Spaces of Recognition

An analysis of Somali-Danish associational engagement and diasporic mobilization

Nauja Kleist

Ph.d.-afhandling Sociologisk Institut

Spaces of Recognition An analysis of Somali-Danish associational engagement and diasporic mobilization

Ph.d.-afhandling Sociologisk Institut Københavns Universitet

Indleveret til bedømmelse: December 2006 Offentligt forsvar: April 2007

Hovedvejleder: Lars Bo Kaspersen, Sociologisk Institut Bivejleder: Ninna Nyberg Sørensen, Dansk Institut for Internationale Studier

Bedømmelsesudvalg: Peter Gundelach, Sociologisk Institut (formand) Peggy Levitt, Sociology Department, Wellesley College Anita H. Fábos, School of Social Science, Media and Cultural Studies, University of East London

Omslag: Klavs B. Thomsen Opsætning: Anne-Lise Schulze Andersen

ISBN-13: 978-87-7296-242-9 ISBN-10: 87-7296-242-9

2 Table of contents

PART I DEPARTURES...... 9

Chapter 1 Introduction...... 11 Puzzles and concerns...... 12 Research questions ...... 17 Processes of inclusion and transnationalism ...... 18 Integration or inclusion? ...... 19 Transnational involvement...... 20 Recognition and positioning ...... 22 Contributions of the study ...... 23 Organization of the study ...... 24

Chapter 2 The theoretical journey. Locations and destinations ...... 27 Constructed and co-constructed social realities...... 27 Situated knowledges ...... 29 Three mappings ...... 31 Simultaneous engagement...... 33 The transnational turn ...... 33 Modes of incorporation and transnationalism...... 36 Associations in Denmark ...... 38 Migrant associations in a transnational perspective...... 41 Fields and social space...... 43 Diaspora ...... 46 Between transnational communities and diasporic identification...... 47 A Somali ‘diaspora’? ...... 48 Diaspora as claim and aspiration ...... 49 Recognition ...... 52 Between intact identity formation and participation...... 52 Considerations and questions...... 56 Differentiation and cultural intelligibility...... 58 Chapter resume: Linking the three mappings...... 62

Chapter 3 Directions. Methodological considerations and analytical strategies...... 65 Analytical principles...... 65 Location-work: Defining and exploring ‘the field’...... 65 Asking, looking, listening...... 67 The research site and the process ...... 69 The positioned researcher ...... 74 Producing and processing data ...... 77 The informants...... 78

3 The interviews ...... 80 From fieldwork to analysis...... 81 Chapter resume: From theory to data to analysis...... 83

PART II ACROSS BORDERS...... 85

Chapter 4 Journeys and settlements. Outline of the Somali political and migration history ...... 87 Stories of Somali culture...... 88 Colonialism...... 93 Colonial migration...... 94 ‘I always wanted to be a sailor’: Mohamed...... 95 Independence ...... 96 Oil workers ...... 98 ‘I became another person’: Aisha...... 100 Civil war...... 101 Refugees ...... 104 ‘The war never stopped’: Omar...... 106 Transnational connections...... 109 ‘Some of us were born here, so we think it’s pretty normal’: Halima....111 Continued migrations ...... 113 Chapter resume: Scattered dreams and changing mobility ...... 114

Chapter 5 Ambivalent encounters. Life in Denmark ...... 117 From multi-national state to small nation-state...... 117 Guest workers and refugees...... 120 Somalis in Denmark...... 123 In the eye of the hurricane ...... 126 Ambivalent encounters...... 132 ‘It’s like Denmark has two faces’: Saphia and Fatuma ...... 133 ‘You have to manage your life’: Mahmoud and Abdirizak...... 136 Gender change and gender nostalgia...... 139 Masculinity at risk ...... 141 Identifications and categorizations...... 145 Between strangeness, Danishness and Somaliness...... 148 Recognizing difference...... 149 Chapter resume: Ambivalence and (mis)recognition...... 152

Chapter 6 ‘Making a difference’. Somali-Danish associational engagement...... 155 The Somali-Danish associational field...... 156 Fiery souls: A history of Somali-Danish associations ...... 158 ‘In Denmark, you need to organize yourself to be heard’: Gelle...... 161 Social principles of organization...... 163

4 Gendered and religious spaces...... 163 Clan affiliation ...... 167 Territorialization of loyalties ...... 171 Continuums of identification ...... 174 The Women’s Association ...... 175 Three afternoons in August and September...... 177 Dealing with everyday challenges ...... 179 ‘I was always a nationalist’: Noora...... 182 Learning processes ...... 185 New political culture and opportunities...... 187 Establishing and negotiating positions ...... 190 ‘Ordinary members’ and spokespersons ...... 191 Chapter resume: Platforms for action...... 195

Chapter 7 Mobilizing ’the diaspora’. Transnational involvement ...... 197 The Development Organization of East Africa ...... 198 A meeting between important men ...... 199 Homeland politics ...... 200 ‘Clan is a bridge we need to cross’: Omar...... 203 Principles of governance and identification...... 205 Returning to the desert? The Somscan & UK Cooperative Associations... 208 Organized return ...... 209 Buying land, negotiating lineage ...... 210 Creating a Somali-Scandinavian neighbourhood...... 211 The squatter incident...... 216 ‘The diaspora’ at work ...... 219 Social remittances ...... 221 ‘Walking with two legs’: Salman ...... 224 Chapter resume: Agents of modernity and development ...... 227

PART III HORIZONS ...... 229

Chapter 8 Spaces of recognition ...... 231 Struggles for recognition ...... 232 Walking around Southall ...... 232 Recognizing achievement ...... 233 Repairing gender (dis)order ...... 236 Empowerment or cultural intelligibility?...... 239 Symbolic capital and symbolic struggles...... 242 Processes of inclusion and transnational involvement ...... 245 Between fragmentation and collaboration ...... 245 At the centre...... 247 In the margins ...... 250 Migrant associations and processes of inclusion ...... 251

5 In the name of diaspora...... 255 Between homeland and society ...... 255 Diasporic identification ...... 257 The re-birth of a nation...... 258 Recognizing ‘the diaspora’...... 262 Chapter resume: Spaces of recognition...... 264

Chapter 9 Conclusion...... 265 Struggles for recognition and negotiations of positions...... 265 Destinations, commitments and loyalties of associational involvement ...... 267 Contexts of exit and reception ...... 269 Transnational engagement, diasporic mobilisation and inclusion ...... 271 Moving on: Thinking beyond this study ...... 273

Post-script...... 277

Acknowledgements ...... 279

List of references...... 281

English Summary ...... 309

Dansk resume...... 313

6 Tables of figures

Table 1. Somali refugees in neighbouring contries of asylum, 1990-2003 ...... 104

Table 2. Somali refugees in selected western contries of asylum, 1991-2003 ...... 105

Table 3. Somali asylum applications in selected western countries, 1993-2002...... 107

Table 4. Number of Somalis in Denmark, 1990-2005 ...... 125

Table of maps

Map 1. Colonial and political borders as well as indication of the Somali-speaking population...... 90

Map 2. Political map of Somalia...... 91

PART I

DEPARTURES

9 10 Chapter 1 Introduction

Denmark, February, 2003: “In your opinion”, Abdirizak Ali rhetorically asks me, “what can we do to change our situation in Denmark?” We are having a tea break and Abdirizak Ali, a Somali-Danish teacher and the father of seven children, has come over to chat with me. “Our situation is really difficult”, Abdirizak continues. “We need to do something”, he adds with a serious expression, before he breaks into hearty laughter. Though Abdirizak laughs, his remark is not a joke. He and nineteen other Somali-Danish men and women are participating in a seminar, organized by the Somali Network in Denmark. They have gathered to organize network activities in Denmark to further collaboration and ameliorate the situation for Somali-Danes. He and the other participants are all active in different associations or in local Somali media; some have an academic background, others work as teachers, or are students. Now they are discussing, laughing, and teasing each other at such a rate that my Somali translator re- marks that, “we really like to talk, you know”.

During the afternoon, we go through the agenda, written on the blackboard. It includes ‘bi-cultural identity’, upbringing and education of children and youth, the image of the Somali group in Denmark, reconstruction in Somalia, and media communication. The issues are debated with passion. Abdirizak Ali poses his question again: “What can we do to change our situation?” This question turns up again and again throughout the day in discussions and conversations in the breaks. Omar Said, a business student involved in several associations and local radio, calls for more pan-Somali cooperation. Others talk about reconstruction and repatriation. “We need a coordinating body of reconstruction projects”, Omar says. Little do I know, that in three months I will attend the constitution conference for a transnational umbrella organization together with Omar and several of the other seminar participants. Abdirizak mentions an ambitious project, the Somscan & UK Cooperative Associations, that aims to

11 establish a new neighbourhood in Somaliland1. Neither do I know that in nine months, I will spend two hot and dusty days, visiting the construction site of the Somscan & UK project in Somaliland. For now, my attention is directed towards the issues debated here: Integration. Culture. Education. Tolerance. Dialogue between different Somali-Danish groupings and between Danish society and the Somali-Danes. Using the media! Writing letters to the editor! In- volving the local associations! Don’t spread negative views in Danish society – respect it! Stop thinking about the clan! Cooperate! Don’t just talk – do something!

The meeting is recorded and later broadcast on local Somali radio stations in Denmark. I write the seminar report in Danish. A Somali Internet feature with photos of the participants is published on a Somali homepage, resulting in much interest from other Somalis, it appears. When I later talk to the organizer, he tells me that they have received responses from other Somalis all over the world. They are enthusiastic about the ‘Danish diaspora initiative’ and praise them for their reconciliation efforts. Furthermore, the questions brought up at the seminar are shared concerns. “Many Somalis in the diaspora”, he explains, “worry about the same question of how to manage their lives in a new society, as well as the reconstruction of Somalia”. This question is the heart of the matter of my dissertation.

Puzzles and concerns In February 2003 I start fieldwork for my dissertation, spending a day with the Somali Network in Denmark, listening to their discussions, writing the seminar report and later discussing the seminar with the organizer. This is not my first work among Somalis in Denmark2 and I am curious to see how the people I know are doing and what they are up to. In 1999, I carried out fieldwork for my Masters thesis on Danish repatriation politics and Somali senses of belonging

1 I refer to Somaliland to indicate the geopolitical region, corresponding with the self-declared Republic of Somaliland. I do not wish to take any stance on the question of the political recognition of Somaliland. 2 I refer to Somalis in Denmark and Somali-Danes interchangeably. As I come back to later in the dissertation, not all Somalis in Denmark describe themselves as So- mali-Danish. It is thus my categorization, meant to emphasize the possibility of belonging to both places at once.

12 (Fink-Nielsen and Kleist 2000). Now I want to follow up on that study, but with a different perspective: dynamics between processes of inclusion in Danish society and transnational involvement. This overall theme addresses central academic and political debates, often revolving around questions concerning the ability of migrants to involve themselves in their country of origin and still ‘be integrated’, or whether transnational relations and identification impede loyalty to and engagement in the host country. I find that these questions are important to analyse, but I want to approach them from a migrant perspective as the voices of those who are supposed to ‘integrate’ are too often left out of both academic and political debates. The Somali-Danish seminar seems to be a good place to begin this exploration. It is thus my first step into larger ethnographic fieldwork among Somalis in Denmark engaged in associations, taking me not only to various locations in the greater Copenhagen area, but also to Somaliland and London. Below, let me explain the background of the study and why I have chosen to focus on Somalis in Denmark. Since the eruption of the civil war in Somalia in 1988, more than one million Somalis (UNDP 2001, 61)3 have been living outside the country. Somalis are now scattered almost all over the world. Some Somali refugees have followed in the footsteps of pre-civil war migration, such as educational and work-related migration to the former colonial powers of Italy and Great Britain, and some have arrived in hitherto almost unknown destinations such as Denmark. Den- mark is thus a new destination for Somali settlement and only a very few So- mali citizens lived in the country before the 1990s. However, since the end of the 1990s, the Somali population has constituted one of the largest groups of so-called third country residents4 in Denmark, comprising about 16.500 people in 2006, including descendants and naturalized citizens (Udlændingestyrelsen 2006, 55). In Denmark, the Somali group as a whole has been received rather ambiguously. While the Danish acceptance rate of asylum seekers was amongst the highest in the world in the 1990s (UNHCR 2003a), the Danish media, many

3 The 2001 Human Development Report for Somalia refers to Nair, KN.N. & F. Abdullah (1998): Somalia 1997-1998: A Status Report. Nairobi: UNDOS. More than one million people is indeed an impressive number when compared to the ap- proximate number of people living in Somalia of 6.8 million people in 2002 and 7.5 million before the War (UNDP Somalia and World Bank 2003, xii). 4 Third countries refer to non-EU member states. Third country residents are nationals from outside the European Union, residing in an EU member state.

13 politicians and social workers portrayed Somali asylum seekers as bogus and as ‘very difficult to integrate’ (cf. Fadel, Hervik, and Vestergaard 1999; Jørgensen and Bülow 1999; Skak 1998). The marginalized position is emphasized by the high rate of unemployment and social exclusion5. Given the huge internal differences within the Somali-Danish group, there is no such thing as the ‘average Somali-Dane’. Nevertheless, the average employment rate of Somali citizens and descendants in 2005 was 23 per cent for men and 10 per cent for women (Ministry of Integration 2005b, 174)6. This is the lowest employment rate amongst all ethnic groups in Denmark. The majority of Somali citizens in Denmark are thus on social welfare or in employment activation programmes (Bjørn, Pedersen, and Rasmussen 2003, 30), though many men have some educational background (ibid., 49). During my first fieldwork in 1999, many of the informants spoke at length about feeling unwelcome in Denmark and their ambivalent fields of belonging. Given the rather hostile public and political exposure in the middle and late 1990s, this may come as no surprise (cf. Møller and Togeby 1999). Still, it makes me curious about what Somalis in Denmark actually do to deal with their situation. Likewise, it makes me wonder whether and how involvement in Denmark and transnationally is related to positions in Denmark and Somalia. Or in other words, what the contexts of exit and reception mean; i.e. departure from and possible transnational relations to Somalia, as well as arrival in and processes of settlement in Denmark. When I start fieldwork in 2003, the ‘foreigner question’ – udlændingespørgsmålet – has become even more politicized, and asylum and integration legislation has been tightened. Social subordination – whether experienced at the personal level or in relation to the media and political discourse of ‘the foreigners’ and ‘the difficult Somalis’ – is again a key theme in my interviews. Furthermore, a number of the informants explicitly articulate their reflections of life in Denmark in terms of lack of recognition,

5 A large Danish survey of poverty and social exclusion (Larsen 2004) shows that ethnic minorities (measured in the report as people with a non-Danish mother tongue) are especially exposed to social exclusion and poverty. The report does not differentiate between different ethnic groups. Other vulnerable groups include elderly people, retired persons, single mothers and unemployed. Many Somalis in Denmark are unemployed and/or single mothers. 6 For convenience, I refer to the Ministry of Refugee, Immigration and Integration Affairs as the Ministry of Integration.

14 while others more indirectly focus on this issue. Associational engagement is just one theme amongst others in my interview guide at this initial stage of my fieldwork. But I quickly become more curious about this issue. Most of the informants I know from 1999 are active in various associations and I find that several of them have realized – or strive to realize – dreams and ambitions they voiced back then through associational engagement. They talk about ‘making a difference’ and about getting respect for their efforts. It becomes clear to me that the Somali associational life in Denmark is very active, with quite a lot of events and activities focusing on social and cultural life in Denmark. Likewise, there are a number of collective reconstruction initiatives and conferences concerning the role of the so-called ‘diaspora’, i.e. the globally dispersed Somali population. Like other Somalis in the West and the Gulf countries, many Somalis in Denmark are engaged in a substantial remittance economy which links various countries of residence with Somalia. A large majority of the population in both Somalia and Somaliland live below the poverty line and remittances and transnational engagement form crucial contributions to the economy and the survival of many families (see for instance UNOCHA 2005; Horst 2003; Gundel 2002). Most remittances are sent on an individual basis to families to support their livelihoods and, sometimes, to invest in housing and land, but there are also collective reconstruction projects, supported and organized by Somali organizations in Denmark and elsewhere. The transnational dimension makes Somali-Danish associational engagement even more intriguing in relation to the question of integration in Danish society: while associational activities focused on integration are widely supported and seen as a valuable part of social and cultural life in Denmark, transnational activities focusing on the ‘homeland’ are not necessarily encouraged or appreciated. At first, I am very impressed by the scale and commitment that I witness when I talk to organizers and activists in associations. The seminar referred to above is a case in point. However, the more I talk to the informants, the more I realize that Somali associations are divided and sometimes contested. Some activists I know from 1999 are ‘missing’ from certain activities and others seem to have stopped collaborating with each other. The principles of organization guiding migrant associations and associational engagement are more complicated than I first assume, and seem to be organized around notions of gender, status and education – as well as the political orientation towards Somalia and

15 Somaliland. It seems that the Somali-Danish associational scene is not only an active one, but also a political one with transnational dimensions. I thus become more curious about processes of recognition and positioning in relation to associational involvement, both in terms of ‘making a difference’ as well as in relation to conflict. I am taken by further surprise when many of the more engaged Somali migrants I talk to articulate their engagements in diasporic terms7. Phrases such as ‘the diaspora should do something’ or ‘Somalis in diaspora’ have entered the vocabulary of Somali migration as a way of explaining the position and situation of Somalis abroad, not only articulated by researchers, but also by migrants in Denmark and elsewhere in the West, by actors in Somalia and Somaliland, and by western and Somali policy makers. ‘The diaspora’, it seems, has become a way of categorizing transnational engagements and politicizing and mobilizing moral commitment. While I had found inspiration in diaspora theories in my Masters thesis, the concept was not part of the informants’ own vocabulary to describe their transnational engagement in 1999. However, while the informants do use the term to refer to the globally displaced and dispersed Somali population, they mainly do so in relation to political or transnational involvement and mobilization. I am thus forced to reconsider the implications of the concept of diaspora again. Given that ‘the diaspora’ is a broad and inclusive collective identity category and that Somali-Danish local and transnational engagement seem to be both flourishing and divided, I become curious of how and why activists mobilize in the name of diaspora. Somali-Danish associational engagement thus constitutes an interesting case, I think. Firstly, the position of Somali-Danes on the whole is a marginalized one, turning one’s attention to questions of recognition, positioning, and processes of inclusion and exclusion. Secondly, there is a vivid Somali remittance economy linking Somalia and Somaliland with the so-called ‘diaspora’ through different kinds of transnational practices, relations and transfers which are both expected and, indeed, needed. Thirdly, there is large number of Somali-Danish associations which both address processes of inclusion in Denmark and recon-

7 Generally speaking, diaspora studies focus on dispersed populations that are expelled or displaced from a real or imaginary homeland, theorizing the relationship between belonging, identity and place (cf. Clifford 1997; Cohen 1997) as well as narratives and identification (Axel 2004).

16 struction in Somalia, constituting a setting for socially and politically engaged Somali-Danes to enact and display agency and responsibility, as the vignette above indicates.

Research questions As mentioned above, the main theme of this dissertation is the relationship between processes of inclusion in Danish society and transnational involvement. I explore this theme through a case study of Somali-Danish associational engagement with a specific focus on struggles for recognition and negotiations of positions within and across borders. The study is based on ethnographic fieldwork among Somali-Danish associational key persons and organizers in Co- penhagen, and supplemented with fieldwork in Somaliland and London. It thus focuses on a specific group, not the general Somali population in Denmark, as the Somali-Danish informants are among the most well established Somalis in Denmark in terms of employment and education. Furthermore, let me also make clear that the associations included in the study either focus on social and cultural issues in Denmark or reconstruction in Somalia or Somaliland – or, as we shall see later, both. I do not include any religiously-defined associations in the study. This delimitation was necessary both because of problems of access to Somali Muslim associations and because a focus on Islam8 would require different theoretical and analytical perspectives in this already quite complex study.

The study is organized around the following research question: How do Somalis in Denmark negotiate positions and struggle for recognition in Danish society and transnationally through associational engagement?

8 In a report prepared for the now abolished Board for Ethnic Equality, it is estimated that about one third of Somali associations are religious or cultural (Møller and Togeby 1999, 45). One of the Somali-Danish associational key persons estimated that almost half of the Somalis in Denmark regularly join some kind of religious activity, though they might not be member of a religious association as such. The importance of Somali religious associations thus seems to be considerable.

17 In order to make this question both more concrete and more theoretically informed, I ask three supplementary questions. 1. Why and how do Somali-Danes involve themselves in associations – and towards which destinations, commitments and loyalties are they oriented? 2. To what extent and how do the contexts of exit and reception shape their engagement? 3. How does transnational engagement and diasporic mobilization relate to processes of inclusion in Danish society?

These questions have both empirical and theoretical dimensions. I have already presented some of the empirical background to Somali-Danish associational engagement and go into more detail with the theoretical aspects in the next chapter. Below, I unravel the relationship between processes of inclusion and transnationalism a bit further to spell out the sociological and political relevance of the dissertation and its contributions.

Processes of inclusion and transnationalism Focusing on the relationship between processes of inclusion and transnationalism, my dissertation is situated in a field which has both sociological and political aspects. Let me start by locating the dissertation in relation to the Danish political context of migration and migrant incorporation. This is important, not only because it is part of the context for the study, but also because (at least some of) the situation in Denmark reflects the situation in other western welfare states. Like in many other countries, migration has become a highly politicized, debated, and contested issue in Denmark. Migrants and minorities’ incorporation in Danish society, their loyalties to Denmark and Danish norms and values are considered increasingly important, and debates around these issues have received massive political and public attention. In Denmark, these debates are subsumed under the term integration, and though there is little reference to what integration is and how it should be achieved, it is broadly agreed that it is necessary (cf. Mortensen 2000). Recently, adherence to Danish values has been incorporated in official Danish integration policies, with tests in Danish culture and history as well as a ‘declaration of integration and active citizenship in

18 Danish society’ that applicants for a permanent residence permit must sign (Ministry of Integration 2006; 2005b). In other words, integration is a buzz- word, repeated over and over again, not least in relation to whether and how the presence of Muslim third country residents and their descendants challenge the future of the Danish nation-state and welfare state (cf. Hervik 2002; Necef 2001). The question of ‘the foreigners’ has indeed become a key political question, dividing the Danish population and politicians. In spite of – or because of – this debate, there is no agreement about what the concept of integration implies, not least because the term is often used to refer to both the process and the goal (Emerek 2003; Ejrnæs 2002; Mikkelsen 2001). Still, the main concern of Danish integration policy is self-provision, and increasingly also acceptance of Danish values. While the labour market and public schools are considered the main keys to integration, associational activities are also emphasized for having integrative potential. Associations – foreninger – are often considered something specifically Danish: a cornerstone in Danish civil society and culture, a means for exercising active citizenship, and an opportunity for creating social networks (Minister for ligestilling 2005; Ministry of Integration 2003). There are and have been considerable opportunities for obtaining public funding for projects undertaken by associations, especially if they are considered to further integration in Denmark. Still, in spite of this interest, there has been relatively little research on migrant associations in a Dan- ish context (exceptions include Dahl and Jakobsen 2005; Mikkelsen 2003b; 2002; Hjære and Balslev 2001; Hammer and Bruun 2000). This dissertation addresses this lack of research, contributing to our knowledge of migrant associations in Denmark. Furthermore, it does so both in relation to involvement in Denmark and transnationally.

