMLN. Rarely has literature on refugees and exiles extended its focus to consider the wide range of non-political transnational activities that are so central to most studies of transnationalism amongst other migrant communities. In either excluding or exceptionalising refugees, transnational studies are building upon a long-standing trend in international migration studies, which is probably unsurprising as so much of what is sometimes depicted as the 'new' study of transnational migrants actually has clearly identifiable roots in earlier studies of labour migrants (Vertovec and Cohen 1999). Traditionally, a clear distinction has been drawn in the study of international migration between refugees and labour migrants, with the former representing the political, and the latter the economic, outcome of global systems and interactions. An alternative ISSN 1369-183X print/ISSN 1469-9451 online/01/040615-20 © 2001 Taylor & Francis Ltd DOI: 10.1080/13691830120090412 Carfax Publishing 616 N. Al-Ali, R. Black and K. Koser distinction casts refugees as the involuntary, and labour migrants as the voluntary, aspects of international migration. However, in continuing to tread this well-worn path, transnational studies run the risk of ignoring a range of relatively recent approaches within international migration studies that have begun to question the validity of the distinction between refugees and other migrants. Without underestimating the crucial legal distinction between refugees and others, nor the social implications of that legal distinction, a number of studies have interrogated the conceptual bases upon which the distinction continues to be made. For example, it may not be accurate to distinguish so straightforwardly between political and economic migrants because of the need to recognise that almost all migrants in reality move for mixed motivations, including social reasons (Koser 1998). Similarly, the distinction between forced and voluntary may belie the realities of migration decisionmaking: for at least some refugees, moving is only the start for a range of possible options in terms of destination and status, whilst for at least some labour migrants international migration results from a corporate imposition rather than an individual choice. This article builds upon the conceptual blurring of the distinction between refugees and labour migrants in international migration studies to extend the focus of transnationalism to the case of refugees. It draws on our current research on the mobilisation and participation of transnational exile communities in post-conflict reconstruction (Al-Ali et al. 1999) funded by the UK's ESRC Transnational Communities Programme.1 Our research seeks to identify and analyse the contributions - social, cultural and economic as well as political -that refugees can make in their countries of origin without permanently returning there. It compares the cases of Bosnian refugees in the UK and the Netherlands and Eritrean refugees in the UK and Germany.2 Through the development and discussion of a series of typologies, this article focuses on two aspects of our research findings. The first typology describes the range of transnational activities in which these two refugee groups currently engage. Although differences are found both between the groups and between the various study countries, the main conclusion is that they engage in a broad range of activities far beyond political activism alone. The second typology focuses on the capabilities of refugees to engage in these various activities, and identifies a range of obstacles, not least those relating to problems of labour market integration and social representation in host countries associated with their refugee status. The implication for the preceding debate on the distinction between refugees and labour migrants is somewhere towards a middle ground. On the one hand, we suggest that refugees can and do engage in a wide range of transnational activities, and thus go some way towards undermining the conceptual distinction that has left them largely excluded from transnational studies. On the other hand, there are limits to their transnationalism, sometimes imposed by their legal status. Refugees, return and reconstruction Another reason why a sharp distinction between refugees and economic migrants is conceptually improbable is variety within these two migrant types. The empirical focus of our research project on post-conflict reconstruction determines that we have focused attention on two refugee groups in a particular situation Refugees and transnationalism 617 - namely that the war from which they fled is over, and that on the whole they now have or expect to obtain permanent residence rights in their host countries. This is a different situation, for example, from many Haitians and Salvadoreans in the USA, who have either temporary status or are staying illegally in the USA. Such differences in legal status probably go some way towards explaining why evidence of transnationalism in our case studies is so much more extensive, especially in the case of Eritreans who have secure residence status. There are sound reasons to explain why the study of transnationalism in a refugee context has focused almost exclusively on specific case studies of political activism. It seems sensible to assume that where they have any resources, refugees will mobilise and target these on overthrowing the regime from which they have been forced to flee. Once that aim has been achieved, the spotlight of transnationalism has probably shifted away from refugees for three reasons. First, it is generally assumed that most refugees will return to their country of origin, to consolidate the process to which they have contributed from abroad (Koser and Black 1999). Second, where they do not return, attention has normally focused on their integration in host countries and their transition into 'ethnic minorities' (Portes and Rumbaut 1990). Finally, and as a result of their decision not to return, it is often assumed that governments in their country of origin effectively 'disown' those left abroad. These assumptions belie far more complex realities for both Bosnians and Eritreans. In the case of Bosnians, repatriation from European countries to Bosnia and Herzegovina has been limited since the end of the war in 1995, with most of those who have repatriated being effectively required to return from Germany and Switzerland. Thus, of a total Bosnian refugee population of nearly 1 million in late 1995, it was estimated that just 336,000 had returned by July 1999, of which nearly half returned as part of the Government Assisted Return Programme (GARP) in Germany. Of the refugees we interviewed, most are not planning to return in the near future, although many expressed homesickness and the wish to return at a later stage. The vast majority of the approximately 25,000 Bosnians in the Netherlands were granted refugee status shortly after their arrival and are now eager to obtain Dutch citizenship. Most of the estimated 7,000 Bosnians in Britain are hoping to have their temporary protection extended into the right to reside permanently within the country. The reasons for the relative low numbers of returnees are complex. Difficulties concerning the Dayton Peace Agreement's stipulations concerning democratisa-tion, inter-ethnic co-operation and the protection of minority rights are particularly significant for those refugees who would constitute minorities upon return. However, a large number of refugees seem to be more worried about the ongoing economic crisis, rampant unemployment, and dependence on international humanitarian assistance. The lack of housing, inadequate healthcare and problems related to education are other frequently mentioned factors which prevent Bosnians from returning, as is a perception of resentment towards refugees seen as having betrayed their home country by leaving during the war. Despite these