Integration or inclusion? To write about integration, however, is to open a Pandora’s Box of political and normative claims and speculations, and I use the term with some caution, not to say concern. The term integration is “extremely fuzzy, with no fixed, readily agreed meaning” (Grillo 2001, 11), often used as a synonym for assimilation. So why refer to it at all? I do so because I find that the questions and debates about integration are important both to discuss and to nuance. It is a term that makes people ‘open their ears’. To use it is to enter a debate, with which I do

19 not necessarily agree, but which I nevertheless find is important to contribute to. Still, as the term is so politically and normatively loaded, I find it is difficult to use it as an analytical concept. I will use the concept processes of inclusion, rather than integration, to refer to processes of involvement between groups of actors, rather than the underlying notions of societal harmony and identification which the concept of integration often denotes (Mortensen 2000). Furthermore, as Larsen points out, integration does not necessarily imply equality between groups; some groups can be ‘integrated’ in ways which exclude others (Larsen 2004, 71; cf. Daly and Saraceno 2002). Theoretically, the concept of integration is often related to functionalist ideas of society as a whole into which different elements – for instance migrants and minorities – must be incorporated to avoid disintegration and disharmony (Ejrnæs 2002, 5; cf. Andersen 2003; Mortensen 2000). The focus is thus on the maintenance of social order. By focusing on processes of inclusion, I want to avoid these associations. However, it is impossible to avoid the concept of integration all together. As indicated above, it is a keyword in political and academic discussions on migrants’ inclusion and therefore used a lot. When I refer to integration, it is with reference to these discussions as well as to Danish politics and debates concerning migrants’ inclusion. To avoid confusion, I will state explicitly if I refer to political debates or use the concept as a sociological term. In the latter case, I use it to refer to general societal processes of furthering cohesion through identification with specific political principles or ideas of culture (Mortensen 2000; Habermas 1994). When I refer to the overall question of different patterns of migrant adaptation to and inclusion in their countries of residence, I will use the term incorporation. Processes of inclusion can thus be seen as a more specific aspect of this.

Transnational involvement Reflections on processes of inclusion versus integration are especially important to bear in mind in a Danish context where migrants’ transnational practices are generally ignored or considered disturbing to their integration (as it is termed in Denmark) – for instance in relation to so-called corrective education

20 trips. While there certainly is transnational research undertaken in Denmark9, research on transnational aspects of migration in a Danish context is still a relatively new field (cf. Emerek 2003; Mikkelsen 2002), and there are only very few works focusing on linkages between transnationalism and processes of inclusion in Denmark (see Østergaard-Nielsen 2002 for an exception). In an international perspective, however, there is considerably more interest in migrants’ transnational practices. Since the early 1990s, an influential and elaborate literature on migrants’ transnational practices has developed (Schiller, Basch, and Blanc-Szanton 1992), focusing on, among other things, social relations, identities, remittances, and institutions across borders. Studies of the relationship between integration and transnational engagement suggest that there is no contradiction between the simultaneous involvements in the country of residence and origin (cf. Levitt and Schiller 2004; Portes 2003; Itzigsohn and Saucedo 2002). However, these studies – as well as many other transnational ones – are based on American or other multicultural contexts which are quite different from the Danish case (see also Al-Ali and Koser 2002b). By focusing on a migrant group in Denmark, this dissertation addresses this lack of research. While inclusion in Danish society and transnational practices might seem to be two distinct and well-separated fields, I will investigate if this is really the case in a Danish context. If we turn the analytical gaze from the logic of the nation-state to a focus on migrants’ lives, the simultaneous involvement in Denmark and transnationally might well be part of migrants’ everyday lives, priorities and, as the seminar organizer said, their worries. I thereby situate the study in what sociologist Peggy Levitt and anthropologist Nina Glick Schiller have termed “a larger intellectual project, undertaken by scholars of transnational processes in a variety of fields, to reformulate the concept of society, such that it is no longer automatically equated with or confined by the boundaries of a single nation- state” (2004, 1003-1004). Ultimately, such a reformulation would imply recon- sidering the notions of integration and inclusion as well.

9 For instance at the Danish Institute for International Studies (DIIS) and the Acad- emy for Migration Studies in Denmark (AMID) as well as smaller research envi- ronments at University of Copenhagen and Roskilde University Centre.

21 Recognition and positioning There are many ways to approach the overall relationship between processes of inclusion and transnational involvement, and my particular focus is on struggles for recognition and negotiations for positions in Somali-Danish associational engagement. The vignette from the seminar and my fieldwork reflections suggest that Somali-Danish associations are organized and institutionalized responses to the obligations that Somalis in Denmark deal with, both within and across borders. But there are more things going on. As I also indicated, the associational scene is a divided and political one – with regard to political orientation, gender, clan etc. Different actors seem to struggle for different aims and purposes, sometimes collaborating, sometimes not. While some, as Omar Said at the seminar pointed out, emphasize the need to collaborate, others prefer to establish their own associations. And while probably all would agree with Ab- dirizak Ali when he says that ‘we need to do something’, they might not agree on what should be done. In this dissertation I thus explore what activists in Somali-Danish associations are struggling for. More precisely, my interest is in the associational engagement, meaning that the analytical unit comprises individuals and their social relations and networks, and not the associations as such. What especially interests me is how organizers and activists use associations, and for what purposes. This focus broadens the analytical attention to the informants’ lives before and outside the associations – in Somalia and in Denmark (and possibly elsewhere) – and how they respond to the challenges and opportunities they face. Towards where and to whom are the loyalties and commitments in the associational engagements directed, and how? Do the informants use associations to negotiate social positions; do they establish, repair or regain social esteem and respect; or do they take the opportunity to negotiate and challenge former or existing social hierarchies? And do they seek recognition in Danish society from other Somalis in Denmark, Somalia, or the so-called diaspora? These are questions that I explore further in the dissertation. Recognition and positioning also have theoretical aspects which I will come back to in more detail in the next chapter. Recognition has become a central theme in recent sociological and political discussions, where intense debate focuses on how struggles for recognition should be understood and analysed in relation to identity politics, self-realization, and societal participation (e.g. Fraser

22 and Honneth 2003; Taylor and Gutman 1994). Recognition has thus become a keyword in relation to contemporary social struggles. At the same time, I am curious about the negotiations of positions –positioning – that might take place as part and parcel of such struggles. Broadly speaking, positions refer to location and standing within social spaces or discourses. I am especially inspired by social constructivist work on the discursive and performative establishment and negotiations of identification and subject positions (Davies and Harré 1990; Hall 1990). Another analytical aspect of the term recognition is thus (re-)identification: being recognized or recognizable as something or somebody (Butler 2004; Ahmed 2000). Positioning thus refers to how notions of, say, ‘proper’ masculinity, femininity, ‘Danishness’, or ‘Somaliness’ are articulated and enacted in various ways which might stabilize or challenge their content in different contexts. With this analytical duplicity, I link the discussion of recognition with positioning.

Contributions of the study The literature on Somali migration has expanded significantly while I have carried out my research. As the focus of literature varies a great deal, it does not make much sense to provide a literature review of Somali migration studies here10. Rather, I will present the literature as it becomes relevant in relation to the theoretical reflections or empirical analysis. Main themes in the studies focus on ethnicity, gender, clan and Islam in relation to resettlement in western countries of residence (e.g. Fangen 2006a, 2006b; Kallehave 2003; Farah 2000; Griffiths 2002; McGown 1999), community involvement and associational engagement (e.g. Hopkins 2006; Griffiths 2002), and diasporic and transnational involvement (e.g. Horst 2004; 2003; Montclos 2003). The specific contribution of this dissertation is the development of an analytical approach which connects the discussion of processes of inclusion with transnational engagement, and which theoretically combines diaspora, recognition and positioning theory. I am thereby able to shed light on struggles for recognition and negotiations of positions in relation to Danish society, transnational involvement, and diasporic mobilization. I find that this perspective is

10 The conference proceedings of the International Congresses of Somali Studies prove a good measure of the rich variety of Somali Studies (Lilius 2001; Adam and Ford 1997).

23 both adequate and fruitful in relation to migrants’ lives, and that it is under- explored in existing research. The dissertation thus offers both new theoretical suggestions and empirical insights. I start this exploration through the case study of Somali-Danish associational engagement within and across borders. As I hope to show, the combination of perspectives throw new light, not only on each other, but also on questions of inclusion and associational engagement. I thereby wish to offer some new ways of understanding and conceptualizing contemporary migration and the relationships between processes of inclusion and transnational involvement. This perspective is necessary and relevant in relation to both academic and political discussions. More specifically, the dissertation contributes to the research on migrant associations, diasporic mobilization and gender relations, as well as to the growing interest in migrants as agents of development. Finally, it contributes to our knowledge of Somali migration, an issue which is receiving substantial policy interest in Denmark and elsewhere.

Organization of the study I have divided the dissertation into three parts. In Part I: Departures, I present the theoretical and methodological background that informs and organizes the study. In chapter two, I introduce and discuss the theoretical mapping around which I approach and analyze the data: simultaneous engagement, diaspora, and recognition theory. In chapter three, I present my methodological approach: how I have ’translated’ the theoretical reflections into analytical principles for the production of data and its analysis. In this chapter, I also present my fieldwork and the empirical material in more depth. Readers less interested in theoretical and methodological reflections can go directly to chapter four.

Part II: Across Borders contains four chapters. In chapter four, I analyze the context of exit: Somali history with emphasis on migration from the colonization to the present time, focusing especially on the civil war and transnational involvement. In chapter five, I take a closer look at the context of reception: Somali migration to and settlement in Denmark. To analyze recognition and positions, I also explore how the informants articulate their lives in Denmark and the problems they face – not the least in relation to break-up of families and gender relations. In chapter six, I turn to Somali-Danish associational engage-

24 ment. I analyze the social principles of organization guiding them to analyze negotiations of positions and struggles for recognition, as well as how associations form platforms for actions, and from which positions and recognition can be established and negotiated. I continue these discussions in chapter seven with specific focus on transnational involvement and diasporic mobilization: claims and identification in the name of the diaspora.

Finally, in Part III: Horizons, I return to the theoretical aspects of the research questions posed. In chapter eight, I investigate how processes of recognition, positioning and cultural intelligibility play together in different ways. Likewise, I discuss the relationship between transnationalism, diasporic mobilization and identification, and processes of inclusion. I conclude the entire dissertation in chapter nine. Finally, I include a short postlude where I follow up on the informants, sketching the continuous journey in time and space.

25 26 Chapter 2 The theoretical journey. Locations and destinations

Theory is meant to orient, to provide the roughest sketch for travel (Haraway 2004, 63)

Donna Haraway suggests that theory is an orientation device in explorations of the world – a “mapping exercise and travelogue through mindscapes and land- scapes” (ibid.). As the organization of this dissertation bears witness to, I like to think of research as a journey. The travel metaphor is used not only to reflect the movements, border crossings, and shifting horizons involved in migration and transnational involvement; it also emphasizes the process aspect of research, of learning along the way, of journeying through familiar and unknown terrain. To navigate, the traveller – the researcher – needs to locate herself in the same way as she needs maps and directions. In this chapter, then, I present and discuss the theories which have oriented my work; which have inspired me, guided me to destinations, and among which I strive to create new theoretical dialogues. It is thus, as Haraway writes, a mapping of the theories which informs the way I have constructed the research questions and the data, and shaped my analytical gaze. In this chapter, I strive to clarify how the theories function as directions, and which considerations and questions they raise. I start by positioning the research in relation to the overall question of sociology of science, i.e. the nature of knowledge.

Constructed and co-constructed social realities In the SAGE Handbook of Qualitative Research, social scientists Egon Guba and Yvonna Lincholn argue that there has been a “distinct turn of the social sciences toward more interpretive, postmodern, and critical practices and theorizing”, or in other words, a “non-positivist orientation” (2005, 191). I position my dissertation as a part of this turn, in what could be termed a moderate social constructivist position, focusing on “constructed and co-constructed realities” (ibid., 193, my emphasis). Stressing co-construction, I want to emphasize the

27 dialectical and interaction dimension of social reality. This position fits well with my overall interest in processes of inclusion, positions and recognition, which are of an inherently social and inter-subjective nature, as well as with my emphasis on the contexts of exit and reception which focus our attention on both individual life circumstances and overall structures. My position should not be confused with ‘vulgar’ or radical social constructivism, where individuals ‘just’ invent or construct the world as she or he pleases in terms of sheer in- tentionality (cf. Andersen 2005; Collin 1998). The world is more complicated than that. As Professor of Law and black feminist Kimberlé Crenshaw has termed it, “to say that a category such as race or gender is socially constructed is not to say that that category has no significance in our world” (1994, 112; cf. Staunæs 2004, 83). Gender and ‘race’, to take this example, are categories with histories and consequences; however, these consequences are not set in stone, but change and are negotiated over time and space, sometimes being reproduced and reinforced, and sometimes challenged. Furthermore, as I will get back to in more detail later in the chapter, they are doings, not inner cores, always done “with and for another” (Butler 2004, 1). That social realities and social categories are constructed and co-constructed thus does not mean that they are freely chosen. Rather, they are embedded in political, economic and cultural circumstances. Neither are life trajectories the result of the constructions of the individual alone, but are also situated in circumstances which the individual has not chosen – into which he or she has been ‘thrown’, so to speak – and which do not disappear because she or he ignores them (cf. Bourdieu and Wacquant 1992, 97). Being born in Somalia, a poor country governed by a dictator and then thrown into chaos by civil war, is a case in point. Still, there is no independent or inner force, which determines the ways individuals make sense of their lives and, for example, their possible experiences of the Somali civil war or the asylum system in Denmark. Rather, this is a question of individual and inter-subjective meaning-making processes, both embedded in structural contexts as well as in personal biographies and life circumstances. Therefore, while I do not subscribe to ontological realism, I do emphasize that there are power hierarchies and overall economic and political circumstances that the individual (normally) cannot change. In this sense, I follow sociologists James A. Holstein and Jaber F. Gubrium’s insistence on inter- pretative practices, which overcome the divide between an exclusive focus on

28 either construction processes of social reality or on their structural embeddedness. Holstein and Gubrium strive to overcome this duality through a simultaneous focus on “the constellation of procedures, conditions, and resources through which reality is apprehended, understood, organized and conveyed in everyday life” (2005, 484). They continue:

Interpretive practice engages both the hows and the whats of social reality; it is centered in both how people methodologically construct their experiences and their worlds, and in configurations of meaning and institutional life that inform and shape their reality-constituting activity […] The dual concern not only makes it possible to understand the construction process but also foregrounds the realities themselves that enter into and are produced by the process” (ibid., 484, emphasis in original).

The double focus is thus an aspiration to avoid the radical versions of social constructivism where context and constraints are ignored or, in the opposite extreme, to avoid structural determinism or empiricism (cf. Haraway 1991). The question is thus what the consequences of this position are for the production of knowledge.

Situated knowledges On the epistemological level, I follow the constructivist position that knowledge is always historically and socially mediated and constructed. Meaning and ‘truth’ are temporary products of ongoing struggles for stabilizing some visions and perceptions of the social world – and of what counts as legitimate knowledge and research (cf. Guba and Lincoln 2005; Jørgensen and Phillips 1999; Laclau and Mouffe 1985). There is no definite and objective truth to which the researcher has privileged access through means of objective procedures, no “well-polished Cartesian mirror of the mind” (Smith and Hodkinson 2005, 916), no ‘God trick’ (Haraway 1991). What are the consequences for validity in relation to this epistemological position? Validity is, as Patti Lather writes, “the power to determine the demarca- tion between science and not-science” (2001, 243). However, with the non- positivist and social constructivist turn, there are no neutral or universal criteria. This means that the positivist validity criterion of correspondence and its focus on a ‘match’ between the researcher’s categorizations and ‘objective truth’ is challenged (Lather 2001, 244; Kvale 1997, 234-235). Rather, I propose three criteria, which I find are relevant for assessing validity in qualitative, construc-

29 tivist studies like mine: craftsmanship, reflexivity, and fruitfulness. The first criterion of craftsmanship (Kvale 1997, 236-239) refers to the credibility of the study (Charmaz 2005, 528; cf. Richardson and St. Pierre 2005, 964) or adequacy (Bech 2005, 26). This criterion relates to the quality and adequacy of the data production and the analytical strategy, familiarity with the settings, and whether or not the researcher makes a persuasive and well-argued analysis. The criterion of reflexivity demands that the researcher critically investigates the implications of the knowledge practices and techniques employed, including reflecting on the positions of the researcher and the researched. Or in Richard- son’s words, is there “adequate self-awareness and self-exposure for the reader to make judgments about the point of view?” (2005, 964). Finally, the criterion of fruitfulness (Bech 2005, 26; Richardson 1997, 964) emphasizes that ‘good research’ should not only be adequate and well-made, but also inspiring, spurring new ways of thinking about the problems examined, and/or holding potential for contribution to social change or action. To be sure, epistemological constructivism thus does not surrender to relativism in the sense that ‘anything goes’ or that methodological considerations are irrelevant (cf. Andersen 2005). Rather, the impossibility of theory-free and value-free knowledge is, as Smith and Hodkinson express it, “nothing more or less than recognition of our human finitude. It is not something to be tran- scended; rather it is merely something with which we, as finite beings, must learn to live” (2005, 922). The consequence of this epistemological position is a focus on location, situatedness, and partiality, not a rejection of endeavour, rig- our, and care in the research process, as I have just argued. Rather, I follow Donna Haraway’s (1991) insistence on situated knowledges, making explicit the specific locations and partial claims of the researcher. Knowledge is always partial, always incomplete, always articulated from somewhere. Situated knowledge might seem like a modest understanding of research, compared to classic empiricist science as well as Bourdieu’s (and Weber’s) aims of establishing models which include both the immediate or phenomeno- logical understanding of a field as well as the underlying “objective truth” (Bourdieu and Wacquant 1996, 231, my emphasis; Bourdieu and Krais 1997). My ambition is to offer a theoretically informed analysis – not as the only possible ‘truth’, but as a qualified and inspiring one. Though I do not claim uni- versalism, I do strive towards empirically and theoretically grounded research

30 with analytical suggestions that may not only shed light on the actual subject matter, but also highlight other related problems. Claiming not to know everything is not the same as not to know anything at all. In the words of Haraway,

So I think my problem and ‘our’ problem is how to have simultaneously an account of radical historical contingency for all knowledge claims and knowing subjects, a critical practice for recognizing our own ‘semi- otic technologies’ for making meanings, and, a no-nonsense commitment to faithful accounts of a ‘real’ world (ibid., 187, emphasis in origin).

To me, Haraway’s triple insistence of historical contingency, self-critical and reflexive knowledge practices as well as the commitment to making faithful accounts of a ‘real’ world, as she writes, remains one of the most accurate and inspiring aspirations of what research should strive to be. It matches the criteria I presented above, emphasizing the importance of positioning and commitment. “Faithful accounts of the ‘real’ world” is not recourse to realism, but to analysis of the socially constructed and co-constructed life worlds, embedded in power relations and structural contexts, and analyzed from specific theoretical perspectives and social location. In the remaining part of the chapter, I map the theoretical perspectives which inform the dissertation, and in the next chapter, I reflect upon my social location in relation to the production of data and the analysis.

Three mappings I have divided the chapter into three theoretical mappings: simultaneous engagement, diaspora, and recognition. Each one refers to different aspects of the research questions, but there is a focus on positioning running through all three of them. Firstly, I present the transnational perspective which makes migrant engagement within and across borders appear as a relevant problem to study in the first place (cf. Bourdieu and Wacquant 1996, 220). Though the dynamics between migrant transnationalism, inclusion, and incorporation is an under- researched issue in the Danish context, it feeds into a large international literature on transnational practices and (migrant) associations which I introduce. Secondly, I introduce diaspora studies with their focus on the overall relationship between loyalties, identification and involvement – as well as diaspora as political claims and mobilization. Thirdly, I turn to recognition theory to present a different set of theoretical discussions to analyze and challenge the em-

31 pirical material in ways which the transnational and diaspora perspectives do not necessarily offer. While transnational and diaspora studies are often connected (and sometimes confused), their combination with recognition theory is still in the making, offering new theoretical perspectives to migrants’ involvement within and across borders. The purpose of the rest of the chapter is three-fold. Firstly, it serves as a literature review of the theoretical discussions which inform the dissertation. Sec- ondly, I construct my analytical approach through presenting and positioning the theoretical perspectives which inform and guide the dissertation – mapping the locations I am writing from. I ask a range of questions along the way, ‘thinking aloud’ so to speak, to show how the theoretical reflections inspire the analysis. Thirdly, through the combination and juxtaposition of these three perspectives, I provide the basis for new theoretical discussions in the analysis, relating the case study to the theory. Finally, let me note here that I use the term migration and migrants to include different practices of movement and different categories of people who move or who have moved (cf. Castles and Miller 2003; Sørensen, van Hear, and Engberg-Pedersen 2002; Brettel and Hollifield 2000). In my usage, refugees, asylum seekers, family reunified, migrant workers, naturalized citizens etc. are all included in the overall category of migrants. I refer specifically to the legal categories when this is relevant, for instance in relation to rights. Categories such as refugee and asylum seeker, for example, are of a legal, not existential, nature, reflecting rights in the international and national immigration and asylum regimes, but often also a good deal of coincidence, luck or misfortune11. That said, let me hasten to emphasize that there are of course different circumstances, motivations, and indeed emergencies, which differentiate migrants. I do not by any means want to downplay or ignore the suffering and struggles caused by war, humanitarian catastrophes, poverty and desperation, such as for instance those due to the Somali civil war; indeed most of the Somali-Danish informants in the study are recognized refugees. The point is that migrants fleeing from civil war are not only defined by their flight and need for asylum: they are also people with different economic resources, family-related obligations, and personal ambitions. It is thus necessary to employ a perspective which can

11 Change in legislation, for instance, has consequences for who can become recognised as refugees, or can naturalize etc.

32 include these experiences and circumstances and avoid the unfortunate division between refugee studies and diaspora and transnationalism studies (cf. Wahl- beck 2002). This is thus yet another reason for combining theoretical perspectives.

Simultaneous engagement In this first mapping, I present the theoretical perspectives of transnational studies vis-à-vis processes of simultaneous connection, incorporation scenarios and migrant associations. I am especially inspired by two overall sets of literature. On the one hand, by sociological and anthropological studies of migrants’ everyday social relations and identifications across borders, focusing on socio- cultural and socio-political aspects of migrants’ lives. The concepts of social space, field, simultaneity, and positions are central here. On the other hand, I am inspired by political sociology and political science work on associations. Here, questions of incorporation scenarios and social capital are central. I use the term simultaneous engagement to bring together these two perspectives.