various obstacles to return, Bosnian refugees have increasingly sought links and contact with their country of origin. Both as individuals and as members of community organisations, there have been few yet growing attempts to get involved in ongoing developments within Bosnia, i.e. sending financial remittances, organising cultural and social events, and regular visits to see friends and family. So far, the Bosnian government has not actively encour- 618 N. Al-Ali, R. Black and K. Koser aged these emerging transnational ties; if anything, it has added to the resentment towards refugees, especially in the party-dominated media. There has been an even lower rate of return among Eritrean refugees in Europe, whose country has been independent and at peace for longer than Bosnia, since 1993 (though war re-started between independent Eritrea and Ethiopia in 1998, resulting in defeat for Eritrea 2 years later). In contrast to the situation of many Bosnians, few of the obstacles to return for Eritreans have been political. Instead, like many Bosnians, respondents have cited a range of mainly economic, social and cultural reasons for their decisions not to return, ranging from the lack of employment opportunities in Eritrea, through social discrimination against single women, to a desire to complete children's education in Europe (Koser 1999). Similarly, for most Eritreans obstacles to return have been complemented by the assumption of permanent residence rights in their European host countries, mainly through the granting of full refugee status. Perhaps even more clearly than in the case of many Bosnians, for the majority of Eritreans their decisions not to return have not arisen from a desire to get away from the troubles of their home country. Indeed, most of the respondents still have close family members living in Eritrea, to whom many send remittances on a regular basis, and whom a number also visit from time to time. Furthermore, the engagement of many Eritrean exiles with their home country has been heightened over the last 2 years in the context of the renewed conflict with neighbouring Ethiopia, as a result of which the Eritrean government has taken a series of initiatives to raise funds among the diaspora (Koser 1999). It is also worth emphasising that there is nothing especially unusual about the evolution of the Bosnian and Eritrean communities since the end of the wars from which they fled, and this point is made to emphasise the wider applicability of the findings that follow. Indeed a very similar set of processes was found among Haitians and Salvadoreans in the USA. Even after Duvalier's son was deposed in 1986, many Haitians did not return permanently, instead consolidating their links with Haiti from abroad. In 1991 Aristide declared Haiti's Tenth Department' to add the Haitian diaspora to Haiti's nine territorial divisions, and spoke of the 'bank of the diaspora' through which Haitians abroad would provide financial assistance to Haiti (Richman 1992). In the case of Salvadoreans, return was also limited; instead Salvadoreans abroad sought to influence the course of change at home through community-based social justice organisations, home-town associations and community development projects. Beyond these specific case studies, the circumstances of both the Bosnian and Eritrean communities in the post-conflict era dovetail with wider findings on the elusiveness of refugee repatriation even after the end of conflicts (Koser and Black 1999); that refugees send home economic remittances (Diaz-Briquets and Perez-Lopez 1997); and that nation-states increasingly view communities in exile as legitimate constituencies (Basch et al. 1994). Transnational activities There is no clear and accepted existing typology of what constitutes a 'transnational activity' and what does not. Here, our intention is not to provide a hard and fast typology, but rather to draw on empirical evidence of what appear to be transnational features amongst our study communities, with a particular focus on activities that might be seen as contributing to 'reconstruction' of the Refugees and transnationalism 619 Table 1. Categorisation of individual and community activities by type and geographical focus Home country focus Host country focus Economic • Financial remittances • Charitable donations • Other remittances (e.g. • Donations to community medicine, clothes) organisations • Investments • Charitable donations • Taxes • Purchase of government bonds • Purchase of entry to government programmes Political • Participation in elections • Political rallies • Membership of political • Political demonstrations parties • Mobilisation of political contacts in host country Social • Visits to friends and family • Membership of social clubs • Social contacts • Attendance at social • 'Social remittances' gatherings • Contributions to newspapers • Links with other organisations circulated in home country (e.g. religious and other refugee organisations) • Contributions to newspapers • Participation in discussion groups (e.g. Internet bulletin boards) Cultural • Cultural events including • Events to promote culture (e.g. visiting performers from the concerts, theatre, exhibitions) home country • Education Source: Field data (1998-1999). home country. Thus Table 1 identifies the main activities found amongst the study communities, distinguishing between economic, political, social and cultural activities. The table combines findings from both communities, but, as analysis later in this section shows, certain activities are more prevalent among one community than another. The table also combines activities that take place at individual, family and community levels. One reason is because these distinctions are often blurred -many Eritreans, for example, have made donations to relief efforts in Eritrea both on an individual basis via collections centred on the local community, and also via collections focused more widely on the entire Eritrean populations in both the UK and Germany. In other cases the distinctions are more meaningful - for example cultural events among the Bosnian communities tend to be organised at a community level. Nevertheless, a second initial conclusion that can be drawn from the table is that within both communities there is a wide range of activities, and therefore that for individuals within each community there exist multiple opportunities to contribute towards reconstruction in their home countries. In developing a simple empirical typology of activities, Table 1 distinguishes 620 N. Al-Ali, R. Black and K. Koser between those activities that are focused on the home country and those focused on the host country. Probably the most obvious activities that can contribute towards reconstruction in Bosnia or Eritrea are those with a direct impact in either country - for example investments by refugees in land or businesses. At the same time, activities that sustain the society and culture of the home country within the exile community are considered by both communities to be equally important in shaping the future of the home country. For example, amongst many Bosnian and Eritrean refugees there is a strong conviction that children born in host countries should learn their mother tongue, and share a national consciousness. Examples of transnational economic activities Individual remittances. Edina3 and her husband Namik, two Bosnian Muslims from Brcko, came to the UK in 1993. They both had to leave their respective parents in their home town. Since their arrival they have been sending small sums of money, varying between £50 and £100, every 2 or 3 months. 