The transnational turn With a focus on migrant involvement in Denmark and transnationally, my study is constituted through a transnational optic. I define transnational involvement as practices and relations of non-state actors, which span national borders. The transnational perspective offers an overall perspective to analyze migration as a relational process with a focus on relations, practices, and institutions spanning national borders, rather than the clearly delimited and separated notions of emigration and immigration. Transnational studies thereby have a potential to refo- cus the research attention and imagination (Al-Ali and Koser 2002a, 2), making new ways of conceptualizing empirical phenomena possible. This study, for instance, would simply appear impossible or irrelevant in an approach where migrants’ engagement outside the nation-state is ignored. Transnational studies can thus be seen as corrections to what Andreas Wim- mer and Nina Glick Schiller have termed methodological nationalism to refer to “the assumption that the nation/state/society is the natural social political form of the modern world” (2002, 302). The consolidation of the nation-state system has shaped ways of perceiving and receiving immigration, which in turn “have influenced, though not completely determined social science theory and meth-

33 odology, and, more specifically, its discourse on immigrant and integration” (ibid., 301-302). The challenge is to move beyond the container theory of society as corresponding to the territory of nation-states, naturalizing and territorial- izing the social science imaginary (ibid., 307, see also Beck 2000)12. Rather, we need an analytical move beyond nation-state boundaries. The end of the Cold War was a turning point in the social sciences, leading to an increasing interest in globalization and new ways of conceptualizing migration (Wimmer and Schiller 2002, 321). As is often mentioned, the global re- structuring of capital and the following reorganization of work and production as well as the development in communication and transport technologies have facilitated and accelerated migration (e.g. Castles and Miller 2003; Guarnizo and Smith 1998). From the 1990s, a growing and influential literature on transnational practices has developed13. The ‘transnational turn’ takes as its point of departure that migration must be analyzed as an ongoing process, rather than emigration and immigration as two separate fields of analysis. In Nations Un- bound, one of the first works on transnational migration, transnationalism was defined as “the processes by which immigrants forge and sustain multi-stranded social relations that link together their societies of origin and settlement” (Basch, Glick Schiller, and Szanton Blanc 1994, 7). This particular focus on migrants’ social relations has been supplemented and challenged in a range of other works, not least in relation to the fact that both migrants and their counterparts in the country of origin and elsewhere – relatives, business partners, political collaborators etc. – take part in these processes (van Hear 2002b; Levitt 2001, 198). Many transnational studies – and indeed most of the first ones – take as the point of departure Central and Latin American work-related migration to North America, often focusing on the homeland-host country axis (e.g. Itzigsohn and Saucedo 2002; Goldring 2001; Schiller and Fouron 2001; Portes, Guarnizo, and

12 As, among others, Benedict Anderson (1998; 1994; 1991) has shown, the rise and spread of nationalism was indeed a trans-border process. The term methodological nationalism thus relates to the territorially bounded aspects of nation-states and might be termed methodological nation-state territorialism. 13 I am not going to provide a state-of-the-art review of transnational migration literature here. The transnational literature is by now so huge that such a task would require much more time and effort than is suitable for my purpose here. See the special editions of International Migration Review 2003, 37(3), Ethnic and Racial Studies 2002, 22(2) and Global Networks 2001, 1(3) for more details.

34 Landolt 1999; Schiller 1999; Smith and Guarnizo 1998; Sørensen 1995; Basch, Glick Schiller, and Szanton Blanc 1994). However, as Al-Ali and Koser have pointed out (2002a), studying transnational practices of refugees and taking as the point of departure regions other than America raises new theoretical challenges and questions. It is only in recent years14 that the transnational perspective has been more systematically applied to European (and other) countries of destination and to studies of refugees (see for instance Faist 2004; Horst 2003; Al-Ali and Koser 2002b; Al-Ali, Black, and Koser 2001; Østergaard-Nielsen 2001). There is by now a huge literature examining various kinds of transnational phenomena such as identity formation, family relations, religion, remittances, development and transnational political mobilization. Though the ‘transnational turn’ has been enthusiastically received, it has also been criticized. One of the criticisms directed towards many transnational studies is an exaggeration of the scope and importance of transnational activities, as well as an overextension of the concept (Portes 2003; 2001; Heistler 2000; Schiller 1999), which therefore leads to a risk of it becoming “an empty conceptual vessel” (Guarnizo and Smith 1998, 4). Transnationalism, these critics warn, should not be seen as the new doxa of migration studies15; it should not be presupposed, but should be viewed as one possible outcome among others. Furthermore, it has been further argued that in some – maybe most – cases, it is the minority of migrants who do more than remit money to their relatives every now and then (Dahinden 2005; Portes 2003, 876-877; Itzigsohn and Saucedo 2002). This is an important critique, which emphasizes the importance of specifying the degree – and indeed existence or non-existence – of transnational involvement. To accommodate this critique, I am including both associations oriented towards issues in Denmark and transnationally oriented associations which support reconstruction projects in Somalia. I thereby want to avoid con-

14 The Transnational Communities Programme in the UK (www.transcomm.ox.ac.uk 2006), running between 1997 and 2003, as well as the EU-funded Network of Excellence on International Migration, Integration and So- cial Cohesion (www.imiscoe.org 2006), initiated in 2004, are two examples where European perspectives are central. 15 Likewise the claimed novelty of transnational practices – especially found in the first, enthusiastic works – has been criticized and refuted (see for instance Schiller 1999).

35 fusing the evidence of transnational practices with their significance (cf. Portes 2001). A second criticism cautions against conceptualizing migrants and their transnational practices as unbounded and deterritorialized actors and phenomena (Guarnizo and Smith 1998). This critique is especially relevant in relation to the early works of transnationalism (e.g. Basch, Glick Schiller, and Szanton Blanc 1994) and in some cultural studies (e.g. Appadurai 1996). Transnational actors and actions are not free-floating, sociologists Luis Eduardo Guarnizo and Mi- chael Peter Smith warn. They write:

[T]ransnational actions are bounded in two senses – first, by the understandings of “grounded reality” socially constructed within the transnational networks that people form and move through, and second, by the policies and practices of territorially-based sending and receiving local and national states and communities (1998, 10).

Guarnizo and Smith thus call for transnational studies to be located and histori- cized (ibid., 29) on both the ‘indigenous’ and structural levels. Their call for location is echoed by other studies, which emphasize the importance of studying the context of exit and reception (Portes 2003; Itzigsohn and Saucedo 2002). I share this view. I present the Somali context of exit in terms of the Somali political and migration history in chapter four and the context of reception and settlement in Denmark in chapter five. In the remaining part of this first mapping, I now turn to two more specific parts of the transnational literature: modes of incorporation and migrant associations.

Modes of incorporation and transnationalism As indicated above, transnational studies can be seen as a correction to the Chi- cago School conceptualization of migrant assimilation as a naturally occurring process (e.g. Park 1928). Assimilation was the dominant sociological paradigm from the 1920s to the late 1960s when theories started to emphasize the structural and societal aspects of incorporation (Heistler 2000). At the same time, assimilation was challenged by political models of multiculturalism and integration, where the latter has been characterized as striving towards “‘equal oppor-

36 tunity, coupled with cultural diversity, in an atmosphere of mutual tolerance’” (Jenkins in Grillo 2001, 11)16. Transnational studies have challenged models of assimilation and the idea that migrants – gradually or immediately – cut relations and identification towards their country of origin. Following Levitt and Schiller, “incorporation of individuals into nation-states and the maintenance of transnational connections are not contradictory social processes” (2004, 1003). Rather, they propose the term simultaneity of connection as a median point in migrants’ simultaneous transnational attachment and incorporation processes, rather than opposite bina- ries. They write:

Instead, it is more useful to think of the migrant experience as a kind of gauge which, while anchored, pivots between a new land and a transnational incorporation. Movement and attachment is not linear or sequen- tial but capable of rotating back and forth and changing direction over time. The median point on this gauge is not full incorporation but rather simultaneity of connection. Persons change and swing one way or the other depending on the context, thus moving our expectation away from either full assimilation or transnational connection but some combination of both (2004, 1011).

I will use this term as a way of bringing together various experiences and forms of simultaneous engagement and connections with a focus on the possible relationship between processes of inclusion and transnational relations; as a way of overcoming the divide between either assimilation or segregation. This focus thus displaces the classic sociological question of: ‘What happens to migrants in the receiving countries?’ (Heistler 2000, 77) to: What happens to migrants in the receiving countries and to their possible transnational attachment, and how do such processes interact? Such a reformulation brings to fore questions of how migration, incorporation and transnational practices relate to and shape each other, instead of asking when transborder relations stop and assimilation begins, or simply ignoring transnational practices. Recent surveys of migrant transnational activities indicate that incorporation and transnational involvement is not a contradiction (Snel, Engbersen, and Leerkes 2006; Portes 2003; Itzigsohn and Saucedo 2002). In a survey of 300 migrants in the Netherlands, it is argued that transnational activities do not im-

16 Grillo refers to the so-called ‘Jenkin’s formula’ as formulated by British politician Roy Jenkins in the quote.

37 pede integration, contrary to widespread belief in that country and in Denmark (Snel, Engbersen, and Leerkes 2006, 287). On the contrary, migrants of diverse nationalities, education, and employment situations were engaged in different kinds of transnational activities, especially political and socio-cultural ones. However, the authors conclude that, although “generally speaking, transnational involvement does not constitute an impediment to successful integration into Dutch society, the situation for specific marginalized migrants groups may actually be quite different” (ibid. 305). Their conclusion thus emphasizes that it is necessary to pay particular attention to the concept of exit and reception of specific groups (cf. Portes 2003; Itzigsohn and Saucedo 2002). It also raises the question of what the marginalized position of Somalis in Denmark means for their transnational engagement via associations. Associations thus constitute a particular context within the broader context of reception. In the following, I outline this context and how migrant associations have been conceptualized in relation to processes of inclusion and incorporation.

Associations in Denmark Voluntary associations – foreninger – have a long history in Denmark. Associa- tions are regarded as a cornerstone of Danish democracy and civil society, and have played an important role in the development of Danish society and the welfare state. Danish sociologists Lars Bo Kaspersen and Laila Ottesen (2001) argue that the development of associations first furnished the common patriotism of the latter part of the 18th century. In the 19th century, when patriotism turned into nationalism, associations furthered and strengthened nationalist ideas and the notion of a Danish people. In the Constitution of 1849, citizens were granted freedom of association: to join and form associations for any law- ful purpose without prior consent (ibid.; Gundelach and Torpe 1999, 70; Sco- cozza and Jensen 1994, 230). Since then, a huge number of organizations and associations have been established in almost all spheres of society, including sports, banking, professional organizations, politics, religion, culture, education, leisure time activities, animal protection etc. Associations are commonly defined as “a group of people with common interests” (Ministry of Integration 2003, 106) and associational life is and has been encouraged and economically supported by the Danish state to support the civil society, including a ‘democratic education’ and folkeoplysning – literally ‘enlightenment of the people’.

38 These ‘democratic roots’ are mirrored in the usual organization of Danish associations, with a statute established at a statutory meeting, an annual general meeting with a report of the past year, and the democratic election of the executive committee. These conditions must be met for associations to receive any kind of public support. In many ways, associations can thus be seen as ‘mini-democracies’ and as a central way of participating and gaining influence in Danish democracy and society (Torpe and Kjeldgaard 2003; Kaspersen and Ottesen 2001; Mikkelsen 2001, 112ff). Furthermore, it is often claimed that associations have an important bridging function in Danish society in the sense of providing citizens with forums which transcend social differences and positions. Danish research on associations suggests that associations further two overall processes of incorporation: between different groups of citizens, and between citizens and the political system (Gundelach and Torpe 1999, 77-78). According to this view, associations may enhance societal cohesion and inclusion insofar as their members interact with the surrounding society rather than isolate themselves in segmented subcultures. The idea that associational engagement and, more broadly, civic engagement can further societal collaboration and trust has been most explicitly articulated by political scientist Robert Putnam in his concept of social capital (2000; 1995; cf. Gundelach and Torpe 1999). Putnam defines social capital as “features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefits” (1995, 2). It is thus a collective good, practised and produced by civic engagements such as voluntary associations. According to Putnam, the higher the ‘stock’ of social capital is in a given community, the smoother, the more trusting and more efficient the social interactions of the community17. Gundelach and Torpe, however, posit that the role of associations is changing in Danish society. In contrast to earlier times, they claim, associations no

17 Putnam’s work has both been much celebrated and much criticized. According to Portes (2000; 1998), Putnam’s (earlier) work suffers from a logical circularity in that he does not separate the sources and effects in his analysis, i.e. collective social capital and the effects on civic engagement. Or, in other words, it cannot be settled whether members in associations are more tolerant and open because of their engagements, or if they joined associations because they were already open, trusting etc. (cf. Stolle 2001). Likewise, Portes criticizes Putnam and other proponents of the concept for ignoring the possible negative functions and effects of social capital.

39 longer bridge the social gap between different groups in society (1999, 89). Likewise, many new associations are inward-looking and disengaged from public and political life (ibid., 85). Gundelach and Torpe thereby point to what Put- nam has termed the bridging and bonding dimension of social capital. Bridging social capital refers to networks of organization which function across social institutions and groups. It is good for creating linkages – or bridges – to “external assets and for information diffusion” (2000, 22). Bonding social capital, on the other hand, is inward looking, focusing on the reproduction and support of established groups, bonding people with others like themselves, and thereby furthering closed and inward-looking social networks; thus it does not possess the potential for societal inclusion. Putnam thereby accommodates some of the critique that he has been overly celebrating the positive dimensions of social capital (Portes 1998). The ambivalent nature of associations has become a fre- quently discussed issue in Denmark, not least in relation to radical political and religious groups. Kaspersen and Ottesen, on the other hand, do not share the worries of societal fragmentation due to the possible bonding character of migrant associations. Rather, they call attention to a more procedural liberalist vision of society in which the only necessary political consensus is adherence to the Danish constitution. Furthermore, they claim that the function and integrative potential of associations is overestimated in the first place. They write:

We claim that it is a myth that voluntary associations assembled people across social and cultural barriers. Moreover, we claim that the key aspect of the Danish version of associative democracy is the democratic structure and function of the associations. The educational aspect of participating in an association where each member to some degree had to relate to democratic procedures is far more important than the cross- cultural or cross-class aspect (2001, 128).

For these authors, it is thus the democratic learning processes of associational engagement and not the possible social and cultural processes of inclusion which are the most important effect and biggest virtue of associational life. They thus place minor emphasis on the function of bridging capital in associations. Still, Kaspersen and Ottesen support the idea that migrant associations can contribute to processes of societal inclusion, not because of their claimed capacity for furthering cross-cultural and cross-class incorporation, but because of their potential for political inclusion.

40 Migrant associations in a transnational perspective The relationship between associational involvement and processes of inclusion is thus a central theme in Danish sociological research on organizations, focusing on the bridging-bonding potentials of associations as well as democratic learning processes. Still, with the exception of the work of political scientist Flemming Mikkelsen, only relatively little has been written specifically about migrant associations in Denmark (Mikkelsen 2003b; 2002; 2001; see also Østergaard-Nielsen 2002; Hjære and Balslev 2001; Hammer and Bruun 2000; Togeby 1999). I define migrant associations as associations formed by and for a particular ethnic or national migrant group, what Mikkelsen (2003a) terms ethno-national associations (cf. Cordero-Guzman 2005, 894). Mikkelsen argues that ethno-national migrant associations are mostly occupied with issues regarding everyday life in Denmark, rather than transnational activities, and that involvement in associations increases migrants’ political engagement. He concludes that migrant associations constitute an important tie between Danish society and ethnic minorities and hold a central position in the incorporation efforts (2003a, 165; cf. Østergaard-Nielsen 2002). Mikkelsen thus agrees with the positive evaluation of migrant associations in relation to processes of inclusion. Apart from Mikkelsen and Østergaard-Nielsen, there is, to my knowledge, hardly any research on migrant associations in Denmark that includes a transnational perspective. Internationally, migrant associations have received more attention, for instance in relation to hometown associations18 (Caglar 2006; Moya 2005; Levitt 2001; Landholt, Autler, and Baires 1999), the impact of the political opportunity structures19 in the country of residence (Hooghe 2005; Vermeulen 2005), and the relationship between migrant associations and the migration process (Cordero-Guzman 2005; Layton-Henry 1990). It is the latter literature which is

18 Hometown associations (HTAs) are organizations made by and for migrants from the same town or parish in the country of origin “who congregate primarily for social and mutual-aid purposes” (Caglar 2006). 19 In a recent special issue on migration associations of the Journal of Ethnic and Migration Studies, Schrover and Vermeulen emphasize the importance of group characteristics as well as the local opportunity structure in an analysis of migrant associations (2005, 828). The term political opportunity structure refers to the “consistent – but not necessarily formal, permanent, or national – signals to social or political actors which either encourage or discourage them to use their internal resources to form social movements” (Tarrow 1996 quoted in Hooghe 2005, 978).

41 of most interest to me here. Political scientist Zig Layton-Henry suggests that migrants’ transnational associational engagement is a temporary phenomenon20, which nonetheless promotes processes of societal inclusion in the country of residence. He presents an ideal-typical phase model, following a linear temporal development moving from home country orientation, through a middle phase with dual orientation, towards country of residence orientation (1990, 103). Ac- cording to him, “paradoxically, [migrants’] desire to defend their ethnic and identity and community forces them to participate more in the policy-making processes” (ibid., 102). In his view, then, there is no contradiction between homeland and transnational orientation as it paves the way for inclusion in the country of residence. His model thus supports the overall positive evaluation of the potential for political inclusion via associational engagement – albeit in a longitudinal perspective – as suggested by Kaspersen and Ottesen. Still, Layton-Henry’s model supports the overall assumption inherent in assimilation theory that migrants inevitably give up any transnational orientation and that this gradual reorientation makes migrant organizations central in the incorporation processes. Thus, returning to the metaphor of the migrant process as a gauge with a median point of simultaneous connection, we can see that though Layton-Henry assumes that migrant associations do go through a median point of dual orientation, he still presumes that this simultaneous orientation gradually disappears. Over time, migrants ultimately ‘swing’ towards the country of residence. This is in contrast to Cordero-Guzmán, who suggests a more dynamic relationship. Based on an analysis of migrant associations in New York City, Cordero-Guzmàn claims that migrant associations are “involved in all stages of the immigration and adaptation process” (2005, 908), in the reconstruction of social ties, in community formation, in representation in politics, and in transnational linkages. While he thereby also emphasizes a tem-

20 Writing before the ‘transnational turn’, Layton-Henry does not employ the vocabulary of transnationalism, but refers to home country orientation. His model is explicitly based on one type of migration pattern: young single male labour migrants, initially remitting money to their families in the country of origin and establishing social clubs with fellow nationals, until they establish themselves with their families in the country of residence and attain new associational needs and priorities. This scenario was indeed typical of much labour migration to Western Europe in the late 1960s and early 1970s (e.g. Castles and Miller 2003), but does not necessarily reflect all migration processes.

42 poral development, the difference is that for Cordero-Guzmàn incorporation does not require a gradual orientation away from the country of origin. There are thus two stances in the literature on associations in relation to processes of inclusion. On the one hand, following Putnam, an emphasis on associational engagement as having the potential to bridge and connect different groups – or in contrast – to further bonding and segmentation; on the other hand, following Layton-Henry, an understanding that involvement in associations furthers political inclusion in a society. While the issue of transnational engagement is not discussed in the first approach, the question as to whether or not it is a temporal phenomenon is debated in the second. My position is that the orientation of migrants’ associational engagements and the relationship between processes of inclusion and transnational engagement is dependent on empirical analysis. We cannot presume simultaneity of connection, but we can examine whether this is the case. It is my point of departure that simultaneous, transnational involvement and connection is possible. That said, it is clear that we need to develop analytical tools which can capture the relational, spatial and simultaneous dimensions of migrants’ involvement within or across borders. Below, I discuss some analytical concepts which focus on these dimensions.

Fields and social space The main theoretical claim in transnational studies is, as Levitt and Schiller write, that it is necessary to rethink “the boundaries of social life” and the concept of society (2004, 1003-1004) as stretching beyond the borders of nation- states. The lives of (some) migrants – and indeed many other people as well – cannot be satisfactorily analyzed if confined within national borders. We therefore need alternative ways of analyzing social relations. One approach is Bourdieu’s analytical concept of field. Bourdieu defines field as

a network, or a configuration, of objective relations between positions. These positions are objectively defined, in their existence and in the de- terminations they impose upon their occupants, agents or institutions, by their present and potential situation (situs) in the structure of the distribution of specifies of power (or capital) whose possession commands access to the specific profits that are at stake in the field, as well as by their objective relations to other positions (Bourdieu and Wacquant 1992, 97).

43 Bourdieu thereby focuses on how social actors engage in games of positioning vis-à-vis each other, struggling for various kinds of capital – be it economic, social, cultural or symbolic (cf. Bourdieu 1986). He thus emphasizes the relational nature of a field as well as its game-like nature, comparing it to an arena of conflicts and struggles (Bourdieu and Wacquant 1992, 98), focusing the research on the principles of organization and positioning, on the kinds of capital that are at stake, and on the possible conflict within a field. We may question whether Somali-Danish engagement constitutes a field in Bourdieu’s sense, and if so, what the involved actors struggle for. To answer this question, however, presupposes that the concept can be employed in a transnational context. Ac- cording to Levitt and Schiller, this is not yet the case: the concept of field has not been sufficiently theorized within a transnational perspective. Instead, they propose the term social field as

a set of multiple interlocking networks of social relationships through which ideas, practices, and resources are unequally exchanged, organized, and transformed […] National social fields are those that stay within national boundaries while transnational social fields connect actors through direct and indirect relations across borders (2004, 1009; cf. Goldring 1998).

Schiller and Levitt’s usage of the concept is thus broader than Bourdieu’s notion. A major difference is that their concept does not imply that the actors involved are ‘unified’ in their struggles for obtaining economic, social, symbolic (or other kinds of) capital within this field, following some unwritten game-like rules. Rather, their concept bears resemblance to geographer Doreen Massey’s notion of space21. Massey argues that space is of an inherently social and dynamic nature; the spatial is ”social relations stretched out” in “an ever-shifting social geometry of power and signification” (1994, 2, 3). She thereby emphasizes the shifting, possibly asymmetrical power relations in social space with the notion of ‘power geometry’. This implies

the existence in the lived world of a simultaneous multiplicity of spaces; cross-cutting, intersecting, aligning with one another, or existing in relations of paradox or antagonism. Most evidently this is so because the so-

21 Political scientist Thomas Faist also employs a notion of transnational social space as ”the combinations of ties and their contents, positions in networks and organizations, and networks of organizations that can be found in at least two geographically and internationally distinct places” (2000, 197).