'Edina is very good at saving money in the house', says her husband, 'she even started to bake her own bread'. Sometimes the couple fight over whose parents to support at a specific point. Namik says: We try to help them [their respective parents] financially as much as we can. My parents were forced to flee from their house and now they have to rent an apartment even though they used to own a big house. It is very humiliating to see that your parents live in poverty now and that you can do so little to help them. Edina's father, on the other hand, suffered a stroke a few years ago and is dependent on expensive medicine. He is also supported by Edina's brother, Ahmed, who has been in Sweden since 1994. Like Edina and Namik, the majority of Bosnians interviewed in both the UK and the Netherlands have been sending financial remittances to Bosnia. Money is generally sent to close family members, most notably parents and siblings. It was impossible to obtain concrete figures, but it was often stated that financial remittances were a direct response to the basic needs of the refugees' relatives, including food, housing and medicine. A similar pattern was found among the Eritrean respondents, 35 (of a total of 44) of whom reported sending money to relatives still living in Eritrea. There were only three respondents who said they had family in Eritrea but had not sent home remittances - and in each case the reason was a lack of surplus money. Besides responding to specific needs, remittances also have wider impacts in the Eritrean context. First, in the absence of a social welfare system, they provide a crucial 'safety net' during periods of shortfall in Eritrea. For example, as a result of the current conflict, many young men and women are absent from their families, thus withdrawing a source of income. Remittances are one way of filling the gap. Second, local communities in Eritrea are often required to contribute part of the cost for community projects, and money remitted from relatives abroad is often directed towards these contributions. Collective remittances. Bosnian community and charity organisations regularly ask for donations among refugees to send money to Bosnia. In the UK, several Bosnian organisations have been involved in organising numerous charitable Refugees and transnationalism 621 donations with the aim of supporting orphanages and individual orphans within Bosnia. Occasionally, The Bosnia & Herzegovina Refugee UK Network (established in 1996), an umbrella organisation of about 15 community organisations, has called for donations among its members. The support of hospitals in general, and particularly those which help war invalids, was mentioned as an additional objective among some Bosnian community leaders in the Netherlands. Occasionally, donations aim at assisting individuals or collective projects of Bosnian refugees within the host country. Individual assistance can often be found within families and among friends, especially along gender and generational lines, whereby those who have found employment and have secured a regular income tend to support those who live on income support. Collectively, members of organisations might be asked to contribute financially to the establishment or continuation of certain projects, such as film and social clubs, Bosnian language classes for children, and cultural events. Bosnian Islamic organisations in the UK and the Netherlands regularly collect the obligatory religious 'poor-due' (zakat) from the community and redistribute the money according to specific needs. 'Taxes'. All Eritreans in exile are asked to pay 2 per cent of their monthly income directly to the government. This rate applies across all social categories, and includes the unemployed. Although this payment is not compulsory, most respondents consider it their 'duty' as Eritreans to meet the payments. These payments have continued a tradition of contributing amongst Eritreans in the diaspora. Since the start of the most recent conflict with Ethiopia, Eritreans in the diaspora have been asked to increase their contributions. In the UK, Eritreans are now asked to pay the sum of £1 per day, plus a one-off annual payment of £500. In Germany, by means of a one-off payment, the monthly payment for December 1998 was raised from 2 per cent to 10 per cent; in April 1999 there was a request for a further one-off payment of DM1,000, and for an additional DM30 per month. These payments are not only contributing towards the cost of the conflict, but also towards providing emergency relief for victims of the conflict. In addition to direct requests for payments, the Eritrean government has also devised a number of more innovative ways of raising money from the diaspora. In 1999, government bonds were issued for the first time in Eritrea. The Economic Advisor to the President estimated in August 1999 that expenditure on bonds already amounted to some US$30 million in the USA, US$20 million in Europe, and US$15-20 million in the Middle East. As is indicated below, now that the conflict with Ethiopia seems finally to be over, there are indications that at least some Eritreans are losing their motivation to contribute to the state any longer. Examiples of transnational political activities Direct participation in the political process. Participation in elections in both 1996 and 1998 was found to be relatively low among the Bosnian respondents. In contrast, every Eritrean respondent who was eligible to vote reported having participated and voted positively in the April 1993 referendum for independence. This corresponds with estimates that over 90 per cent of all eligible voters in the Eritrean diaspora participated in the referendum, in which over 99 per 622 N. Al-Ali, R. Black and K. Koser cent voted in favour (Styan 1993). A further political process that incorporated the diaspora was the drafting of the constitution of Eritrea. Eritreans in the diaspora were involved in three main ways. First, elected representatives of the diaspora served on the Executive Assembly of the Constitutional Commission. Second, extensive consultation took place at each stage of the drafting of the constitution. Finally, the diaspora was represented during the ratification of the constitution in 1997. One of the requirements of the constitution is the holding of constitutional elections in Eritrea. These have not yet taken place, and for the time being Eritrea remains a single-party state. The burgeoning presence of Eritrea on the Internet provides an interesting channel for members of the diaspora to contribute towards reconstruction in their home country. One web page makes a request for donations to plant trees in the National Martyr's Park outside Asmara, which includes a monument, a museum and forest trails, and commemorates martyrs from the independence struggle.4 Fifteen Nakfa (or US$3) pays for one tree to be planted. White, silver or gold certificates are issued to contributors depending on how many trees they sponsor. Political affiliations. None of the Bosnian respondents stated that they were members of a political party. However, political affiliations could easily be detected in discussions among refugees and were also apparent in some of the community centres visited. Political affiliations are frequently subject to contention and conflict within community organisations and among individual refugees. While there was no direct evidence of activism in opposition parties, several of the economically successful refugees interviewed stressed that investments in Bosnia need to be preceded by political stability and expressed their interest in becoming active in opposition parties. Similarly, there was little evidence of formal opposition among the Eritrean communities. Opposition parties do exist, but they are splintered and have very little political influence in Eritrea. An example of how the direct revealing of political affiliations can prove problematic is provided by a Bosnian labour migrant who established the Bosnian Association in the Netherlands at the beginning of the war. At the time that the research was carried out he was known to be a supporter of the ruling party and of the former President Izetbegovic, whose picture is apparent for everyone to see. During the past years, the majority of people