44 cial relations of space are experienced differently, and variously interpreted, by those holding different positions as part of it (ibid., 3).

Massey thereby turns our attention to the simultaneity of spatial scales, to “the coexistence of social interrelations and interactions at all spatial scales, from the most local level to the most global” (ibid., 264). Space is thus not neutral and static, but socially constituted as “a moment in the intersection of configured social relations” (ibid., 265, my emphasis). This notion of social space brings to the fore an analysis of the positions of migrants engaged in transnational relations, simultaneously relating to (at least) two sets of political, economic and socio-cultural contexts – and thereby possibly shifting positions. This would seem to complicate Bourdieu’s notion of a configuration of objective relations between positions. As Levitt and Schiller write,

individuals [may] occupy different gender, racial, and class positions within different states at the same time. Recognising that migrant behaviour is the product of these simultaneous multiple statuses of race, class, and gender makes certain social processes more understandable (2004, 1015, my insertion).

The question of differential but simultaneous positions in transnational space thus complicates the questioning of positioning. Sociologist Avtar Brah has used the term the politics of location to refer to the question of shifting positions in social spaces. She defines this concept as “locationality in contradiction – that is, a positionality of dispersal; of simultaneous situatedness within gendered spaces of class, racism, ethnicity, sexuality, age” (Brah 1996, 204, emphasis in original; cf. Mahler and Pessar 2001, 445-446). In my view, however, ‘positionality of dispersal’ does not necessarily imply contradiction: positions might as well as reinforce as contradict each other. Whether the contradiction or reinforcement – possibly interchanging – is the case is up to empirical analysis. Still, Brah’s insistence on the simultaneous situatedness within different spaces is a useful one, which not only emphasizes simultaneity, but also shifting and dispersed positions. In contrast to Bourdieu’s notion of field, the positions of actors are seen as unstable and changing, as they may relate to more than one overall context at the same time. In many ways, the notions of social field and space are thus almost identical, in that they focus on social relations and positions. To avoid too much confusion, I reserve the term field in Bourdieu’s sense as a network of positions of a

45 game-like nature, characterized by the same norms and regularities. Following Levitt and Schiller, I reserve the term transnational social field for fully developed interlocking networks of social relationships spanning national borders regularly and intensely. However, not all social networks and positions constitute a part of a regular, game-like field and not all transnational relations develop into transnational social fields. Rather, these two concepts refer to specific social phenomena. I will use the term social space as a broader, more inclusive concept (cf. Mahler and Pessar 2001, 444). Social space refers to intersections of configured social relations, as Massey termed it, turning our analytical attention to simultaneity and multiplicity on the local and global scale. In contrast to Bourdieu’s notion of field, it contains both order and chaos (Massey 1994, 265). Both the concepts of field and social space can be transnational or national in nature, depending on the actual case. These concepts thus focus the research on relations, locations, spatiality, and simultaneity. They thereby help in overcoming methodological nationalism (or, as it were, methodological nation-state territorialism). Still, the overall transnational perspective can be criticized for over-emphasizing the homeland-host country axis at the expense of potentially multi-local transnational social fields. One of the strengths of the concept of diaspora, which I now turn to, is precisely the focus on multi-locality and dispersion.

Diaspora The second theoretical mapping revolves around the concept of diaspora, which is propagating in migration studies and in relation to political struggles. On the one hand, scholarly contributions on the nature of diaspora have extended the understandings of the terms and its theoretical implications. From being mostly employed in religious studies, referring to the Jewish experience, diaspora is now widely employed to refer to dispersed migrant and ethnic populations or, especially in cultural studies, to a conceptualization of a cosmopolitan and un- autochthonous consciousness. On the other hand, the concept of diaspora has also entered the realm of politics. ‘Diasporas’ are seen as central in relation to a range of issues, from struggles for political recognition of nation-states over identity politics to transnational mobilization of development and reconstruction projects. There is a growing body of not only academic studies, but also of policy reports about ‘the diaspora’ as well as organizations and groups, who

46 identify themselves as ‘diasporas’. Often the political and policy usages of the concepts of diaspora draw on notions of collectivity, shared ethnicity, and stable identity that the theoretical understandings of diaspora tried to challenge in the first place. To avoid confusion between these different usages, let me start by clarifying how I employ the concept of diaspora. I refer to ‘the diaspora’ or ‘diasporas’ as a noun, to indicate that the concept is employed as a designator of a group of persons or a people. I use the inverted commas as a reminder of this aspect. I refer to diaspora (in the indefinite singular) as a theoretical nodal point to indicate theories and studies of the concept of diaspora.

Between transnational communities and diasporic identification The term diaspora is intimately connected to the Jewish expulsion from Jerusa- lem and has a strong religious dimension of loss, suffering and, in due course, return and final gathering, constituting “a Jewish theology of exile” (Baumann 2001, 2). Since the 1960s, however, the concept of diaspora has gained a wider usage, referring to other histories of forced expulsions, such as the Greeks and Armenians (Anthias 1998), and the ‘Black Atlantic’, the slaves and their descendants (Gilroy 1993). In the 1990s, diaspora studies started to proliferate in the social sciences and in cultural studies, when a range of enthusiastic articles, books and journals appeared on the topic (e.g. van Hear 1998; Cohen 1997; Brah 1996; Clifford 1994; Gilroy 1994; Safran 1991). Though these works do not necessarily agree on the implications of the term, they share an understanding of diaspora as dispersion from an existing or imaginary centre or homeland. As I outline below, some studies focus on ‘diasporas’ as dispersed populations, while others employ the concept to conceptualize conditions for identification (cf. Anthias 1998). William Safran, one of the pioneering proponents of the (in the early 1990s novel) usage of diaspora, defines the concept as expatriate minority communities that share some of following characteristics: 1) dispersal from an original ‘centre’ or homeland, 2) a collective memory or myth about the homeland, 3) the experience of partial alienation in the host society, 4) an idea of eventual return to the ‘homeland’, 5) collective commitment to the maintenance of the homeland, and 6) continued relations with the homeland and solidarity as well as an ‘ethnocommunal consciousness’ between the scattered populations (1991, 83-84). Sociologist Robin Cohen (1997) extends Safran’s ideal type with no-

47 tions of imperial, trade, and cultural ‘diasporas’ as well, thereby employing a very broad and inclusive usage of the term. In the words of Tölölyan (1991), “diasporas are the exemplary communities of the transnational moment” (quoted in Clifford 1994, 303); an understanding which fits well with the original Greek meaning of the word as a scattering of seeds or to ‘sow over’ (Bauman 2001; Cohen 1997)22. Broadly speaking, Safran and Cohen (and many others) thereby conceptualize diaspora as certain kinds of populations, as ‘diasporas’ (see also van Hear 1998). Other studies, however, focus on diaspora as a condition for identification (Brah 1996; Clifford 1994; Gilroy 1994; 1993). These studies also bring to the fore the centrality of collective (stories of) displacement and trauma (especially Gilroy 1994; 1993; see also Brah 1996), but are more concerned with how experiences of displacement relate to identity formation and the constructions of “home away from home” (Clifford 1994, 302). According to anthropologist James Clifford, diaspora discourses articulate a specific cosmopolitanism, which is in “a constitutive tension with nation-state/assimilationist ideologies” (ibid., 308). ‘Diasporas’ are, in this sense, the counter-evidence of assimilation theory. Furthermore, this approach to diaspora highlights processes of identification, mobilization, and belonging.

A Somali ‘diaspora’? The dispersion of Somalis around the world and the extended transnational relations and remittance economy has indeed led scholars, policy makers and So- malis themselves to speak about a ‘Somali diaspora’ since the late 1990s. In Voices of the Somali Diaspora, Somali writer and intellectual Nuruddin Farah (2000) weaves together his own migration story with interviews with Somalis in Tanzania, Kenya, Italy, the UK, Switzerland, Sweden and Canada to understand and describe the experiences and thoughts of flight and exile. Likewise, political scientist Marc-Antoine Pérouse de Montclos (2003) uses the concept

22 As Martin Baumann notes, however, the Greek usage of the term as ‘a scattering of seeds’ had disastrous connotations, associated with processes of dissolution and decomposition (2001, 8). The Greek usage is thus quite different from the sometimes rather celebrative contemporary versions of diaspora. This should not lead us to sombre scenarios of the dissolution and moral decay of dispersed populations (cf. Malkki 1992), but it does serve as a sobering reminder not to overly celebrate diaspora.

48 of diaspora to refer to the dispersed Somali population. Furthermore, a number of works have appeared that focus on Somali populations in just one country or city. They include a study of ‘homemaking processes’ among Somalis and Su- danese in Norway by Sudanese-Norwegian anthropologist Mounzoul Assal (2004); Dutch development researcher Cindy Horst’s analysis of survival strategies among Somali refugees in the Dadaab refugee camps in Kenya (2003) and in Minnesota (2004b); British refugee researcher David Griffiths‘s comparison of Somalis and Kurds in London (2002); and Canadian scholar Rima Berns McGown’s (1999) study of Somali religious practices in Toronto and London. Generally speaking, these works focus on ‘diasporas’ as expatriate populations, thereby following the approach of Safran and Cohen, though some of them also relate to identity issues regarding diasporic identification and mobilization. Furthermore, the concept of diaspora has travelled into circles of policy makers dealing with Somali affairs. A recent example is the USAID funded pilot programme, which examined the potential of strengthening the health sector in Somalia through co-operation between ‘the diaspora’, businessmen, and existing health facilities (Kent, Hippel, and Bradbury 2004). In a similar vein, the Nordic Governments’ Joint Consultation Group for Refugees explored the transfer of competences and human resources of the ‘Somali diaspora’ at an international seminar in 2002 (NSHF 2002) and the UNDP has launched the QUEST initiative to encourage ‘diaspora professionals’ to return in 2004 (UNDP 2004). Similarly, Somali-Dutch political scientist Awil A. Mohamoud (2005) has analyzed the development potential of so-called ‘diaspora professionals’. Also in these understandings, ‘the diaspora’ refers to the Somali population abroad, particularly Somalis in the West and the Gulf Countries, who are educated and who regularly remit money to family members and reconstruction projects. The policy-related approaches thus employ diaspora to point to a population with potential resources, indicating expectations both of economic support and of continued loyalty towards Somalia. The main interest here is transnational practices and obligations.

Diaspora as claim and aspiration These and other diaspora studies offer important methodological and theoretical lessons. First of all, they show how migrants may maintain transnational prac-

49 tices and fields of belonging that include a multitude of localities. This helps overcome the bias of methodological nationalism as well as the homeland-host country bias in some transnational studies. Secondly, diaspora studies stress the necessity of studying across categories, to include refugees, labour migrants, family-reunified, business people etc. (e.g. Crisp 1999b). Finally, some diaspora studies challenge problematic notions of essentialism and autochthony, pointing to alternative ways of conceptualizing home and belonging. These are all important insights, challenging us to focus our research across national and categorical borders. Still, there are also a number of pitfalls – not least in relation to the tendency sometimes to exaggerate the focus on origin and downplay internal differentiation. Anthropologist Brian Keith Axel (2004) and sociologist Floya Anthias (1998) have criticized the tendency to understand ‘diasporas’ as determined by the dispersal from ‘an original homeland’ as a privileged point of origin. Such a presumption situates ‘diasporas’ within a framework of origin and almost primordial ethnicity which, in the words of Anthias, nurtures a “tendency to reinforce absolutist notions of ‘origin’ and ‘true belonging” (1998, 577). According to Axel, one of the consequences of this presumption is that the “place of origin is revealed, perhaps by default as the context of diaspora” (2004, 28, my emphasis); diaspora becomes embedded in a framework of origin and ethnicity. Both Anthias and Axel emphasize that such an understanding offers a simpli- fied picture of who and what is seen as constitutive of ‘a diaspora’. As feminist researchers, among others, have noted, ‘diasporas’ are often presented as homogeneous, disregarding internal differences, and without specifying the particularities of claims made in the name of diaspora. Rather, diasporic identification and mobilization should be viewed as ongoing processes of collective identification, indicating the importance of analyzing the struggles for fixing collective identity (Brah 1996). Again, it is useful to remember Brah’s notion of the politics of location, as mentioned in the first theoretical mapping, in order to complicate and nuance ideas of ‘diasporas’. Axel, however, takes a step further, claiming that the ‘origin’ or homeland is not what constitutes a diaspora, but is on the contrary constituted by the people claiming it. He thus turns our attention to processes of diasporic identification and mobilization, suggesting that diaspora “may be understood more produc- tively as a globally mobile [potential] category of identification” (2004, 27, my

50 insertion), thereby shifting the focus to questions of who claims to be the subject of diasporas and the everyday conditions for such an identification. Along similar lines, anthropologist Thomas Blom Hansen criticizes the “heroic connotations” which have become associated with diaspora as the term has “become a keyword in political strategies of governments, businessmen and cultural entrepreneurs who encourage long-distance patriotism and cultural purification of expatriate populations” (2005, 165-166). Rather, Hansen suggests that the concept should be used to indicate

an aspiration, a fleeting, contested – at time times important – form of imagination that may (or may not) succeed in providing an effective framework of interpretation of identity […] to describe such sentiments and identities that establish meaning through imaginary and practical links with a (lost) homeland, or point of origin (2003, 34; cf. Turner 2007).

Axel and Hansen thus displace the central ideas of dispersion and origin with a focus on processes of identification and mobilization. I find this an important and inspiring approach, emphasizing the need to distinguish between different usages of the concept. I thus supplement my initial distinction between diaspora and ‘diasporas’ with the adjective diasporic to describe the aspirations and characteristics of persons or groups claiming or claimed to be ‘a diaspora’. In this analytical framework, diaspora is thus transformed from a category that tells something about a population to one that tells something about the mobilization processes, discourses, and identifications claimed by – or attributed to – groups who identify or are identified as ‘a diaspora’ or ‘in diaspora’. Thus, rather than dismissing the concept of diaspora, this analytical turn makes it interesting from another angle, as a concept of a political nature that might at once be claimed by and attributed to different subjects and groups. These are the kinds of processes that are especially relevant and important to this dissertation. As already mentioned, the concept of diaspora is already part of the vocabulary describing Somali migration and transnational practices. I will therefore explore such claims to see if the Somali-Danish informants articulate a diasporic consciousness: How do they articulate notions of homelands, belonging, and solidarity? What and where is it? Where are their commitments and loyalties directed towards? What and who is mobilized and identified in the name of ‘the diaspora’? The purpose of these questions is thus not to determine if the dispersed Somalis in the West and more specifically in

51 Denmark form ‘a proper diaspora’, but to analyze how they employ the concept, and whether and how diasporic mobilization takes place. Theoretically, I thus follow the analytical track outlined by Brah, Axel, and Hansen, focusing on diaspora as claim and processes of mobilization and identification. Or, in other words, why do (some) Somalis categorize themselves as (part of) ‘a diaspora’, and what kind of positions does diaspora as an identity marker make possible? These questions lead us to identity politics and struggles for recognition, and thus to another of the central aspects in my problem formulation: how So- malis in Denmark struggle for recognition and negotiate positions in Denmark and transnationally.

Recognition The third theoretical mapping focuses on the concept of recognition. In the Ox- ford Advanced Learners’ Dictionary, recognition is defined in the following way:

1 the act of remembering who sb is when you see them, or of identifying what sth is […] 2 the act of accepting that sth exists, is true or is official […] 3 (for sth) public praise or reward for sb’s work of action (2000, 976).

Recognition is thus an ambiguous term with several meanings. This ambiguity gives it an especially interesting analytical potential, which I will elaborate below. I focus first on the third dimension in the dictionary definition of recognition for something, and then the first dimension of identification of somebody or something23. Together these two dimensions make up my concept of recognition.

Between intact identity formation and participation Sociological and moral philosophical recognition theory was (re)introduced in the early 1990s by, among others, philosopher Charles Taylor (Taylor and Gutman 1994), sociologist Axel Honneth (2006 [orig. 1992]; 2004), and political scientist Nancy Fraser (2000; 1995). The two latter thinkers have recently engaged in a critical, transatlantic dialogue about their diverging views (Fraser and Honneth 2003) and I will mainly focus on them here, as they constitute two

23 I thus leave out the second dimension of accepting something as true or official.

52 extremes in the recognition debate. According to Nancy Fraser, the ‘recognition turn’ signals the transition from an overall focus on redistribution and economic inequality to recognition of cultures and identity. This ‘turn’ makes claims made in the name of culture and identity central in contemporary social and political struggles, as well as in social science and moral philosophical research. The proliferation of diaspora studies and claims can be seen as a part of this development. However, while Honneth and Fraser agree that recognition has become a key concept in contemporary social struggles, they sharply dis- pute the theoretical and political implications of this. Let us turn first to Hon- neth. German sociologist Axel Honneth takes as his departure Hegel’s early Jena writings. Combining Hegel with social psychologist George Herbert Mead and psychoanalyst Donald W. Winnicott, he develops a theory of inter-subjective and dialogical identity formation through recognition by peers (2006). Honneth thereby claims that individual self-realization always take place through social interaction, which he relates to the social integration of society. His theory thus has both sociological and psychological dimensions and roots. Furthermore, Honneth analyzes the genesis of modern identity, arguing that “the establishment of the liberal capitalist societal order is to be described as a process of differentiation by three spheres of recognition” (2004, 351)24. These three spheres of recognition include love, the principle of need; law, the principle of equality; and finally social esteem, the principle of merit in one’s achievement. Accord- ing to Honneth, individuals need to be recognized in all three spheres to enable full self-realization and human flourishing. He sums up his theory in the following way: My basic concern was to defend the thesis that the normative expectations subjects bring to society are oriented toward the social recognition of their capability by various generalized others. The implications of this

24 Taylor also analyzes the “modern preoccupation with identity and recognition” (1994, 26). Both he and Honneth thus offer an analysis of the historical development of modern subjectivity. They both analyze the collapse of social hierarchies as the basis of honour, and the emergence of individualized identity formation, as based on ideas of authenticity, universal rights, and merit-based achievement. Hon- neth, however, criticizes Taylor for providing ”a highly misleading chronology”, which claims that the struggles for legal equality in liberal-capitalist societies have been replaced by social struggles for recognition in an identity-political way (2003, 122).

53 moral-sociological finding can be further developed in two directions: the first concerning the subject’s moral socialization, the second concerning the moral integration of society” (2003, 173).

Honneth thus relates these two levels, claiming that the goal of modern societies is to ensure recognition and thereby full self-realization of its members as “the real aim of equal treatment in our societies” (ibid., 177). Or, in other words, recognition is the means to enable intact identity formation on the individual level as well as social inclusion and equal opportunities on the societal level in all three spheres of recognition and society. Furthermore, Honneth relates this normative goal to social and political struggles. Based on the three recognition spheres, Honneth distinguishes three corresponding types of misrecognition (2006, 175ff): physical assaults on the bodily integrity such as torture or rape; deprivation of rights, meaning that groups or individuals are denied the opportunity of becoming equal citizens or peers and, thirdly, degrada- tion and devaluation of individual or collective ways of life, such as ideas of ‘culture’ or religion. Following Honneth, contemporary social struggles are based on experiences of misrecognition. He writes:

What motivates individuals or social groups to call the prevailing social order into question and to engage in practical resistance is the moral conviction that, with respect to their own situations or particularities, the recognition principles considered legitimate are incorrectly or inade- quately applied (2003, 157, emphasis in original).

For Honneth, then, misrecognition distorts self-realization and identity formation, with the implication that such disrespect, harm, and humiliation must be combated at all levels. Misrecognition is also the driving forces behind present- day social and political struggles that he categorizes as struggles for recognition and, in consequence, as related to questions of identity formation and self- realization. Analyzing struggles for recognition within this perspective thus raises questions regarding individual and collective experiences of respect or harm, especially in relation to identity formation and social esteem. Like Honneth, Nancy Fraser25 also focuses on social struggles in relation to the question of recognition. Taking departure in the political situation of “post- Fordism, postcommunism, and globalization” (2003, 90), Fraser seeks to under-

25 As Catherine Holst (2005) points out, Fraser has moved from exercising imma- nent to moral philosophical critique in her later work (Fraser and Honneth 2003), where she focuses on justice and participation as a moral claim.

54 stand the proliferation of struggles and claims for recognition in contemporary social movements. Fraser’s starting point, then, is to analyze claims for recognition, strongly criticizing Honneth (and Taylor’s) ‘identity model’ and the tendency to “transpose the Hegelian recognition schema onto the cultural and political terrain” (ibid., 109). Rather, Fraser wants to deconstruct and transform power relations and institutions that further social status26 subordination and impede participation. She thus situates her analysis at the political and collective level, not the psychological. One of Fraser’s central points is that struggles of recognition tend to over- look and reject economic inequality which, in her view, needs to be dealt with in terms of redistribution (2003; 2000; 1995). Rather than Honneth’s three dimensions of recognition as encompassing all kinds of contemporary social struggle, including redistribution, Fraser argues for a model of perspective dualism, where recognition and redistribution constitute two separate, but equally necessary, analytical perspectives on social struggles. According to her, many social struggles have both dimensions and thus cannot be reduced to primarily cultural or economic struggles. Still, as Honneth rightly notes (2003), the economy-culture dualism is insufficient. What about rights, what about politics? While Fraser does introduce a third dimension of politics and representation in Redistribution or Recognition?, thereby addressing decision-making procedures and political exclusion as well (ibid., 68), this attempt is never fully developed. Rather than focusing on identity formation, Fraser proposes a ‘status model’, where recognition is rethought as a question of social “status of individual and group members as full partners in social interaction” (2000, 113). Misrecogni- tion, then, becomes a question of social subordination, not the depreciation of group identity. It is the “institutionalised patterns of cultural value that impede parity of participation” which must be addressed in Fraser’s theory of justice (ibid., 115). She writes:

If and when such patterns constitute actors as peers, capable of participating on a par with one another in social life, then we can speak of reciprocal recognition and status equality. When, in contrast, institutionalized patterns of cultural value constitute some actors as inferior, excluded, wholly other, or simply invisible, hence as less than full partners

26 Fraser uses the notion of status to refer to “a cultural order of recognition” (Dahl, Stoltz, and Willig 2004, 377) in relation to prestige and social esteem.

55 in social interaction, then we should speak of misrecognition and social subordination (2003, 29, emphasis in original).

Fraser thereby suggests a ‘non-identitarian politics of recognition’, which “avoids the view that everyone has an equal right to social esteem” (ibid., 32). In contrast to Honneth’s notion of social inclusion “as a process of inclusion through stable forms of recognition” (ibid., 173), Fraser’s objective is parity of participation (ibid., 36). Pushed to the extremes, we might say that while Hon- neth’s approach to recognition is affirmative, Fraser’s is transformative. Though they are both dealing with social struggles and justice, Honneth con- centrates on identity formation and self-realization, while Fraser focuses on status and societal participation. Her work thereby turns one’s attention to how institutionalized patterns of social subordination impede participation. The recognition perspective has by now been employed in a number of studies. Fraser’s approach has inspired several studies on care work and equality (L. Hansen 2005; Dahl 2004; Macdonald and Merril 2002), while Honneth has inspired, for instance, analyses of work (Willig and Petersen 2004) and refugee and immigrant rights (Gosh 2005). Still, to my knowledge, recognition theory has not been applied to a transnational or diaspora perspective. As earlier indicated, such a theoretical combination is one of the objectives of my dissertation. Below I position myself in relation to recognition theory and to how it inspires my analysis.

Considerations and questions To me, one of the most important points in Honneth’s (and Taylor’s) recognition theory is the insight that the self is dialogically constituted through interaction with other people. That identity is not determined by an inner core, but socially constituted, is a central and widely shared insight in sociology and social psychology. Honneth’s recognition theory is thus both a correction to notions of authentic identity formation as well as to ideas on subjectivity as exclusively discursively constituted. Furthermore, I find it sensible to regard recognition as a crucial driving force for social struggles and to strive for equal rights and the avoidance of harm. Indeed, the question about struggles for recognition is central in my problem formulation. In this sense, I share Honneth’s overall prem- ise.

56 However, like Fraser, I find Honneth’s insistence on recognition and self- realization as the fuel for contemporary social struggles and the goal and measure for justice to be problematic. Still, Fraser’s two-dimensional approach is reductive, in spite of her attempts to include a political dimension. Fraser’s force is, in contrast to Honneth, her insistence on the complicatedness of social identities and the inherent conflicts between different claims for recognition. Be- cause, as she rhetorically asks, what should we do “in cases where esteeming the labour contributions of some entails denying equal citizenship to others” (2003, 228)? To regard identities as unified and harmonious is a mistake. In the words of philosopher Seyla Benhabib27, the idea of universal inclusion and seamless interaction is a truly idealistic one (2002, 57). Rather, Fraser and Ben- habib focus on the political aspects of struggles for recognition – not least concerning the relationship between the individual and collective levels. This question is especially urgent in relation to identity political claims made in the name of gender, nationality, ethnicity – or diaspora. Writes Benhabib:

It is both theoretically wrong and politically dangerous to conflate the individual’s search for the expression of his/her unique identity with politics of identity / difference. The theoretical mistake comes from the homology drawn between individual and collective claims, a homology facilitated by the ambiguities of the term recognition. Politically such a move is dangerous because it subordinates moral autonomy to movements of collective identity (2002, 53, emphasis in original).

Benhabib thereby redirects attention to what Fraser has termed ‘the problem of reification’; to ‘freeze’ or affirm ideas of collective, cultural identity models, rather than deconstruct and transform them (Fraser and Honneth 2003, 91; cf. Hall 2000; Appiah 1994). In many ways, this critique runs parallel to Anthias’ critique of diaspora studies to “reinforce absolutist notions of ‘origin’ and ‘true’ belonging” (1998, 577) and Brah’s (1996) emphasis on analyzing the fixing of collective identities in claims and narratives of diaspora. Benhabib further calls for less focus on what groups are and more on “what the political leaders of such groups demand in the public sphere” (2002, 16, emphasis in original). She thereby turns the focus to the “processes through which existing social and cultural cleavages are transformed into political mobilization” (ibid., 17), concentrating on the establishment of boundaries between ‘us and them’ (cf. Mouffe

27 Benhabib refers to Taylor. Still, in my view, Honneth shares the same problem.

57 1997). We therefore need to turn to the establishment and negotiations of social positions: to the question of differentiation and cultural intelligibility.

Differentiation and cultural intelligibility The second aspect of recognition relates to the act of remembering who somebody is or identifying what something is, as defined in the Oxford Advanced Learner’s Dictionary (2000, 976). Here I take a first step in a larger theoretical discussion with the aim of a developing an analytical approach with which I can analyze both struggles for recognition and positioning. Bronwyn Davies and Rom Harré define positioning as “the way in which the discursive practices constitute the speakers and hearers in certain ways and yet at the same time is a resource through which speakers and hearers can negotiate new positions” (1990, 62). They thus emphasize the double aspect of positioning as something that actors do and that is ‘done’ to them, thereby constituting them in certain ways and with certain qualities. Positioning is thus a relational phenomenon, relating to categories such as gender, racialization, social class, age etc. In my view, however, positioning is a broader phenomenon, which also relates to other aspects of identity work than the ‘usual’ categories, I have just mentioned. Claims and identification made in the name of diaspora, for instance, are also examples of positioning in social spaces, of identity work. In the words of cultural analyst Stuart Hall:

Cultural identities are the points of identification, the unstable points of identification or suture, which are made, within the discourses of history and culture. Not an essence but a positioning. Hence, there is always a politics of identity, a politics of position, which has no absolute guaran- tee in an unproblematic, transcendental ‘law of origin’ (1990, 226, emphasis in original).

Negotiating positions is thus negotiating identification and doing identity work, as “a ‘production’, which is never complete, always in process […] always ‘in context’, positioned” (ibid., 222, emphasis in original). It is about constituting oneself and others as recognizable – or unrecognizable – in certain social spaces and discourses. Cultural and feminist analyst Sara Ahmed has reflected upon the identification aspect in terms of recognizability. Drawing on Althusser’s notion of interpellation or hailing – ‘hey you there!’ – she suggests that, “subjects become differentiated at the very same moment they are constituted as such” (2000, 23). She continues:

58 the subject is not primarily constituted by being recognised by the other, which is the primary post-Hegelian model of recognition (see Taylor 1994). Rather, I am suggesting that it is the recognition of others that is central to the constitution of the subject […] The subject is not, then, simply differentiated from the (its) other, but comes into being by learning how to differentiate between others (ibid., 24, my emphasis, round brackets in original).

Ahmed thus turns our attention to the act of recognizing and being recognized as somebody or something – be it stranger (which is Ahmed’s example), Dan- ish, Somali, man, woman. It is not my aim here to go into a discussion over whether recognition by the other or of others is more important to subject constitution. To me, both aspects are central to my analytical purpose of exploring positioning and differentiation. What I find interesting in Ahmed’s suggestion is firstly that recognition does not necessarily imply a positive status, and secondly, that positioning is not only a discursive or conversational phenomenon, as Davies and Harré (1990) proposed, but also has performative and embodied aspects. Rather, Ahmed emphasizes that recognition and positioning operates through “a visual economy: it involves many ways of seeing the difference” (2000, 24, emphasis in original). We thus need to extend the notion of positioning with performative and embodied aspect as well. Here I turn to philosopher Judith Butler and some of her reflections about identity work in relation to how subjects become constituted and recognized as somebody (2004; 1999). Butler focuses on gender, but I will take the logic a bit further. In her very influential book Gender Trouble (1999), Butler reckons with the idea of gender as an essence, troubling and unpacking the relationship between gender, sexuality and sex. A central point is that gender is performative and ‘done’, thereby “constituting the identity it is purported to be” (ibid., 33). In this sense gender and identity are ‘productions’, as Hall also stated; there is no ‘real thing’. Furthermore, as Butler writes in Undoing Gender, “one does not “do” one’s gender alone. One is always “doing” with or for another” (2004, 1). In both books, Butler is thus preoccupied with what it means “to do and undo norms of sexual and gendered life” (ibid., 1). Without going into – or accepting – Butler’s entire theory, I am inspired by her focus on performativity, borrow- ing her particular idea of cultural intelligibility, which I will employ as a broader analytical concept in my study. Let us therefore take a closer look at this notion. In Gender Trouble, Butler claims that ““persons” only become in-

59 telligible through becoming gendered in conformity with recognizable standards of gender intelligibility” (1999, 22, my emphasis). She goes on to ask:

To what extent do regulatory practices of gender formation and division constitute identity, the internal coherence of the subject, indeed, the self- identical status of the person? To what extent is “identity” a normative ideal, rather than a descriptive feature of experience? And how do the regulatory practices that govern gender also govern culturally intelligible notions of identity? In other words, the “coherence” and “continuity” of “the person” are not logical or analytical features of personhood, but, rather, socially instituted and maintained norms of intelligibility (1999, 23, emphasis in original).

For Butler, then, identity is an effect of norms of intelligibility, turning the attention to ‘matrixes’ and ‘domains’ of intelligibility. While Butler focuses specifically on gender, I think that her notion of intelligibility can be appropriated towards other aspects as well, in the sense of being intelligible or recognizable as something or somebody. Doing gender – or more broadly doing identity – is a social activity. In this sense, Butler agrees with Honneth and Taylor’s focus on the dialogical identity formation. Butler, however, complicates the post- Hegelian approach to recognition. She writes:

The terms by which we are recognized as human are socially articulated and changeable. And sometimes the very terms that confer “humanness” on some individuals are those that deprive certain other individuals of the possibility of achieving that status […] The human is understood differentially depending on its race, the legibility of that race, its morphology, the recognition of that morphology, its sex, the perceptual verifica- tion of that sex, its ethnicity, the categorical understanding of that ethnicity (2004, 2, my emphasis).

Certainly, both Honneth and Taylor would agree that the terms of recognition are historically produced. Butler, however, introduces a power perspective in relation to the differentiation of recognition, reminding us that cultural intelligibility is not always desirable. “I may feel that without some recognizability I cannot live”, Butler writes and continues, “But I may also feel that the terms by which I am recognized make life unlivable” (ibid., 4). Cultural intelligibility is not necessarily a liberating or ‘positive’, as Ahmed also emphasized in her reflections about recognizing strangers. This emphasizes the importance of the social space in which recognition and positioning take place. As Danish psychologist Dorte Staunæs points out, intelligibility takes place in a context, not in a stable, unified, unequivocal culture. She has therefore suggested the term

60 contextual intelligibility (2004, 66). I agree with this correction. Still, the main point is that “recognition becomes a site of power by which the human is differentially produced” (Butler 2004, 2). According to Butler, intelligibility is a product of recognition. However, we might also say that recognition (in the sense of acknowledgement and public praise) presupposes intelligibility, or rather, perhaps, that recognition and intelligibility are connected in a dialectical relationship. This aspect of recognition as intelligibility is thus related to the establishment and negotiation of positions, emphasizing the enactment of social categories and identity work. Butler’s thoughts – as well as ethnomethodological studies on the performative nature of social categories (West and Zimmerman 1987; West and Fen- stermaker 1995) – have inspired a range of studies on social positions and identity work. These studies focus on the ‘doing’ of social categories – both in the sense of enactment and articulation – in ways that might be recognized as masculine, feminine, youthful, mature, Danish, foreign etc. (e.g. Moser 2004; Staunæs 2004; Søndergaard 1996). Likewise, studies of intersectionality28 focus on how social categories interact and intersect with each other, sometimes reinforcing, sometimes overshadowing each other. The work on intelligibility, performativity, and intersectionality then offers another perspective on the question of recognition and social positions that inspire my work. To emphasize the aspect of performance, activity and ‘doing’, I will use the term modes of differentiation to point to social categories such as gender, ‘race’, ethnicity, age etc. Modes of differentiation are, in other words, part and parcel of processes of recognition and intelligibility – enacted, negotiated and made meaningful through social processes and interactions. It is therefore of special interest to me to discover how modes of differentiation are articulated, practised, and mobilized: how they are negotiated and recognized in different social spaces.

28 The term intersectionality has mainly been developed within feminist and anti- racist studies, examining the structural subordination of certain groups (Collins 1998; Crenshaw 1994). Such studies are now supplemented with others at the micro level and the in situ (re)constructions of categories (Staunæs 2005; 2003; Sønder- gaard 2005; Lykke 2003), while other studies maintain the focus on more structural dimensions of inequality (de los Reyes and Mulinari 2005). I here use the term to refer to the broad insight that social categories are always acting in concert – to paraphrase Butler – that gender, ‘race’, ethnicity, social class etc. are interrelated in different ways.

61 To summarize, my usage of recognition has a double dimension, referring on the one hand to self-realization and participation in relation to social struggles and, on the other, to intelligibility and positioning. While the former relates to the ‘recognition turn’ in Honneth and Fraser’s versions, the latter is inspired by Butler and others involved in feminist work on intelligibility, differentiation, and positioning. In contrast to Honneth and Fraser, however, I do not want to establish now which understanding of recognition is the most adequate. This must be established through an analysis based on empirical data. Though I agree with Fraser and Benhabib that societal participation, rather than intact identity formation, is the more promising moral basis for justice, I will employ a more open approach to analyze Somali-Danish struggles for recognition and negotiation of positions. Rather, my discussion of recognition generates questions about the data and the analysis, such as: How do Somali-Danes articulate their experiences and positions in Danish society and transnationally? What positions do they strive to establish and negotiate? What drives their associational engagements? Are they explicitly fuelled by issues of recognition and misrecognition and, if so, in relation to which dimensions? And, inspired by the transnational perspective, in which social spaces are their struggles for recognition and negotiation of positions embedded in and directed towards? I will explore these questions throughout the dissertation. But first, I will shortly summarize where the theoretical journey has brought me.

Chapter resume: Linking the three mappings In this chapter, I have presented the three theoretical perspectives which inform the dissertation: simultaneous engagement, diaspora, and recognition. I have thereby attempted to situate my meta-theoretical location, following Haraway’s ideal of ‘situated knowledge’. The overall purpose was thus to provide an overview – a mapping – to develop my analytical concepts and combine transnational and diaspora studies with recognition theories. My analytical approach is constituted through this combination. The first mapping of simultaneous engagement took departure in the transnational turn, focusing on the relationship between processes of inclusion and associational engagement. The analytical concepts in this mapping relate to spatiality and simultaneity, rather than methodological nationalism (or methodological nation-state territorialism) and assimilation scenarios. I presented Bourdieu’s concept of field as a configuration

62 of regular networks and positions, Levitt and Schiller’s notion of transnational social field as fully developed interlocking networks of social relationships spanning national borders, and finally, inspired by Massey, social space as a broad analytical concept, referring to intersections of configured social relations on the local and global scale. The second mapping concentrated on diaspora studies, which also bring to the fore a focus on spatiality, simultaneity, and dispersion. The diaspora perspective is especially interesting due to its importance both in studies of Somali migration and in how Somalis position themselves. Inspired by Axel and Han- sen, however, I suggested conceptualizing diaspora as claim and a framework of interpretation, rather than dispersed migrant populations. I distinguish between three usages of the concept: diaspora theory, ‘diasporas’ as claims of dispersed populations (constituting an object to study, rather than to take departure from), and the adjective diasporic to refer to mobilization and identification processes, referring to a real or imagined homeland. Furthermore, I emphasized – like in the first mapping – the importance of paying attention to attempts of fixing collective identities, recalling Brah’s concept of the politics of location, addressing the simultaneous, shifting, and perhaps contradictory, positions in social space. The third mapping revolved around theories of recognition. I first presented the diverging understandings of recognition by Honneth and Fraser as self- realization and parity of participation before I turned to feminist reflections on performativity and differentiation, inspired by, among others, Ahmed and But- ler. Bringing together these thinkers, I developed a two-dimensional analytical concept of recognition. One dimension relates to recognition as self-realization and participation, focusing on social struggles for recognition and identity politics. The other concerns recognition as intelligibility in terms of processes of differentiation and positioning that make persons recognizable as somebody or something. To emphasize this aspect, I further termed social categories such as gender, ethnicity etc. as modes of differentiation. Brought together, the analytical concepts developed here point to my overall interest in processes of positions and recognition in social spaces. They emphasize the necessity of focusing on both a spatial and temporal scale in order to analyze (possible) transnational, simultaneous involvement. In the next chapter, I turn to the methodological dimensions of this position: how I ‘translate’ the

63 theoretical perspectives and analytical concepts in the production of data and the analytical strategies.

64 Chapter 3 Directions. Methodological considerations and analytical strategies

The way in which we know is most assuredly tied up with both what we know and our relationships with our research participants (Guba and Lincoln 2005, 209, emphasis in original).

In this chapter I present the analytical principles which guide the data production and analysis to reflect upon the way I know, and thereby the way that I answer the research questions. I started this exploration in the previous chapter, where I discussed the theoretical perspectives. The questions now focus on how to produce and analyze data which I can use to explore my theoretical interests and empirical curiosity.

Analytical principles As the reflections about simultaneity, transnationality, social space and the politics of location indicate, the question of the research and fieldwork sites is not a simple one. Most sociological and anthropological fieldwork studies have moved away from ideas of holistic, ‘integrative’ ethnography (Baszanger and Dodier 2005), based on studies of ‘complete’ cultural communities. Rather, notions of fieldwork have been supplemented with ‘homework’ (Koptiuch 1997) – studying ‘home’ society, rather than faraway places. Likewise, the reflections by Bourdieu and others have challenged bounded notions of a field. Further- more, in a transnational and diaspora perspective, the question of fieldwork is a particularly thorny issue: in which of the many possible localities within transnational or diasporic networks should fieldwork be carried out?

Location-work: Defining and exploring ‘the field’ Ideally, studies of transnational practices require multi-sited fieldwork, whether the aim of the research project is to follow people or things (Marcus 1995). No matter which definition of transnationalism is used, the perspective always re-

65 lates to practices or relations between at least two national localities: this is the very core of the concept. Therefore, the ideal method and analytical strategy should include fieldwork at both or more ’ends’ of the transnational social space. However, multi-sited fieldwork is not the answer for all methodological problems. First of all, what and where is the site or the field? Taken together, the associational involvements of the informants include a myriad of localities, and the question is which localities to include and which to leave out. Secondly, there is the question of contexts and localities. As emphasized in chapter two, transnational practices are not free-floating or deterritorialized, even if they are mobile. Rather, transnational actors are situated in different settings, even if those settings are temporary. Thus the location of transnational practices not only relates to the structural conditions, but also to the localities where transnational actors actually live, where they are involved, or between which they move. Needless to say, fieldwork will always be partial; there is no such thing as capturing it all. Anthropologists Akhil Gupta and James Ferguson have suggested the use of the term ‘location-work’ in a discussion of the concepts of ‘the field’ and ‘fieldwork’ in anthropological research. They propose working “with an atten- tiveness to social, cultural, and political location and a willingness to work self- consciously at shifting or realigning our own location while building epistemological and political links with other locations” (1997b, 5, emphasis in original). Location is thus not to be conflated with a methodological return to studying discrete local and bounded communities. Rather, one might say that location- work is the analysis of how different actors (including the researcher) are located and positioned within the field – including the contradictions and ambivalence of changing positions in different contexts. Gupta and Ferguson suggest a focus on “shifting locations rather than bounded fields” (ibid., 38, emphasis in original). This point precisely reflects the multi-dimensionality of location within different contexts. Location-work is thus a particular aspect of fieldwork and the analytical process, focusing the analytical attention on how social relations and modes of differentiation cluster together and how they are recognized (in the double sense of acknowledgement and cultural intelligibility) and/or mobilized. This analytical principle thus relates directly to the main research question of recognition and positioning, as well as to Donna Haraway’s call for

66 “politics and epistemologies of location, positioning, and situating” (1991, 195). Like Gupta and Ferguson, Haraway asks for critical, mobile positioning: to change the focus of the researcher, and to research positions in reflexive and self-critical ways, both in relation to the production and to the analysis of data. This aspect of location-work is thus methodological in nature, relating to how research is produced, to the ways we know, as Guba and Lincoln pointed out. Similarly, location-work relates to analyzing how the informants position themselves and each other. It thereby links to the theoretical perspectives presented in chapter two, as positions and locations are central aspects in all three theoretical mappings. Location-work thus also implies the act of analyzing the politics of location (Brah 1996), i.e. the informants’ simultaneous situatedness and positionings in social spaces. It is thus an analytical strategy with a double focus on both the social locations and positions of the researcher and the informants. It thereby turns our attention to the dynamics of positioning that all social actors are embedded in. As an analytical principle, location-work implies an attention towards the (possibly multiple, perhaps contradictory) positions that actors occupy or relate to. The principle of location-work fits well with the concept of field and social space as introduced in chapter two. The focus on shifting locations implies attention to how positions and modes of differentiation are articulated and practised in different contexts and by different actors. Location-work is further complicated in a transnational analysis. As Levitt and Schiller write, in transnational social fields, individuals may occupy or relate to (possibly) different, but simultaneous social positions and social statuses (2004, 1015). They might have multi-local frames of reference, positioning them differently in Danish or Somali contexts. However, as Fraser (2003, 59) reminds us, the absence of a stable and singular status group order is a shared condition in late modern society. We therefore need to complicate contexts, positions and locations even further. Location-work thus implies a focus on how different actors (including the researcher) are positioned within the field: the multi- dimensionality, contradictions and ambivalence of changing positions in different contexts.

Asking, looking, listening Methodologically, my interest in contexts, simultaneity, and positions means that I have asked, looked and listened for simultaneity and the spatial dimen-

67 sions of associational engagement, as well as for positioning during the production and analysis of data. As sociologist Les Back puts it, I have actively listened for certain kinds of stories and practices (2005). I brought the focus on transnationality, simultaneity and positions with me into the fieldwork right from the start, while remaining open to other elements and significances (Baszanger and Dodier 2005). I have thus been on explicit lookout for how, to where and with whom the informants relate their involvement and positions, thereby exploring spatiality and simultaneity. More concretely, this means that I have endeavoured to include a focus on both the families and the social backgrounds of the informants, their life trajectories, their perceptions of life in Denmark, as well as on their associational engagement. To analyze positionings requires data about social relations and ideals, articulated and practised from different contexts and viewpoints; it requires that informants talk about themselves and talk about others in relation to both ideals and practices. I have therefore enquired about ideals, events and associational actors, asked about specific situations and, not least, listened to the sometimes diverging, sometimes coinciding stories and opinions concerning life in Denmark, the developments in Somalia and Somaliland, transnational engagements, gender relations, politics, specific events etc29. Carrying out participant observation, interviewing and engaging in informal conversations several times with the same informants made such location-work possible. Furthermore, due to multi-sited fieldwork and my inquiries about transnational practices, I had an analytical focus on simultaneous contexts, keeping in mind the transnational perspective and the “simultaneity of spatial scales” (Massey 1994, 264). I thereby made an effort to be sensitive as to when and how struggles for recognition and negotiations of positions relate to several simultaneous contexts at once and, when they do not, to show an awareness of when associational involvement has transnational dimensions and when it is mainly focused on life in Denmark.

29 Let me emphasize that while I have often heard informants talk about other informants, I have not disclosed any personal information that have been confided to me during interviews and conversations. Or in other words, I have listened to gossip, but I have not gossiped myself.