visiting the association have been Bosnian Muslim refugees. Yet, many Bosnians who do not visit the association perceive the organisation to be exclusively Bosniak (Bosnian Muslim). Several of the younger refugees who frequently visit the association and participate in its activities have recently challenged its President and original founder, and have called for a multi-ethnic organisation which is independent from any political party and open to Bosnian Croats and Serbs. Political mobilisation in host countries. Another way that other diasporas have been found to influence the politics of their home countries is by bringing pressure to bear on the government of their host country (McDowell 1997). Little evidence for such activities was found among either the Bosnian or the Eritrean respondents. Besides having a relatively small numerical presence in both countries, several key informants expressed the view that while the Eritrean community is well organised internally, it is not organised in terms of external Refugees and transnationalism 623 affairs. In March 1999 peace demonstrations were held in the German cities of Bonn, Stuttgart, Frankfurt and Mannheim, one aim of which was to raise the profile of the conflict against Ethiopia. However, these were reported to have been poorly organised and poorly attended (about 200 Eritreans marched in Frankfurt), and to have received virtually no media coverage. Examiples of transnational social activities Maintaining contacts with family and friends. The vast majority of Bosnians and Eritreans interviewed have regular contacts with friends and family in Bosnia and Eritrea. In both cases telephone is the main means of communication. No Eritreans, and only a few Bosnians have used the Internet to maintain contacts, primarily due to a lack of access in both home countries. For Eritreans, the intensity of contacts appears to have increased during the recent conflict. Over half of the Bosnian respondents in each host country reported having visited Bosnia for at least a short period of time. Similarly, 29 of the 44 Eritrean respondents reported having visited Eritrea in the past 2 years, and 12 were planning at the time of the interview to visit soon. Amongst both communities, these visits often take place during summer months, especially for those families with children. Their motivations to visit were also similar, arising from social obligations to family members, specific events such as a death in the family, or bureaucratic matters such as those relating to housing or land. A number of Bosnians in particular felt ambiguous or were even opposed to visiting Bosnia, but felt compelled due to such obligations. Their experiences of these visits varied greatly. Most people described reunions with their families as positive experiences, but some people mentioned resentment and envy by those who remained in Bosnia. It is important to stress that such visits, although seen by some policy-makers as a step to permanent return, are as often a key deciding factor in decisions to remain outside the country. For example, Maja and Ferid, a mixed-marriage couple in their 40s, arrived in the Netherlands in 1993. Both left behind close family and friends in their home town, Sarajevo, with whom they kept close contacts throughout the war. Maja, more so than Ferid, told of how she missed her family and for a few months even went into depression. While she wanted to visit her family as soon as possible, Ferid was hesitant, mainly because of a fear of resentment by those who did not leave, but also because of changing attitudes towards mixed marriages. In 1996 they went back for the first time and both felt ambiguous about their first visit. They were overjoyed to be reunited with their families, but were hurt by some comments by neighbours and friends who accused them of 'taking the easy way out' and escaping during the war. They both went back again in 1998, and will probably return again, but they have given up their idea to return permanently. Even more negative was the experience of Kelima, a Muslim woman in her 50s, and her husband Ismet, who went back to their native Bosanski Brod in 1998. It was difficult to enter the Bosnian Serb Republic, and Kelima described how she bribed a border guard to let her through. The couple had worked in the local post office for almost 30 years and recognised many of their colleagues when they visited their former work place. Yet Kelima remembered with tears that her colleagues shouted at her and warned her never to return: '... otherwise there will be blood up to your knees'. A neighbour greeted them with: 'get the 624 N. Al-Ali, R. Black and K. Koser hell out of here', and people in their favourite cafe turned their backs on them. For Ismet and Kelima the experience was so chilling and traumatic that they decided to sever all social ties with people from their home town in Bosnia. 'Social remittances'. It is not only people who travel between countries, but also ideas, values and cultural artefacts. These latter have been described as 'social remittances' (Levitt 1998). Several Bosnian intellectuals and artists stated that they aimed to produce writing and art, which could be distributed in both the host country as well as in Bosnia. Some journalists continue to work on a freelance basis for the Bosnian media, but others have either changed their profession or tried to establish themselves in the host country. Those who continue to write for newspapers, or work for either radio or TV, stressed their aim to promote ideas of tolerance, a multi-ethnic Bosnia, democracy and freedom of speech. This was also true of artists, such as writers and painters, who are concerned with changing ideas in both Bosnia as well as their host country. Little evidence has been found of more formal channels for the transmission of 'social remittances' among either study community. In the Eritrean context, the greatest potential surrounds DEHAI, which is probably the leading web site for the burgeoning 'virtual' Eritrean diaspora. DEHAI provides current news on Eritrea, links to other relevant web sites, and a bulletin board for discussion. It provides one example of how the construction of a new Eritrean nation is being discussed, and at times contested, by the diaspora - in this case in a 'virtual' environment. Limited Internet access in Eritrea limits the extent to which discussions are impacting upon that country directly, but they may well be shaping the ideas of the diaspora, which both formally - for example through elections - and informally - through correspondence with friends and relatives in Eritrea - have a role to play in shaping the future of the country. Social activities in host countries. Within the UK and the Netherlands, a large number of Bosnians seem to socialise mainly with other Bosnian refugees. Social visits are described as an essential part of Bosnian culture and are held in great esteem among most people interviewed. However, several people stated that, due to different lifestyles in either the UK or the Netherlands, the wide distribution of refugees across the country (particularly relevant for the UK where 6,000-7,000 refugees live in 55 cities), costs of transportation, widespread emotional instabilities and even depression, social visits are much less frequent in the host country than in Bosnia. Social clubs and community organisations exist in both the UK and the Netherlands. They are generally perceived to be important venues for social gatherings, for maintaining links with other refugees and for accessing information about issues related to the host country and Bosnia. Some organisations have started to provide premises where people can socialise spontaneously in the form of social clubs or 'cafes', while others organise more formal gatherings. Several organisations have also established contact and close relationships with other non-Bosnian refugee groups. These exchanges and encounters