68 The research site and the process The next question, then, is what these reflections mean for the delimitation of the research site and the production of fieldwork. Due to my interest in processes of inclusion and transnational involvement, I delimitated the research site to Somali-Danish associational engagement, which takes place in Denmark as well as transnationally, mainly – but not only – in Somalia and Somaliland. The research site thus takes departure in Greater Copenhagen but stretches to other localities as well. In this sense, I have carried out multi-sited fieldwork (Marcus 1995), following (part of) the associational involvement and relations from Copenhagen to Somaliland and London as two central destinations. I have thus not done fieldwork of the transnational practices in their entire transnationality or multi-locality so to speak (cf. Hannerz 2001, 18). Rather, the research site corresponds to what could loosely be termed the Somali-Danish associational field, feeding into larger transnational social fields where I have focused on particular localities. However, as Bourdieu (1996a) emphasizes, field is an analytical concept which should be carefully constructed and analyzed; it is not a discrete, easily delimited, geographically distinct entity. A part of this kind of fieldwork is thus also to delimit and explore ‘the field’ along the way, emphasizing the process aspects of research. I structured this ‘field exploration’ into three stages, which were partly overlapping and combined with other research activities, such as theoretical read- ings and initial coding and analysis. This tripartite process was partly a result of critical and mobile positioning, as Haraway and Gupta and Ferguson termed it, and partly of luck and coincidence, emphasizing the need not only for strategic planning, but also of seizing opportunities along the way. Fieldwork started in Copenhagen. In the first stage, ranging from February to August 2003, I contacted a number of the informants I knew from 1999, including Mohamed Gelle, the organizer of the Somali Network in Denmark. Gelle invited me to a seminar in February as well as to an international Somali conference held in Copenhagen in May (that I analyze in chapter seven). At these events, I observed, and wrote elaborate field notes as well as the seminar reports. In this initial stage, my aim was to establish and re-establish rapport and get a general feeling of ‘what was going on’, using the seminars to make contacts and observe the activities. From earlier fieldwork, I had learned that seminars offer good venues for establishing contacts, as the participants usually have surplus

69 energy and engagement and therefore might constitute potential informants, even if the subject matter is not wholly related to their organizational activities. This time, I quickly became curious about associations and decided to make them central to my dissertation. I thus started to carry out an overview mapping of Somali-Danish associations, interviewing key persons about the Somali- Danish associational landscape, its history, positions, and scope. Likewise, I carried out in-depth interviews about the seminars I had participated in, enquir- ing about the participants, the discussions, and the commitments and destinations of the activities. The first stage, then, can be described as an orientation and mapping phase to enable me to include different positions and locations in the remaining part of the data production. In the second, more focused stage, I also combined interviews and associational observations. This stage took place from August to December 2003, also in Greater Copenhagen. I conducted semi-structured qualitative in-depth interviews, loosely structured around an interview guide that I adjusted from time to time. In all interviews, I noted basic socio-economic indicators such as age, education, employment, length of stay in Denmark, and family and socio- economic background from Somalia and in Denmark. The interviews focused on the informants’ associational activities as well as on their life histories, family relations, and perceptions of life in Denmark. As I already knew ten of the Somali-Danish informants from my fieldwork in 1999, their background and life histories were familiar to me. In most of the other cases, I conducted at least two interviews with the informants, where one focused on associational involvement and the other on life histories, and/or I supplemented the interviews with informal conversations. Generally, I tried to establish more regular relationships with the informants, meeting and chatting informally in different contexts because, as Kathy Charmaz notes, “multiple visits over time combined with the intimacy of intensive interviewing do provide a deeper view of life than one-shot structured or informational interviews” (2005, 529). This was possible with about half of the Somali-Danish informants, and out of them, I developed close relationships with six informants. Furthermore, as several of the informants were key persons in the Somali-Danish associational scene, I also ran into a number of them when I had not planned to, for example at seminars or in my neighbourhood in Copenhagen where there are quite a lot of So- mali clubs and shops.

70 I let the informants decide where it would be most convenient to meet, which was usually in their home, in associational premises, or in my office. At other times, however, we met in my home, at their workplaces, in a community house, or in cafes. Most of the women preferred to meet in their homes, while many, but not all, of the men found it more convenient to meet outside their homes, perhaps in my office or in another more neutral place. This can be seen as reflecting gendered space where women are connected to the domestic sphere and men to public space (cf. Farah 2000), but it might also reflect a combination of small apartments and big families, as many of the male informants had more children than the female ones. In many cases, the interviews were combined with a social visit: drinking tea, exchanging news, and sometimes sharing a meal. Apart from sheer pleasure, this also offered the opportunity to talk about things other than those included in the interview guide, thus furthering confidence and enabling me to observe the context and social positions the informants were a part of (cf. Charmaz 2005, 529). Such observations and sociality took place when I spent time with the informants during interviews, at associational events, in trains and cars to and from such events, at cafes, or at parties. In addition, I arranged to attend a Somali language course together with Peter Hansen, a colleague also studying Somali migration. While I never managed to learn more than a very few basic Somali sentences, the language training developed into general discussions about Somali culture, politics, family and kinship structure, which were very interesting and useful as a kind of background knowledge to Somali daily life and as stories of Somali cul- ture30. I also started to carry out regular participant observations in two associational sites, selected due to their different organization, settings and political lo-

30 The concept of culture has been deconstructed and thoroughly criticized within more recent academic work (.e.g. Amit-Talai 1997; Gupta and Ferguson 1997a; Olwig and Hastrup 1997). In accordance with my social constructivist position, I share a critical perception of the concept of culture and therefore focus on the stories and claims of culture as frameworks of interpretation and acts of mobilization. I thereby operate with a narrative view of culture (Benhabib 2002, 4), as “a horizon that recedes each time one approaches it” (ibid., 5), not as a unified, holistic and unequivocal collectivity. Stories of culture in Benhabib’s sense thus constitute a framework of interpretation at both the individual and collective level, which are interesting and necessary to analyzes, as stories and claims of culture are often employed as explanations and reasons of behaviour.

71 cation: the Somali Network in Denmark and what I here call the Woman’s As- sociation. In August and September 2003, I spent one day a week in the Somali Network’s office facilities in the Danish Refugee Council, helping out with applications, seminar reports and other written material, as well as talking with Gelle. At the same time, I spent one to two afternoons a week as a Danish language training volunteer in the Women’s Association, practising Danish with some of the members and hanging out afterwards. These two parallel series of observation gave insights into very different aspects of associational life and locations. Spending time in the Somali Network, which has as its objective to initiate, support, inform, and coordinate Somali activities and associations, meant that I mostly discussed issues both concerning coordination of and collaboration between Somali migrant associations, and in relation to NGOs, ministries and other stakeholders in Denmark, Somalia and elsewhere. It was in many ways a location between different stakeholders and associations. Furthermore, the Somali Network is pan-Somali, including associations and activities from all over the Republic of Somalia and is explicitly for its unity – both in terms of cooperation between Somalis in Denmark and in relation to the political situation in Somalia. It was thus also a specific political location. In the Woman’s Association, I helped out in the Danish classes. These were attended by between two to twelve Somali women, who had come to practise Danish or get practical help or advice from the volunteers. Apart from me, there were some other Danish volunteers present, and usually one or two Somali organizers from the association. It was sometimes very busy and sometimes very quiet. Here I could observe associational activities in practice, speaking both with members about issues concerning their daily lives, and discussing the particular activities of the association with the organizers. In addition, as the Woman’s Association is defined by a regional affiliation and in favour of the region’s political independence, it also had a specific political location in relation to the question of Somali politics and other associations. Given that I spent a lot of time both with Gelle in the Somali Network and an organizer in the Woman’s Association, both series of observations gave me the opportunity to ask about and discuss issues I was curious about in relation to my fieldwork (which were quite a lot). In this second stage, I thus explored Somali-Danish associational engagement through interviews and observations, focusing on three sets of connections: between associational engagement and the infor-

72 mants’ life trajectories and present situations, between associational engagement and transnational involvement, and between associational engagement and positioning. In the third stage of data production, I supplemented my Danish fieldwork with supplementary and contextualizing fieldtrips (cf. Kleist 2004b) to analyze the transnational involvement that the informants were engaged in. I spent two weeks in Hargeisa, the capital of Somaliland, in October 2003 and, during a research period at the Sussex Centre for Migration Research, I went several times to London in March 2004. My aim with these two supplementary fieldtrips was three-fold. Firstly, I wanted to get an insight into these localities because of their centrality in the lives of many of the informants. Secondly, I wanted to follow transnational engagements and relations through interviewing associational collaborators, relatives of the informants, project employees, government officials and other stakeholders, and through visiting projects. Thirdly, I wanted to contextualize life in Denmark – including associational engagement – through a focus on how life in Denmark, and more generally the West, gender relations and ‘the diaspora’, are articulated from a Somaliland perspective. Location- work was thus part and parcel of the fieldwork design, not only through the inclusion of different people and associations, but also through locating me at different ends of the transnational networks. Somaliland and London are two important destinations in the transnational relations, engagements and imagination of many the informants, even if they obviously are not the only ones. A number of the informants have relatives in Somaliland as well as in the Somscan & UK project, which I mentioned in the vignette, where houses are being built in Somaliland. Furthermore, due to practical possibilities including both the security situation and the fact that my colleague Peter Hansen was already in Hargeisa and had generously offered to help me out, Somaliland seemed the most obvious place to go. In Somaliland I carried out fieldwork together with Peter, visited projects and institutions supported by Somalis abroad, including the Somscan project, and carried out interviews with returnees from Denmark and with one of the Somali-Danish informants who was living in Hargeisa at that time. We also interviewed personnel from money transfer companies, the UN and NGOs, and government officials, including Abdullahi Hussein Iman, the Minister of Reconstruction, Repatriation and Rehabilitation. I met family members of Somali-Danish informants, and fi-

73 nally, of course, I was hanging out, seeing Hargeisa, taking lots of photos, driving around with Peter in his car, and having a good time. On my return to Co- penhagen, I used my photos, my trip and my new knowledge of Somaliland politics and daily life to qualify my interviews and interaction with the informants. London is another hub for Somali transnational relations and, furthermore, an important centre in the hierarchy of destinations among many Somalis. It also proved an interesting place to carry out fieldwork and to follow up on relations and associations. Like in Somaliland, I interviewed relatives and associational collaborators, including a Somali-British Somscan & UK organizer. I followed a hearing in Parliament, and I was also present at a big pro-Somaliland demonstration in the heart of London, as I come back to in chapter eight. In addition, I carried out some joint fieldwork with Peter. London marked the conclusion of my fieldwork. Though it was difficult, I had to stop producing data and start analyzing the results in more depth.

The positioned researcher Following Gupta and Ferguson’s location-work (1997b) and Haraway’s notion of “politics and epistemologies of location, positioning, and situating” (1991, 195), the question of my position in the research sites is a crucial one. Or rather, positions. Because, as the reflections about location-work, situated knowledge and intersectionality indicate, there is no simple answer. There is embodied location as well as strategic, mobile positioning during fieldwork. As anthropologist Cathrine Hasse writes, we are always participants in the social space within which we do fieldwork and observe: we are always interpreting as well being interpreted (2000, 43). My embodied and social position – white Danish middle-class woman in her mid 30s, without any particular religious conviction, married with two kids, university-based PhD student and non-Somali speaker – constitutes the backdrop of my location. There was no way that I could ‘pass as’ or be ‘taken for’ one of the informants myself (Grünenberg 2006; Fortier 2000); my participation was always visible. However, that said, I will argue that social location does not determine what we see or how we see it. Rather, social location is relational and contextual and, as argued in the theoretical chapter, it changes which modes of differentiation that are the most apparent – whether

74 gender, age, nationality, racialization etc. This was also apparent in the ways that I was positioned during the fieldwork. Some of the Somalis in Denmark whom I approached seemed to perceive me as a somewhat ambiguous representative of Danish society, whose business of asking personal questions was too much like social workers, official authorities, and nosy journalists. Due to the rather hostile political and media exposure of Somalis in Denmark, the position as Danish researcher does not necessarily further rapport and confidence, and I cannot blame those who found that they had already answered more than enough questions about their personal affairs. In some cases, then, my attempts to establish contact and participation were refused, and in other cases, the interaction remained very cautious. More than once, I found that I could not get through to informants because the family members answering the phone simply denied their existence. My guess is that many were guarded and insecure about their situation and rights in Denmark (cf. Møller and Togeby 1999) and therefore preferred to avoid unnecessary interaction with (someone perceived as) a Danish official/social worker/reporter- /researcher (cf. Adler and Adler 2002). The point is that establishing rapport and confidence were extremely important aspects of the research and, in some cases, this was difficult or impossible to achieve. In most cases, however, I encountered hospitality, friendliness, confidence and willingness to spend time with me, even if many of the informants were extremely busy. Several of the Somali-Danish informants emphasized how they found it important to share their opinions, that I spoke with them, rather than about them. Sometimes I was explicitly positioned as a Danish listener to whom experiences of racism or discomfort could be passed on so that a wider public would eventually hear about them. A few times, I functioned as a chan- nel of frustration (cf. Fangen 2006b), but most people emphasized that their critique was not personal, but rather directed at the societal and political level. My ‘Danishness’ could thus both disable and enable communication. Other times – especially among the politically engaged – I became a potential ally who might offer a voice in favour of their association or in relation to Somali political issues. While I always stressed my impartiality in relation to Somali politics, I did try to help with non-political issues if I was asked to, writing seminar reports, helping out with applications, or giving practical advice for example. Thus, while I ‘used’ the informants for my dissertation, some of them also

75 ‘used’ me for certain purposes or had different interests in talking with me. However, let me emphasize that I do not regard such interests as disqualifying in any way, or less ‘worthy’. It is my genuine impression that most of the informants agreed to participate because they were committed to their projects and wanted to share their opinions and/or do me (or the person who introduced them to me) a favour. And let me emphasize that I am deeply grateful for that. The question about nationality and identification was thus important but certainly not the only one of relevance. Being a trustworthy person was perhaps the most important position in terms of gaining access and carrying out fieldwork. Due to my interest in positioning and relations – as well as the importance of trust in establishing rapport and confidence – I used ‘the snowball method’, going through different existing networks and associations. This was facilitated by my previous fieldwork, as I already knew a number of Somali- Danes and associational key persons, who introduced and recommended me to others because, as some of them said, ‘they already knew me’. Apart from fa- cilitating contacts, this also enabled a longer time perspective on the development of positions in the Somali-Danish associational landscape. Still, establishing contact through associational key persons meant that I went through specific locations, which I – partly due to my previous fieldwork, partly due the mapping of associations in the first fieldwork stage – did as conscientiously as possible, through differentially positioned entry points and gatekeepers. Thus, while a lot of the informants knew (and knew of) each other, they were not part of the same networks, friendships or associations, and certainly did not share the same opinions about quite a lot of subjects. The routes of access thus went through different key persons, different associations, and different political orientations. Getting access through and to different people and positions brought more locations than ‘white Danish female researcher’ into play. As Somali-American sociologist Abdi Kusow argues, the question of insiders and outsiders is much more complicated than the issue of belonging to an ethnic or national group or not. Discussing his own experiences of negotiating access to Somalis in Canada, having immediate access to most Somali men, but also positioned as a “suspicious insider” in relation to politics or culturally sensitive issues (2003, 594), Kusow concludes that research “status emerges from the interaction between the researcher and the participants as well as the social and the political situation within which the interaction occurs”

76 (ibid., 597). In this dissertation, the social and political situation relates both to the reception in Denmark, Somali politics, and gendered norms of behaviour. One of the consequences of my position as a non-Somali is that I do not have a clan31 and therefore am not a part of clan-related trust and distrust. Likewise, my insistence on not taking any particular stance in relation to the question of Somali politics was readily accepted, even though there were certainly a number of attempts to persuade me for or against the legitimacy of Somaliland. Fur- thermore, as a non-Somali woman I had access to woman’s associations, which a man would not have had (cf. ibid., 597). Similarly, as a non-Somali woman I could, to a certain degree, participate in events and spend time with Somali men as the only woman, which, according to some of the female informants, might have been difficult or very uncomfortable for them. The fact that I have a family – and usually talk about it at length, even during fieldwork – probably helped my ‘respectability’ in such situations. I did, however, refrain from going to Somali clubs exclusively frequented by men. Doing joint fieldwork with Pe- ter, a man, further helped overcome my gendered location in Somaliland.

Producing and processing data I thus selected the informants and research sites to include different locations and positions. Below, I present the informants and the interviews I carried out and specify how I produced and processed the data gathered. The key criteria of selection of informants in Denmark were, firstly, associational engagement in a Somali ‘ethno-national’ migrant association and, secondly, diversity of positions in relation to gender, political orientation, regional affiliation in Somalia, family position, educational background, employment, and length of stay in Denmark. In Somaliland and London, the criteria of selection were based on the principle that the informants should be related to the informants in Denmark, engaged in one of the associations supported by Somalis in Denmark, or be stakeholders able to contextualize and put associational engagement and transnational practices into perspective. I included fifteen Somali-Danish associations in total, of which nine had activities focusing on processes of inclusion in Denmark, and six were supporting reconstruction projects in Somalia or Soma- liland or were engaged in other kinds of transnational practices. I present the as-

31 Clan refers to the patrilineal lineage system into which all Somalis are born. I discuss clan in much more detail in chapters four and five.

77 sociations in more detail in chapter six, which focuses on Somali-Danish associational engagement. Let me now introduce the informants.

The informants During fieldwork I interviewed, observed and engaged in numerous informal conversations, speaking to a large number of Somalis in Copenhagen, Hargeisa and London. Likewise, I discussed my project with many other people, especially in Copenhagen, such as project employees, social workers, students, and other researchers. All these different people have contributed to the study and my reflections in some way or other. However, as the study is carried out from the migrant perspective, I have delimited the category of informant to Somalis in Denmark – and affiliates in Somaliland and London – with whom I have carried out a formal interview and/or spent a lot of time. This adds up to a total of 50 informants: 32 men and 18 women. In Denmark, I interviewed and spent time with 14 women and 15 men; in Somaliland three women (of whom one was also a Somali-Danish informant) and 11 men; and, finally, in London two women and six men. The gendered bias is due to the fact that most of the interviews in Somaliland were with male returnees, officials or project workers. The age of informants ranged from 18 through to the late 60s, though most of them were between 35 and 55 years old. With one exception, all the Somali-Danish informants had come to Denmark as asylum seekers or family re-unified as adults or young people. The majority of Somali-Danish informants were chairpersons, organizers, or other kinds of key persons in the fifteen associations. Out of these, about half were engaged in transnational reconstruction associations and almost all of them were contributing to collections or remitting money on an ad hoc basis to their relatives. While these people constitute the main informants, I supplemented them with interviews and informal conservations with others who were not connected to associations at all. These included two men, who both, however, frequented Somali clubs, and four women, who were not and did not want to be part of any Somali association. Though the informants’ backgrounds in many ways were diverse, almost all of them shared the same middleclass background in Somalia; a few came from upper-class and very politically influential families, while there were also some with considerably more modest backgrounds. This does not mean that the informants lived wealthy lives in Den-

78 mark, as they had practically all lost their livelihoods, possessions and possible wealth due to the civil war. What it means is that the vast majority of informants had at least a high school degree from Somalia, and several had pursued further education, either in Somalia or elsewhere. Several of the informants who had come to Denmark as adults had worked as professionals in Somalia or elsewhere; a few of them had long successful careers behind them. However, as it is very difficult to use and transfer educational and work-related qualifications to Denmark, none of the Somali-Danes I have met had been able to directly use their professional qualifications from Somalia in Denmark32. Most of those who came with professional training had supplemented – or rather substi- tuted – it with a Danish education, or they had started to do so. The large majority of informants thus hold jobs as bilingual teachers, translators, social workers, or healthcare workers. A number of the younger informants were studying and just a few were unemployed. At the time of fieldwork in Denmark, five of the female informants were working, seven were studying and two were unemployed. Among the male informants, seven were working, two were studying and six were unemployed. These numbers are thus quite different from the average Somali employment rate, as we saw it in the introduction, which totals 23 percent for men and 10 percent for women (Ministry of Integration 2005b, 174). The informants were, in other words, not statistically representative of all Somalis in Denmark. Rather, they were among the more privileged and doing relatively well in Danish society in terms of employment and socio-economic position. Or in other words, in spite of my efforts to include different positions, most of the informants shared – at an overall level – a middle class background, even if they were divided by political orientation concerning the question of Somalia and Somaliland. There is thus a certain bias in the material. At the beginning of the fieldwork, all the Somali-Danish informants lived in Greater Copenhagen, though some of them actually moved away from Den- mark during or after the period of research, and a couple of them went back and forth to Somaliland. Also, all of them had relatives living in other countries, not only in Africa, but also in Europe and/or North America and in Asia: they were thus part of extended transnational families. In terms of family background in Denmark, the majority of informants had children and some had grandchildren.

32 Exceptions were people who had either studied outside Somalia and/or taken a similar educational course in Denmark.

79 A number of the women and men were divorced; some had remarried, others were living alone and others again found new partners during the time of fieldwork. The informants came from all over Somalia and they had all lived in Denmark since the early or middle of the 1990s, although a few of them had come much earlier and were, in fact, among the very first Somalis in Denmark. The length of stay in Denmark and educational background also meant that most of the informants spoke Danish well. In a few cases, however, we spoke English together, and I used a translator in one case. Language is thus another issue that biases the data I have produced, meaning that I have not included the most isolated groups of Somalis in Denmark in the study.

The interviews I have combined qualitative semi-structured interviews, informal conversations, and participant observations in all three stages of the fieldwork. I carried out 53 interviews altogether, of which about half were tape-recorded; the other half were based on notes, which were primarily interviews carried out in Somaliland and London and with Somali-Danish informants who felt uncomfortable about the tape recorder. Two of these sessions were focus group interviews. Likewise, there were several people present during a few of the other interviews, but the majority were individual interviews. The interviews typically lasted between one and two hours. I carried out two or more interviews with 12 of the informants, typically one about associational engagement and one about their family relations and life history. The aim of the life histories was to analyze the trajectories of physical and social mobility of the informants and to allow them to “define and explicate to others their own fields of belonging and identification and how these fields are articulated with the socio-cultural and physical boundaries which they experience in their everyday life” (Olwig 1999, 29). Life histories thereby make it possible to situate the struggles for recognition and negotiations for positions into a wider perspective. Research assistants transcribed the tape-recorded interviews verbatim, but I listened to all the transcribed sessions afterwards, adding and correcting details in the transcripts and getting the interviews ‘under my skin’ so to speak. In the case of note-based interviews, I typed my hand-written notes as quickly as possible after the interview, adding as many details as I could. I also wrote detailed descriptions of the interviews and observations, memorizing the context, the

80 atmosphere and the main empirical points. Similarly, I wrote memos of the encounters, telephone calls, etc. and kept separate files for ideas for analysis and interpretation. With the exception of Mohammed Gelle, Musse Sheikh, and the politicians and officials in Somaliland who did not wish for anonymity, I have kept all other informants anonymous, changing their names and personal details to protect their privacy. Likewise, I have changed the names of associations included, with the exception of the Somali Network in Denmark, the Somali Council, and Somscan & UK. To offer context as well as to ensure anonymity, I have pursued a writing strategy of changing between contextualization and decontextu- alization. Thus, in some cases, I present the life story and life situations of the informants in detail, while in others I merely refer to ‘a Somali woman’ without disclosing any further details. When I present detailed descriptions such as life stories, I have – to the extent that it was possible – showed the texts to the informants to ensure that their privacy has been sufficiently protected. Further, I have erased the town of birth of one person and altered other small details, but have not changed any major issues.