can also be found between Bosnian and other refugee intellectuals and artists. The most significant links, however, can be found among Bosnian Muslim and other Muslim organisations. The Islamic Centre in London, for example, is a member Refugees and transnationalism 625 of the Muslim Council of Britain. Other Bosnian Muslims in the UK reported individual contacts with other Muslims, particularly Pakistanis. Examples of transnational cultural activities Musical, artistic and literary events. Cultural activities often overlap with social activities, but one can point to several trends that characterise cultural exchanges and activities. At the core of most cultural activities lies the attempt to maintain cultural links between the refugee community and their home country. While most activities take place within the host country, one could categorise many events as having both a home and a host country focus. This is because many events amongst both study communities are organised by inviting musicians, artists and writers from home countries to perform within the host country. In July 1999 some 6,500 Eritreans attended the largest European Eritrean festival, in Frankfurt. The festival included seminars and speeches, as well as music and poetry recitals. Leading academics, musicians and poets came from Asmara, and it is significant that the festival was funded by the government of Eritrea via the Embassy in Frankfurt, demonstrating the significance placed on the diaspora by the government. National holidays and parties. Culture understood in the broader sense of traditions, customs and values is promoted through parties celebrating religious and secular holidays, listening and dancing to folklore music, and cooking. Just as has been found amongst the Bosnian communities, the Eritrean communities in both the UK and Germany maintain an active cultural calendar. A good example of how the diaspora maintains a cultural link with Eritrea is through the celebration of public holidays that mark significant historical events. The three most important are Independence Day on 24 May, National Martyr's Day on 20 June and the Start of the Armed Struggle on 1 September. Language. Language as an important marker of culture has also gained in significance. Language skills are valued highly with regard to children's education and many Bosnians and Eritreans fear that their children will lose their mother tongue as time goes by. Special efforts are made by some parents to promote native language speaking, reading and writing skills among their children. However, language skills amongst both Bosnian and Eritrean children remain low. At a collective level, several communities in the UK have established Bosnian supplementary weekend schools. The Bosnian Embassy has provided an official curriculum issued in Bosnia, which includes Bosnian language, history, geography, music and art. Currently there are about ten schools in the UK with more than 250 pupils aged 6-16 years attending classes. The teachers are professionals who used to work in Bosnia. Most work on a voluntary basis; some receive a small stipend. The need for more schools was expressed by many respondents, but due to a lack of premises, funds and professional teachers the numbers have been restricted. In the Netherlands, some associations informally organise language or art classes for children. Largely in response to demands from parents, several Eritrean communities also organise 'cultural' lessons, which focus mainly on language training but also include lessons on Eritrean history and culture. In London there are at least 626 N. Al-Ali, R. Black and K. Koser two 'schools', associated with the Orthodox Church and the Catholic Church, which also include a religious element. The language for these lessons is Tigrinya, and several Muslim Eritreans complained that Arabic is not also taught within community schools. Just as within the Bosnian community, language can be a divisive issue. Transnational capabilities Capacity and desire In investigating the capabilities of Bosnians and Eritreans to participate in relief and reconstruction in their home countries, we have found an important distinction between individuals' capacities - or abilities - to participate, and their desire - or willingness - to participate. On the one hand, it is clear that where an individual is unemployed or earns only a low salary, he or she will often have no surplus money to contribute. In this case, unemployment, or a low salary, are factors influencing the capacity of the individual to participate. On the other hand, if an individual is in opposition to the government in the home country, and therefore does not want to support national reconstruction under that government, he or she may choose not to contribute despite being able to afford to. In this case, political opposition to the government in the home country is a factor influencing the desire of the individual to participate. The crucial implication is that the capability of any one individual to contribute to their home country is influenced by a combination of both capacity and desire to participate. A typology of capabilities Building upon this distinction between capacity and desire, Table 2 combines two types of information. At one level it indicates the main factors that influence both capacity and desire - distinguishing broadly between economic, political and social factors. At the same time these factors are qualified, to indicate the particular circumstances in which the capabilities of individuals to participate are found to be increased. Thus, for example, access to savings, a secure legal status in the host country and freedom of movement within the host country, have all been found to increase the capacity of individuals to participate. One advantage of presenting the factors in this qualified way is to stress that they are dynamic. The factors listed relate to the personal circumstances of individuals (such as their contacts with friends and family in the home country), and more contextual circumstances in both the host country (such as the policy of the host government towards refugees) and the home country (such as economic or political stability). Changes can occur in each of these categories. At an individual level, relatives in the home country may migrate or die. The recent conflicts in both the Horn of Africa and the Balkans have threatened to destabilise both countries of origin. In turn, an advantage of emphasising that the factors influencing capabilities to participate are dynamic, is to highlight the role that policy interventions can play in increasing capabilities. Many of the factors suggested to influence capacity in Table 2 are familiar from numerous other studies on the integration of refugees (e.g. Wahlbeck 1999). It is probably no surprise that better integration Refugees and transnationalism 627 Table 2. Factors increasing individual capabilities to participate in reconstruction in the home country Capacity Desire Economic Political Social Employment Savings Access to welfare and pensions from home country Access to welfare and pensions from host country Access to information Access to banking facilities Secure legal status Positive attitude of host government and population towards ethnic-national diasporas Political integration of diaspora by home government Freedom of movement within host country Gender equality Successful social integration in host country Place of origin in home country Financial stability in host country Economic incentives (or lack of disincentives) for remittances and investments in home country Economic stability in home country Secure legal status in host country 'Non-alienating' circumstances of flight Positive attitude of home government towards diaspora Political stability in home country Lack of ethnic/religious discrimination in home country Links with family and friends in home country Links with friends and family in other host