From fieldwork to analysis Though I did not set out to do a grounded theory study, my procedures in producing and processing data in many ways follow the guidelines of constructivist grounded theory as outlined by Kathy Charmaz. Reckoning with the positivist roots of much grounded theory, Charmaz stresses the interactional nature of the research process and the importance of “giving close attention to empirical realities and our collected renderings of them – and locating oneself in these realities” (2005, 509, emphasis in original). Data, in this sense, are not separate from and untouched by the researcher; rather, as “social scientists, we define what we record as data, yet how we define data outlines how we represent them in our work” (ibid., 511, emphasis in original). Data are, in other words, not found, observed, or collected, but produced by the researcher in interaction with the informants. Charmaz calls for rich empirical material and intimate familiarity with the setting. In terms of the processing and analysis of data, she emphasizes initial and analytical coding, constantly comparing data and categories, and paying close attention to meaning-making processes and action, social context and language. These have also been priorities in this study.

81 I started processing and analysing data during fieldwork, initially making notes of the main empirical and analytical points of each interview and observation in order to start developing concepts and themes. After I had produced the vast majority of data, I coded all data in the software data management programme QSR Nvivo, according to type of data (informants, associations, observations etc.), empirical themes (such as associational engagement, life in Den- mark, life in Somalia, transnational practices, gender relations etc.) and analytical concepts (recognition, space, temporality etc.). This exercise both served to provide an overview of the quite extensive data material, and furthered the analysis through the coding process and the concentrated reading of my material. However, I also ‘took breaks’ from fieldwork and, later, data coding, writing papers based on particular events or associations. I thereby ‘opened up’ the data through in-depth analysis of a limited part of the data and through exploring the analytical potential of theoretical perspectives. In this way, I used what sociologist Charles Ragin has termed casing as a methodological step to explore and analyze links between ideas and evidence – between theory and data (1992, 218), striving to produce “theoretically structured descriptions of the empirical world” (ibid., 225). Likewise, as sociologist and poet Laurel Richard- son (2005; 1997) suggests, I engaged in writing as a method of inquiry, which not only serves as a medium of communicating already processed knowledge, but also as a way of getting to know through writing, exploring, and, yes, revis- ing and rewriting. Thus, while case studies enabled me to analyze and study in depth, focusing on particular events and people, the coding of the empirical material ensured that the entirety of data was also analyzed comprehensively with regard to questions of regularity or particularity. I have thus analyzed in depth as well as across cases. The dissertation reflects this double analytical process in that I both include specific case studies as well as the more general analysis. I use quotes as illustrations of the broader analysis, but again, I also include examples which are atypical or particular – and I strive to tell the reader when I do what. I have edited transcripts and note-based interview reports to make quotes fluent, thereby transforming the spoken language to written prose (cf. Kvale 1997). In the case of note-based interviews, this transformation already takes place at the moment of interview, and I have tried to make interview reports as accurate and elaborate as possible. Furthermore, I have translated quotes from Danish interviews into English. This second transformation em-

82 phasizes that although the quotes in the dissertation are based on the interviews, they are also selected, edited, and sometimes translated. In all cases, they are transformed but loyal re-presentations of the interviews. Quotes are thus not naturalistic reflections of a reality, untouched by the researcher, but rather a product of the data production: not invented, but co-produced by me and the informants. I use the following transcript signs in the quotes:

Italics Words pronounced with emphasis … Short break […] Words or sentences omitted [explanation] Explanation of a word or expression

Quotes are used in the analysis as well as in the vignettes. However, I change genre in the vignettes. The main part of the dissertation swings back and forth between realist and confessional tales (Maanen 1988), emphasizing the empirical groundedness as well as location-work, while in the vignettes I try to retell fieldwork in more vivid ways, as an impressionist tale, as Maanen terms it. The vignettes are still based on fieldwork notes and interviews, but I have allowed myself greater freedom in the presentation. In this way I hope to pass on just a bit of the fun, atmosphere, excitement and, basically, human interaction that is always a part of fieldwork – and indeed one of the reasons that I find that it is one of the (academically) most gratifying things to do.

Chapter resume: From theory to data to analysis In this chapter, I have presented the methodological considerations and analytical principles which underpin the dissertation. Following Gupta and Ferguson, I introduced the concept of location-work as a central tenet that places analytical attention to the mobile and multi-local positioning of both the researcher and the researched at the foreground. This methodological concept is related to Haraway’s notion of situated knowledges and Brah’s focus on the politics of location. I have integrated location-work into the fieldwork through two overall principles. Firstly, by looking, listening, and asking for positions, and secondly, through locating myself in different associations and different research sites in Copenhagen, Hargeisa, and London. Similarly, I have reflected upon my own

83 position, and how different social locations came into play during the course of fieldwork, furthering or hindering the relations with the informants. The main bulk of fieldwork was carried out in Copenhagen from where I followed relations and associations to Somaliland and London. Fieldwork was divided into three overlapping stages: a first orientation and mapping stage in Co- penhagen; a second focused stage, where I carried out systematic interviews and a series of observations, also in Copenhagen; and, thirdly, a stage of contextualizing fieldwork in Hargeisa and London. The data material consists of interviews conducted with a total of 50 informants, 32 men and 18 women, as well as a range of observations. The informants were selected according to a criterion of active engagement in a Somali-Danish association, with the majority being associational key personnel and organizers; I also included a few informants who were not active in any associations. Furthermore, I endeavoured to include a variety of positions in terms of gender, age, political orientation in relation to Somalia and Somaliland, etc. Still, most of the informants share a middleclass background from Somalia, indicating a certain bias in the data material. After concluding the main fieldwork, I coded interview transcriptions, field notes, and memos in detail in a software management programme. The coding process allowed me to analyze the rather large amount of empirical material both in depth and across cases, which is reflected in the dissertation through a mix between cases studies and the general analysis. And now it is time to move on to the empirical analysis. In the next chapter, I analyze the Somali context of exit. I pay special attention to Somali migration history, and present four life histories, focusing on the migration from Somalia to Denmark and transnational relations. We are thus going to move across borders.

84 PART II

ACROSS BORDERS

85 86 Chapter 4 Journeys and settlements. Outline of the Somali political and migration history

He was adrift (and so was the Somali nation everywhere) on a tide of total abandon. At least, he kept thinking to himself, staring at the map on the wall, there would be changes in the cartographer’s view on the Horn of Africa. And so, with his felt pen, using his own body, he re-drew the map of the Somali-speaking territories, copied curve by curve, depression by depression (Farah 1986, 97)

***

In the novel Maps, Nurrudin Farah writes about the Somali boy Askar, who is growing up in the Ethiopian Ogaden area, but has to flee with his family to Mogadishu after Somalia’s failed attempt to conquer the area in 1977. Askar reflects a great deal about the significance of maps, which make some rulers and some refugees. In this chapter, I turn from theoretical mappings and methodological directions to an outline of the Somali political and migration history33. This history is also related to maps, political maps of nation-states and their borders. I show two maps. The first, from 1963, shows the cultural borders indicating ‘the Somali-populated territory’ as well as international and (former colonial) boundaries. The second map depicts the national borders of the Re- public of Somalia. These boundaries are not the same. But it is not only political borders in the sand that separate the Somalis. Somalis are renowned for their mobility and are now spread throughout the world. From South Africa to Finland, from China to Canada, from Yemen to Australia, from Somalia to Denmark. I outline some of the aspects of this migration, focusing on how different cohorts of migrants from (what became) the Somali Republic are embedded in a historical framework which relates to different periods and events in Somali history as well as to global and geopolitical developments. The chapter

33 For more detailed studies of the history of the Somali-speaking region and Soma- lia more generally, see Lewis (1994; 2002), Ahmed (1995b) and Bradbury (1997). For an extended version of this chapter, see Kleist (2004a).

87 is based on historical reviews combined with colonial documents and other ar- chival material, such as reports, newsletters and other kinds of grey literature34. I refer to statistics to exemplify tendencies, even if statistics concerning migrants and refugees are often notoriously unreliable (cf. Crisp 1999a; Steen 1993). Furthermore, I present four migration stories to exemplify how migration to Denmark is embedded in larger historical and political changes, which relate not only to civil war, but also to colonization and geopolitical changes35. These stories thus relate fields of belonging and identification to socio-cultural and physical boundaries and movements (cf. Olwig 1999, 29); they also connect the historical account with the Danish context. I start by a brief introduction to ‘master stories’ of Somali culture.

Stories of Somali culture Mobility has been a quintessential part of life in the Somali-speaking region for many centuries and has played an important role in Somali culture and the dominant discourses of ‘Somaliness’ (Horst 2003; Lewis 1994). While there have been trade towns along the coast for centuries and fertile lands in the south, most of the Somali-speaking region is semi-desert and, until a few years ago, nomadic pastoralism and trade have been the main livelihood. Further- more, the Somalis are said to have migrated from Aden to the Horn of Africa about a thousand years ago (Lewis 2002, 18-19). Migration and nomadism are thus part of ‘Somaliness’ and Somali history, as are lineage affiliation and Is- lam. As already mentioned, I employ a narrative view of culture (Benhabib 2002), focusing on discursive repertoires and ‘master stories’, invoked and mobilized for different aims. That said, let me emphasize that I do not want to imply that Somalis have nothing in common, but stress that ‘Somaliness’ and ‘Somali culture’ – like all other ‘cultures’ – is produced, invoked, contested, and negotiated – not given or natural (cf. Hall 1990, 226). The ‘master stories’ of culture thus relate to the discursive repertoire through which the world is per-

34 These include reports from the Colonial Office (1960a; 1960b; 1931; His Ma- iesty's Stationary Office 1928), newsletters from the Anglo-Somali Society (Suleiman 1992; Wilson 1990a; 1990b) and reports from associations and NGOs (Save the Children Fund 1994; Somali Relief Association 1992). 35 The stories are constructed on the basis of several interviews with each of the informants. I have edited the stories, but the words and ways of telling them are as close to the interviews as possible.

88 ceived, negotiated, and contested (cf. Jackson 2002, 18), not least in relation to what has caused change, development – and, especially in the Somali case – conflict. Somali historiography can be divided into two schools or directions. The most influential writer of the Somali history and culture is I.M Lewis, Emeritus Professor in Social Anthropology at the London School of Economics. Lewis has done pioneering (as well as contested) work on Somali issues since the 1950s, especially in what was then the British protectorate of Somaliland in the north-west. For Lewis, the genealogical lineage system, often termed clan, is central to all aspects of Somali culture and society. In this chapter, I refer in particular to two main works by Lewis (2002; 1994), and to a Somaliland country report by social analyst Mark Bradbury (1997) which also draws quite extensively on Lewis. In opposition to Lewis, there is a body of work challenging the centrality of the clan lineage system, criticizing it for being primordial and overemphasizing the importance of clan at the expense of other social condi- tions36 (e.g. Farah 2000; Ahmed 1995b; Samatar 1994b). The contested question of clan is thus one of the major issues in accounts of Somali history and Somali culture. Clan refers to the segmentary and patrilineal lineage system, which structures persons’ kinship according to the father’s line of relatives in descent groups (Barnes 2006; Luling 2006; Lewis 1994; Mansur 1995). According to this system, Somalis are born into a lineage: the child is given a first name and then has his or her father’s and grandfather’s name as family names. Neither women nor men change their names when they marry. Elderly men, politicians and men of status are often renowned for their comprehensive knowledge about clan lineages. I am not going to go into details about specific lineages here or later in the dissertation; what interests me is how clan is used as a principle of organization or mobilized for political purposes. In Somali, this agnatic descent is termed tol – literally that which sews together (Ahmed 1995a, xi) – and the descent groups are often referred to as

36 For a summary of this controversy, see Barnes (2006).

Map 1: Map of colonial and political borders as well as indication of the Somali-speaking population. Touval (1963, 7).

Map 2: Political map of Somalia. Map. No. 3690 Rev. 5. United Nations.

91 reer37 (Lewis 1994, 82ff). The meanings, value and impacts of the lineage system remain highly contested and debated. According to Lewis, tol is “the cornerstone of Somali social relations, providing the individual’s primary group identity and loyalty – or more accurately […] an elastic range of solidarity” (ibid., 82). The relationship between different agnatic groups is regulated by shifting political ‘contracts’, heer, which make alliances between some groups, and which are subject to change when new needs and conditions arise (ibid., 96ff). Lewis describes the kinship system as stable, while the different allegiances are highly unstable, flexible and fluid. The clan lineages are divided into two: the Sab and the Samale. The Sab lineages are comprised of the seden- tary Digil and Rahanweyn clan families in the South, whereas Samale include the clan families of Dir, Isaaq, Hawiye and Darod, who are – or rather used to be – pastoral nomads scattered throughout the Somali-speaking region (Lewis 2002, 4ff). According to common myth, sheikhs Isaaq and Darood migrated from Aden to the Horn of Africa about a thousand years ago, and are said to be related to a cousin of the Prophet (Luling 2006, 473), a relationship which emphasizes both the Muslim heritage and the tradition of mobility.

37 Following Lewis, agnatic unities are most often mobilized at what he terms the ‘diya paying group level’, exogamous groups of up to six generations (Lewis 1994, 83-84). Traditionally, diya paying groups pay and receive diya – blood compensa- tion – on a collective basis, constituting “focal units for cooperation, political action and collective solidarity” (Bradbury 1997, 4). While the overall clan system was much debated among the informants in this study, none of them referred to diya paying groups.

For Lewis, the lineage system is an all pervasive feature of Somali society, which he describes as a “multipurpose, culturally constructed resource of com- pelling power because of its ostensibly inherent character “bred in the bone” and “running in the blood”, as Somalis conceptualize it” (Lewis 1994, 233). Such a view, however, is challenged by other writers; for example, Ahmed Sa- matar criticizes “the master concept of clans as the only significant explanatory factor” (1994b, 6). Likewise, Nuruddin Farah criticizes the overuse of the concept as “a wraparound concept, perhaps one of the most abused tropes in our vocabulary. The clan is seen as both evil common denominator and an explica- tor all actions” (2000, 14). While I share these precautions, both Somali history and the present situation bear witness to the fact that the kinship system cannot be ignored. Therefore, it is important to present the concept and to examine how it has been employed in Somali history in order to understand what it means in the Somali-Danish context38. As will become clear in the next session, the lineage system remains central – for better and often for worse – in modern Somali political history.

Colonialism Up until the opening of the Suez Canal in 1869, the interests of the European colonial powers in the Horn of Africa were relatively limited and consisted of a few different treaties with local rulers39. Following the Berlin Conference in 1885, where the rules of the partition of Africa were laid down, the colonial powers started to divide their spheres of influence in the Somali-speaking region between them. In 1897 the Somali-speaking region was divided between four colonial powers. Great Britain possessed the Protectorate of British Soma- liland (from 1960 the north western part of the Republic of Somalia, from 1991 the self-declared Republic of Somaliland) and the Northern Frontier District (NFD – from 1963 the North Eastern Province in Kenya). Italy colonized La Somalia Italiana (which became the UN-Trusteeship of Somalia under Italian administration in 1950, from 1960 the southern part of the Republic of Soma-

38 For discussions about the meanings of lineage among Somali migrants see Grif- fiths (2002; 1997) or Farah (2000). For an overview of Somali clan family genealogies, see Lewis (1994) or Hagi & Hagi (1998). 39 For more detailed studies of the history of the Somali speaking region and Soma- lia more generally, see Lewis (2002; 1994), Ahmed (1995b) and Bradbury (1997).

93 lia). France possessed La Côte Française des Somalis (from 1977 Djibouti), and the regional colonial power Abyssinia (Ethiopia) colonized the Ogaden and later the Haud areas (still parts of Ethiopia) (Lewis 2002; Hess 1966; Vignéras 1900). As part of the British, Italian and French empires, the Somali colonial states were gradually incorporated into the world economy (Veney 1998) and the two world wars.

Colonial migration Colonization meant both restrictions of and opportunities for mobility. On the one hand, a substantial number of the Somali colonial subjects were nomads, who would wander over distances – including colonial borders – to find pas- tures for their animals or to trade. Mobility and borders thus seems to have been an issue from the beginning of the colonial ventures and the colonial powers tried to govern the ‘Somali tribes’ by restraining their mobility to avoid trouble (Hess 1966, 183; Colonial Office 1960b; 1931). On the other hand, by creating empires and colonial ties, the European powers also established new avenues of migration, both during colonization itself and afterwards, for instance in terms of employment and education. Somalis served in the British and Italian armies, not only in Africa but also in other parts of the world. Likewise, a number of Somalis went seafaring, travelling around the world, including the Far East, the Americas, Australia, Africa and Europe (Wilson 1990a; 1990b; Hess 1966). In the introduction to one of the numerous travel tales from the early 20th century, the author vividly describes this (male) Somali migration.

The Somali wanders afar. You will find him working as deck hand, fi- reman, or steward, on all the great liners trading to the East. I know of a Somali tobacconist in Cardiff, a Somali mechanic in New York, and a Somali trader in Bombay, the latter of whom speaks French, English, and Italian fluently (Rayne 1921, 6).

While some sailors returned to Somalia, others went to major UK ports as sailors or traders, where they turned into established communities (Jordan 2004, 14-21; Griffiths 2002, 77-83; Farah 2000, 98). Farah and others report that the Somali-British community started out as almost exclusively male and that the Somali seafarers and settlers in UK sent money back home as early as the 1880s, thereby indicating a long tradition of transnational linkages. Somali seafarers had their families in the Somali regions or in Aden in Yemen, where

94 many British ships put into port. It was not until the 1960s that Somali women and children joined their husbands on a larger scale in the UK. In Italy, Somali migration seems to have been more limited during the time of colonization; although Somalis went to Italy as traders and students, no established Somali community came into being (Suleiman 1992). Apart from intercontinental migration, a larger number of Somalis migrated back and forth between the Somali colonial states, East Africa, and the Arab countries in relation to trade, pastoralism, family movement, education etc. (Goldsmith 1997, 472; Greenfield 1987). One effect of colonialism is that the mobile livelihoods (Sørensen and Olwig 2002) of Somali pastoralists and traders became international migration as colonial – and later national – borders were established. Somali family networks and clan families might be divided between the British Protectorate, Aden and Ogaden in Ethiopia, or between British East Africa, and Somalia Italiana. All the same, the colonial ‘mother- lands’ became destinations and, as empires, they offered possibilities in relation to trade, military employment and seafaring. These themes are illustrated in the story of Mohamed Abshir, an elderly Somali-Danish man who arrived as one of the first Somali sailors in Denmark in the 1960s. First, however, his route took him to Britain and back again.

‘I always wanted to be a sailor’: Mohamed My father died when I was 8 years old, then I became an orphan. He was a sailor, seaman, and always told me about all these places. I knew the name of Copenhagen, even before I knew of Denmark! He went there as a sailor. I always wanted to be a sailor myself, like him. As a young boy I went to Aden in Yemen and got a false sailor certificate. I then went to Hargeisa and went to the emigration office to get a passport so I could go to Britain. In the papers it said that I was 26 or something and I was just a big boy. The officer looked at me and said that it wasn’t me and sent me away. I then went into the interior for a year to get tougher and look older.

One year later, I came back. I went to the emigration officer again. There was a very tall British man working there. I brought the same false papers. He looked at me from the bottom to the top and I was so much smaller than he was. They looked at my teeth and concluded that I could not be more than 16 or 17 years old. He finally said, ‘these are not your papers, but you are an enthusiastic young man who wants to see the world. You want to be a sailor – can you swim?’. ‘Yes’, I said. I could swim and row a boat. He did not give me a passport, but a special kind of sailor’s laissez-passer papers. I was very happy. I went to Aden again and got a job. After a while, a French passenger ship came to Aden, which was going to Marseilles. I bought my ticket, bought nice suits and

95 quit my job. My friends were going to hold a going-away party for me. But the French Consulate in Aden refused. They told me, ‘This is not a real passport, you cannot go’. What should I do? I had quit my job and everything. I went out to sit on a big rock and thought about my situation. My friends had the party, but I was not there. I decided to go on the boat the clandestine way. I borrowed a rowing boat, rowed out there with one of my friends, and told him not to go back before I was on board. I got on the boat with my suitcase, dressed in my best suit. It worked, they never knew!

There were eleven other Somalis on board, all seamen, and quite a lot of Yemenis. One elderly man who knew my father was also on board. When we arrived in Marseilles I put on my best clothes and disembarked with the other people. I was not stopped. But when we came to a Somali hotel, the staff did not accept my papers and they would not let me sleep there. They told the elderly man that the best thing to do was to take the train to Britain immediately. So we did that. Nothing happened until we arrived in Calais, where the French immigration caught me. Luckily my father’s friend quickly made up a story. He told them that I was from a Greek boat which had left me in Marseilles and that I was going to Eng- land as a British citizen. ‘Bon, allez’, the immigration officer said, and we took the boat to Dover. There I was caught by the British immigration, but since I was from a British Protectorate, I was allowed to enter the country. That was in 1955. I stayed three years in London, where I was involved in a political Somali association. We talked a lot about independence, fighting for our country. I was so involved that I decided to return to the British Protectorate. I went back in 1958.

In Mohamed Abshir’s migration story, the dynamics of colonial rule are bluntly articulated – as are the personal and adventurous initiative that Mohamed un- dertakes to go to Europe. While British colonial officials were in charge of when Mohamed could seek his luck as a sailor, his knowledge of the importance of personal deportment and style of dress when crossing borders, his father’s friends and the immigration legislation in France and Britain were just as crucial in framing his journey to Britain. There, however, his engagement in the political development in the Protectorate was so strong, that he decided to return. Mohamed’s narrative is thus also a story of a transnational life trajectory.