countries Integration within the diaspora in the host country Positive attitudes towards home country Desire to maintain 'national consciousness Source: Field data (1998-1999). tends to empower refugees and increase their capacity to participate; still our findings provide yet another reason to support ongoing efforts in all three host countries to improve conditions for refugees. At the same time the table implies that it is not only host governments that might usefully intervene. Obstacles to achieving many of the conditions shown to influence desire are more directly the responsibility of home governments - including for example removing economic disincentives for remittances, and maintaining democracy. Finally, there are also implications for community organisations in the host countries, relating for example to factors such as social integration and gender equality within the diaspora. Capacity to contribute Employment. As described above, a large number of the surveyed refugees send home financial remittances. Employment, which provides a regular salary and the possibility of savings, is the single most important factor that increases the capacity of refugees financially to assist their relatives. Chances of employment, in turn, have been found among the respondents to be related to factors such as 628 N. Al-Ali, R. Black and K. Koser language skills, education, professional background and experience, the transferability of qualifications, access to training, discrimination and the specific job market situation in the host country. There are also a number of 'subjective' factors that increase or decrease employment possibilities among refugees. Traumatic experiences during the war have often been the cause of depression, insecurity or emotional instability and decrease the capacity to find employment. In other cases, the perceived limbo situation of being neither 'here' nor 'there' can be a paralysing force and prevent refugees from actively seeking employment. A concrete example of the correlation between employment and capacity to contribute is provided from the Eritrean case studies. While every respondent -both employed and unemployed - reported that they regularly pay 2 per cent of their monthly income to the government of Eritrea, several reported that they simply could not afford to make the extra payments currently being requested. On the whole these respondents were either unemployed, or employed on an informal or part-time basis only. Economising and saving. It is worth noting that even some of those Eritreans in full-time employment reported that they were facing difficulties in meeting the additional demands being made by the Eritrean government. To an extent these difficulties arise from individual problems of economising, saving and prioritising expenditure. Several female Bosnian refugees reported that they have managed to save money from their household expenditure by finding bargain shops, baking their own bread and cakes, and buying less meat than is common for their diet despite living on income support. While Bosnians in the Netherlands appeared to be more satisfied with state allowances, some of those in the UK found it difficult to make ends meet. Those with additional income, either due to a war pension from Bosnia or because of employment in the host country, are often faced with a choice between saving money to return, supporting their families in Bosnia, or starting to build up a new life in the host country. Legal status. Even more significant than financial stability in terms of overall links and activities related to Bosnia, is the question of the legal status in the host country for refugees. Despite different policy and legal contexts, Bosnians in both the UK and the Netherlands expressed concern and insecurity with respect to their legal status. This insecurity not only hinders integration and prevents people from actively seeking employment, but also leads to psychological problems, which can cause apathy and non-commitment. There also tended to be a close correlation within the Eritrean community between legal status and employment and level of wages. Despite the range of legal statuses found amongst the respondents, none was legally prohibited from employment as a result of status. On the other hand, the insecurity of some statuses meant that in some cases respondents could only take on short-term jobs, and limitations on movement associated with certain statuses in Germany geographically delimited the job market for others. The lack of freedom to move within Germany also restricts ability to participate in Eritrean cultural events organised in other towns - one respondent reported being arrested in Frankfurt last year when it was discovered that his status restricted him to Berlin. 'Home country factors'. Of specific relevance in the Bosnian context, another extremely significant factor in determining capacities to contribute relates to Refugees and transnationalism 629 whether refugees used to live in an area where their ethnic group is now in a majority or a minority. As the greater number of Bosnian refugees in both the UK and the Netherlands are of Muslim origin, those who come originally from the area that is now under Serb control (Republika Srpska) are less capable of maintaining links with Bosnia. This is particularly true for social relationships, even if people try to maintain contact with displaced relatives and friends who now live in the Federation. While the government of Eritrea has always appreciated the value of the diaspora, it seems that the government is currently engaging with the diaspora much more actively than previously. The most obvious example is the opening of new channels for contributions by the diaspora. In addition, there has been an increasing incidence of visits to important countries of exile for the diaspora by government representatives, and other efforts to encourage the diaspora such as sending videos to community organisations. Gender. A less obvious but nevertheless significant obstacle to the involvement in economic, political and social activities revolves around issues related to gender roles and relations. Focusing specifically on the Bosnian case, it has to be stressed that the effects and consequences on gender relations vary greatly and cannot be delineated in terms of a simple cause and effect. Some women interviewed not only lost their previous home and loved ones, but also lost their economic independence and support networks. Many professional women were unable to find comparable jobs in the host country and were either forced to take up menial work or stay at home with the children. This situation has not only undermined their self-confidence but has also increased their dependence on their husbands. There is also some evidence for increased equality between spouses. This can be partly explained by the promotion of gender equality (in relative terms) in the host countries. At the same time, many non-professional women reported that it was much easier to find work for them than for their husbands. Their financial contribution to the household also gives them greater negotiating power in terms of deciding how the money is spent as well as greater freedom of movement. Desire to contribute Orientation towards the home government. Throughout the interviews, it became obvious that the desire to become involved in any kind of activity related to Bosnia very much depended on the political landscape of what used to be 'home'. Thus, for example, for those Bosnians who come from areas in which they would now constitute a minority (Bosniaks in Republika Srpska or Bosnian Serbs in Sarajevo, for example), the desire to contribute to reconstruction was generally very low. This often reflected both political opposition to the government and negative feelings towards the majority population in that area. Political opposition to the government not only affects those people who would comprise a minority in their home town, but was also significant for people coming from so-called 'majority areas'. Several interviewees stated that their lack of identification with the government and the general political culture prevents them from seeking greater involvement in reconstruction processes. In this context, 'nationalism', 'increased religiosity', and 'lack of democracy' were 630 N. Al-Ali, R. Black and K. Koser mostly mentioned as deterrent factors. For many refugees in mixed marriages, 'a decrease in tolerance' and the failure to implement some of Dayton's stipulations aiming at a multi-ethnic Bosnia have diminished their desire actively to participate in reconstruction. Among Eritrean respondents, political factors were also important influences on the desire to contribute. In particular, a strong distinction emerged between supporters and opponents of the government of Eritrea. This is a more complex division than it first appears, as opposition is spread across a range of groups who base their position variously on religion, regionalism and politics. It was striking that only one respondent, among several who admitted being opponents, said that he does not pay the 2 per cent quota of his monthly income. Others maintained that this contribution is still being spent on relief. They distinguished this payment from the variety of other payments, which they argued were being spent on the conflict, and to which they were less willing to contribute. As alluded to above, however, there are increasingly indications that even those Eritreans who have traditionally supported the government are beginning to lose their desire to contribute to the state. Now that their contributions are no longer needed to 'protect' Eritrea from attack (as they see it), they are beginning to question the government's motivations. They are critical of the government's engagement and withdrawal from the conflict. They resent the increased economic burdens that have been placed upon them by the state. And they are beginning to demand a greater direct representation in Eritrean affairs in exchange for their continued financial support. Information. Due to the extremely transitory nature of both the Bosnian state and its economy, several economic schemes are virtually unknown to the majority of refugees. Within the context of mass privatisation, for example, every single citizen is allocated a certain amount of assets in the form of a voucher. There exist five different vouchers (depending on length and type of previous work within Bosnia) with which people can buy shares in companies. The privatisation scheme also enables Bosnians to buy formerly socially owned flats. Lack of knowledge about the voucher privatisation system at the time of our interviews clearly decreased both capacity as well as desire to invest and financially contribute. In contrast, the Eritrean government has made great efforts to publicise its fund-raising exercises and to provide a range of economic incentives for remittances. The flexibility associated with the newly issued government bonds provides a good example. They can be redeemed in either US dollars or Nakfa - the new Eritrean currency. They can be transferred between individuals as long as the transfer is registered with the issuing office. And in Eritrea, bonds can be used at the Central Bank as credit notes for investment loans. Identification with community organisations. The level of identification with community organisations has been found to vary greatly amongst both Bosnian and Eritrean refugees, and to play a significant part in determining their desire to become involved in activities and events organised at a community level. Amongst Bosnians, the most obvious element influencing identification relates to ethnic background. Most Bosnian community organisations in both the UK and the Netherlands are generally perceived to be run by and cater for Bosniaks. This Refugees and transnationalism 631 was strongly denied by several community leaders who stress their openness to all ethnic groups. In addition to ethnic background, political views and attitudes also play an important role. The majority of those refugees who do not frequent community associations (in both host countries) describe them as nationalist and pro-government while stressing their own opposition to the existing government. Besides political divisions based on positions in the government in Eritrea, there are also indications that there are other divisions within the Eritrean communities in the UK and Germany that are more influenced by the politics of the diaspora. In particular, many Muslim respondents perceived the Eritrean community structure in the UK as being dominated by Christians. Their feeling of exclusion has limited their desire to participate in community-level activities ranging from charitable collections to festivals. Social factors. There is a complex range of other political and social variables which have been found to influence the desire of Eritrean respondents to contribute. These have not yet been subject to systematic analysis, but there are indications that legal status in the host country can be a significant variable, as can be a distinction between those born in the host country and those born in the home country. The correlations are not straightforward, as these factors have tended to combine with others to influence the overall desire of respondents to contribute. A good example of the complexity of these correlations is provided by a focus on the influence of age on desire to participate. Many elderly Eritreans maintain a generalised view that youngsters are not engaged in developments in Eritrea, and have little interest in playing an active part in the future of the country. In particular, they rue what they perceive as a lack of 'national consciousness' amongst youngsters. This has certainly been found to be true for some, who seem to have come to the conclusion that they should focus their attention on life in the host country. It has equally become clear during the research that other Eritrean youngsters are also amongst the most active. Most of the main Eritrean web sites, for example, have been started and are maintained by students, mainly in the USA. Similarly, both of the Eritrean magazines published in Berlin were founded by young Eritreans. Most of these web sites and both magazines have a sympathetic stance towards the current government, and it is interesting that there is a far clearer division between generations in terms of political opposition, which has been found to be largely absent amongst youngsters. Conclusions This article began with the observation that the study of transnationalism has largely bypassed refugees, or in rare cases has focused specifically on their political activities. Proceeding from recent perspectives in international migration studies, which suggest that there may be at best only a blurred conceptual distinction between refugees and other migrants, the paper has attempted to use the type of transnational analysis more commonly found among labour migrants to analyse two refugee groups. In the context of a project which seeks to understand the contributions that refugees can make to post-conflict reconstruction without returning permanently, it has extended the focus from political activities alone to show how refugees can also become involved in a range of 632 N. Al-Ali, R. Black and K. Koser economic, social and cultural transnational activities. At the same time, it has identified a range of obstacles which differentially influence the desire and capacity of the study populations to participate in these activities. Before drawing some conclusions about the links between refugees and transnationalism, a number of reservations about the wider applicability of these case studies are worth highlighting. First, as emphasised earlier in this article, the two refugee groups in question are in a similar but particular situation: namely the war from which they fled ended with a process of national independence; and by and large, they enjoy permanent residence rights in their host countries. However, the latter circumstance at least may not be as unusual among refugees as is often assumed to be the case, whilst the recent struggle over Kosovo provides a further high-profile example of a war ending in at least a proto-independence. Second, the preceding analysis is based on work in progress, and draws on interviews with only relatively small population samples. Even within these small samples, it has become clear through the course of the analysis that transnational activities and identities in both study groups can be more readily identified among certain respondents than others. In this context, it is interesting to bear in mind Alejandro Portes' caution that transnationalism can only really be said to exist where transnational activities and identities involve a significant proportion of the population (Portes et al. 1999). Here, we would wish to draw a distinction between saying that transnationalism exists, and saying that a group or community is 'transnational'. Thus, whilst it is probably premature to speak of either Bosnian or Eritrean refugees as 'transnational communities', it is certainly possible to identify features of 'transnationalism'. On this basis, some further observations about the links between refugees and transnationalism are worth highlighting. The first, and most straightforward, is that our research has established that transnational activities and identities can indeed flourish among refugees. We have found evidence of a range of economic, social, political and cultural activities which suggest that even where they opt to remain in their host countries permanently, refugees do not necessarily extinguish links with their home countries and transform into 'ethnic minorities'; rather they can actively maintain those links and develop a transnational character. This finding has important policy implications, especially in a policy environment that still sees return as the optimum method for enhancing the contribution of refugees to reconstruction (cf. Hallergard 1998). But it also has more conceptual implications for the study of transnationalism. Even if these case studies prove to be exceptions, they are still exceptions that prove a rule, namely that there is no reason why refugees should not be more fully incorporated into the study of transnationalism. Building on this preceding point, what these particular case studies have permitted is to spotlight two communities in a dynamic situation, as transnational features develop among them. As the war from which they originally fled has ended, and as they have achieved permanent residence status in their host country, so both Bosnians and Eritreans have moved from a situation of temporary exile, in which transnationalism has been limited to political activism against regimes in their home country, to permanent exile, where transnationalism has been extended to an active involvement in contributing to reconstruction in their home country. In other words, these case studies allow for the possibility of charting the evolution of transnationalism. The implication of this conclusion Refugees and transnationalism 633 extends beyond the refugee context alone. It is that transnationalism is not a 'state of being', as is sometimes implied by the existing literature, but rather that transnationalism is itself a dynamic process. Notes 1 See www.transcom.ox.ac.uk 2 Fieldwork among Bosnian refugees has been carried out in the UK and the Netherlands, whilst fieldwork with Eritrean refugees has been carried out in the UK and Germany. Informal interviews with about 30 Bosnians and Eritreans of mixed backgrounds (Muslim, Serb and Croat in the case of Bosnians, and Muslim and Christian in the case of Eritreans) were carried out in each receiving country. However, due to the demographic make-up and the level of identification with Bosnia, the majority of Bosnian respondents were of Muslim origin. Participant observation during social events in community organisations and family gatherings added another dimension to the information obtained during interviews. Interviews were also conducted with representatives from host governments, community organisations and relevant non-governmental organisations. Within Bosnia-Herzegovina, about 35 people were interviewed in four different cities within the Muslim-Croat Federation. 3 All names used in this article are pseudonyms. 4 www.primenet.com/ ~ ephrem/Asmara/nmpark/index.html References Al-Ali, N., Black, R. and Koser, K. (1999) 'Mobilisation and participation of transnational exile communities in post-conflict reconstruction'. ESRC Research Programme on Transnational Communities: Interim Report. Bäsch, L., Glick Schiller, N. and Szanton Blanc, C. (1994) Nations Unbound: Transnational Projects, Postcolonial Predicaments and Deterritorialized Nation-States. Amsterdam: Gordon & Breach. Diaz-Briquets, S. and Perez-Lopez, J. (1997) 'Refugee remittances: conceptual issues and the Cuban and Nicaraguan experiences', International Migration Review, 31(2): 411-47. Hallergärd, C. (1998) 'Bosnia and Herzegovina: problems and progress in the return process', Forced Migration Review, 1: 21-6. Koser, K. (1998) 'Social networks and the asylum cycle: the case of Iranian asylum seekers in the Netherlands', International Migration Review, 31(3): 591-612. Koser, K. (1999) 'From refugees to transnationals?' Paper presented at the International Conference on New Approaches to Migration, University of Sussex, 21-22 September. Koser, K. and Black, R. (1999) 'The end of the refugee cycle?', in Black, R. and Koser, K. (eds) The End of the Refugee Cycle? Refugee Repatriation and Reconstruction. Oxford: Berghahn, 2-17. Landolt, P., Autler, L. and Baires, S. (1999) 'From Hermano Lejano to Hermano Mayor: the dialectics of Salvadoran transnationalism', Ethnic and Racial Studies, 22(2): 290-315. Levitt, P. (1998) 'Social remittances: migration-driven local-level forms of cultural diffusion', International Migration Review, 32(4): 926^8. McDowell, C. (1997) A Tamil Asylum Diaspora. Oxford: Berghahn. Portes, A., Guarnizo, L.E. and Landolt, P. (1999) 'The study of transnationalism: pitfalls and promise of an emergent research field', Ethnic and Racial Studies, 22(2): 217-37. Portes, A. and Rumbaut, R.G. (1990) Immigrant America: A Portrait. Berkeley: University of California Press. Richman, K. (1992) 'A lavalas at home/a lavalas for home: inflections of transnationalism in the discourse of Haitian President Aristide', in Glick Schiller, N., Bäsch, L. and Blanc-Szanton, C. (eds) Toward a Transnational Perspective on Migration. New York: New York Academy of Sciences, 189-215. Styan, D. (1993) 'Eritrea 1993: the end of the beginning', in Allen, T. (ed.) In Search of Cool Ground. London: James Currey, 80-95. Vertovec, S. and Cohen, R. (1999) 'Introduction', in Vertovec, S. and Cohen, R. (eds) Migration, Diasporas and Transnationalism. London: Edward Elgar, xiii-xviii. Wahlbeck, O. (1999) Kurdish Diasporas: A Comparative Study of Kurdish Refugee Communities. Hound-mills: Macmillan. 634 N. Al-Ali, R. Black and K. Koser Author details Nadje Al-Ali is Lecturer in the Institute of Arab and Islamic Studies at the University of Exeter. E-mail: n.s.al-ali@exter.ac.uk Richard Black is Reader in Geography and Co-Director of the Sussex Centre for Migration Research at the University of Sussex. E-mail: r.black@sussex.ac.uk Khalid Koser is Lecturer in Human Geography at University College London. E-mail: kkoser@geog.ucl.ac.uk Copyright of Journal of Ethnic & Migration Studies is the property of Carfax Publishing Company and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. Copyright of Journal of Ethnic & Migration Studies is the property of Routledge and its content may not copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.