Independence When Mohamed Abshir went back to the British Protectorate, preparations for independence were busily taking place both there and in the Italian UN- Trusteeship of Somalia. After the defeat of the fascists in World War II, the future of Somalia Italiana was unclear, and the ambitious plans of uniting Ogaden, British Somalia and Italian Somalia into a Greater Somalia failed. In 1950, an Italian administered UN Trusteeship was ratified, and the UN decided

96 that the Trusteeship should be independent ten years later. Italy had started to make preparations, such as educating civil servants and holding elections, and political parties had emerged (Hess 1966, 190-193). In British Somaliland, independence was less prepared. Elections were held for the first time in 1960 and the leader of the biggest party, Mohamed Haji Ibrahim Egal, was appointed Minister for Local Government. In May that year, Egal and a delegation went to London to discuss independence with the Colo- nial Office. One of the declared aims of independence was to unify with the UN-Trusteeship of Somalia, and this goal was agreed upon in London (Colonial Office 1960a). On 26th of June 1960, the British Protectorate of Somaliland ended, and Somaliland was declared independent. Four days later, on 1st of July 1960, the UN Trusteeship of Somalia followed. The two former colonial states were united as the Republic of Somalia with Mogadishu as the capital. The union of the two territories was seen as a first step towards reunion of all five So- mali territories as mirrored in the five-pointed star of the Somali flag. Mohamed Abshir, who had become the local secretary of a political party in Somaliland, recalls:

I was there on 26th of June when we got our independence. I did not sleep for three days; I was so excited. On 28th of June, 24 other people and I went on a lorry to Mogadishu. We arrived just two hours before the Italian and UN flags went down and the new Somali flag went up. After a while, I got a job in Mogadishu as the regional vice-secretary for the party, but due to a lot of conflicts, I ended up being marginalized in the party. Finally, in December 1965, I was fed up and decided to leave Somali politics and Somalia. I first went to Germany and then, four years later, I arrived in Denmark. I got work as a welder and worked for many years.

Mohamed was not the only one to be unhappy about the development of the new republic. In 1961 there was a failed coup in the former British Protectorate. Liberalism and the reproduction of the economic structures inherited from colonialism dominated the first years of independence, and corruption and unemployment soon promoted dissatisfaction with the government. At the same time, Somalia grew increasingly dependent on international aid and became an ally of the USSR (Bradbury 1997, 5). In October 1969 the President was killed and the Chief Commander of the Army, Major General Siad Barre, seized power in a coup. Barre and the new Supreme Revolutionary Council dissolved the Consti- tution and the High Court, prohibited political activities, and announced a re-

97 gime of scientific socialism. Barre launched ambitious development campaigns, declared that ‘the evils of tribalism’ were buried, and prohibited all references to clan and genealogies (Barre 1971, 99; cf. Lewis 1994, 151-152). At the same time, Barre was increasingly manipulating clan affiliations and politics, concentrating power in the hands of members of the clan lineages of himself and of his close family. Still, the dream of the union of all the five Somali territories was alive, though wounded. In 1963 the Northern Frontier District became a part of Kenya, and in 1977 the inhabitants in French Somaliland voted in favour of creating their own independent nation state, Djibouti. The five-pointed star in the Somali flag had lost yet another point. Barre and other nationalists, however, did not give up and invaded the Ogaden area in Ethiopia in the same year. After a few weeks of military success, the USSR and Cuba sided with Ethiopia and Somalia suffered an overwhelming defeat. About 1.5 million people fled Ogaden, the majority to Somalia, but some also to the West. Barre subsequently turned to the US for military help and economic support, which the country received, due to its strategically important location during the Cold War. The militarization of Somalia took off further, resulting in a country full of weapons (Bradbury 1997, 6-7). After the military defeat in 1978, the first armed groups opposed to the regime of Barre emerged, the oppression of intellectuals was reinforced, and a large number of educated people left the country during the 1980s (Ali 1995).

Oil workers While the first Somali asylum seekers appeared already at the beginning of the 1970s, the migration of seafarers, traders and not least pastoral nomads continued after independence. New ways of migration emerged as well: Labour migration to the Gulf, the so-called muscle drain. Somalis have worked in the oil-industry in Saudi Arabia since the 1950s, and from the 1960s, when the oil boom took off, in the Middle East and the Arab peninsula as well. No reliable statistics exist on the number of Somali oil workers; estimates of the total at the beginning of the 1980s are as diverse as between 12,200 (Owen 1985) and 200,000 – 300,000 (Lewis 1994, 122). The variation in estimates is at least partly due to the fact that a large number of oil workers were and remain undocumented. It appears that this migration for the

98 great part has been undertaken by Somalis from the north-western part of the country, the former British Somaliland, and has been largely a male phenomenon (Lewis 1994, 178). The issue of remittance is an important feature of the Gulf migration: seafarers and oil workers transferred a portion of their earnings to their families. Re- mittances played (and continue to play) a key role in the Somali economy. Lewis (1994) estimates that the amount sent by workers in the Gulf countries was two to three times as large as the export earnings of the country. According to an ILO study, Somali workers earned US $700 million in the Gulf States in 1985 and remitted about US $280, that is 40 per cent of their earnings (quoted in Geshekter 1997, 77). From 1976, when the so-called franco valuta system was introduced, oil workers handed over their hard currency to Somali traders, who bought commodities to sell at the Somali market, and then later handed over an agreed amount of cash to the families of the migrant workers (Gundel 2002; Marchal, Mubarak, Buono, and Manzolillo 2000; Lewis 1994). In 1982 the system was banned, but the money transfers continued and evolved into the xawilaad, money transfer companies, which now transfer money from all over the world to the Somali-speaking region. Like the franco valuta system, the xawilaad is based on trust and occurs along clan lines (Hansen 2004, 12-13; Horst 2002; Fink-Nielsen and Kleist 2000, 121-127). This means that the remit- ter and receiver of money are identified by lineage affiliation details, eventually combined with telephone numbers, passport or other kinds of ID, making a money transfer agent joke that it was very hard work to write down all the lineage details. Somalis have gone to the Gulf and Arab countries for reasons other than to work in the oil industry, as the Gulf States also offered employment opportunities for educated Arabic-speaking Somalis. Somalis also migrated to the Gulf to join family members, to pursue religious education, to engage in business, and to get away from the political situation, which became increasingly tense from the end of the 1970s and 1980s onwards. However, as the Gulf Countries have not ratified the Refugee Convention, do not offer any protection and only very limited rights, going to the Gulf cannot be classified as asylum migration. The consequences of moving to the Gulf Countries became clear when the Gulf War broke out. Over two million migrant workers – Somali and other nationalities –

99 were expelled from Saudi Arabia, Kuwait and Iraq (Collinson 1999, 22). Some of them became asylum seekers in the West. Aisha Ahmed was one of them.

‘I became another person’: Aisha The mother of my father was born in Aden. My grandfather married my grandmother and they settled as nomads in Somalia. One day my father got a job in Jijiga as a medical assistant through his uncle. Jijiga was then a part of the British Protectorate [the reserved area], which is now a part of Ethiopia. An uncle took care of my father’s education, as he already spoke Arabic and knew the Koran. He received further education in a secondary school and then went to Aden for more. His uncle could do this because he was a part of the Camel Corps in the Protectorate. Af- ter he came back, my father got trained as an administrator. He worked in different places – the British transferred him all the time; they did not want him to settle. He retired in 1963.

My mother came from a town in the western part of the Protectorate; she is from a merchant family. Her uncle was a trader between Djibouti [the capital of the former French Somaliland and present Djibouti] and this town. My mother and her siblings were split between different families when my grandmother asked my mother to cook for them. My grandmother opened a shop selling sorghum. After that she bought cows and lambs, became a nomad and collected the whole family. My mother’s family became very famous, because they earned a lot of money from trade. One of my uncles went back and forth. He did trade from Iraq. He also became a politician and was elected as a Member of Parliament. When my father and mother married each other, my father worked in my mother’s hometown. I was born in the town district in a place which is now in Ethiopia. My mother had seven children: five girls and two boys. I am the oldest girl.

My father was famous for the fact that he was the first man to send three girls to school! When I was small, we settled in Hargeisa and I went to a Koran school from when I was four to six years old. When I was six years old, I knew the Koran by heart and I could speak Arabic. My sister and I then went to the only girls’ school at that time. We got special permission and started directly in the second grade. Later one of my other sisters also went to that school. Somebody teased my father about his girls going to school, but he always defended us; he always thought that it was better to be educated.

After school I first worked as a clerk. From 1964-69 I worked in an office. I did all kinds of work. In 1969 I married my husband, who was in the administration. I then made the big mistake of giving up my work and went into the administration myself. In 1972 we moved to Moga- dishu. I became a professional secretary and studied public administration, but I never pursued further education as much as I wanted to. In 1977 we went to Kuwait. We could not stay for political reasons. An- other reason was that I wanted my girls to go to school. I worked in different offices dealing with administration and as a secretary. I also worked in different embassies. In Kuwait, women’s work also proved

100 more promising than men’s work, so I had to work and I worked all the time.

In 1990 I came to Denmark as a refugee, due to the Gulf war. My husband stayed. To get asylum I had to teach my children their clan lineage, which I had never taught them before. We needed to be able to give an account of our lineage in order to get asylum. It was horrible to come to Denmark. I was always a very active person and here I became a passive person, another person. I often miss a lot of my inner person, my old self. My friends and family have noticed it too. ‘What has happened to you?, they ask. I worry so much. When I told my social worker about my career, she did not believe me. She did not even bother to read my file.

In many ways, Aisha Ahmed is a remarkable woman. She was among the first girls in the British Protectorate to pursue education and later a professional career. This opportunity must be seen in relation to her family background, where her father was an administrator in the Protectorate and her mother part of a wealthy merchant family. As Aisha explains, it was only children whose parents were in the government or trade who received education, for two reasons: because their parents could afford it, and because they understood the value of secular education. In this way, Aisha was one among very few girls, as secular education was not very widespread or valued during colonization or after independence. Still, she did not manage to find work in Denmark, in spite of a lifelong career and fluency in several languages. Aisha Ahmed’s life history bears witness to the impact of colonization, dictatorship, the Gulf War, asylum in Denmark – and, not least, the civil war in Somalia, making it impossible to go back when she and her children were expelled from Kuwait.

Civil war In 1988 the Somali and Ethiopian governments signed a peace accord to end hostility between the two countries. This recognized Ethiopian control over the Haud areas, meaning that a larger number of Somali refugees were forced to go to Somalia. Furthermore, the Somali National Movement (SNM), an Isaaq- based resistance movement founded in London in 1981 in opposition to Barre’s regime, lost protection (Lewis 1994, 177-219; Silanyo 1986). The SNM in turn attacked and briefly captured the biggest cities in the north-western part of So- malia, resulting in the government troops responding through full-scale assaults on the local population (Bradbury 1997, 11). Civil war was a reality and more

101 than 600,000 people fled to Ethiopia (UNDP 2001, 214; Republic of Somali- land 1994). During the next two years, civil war spread to the rest of the country and in the beginning of 1991, Siad Barre was ousted from Mogadishu. Rebels, clan- based resistance movements and drought led to huge humanitarian crises in the southern and central regions of Somalia. It is estimated that by the end of 1992, between 370.000 and 500.000 people had lost their lives due to violence and hunger, and an even larger number had fled the country: the total is estimated to vary between 800,000 (UNDP 2001, 59) and up to 1.5 million (Bradbury 1997, 1). The large majority went to isolated refugee camps placed in the deserts of Ethiopia, Kenya, Yemen, and Djibouti. As many as two million Somalis were internally displaced in Somalia in 1992, although the number ‘dropped’ to 500,000 in 1994 and 200,000 in 1997 (UNDP 2001, 60; UNHCR 1998, table 1). There are a number of explanations for the tragic civil war, where nomadism and clannism hold a central and contested place (e.g. Farah 2000; Samatar 1991). In the words of Farah, there is no simple reason for the civil war:

If you are asking, what are the features which brought forth the civil war in Somalia, the list is endless: centuries-old injustices; decades-old political feuds; Siyad’s tyrannical state, and its indifference to the ordinary people’s genuine grievances; the nature of post-colonial set-ups. Suffice it to say that these may be some of the reasons why high principles have been pushed aside and why, in their place, people have begun putting their faith in the pornography of a warlord’s rhetoric, holding forth and reciting chapters and verses of clan mythography. Civil wars do not wait for reasons. They erupt, they happen, and may the rest be damned (Farah 2000, 45).

For Farah, then, the ‘clan mythography’ was invoked, while the background of the civil war is a much more complicated matter. I am not going to go further into this debate here; the point is that no matter whether clannism was at the heart of the matter of the civil war, this ‘master story’, as I termed it, was and had been mobilized for a long time with real and horrible consequences. During the civil war, people’s clans might have been decisive in the encounters with armed militias where the right or wrong clan affiliation might result in death, torture, rape, robbery, or survival and security. In consequence, large numbers of people fled to areas where their clan families dominated and they would be

102 safer40. These areas, which they might never have visited before, now became their ‘area of origin’, pointing to a tension between lineage and regional affiliation. This is illustrated with the Somali couplet U dhashay – Ku dhashay, which can be translated as “‘born to (a family/clan) and ‘born in (a place/region)’” (Barnes 2006, 487). I will return to this tension in more detail in chapters six and seven. While the central and southern regions of Somalia dissolved into anarchy and civil war, not much attention was directed towards the north-western part of the country, the former British Protectorate. In May 1991, the victorious SNM declared the Republic of Somaliland independent with reference to the four days of independence before unification with the UN Trusteeship of Somalia in 1960 (Republic of Somaliland 1996; Lewis 1994, 215). A reconciliation conference followed, and a parliamentary system was introduced. In 1993 Mohamed Egal, the former Minister of Local Government of the British Protectorate as well as the first Somali Prime Minister in the Republic of Somalia, was elected as President. Though disturbed by periods of inter-clan fighting in 1994 and 1998, there has been relative peace in Somaliland since 1996. Likewise, the north- eastern part of Somalia declared itself autonomous in 1998 under the name of Puntland, but in opposition to Somaliland, Puntland considers itself part of a future Somali federal state. The rest of Somalia has remained highly unstable. The beginning of the civil war in Somalia coincided with the gradual termi- nation of the Cold War. Somalia lost strategic value for the US and other western countries, which gradually withdrew economic support, credits and aid programmes. With the outbreak of civil war in Yugoslavia and the Gulf War, the attention of the world tended to drift away from Somalia. The US-led UN inter- vention called Operation Restore Hope was a total failure and, since 1991, So- malia has effectively been without a central government (Marchal, Mubarak, Buono, and Manzolillo 2000; Bradbury 1997, 16-17). A number of peace talks and conferences have taken place, but so far no lasting peace agreement has been reached. In 2000 a three-year Transitional National Assembly was established (International Crisis Group 2004), and in October 2004 a Transitional Federal Government (TFG) was elected in Nairobi with former President of Puntland, Abdullahi Yusuf, as President. The TFG relocated to Jowhar in July

40 The Somali-Danish writer Abdirizak Osman (1996) has made a moving and terri- fying account of the civil war in the novel In the Name of our Fathers.

103 2005, as Mogadishu still proved too insecure (UNOCHA 2006). The Parliament had its first session in Baidoa in February 2006. The latest news of political developments (November 2006) is that the Union of Islamic Courts (UIC) took over Mogadishu in June 2006 and further fights for military control of southern and central provinces, spurring a new wave of Somalis fleeing from war and conflict.

Refugees With an estimated 429,000 Somali refugees in 2003, Somalis were ranked by UNHCR as the fifth largest refugee population in the world after Afghanis, Bu- rundians, Sudanese and Angolans41; with a total of 101,765 applications, So- mali asylum applicants constituted the sixth largest group in the European Un- ion in the period from 1994 to 2003 (UNHCR 2005, C.4). While the numbers in the tables below are not necessarily accurate (as statistics of refugees rarely are), they still give an impression of the fluctuations in numbers of refugees from Somalia42.

Table 1 Estimated number of Somali refugees in neighbouring countries of asylum, 1990 – 2003 Year 1990 1992 1994 1996 1998 2000 2002 2003

Number 450, 000 780,000 536,000 500,000 420,000 320,000 294,000 267,000

Source: (UNHCR 2005; 1995; UNDP 2001). The countries included are Kenya, Ethiopia, Yemen and Djibouti.

41 Palestinians are not included in this ranking, as they are covered by another UN mandate (UNHCR 2003c). 42 The number of Somali refugees does not necessarily mirror the actual number of Somalis living outside the countries. First of all, this is an estimate of refugees and people of concern to the UNHCR and thus excludes those who are not categorized as such, including undocumented migrants. Secondly, those registered as refugees might hold several identity cards and might therefore be counted several times. It should also be noted that since the great majority of Gulf countries have not signed the different refugee conventions and protocols, Somali nationals in refugee-like situations living in these countries do not appear in the UNHCR statistics, even if the number of Somalis living there might be substantial.

104 Table 2 Estimated number of Somali refugees in selected western countries of asylum, 1991-2003. Country 1991 1992 1994 1996 1998 2000 2002 2003 US 700 2,400 8,800 18,200 24,400 34,000 28,700 25,400 UK 2,100 4,300 9,000 14,800 19,200 26,900 33,100 36,100 Netherlands - 5,000 10,200 15,800 17,800 19,400 15,700 14,000 Denmark 700 1,400 3,500 6,400 9,000 10,000 9,600 8,600 Sweden 3,000 4,400 7,100 7,500 7,600 7,300 4,500 4,300 Canada - - 17,800 11,400 7,100 5,200 5,500 5,100 Italy 700 700 10,700 10,000 1,100 400 100 100 Total 7,200 18,200 67,100 84,100 86,200 103,200 97,200 93,600

Source: (2002; UNHCR 2005). - no numbers. The numbers are estimates. Not including family re-unified.

As Tables 1 and 2 show, the number of refugees has changed over the years. On the one hand, the decrease in the number of refugees in the neighbouring countries reflects the fact that substantial numbers of Somali refugees have undertaken so-called voluntary repatriation, often UNHCR-assisted. The number of registered Somali refugees in Ethiopia fell by about 210,000 people from 1997 to 2002 (UNHCR 2003a; UNHCR 1998). In 1997 alone, 50,000 Somalis repatriated from Ethiopia (UNHCR 1998, table 6) to the unrecognized Republic of Somaliland; in 2002, the number of returnees was estimated to be close to 30,000 (UNHCR 2003a). In Somaliland, large resettlement areas have formed new neighbourhoods on the outskirts of the largest towns. Generally these areas are very poor and the resources allocated to the reintegration of returnees are extremely limited. On the other hand, the decrease in the number of refugees in neighbouring countries also reflects the fact that substantial numbers of Somali asylum seekers have naturalized or moved further on, for instance to western countries. Still, Table 2 does not reflect the actual number of Somalis in western countries, which is unknown. The UNHCR does not include family re-unified and Somalis do not figure separately in the census in, for example, the UK, where the estimate of Somalis in 2002 and 2003 varies between 95,000 and 250,000 (Harris 2004, 33). What can be settled, however, is that the number of recog-

105 nized Somali refugees (as well as other Somalis) in the West has risen dramatically from 1991 to 2000, when, with the exception of the UK, it started to drop again. Following this table, the number of refugees in the neighbouring countries has never been close to the number of refugees in the West, though this picture probably would look different, if family re-unified were included. Omar Said is one of the many refugees who spent some years in Kenya before he at- tained asylum in the West.

‘The war never stopped’: Omar I grew up on a farm outside a town in the southern part of Somalia. My father was a farmer. After high school I found work at a fish factory as a sales assistant. Danida [Danish International Development Assistance] sponsored the factory and their field manager was Danish. He was the first Dane I ever met. After two years, the civil war started and General Aideed, a warlord, came closer to our area. His clan and our clan are different and war broke out between the two clans. Aideed captured Kis- mayo, the biggest town, and the war went on and on.

So we fled to Kenya. All the time I was thinking that the war would stop, that Somalia would be normal again, that there would be a system. I was waiting for two - three years. We had this farm. So we took all our farming tools with us to Kenya, all the time thinking that we would return. It got more and more dangerous and the police captured us all the time. We had to pay money to them each night otherwise … It got worse and worse and there was no work. I had this friend, who had lived for four years in Stockholm. He recommended that I go there, but it was difficult with my family. In the end, my father and I agreed to sell our tractor to fund an opportunity for me. So I bought a false passport in Nai- robi, because in Kenya you can get everything if you have money. I would have liked to go to Sweden, because my friend lives there, but I was in transit in Copenhagen and I had to ask for asylum here. After a few months in an asylum centre in Copenhagen I was sent to a very small island. We were the first Africans there and sometimes the local people stopped their cars to look at us. After four years I came back to Copenhagen.

My father and I had a dream to return and grow our farm. We didn’t succeed. My father died with his dreams a few years ago, my mother and sisters still live in a refugee camp in Kenya. One of my sisters has a small shop there; when I can afford it, I send money to her and my mother.

That Omar Said should come to Denmark was not planned. He was caught up in the civil war as a young man, driven out of the southern part of Somalia, and ended up in a refugee camp, like thousands of other Somalis. Selling the family tractor gave him the possibility of reaching the West through the acquisition of

106 a false Swedish passport. But due to the Dublin Convention, which dictates that asylum must be applied for in the first EU country the asylum seeker enters (Udlændingestyrelsen 2004a; 2004b), Omar did not make it to Sweden. Desti- nation Denmark was a coincidence – or rather a consequence of the asylum legislation in the EU (Lisborg and Lisborg 2003). In this way, Omar’s story is identical to many other cases. Table 3 of top-ten western countries in terms of Somali asylum applications might thus reflect a certain contingency insofar as the countries where asylum applications are lodged are not necessarily the destination that the asylum seeker hoped to reach.

Table 3 Number of Somali asylum applications in selected western countries, 1993- 2002 Country 1993 1994 1996 1998 2000 2002 1993-2002

UK* 1.465 1,840 1,780 4,685 5,020 6,680 41,625

Netherlands 4.330 5,393 1,140 2,775 2,110 533 25,688

US* 131 114 1,140 2,268 2,364 538 13,532

Canada 2.468 1,989 962 653 753 388 10,887

Denmark* 1.469 2,011 1,514 662 747 391 10,688

Germany* 1.577 896 1,152 1,078 398 203 8,251

Switzerland 2.295 881 700 610 470 387 7,591

Norway 259 251 180 938 910 1.534 7,233

Sweden 733 934 434 228 260 1.107 5,743

France 1,077 217 85 91 109 115 2,077

Source: (UNHCR 2003a). *Figures for Denmark include asylum applications lodged at embassies abroad. Figures for Germany include re-opened cases. Figures for the US and UK refer to cases.

As Table 3 shows, Somalis continued to ask for asylum throughout the 1990s, both in Britain as well as in the Netherlands, Scandinavia, North America and many other western countries. In a kind of parallel movement, family reunification with relatives in the West became another way out of Somalia or the refu-

107 gee camps43. However, as western asylum legislation places severe restrictions on the possibilities of third-country citizens entering Europe, asylum migration to the West is a very costly – and often very dangerous – affair (Carder 2003; Lisborg and Lisborg 2003). The continuing tightening of the EU asylum policies has led some researchers to characterize the EU as a ‘non-entrée regime’ (Kibreab 1999; Chimni 1998). The European non-entrée regime has several implications. One is that the Somalis and other asylum seekers need to have financial resources to be able to go to the West at all. In the case of Omar, this meant the family had to sell their tractor. Other migrants who do not have such opportunities, might get help from relatives already in the West or in the Gulf, so that at least one family member can go (Carder 2003). It goes without saying that the decisions concerning who can go might be very tough to carry out. In many cases, families are separated by the civil war and sudden flight, but in others, separation seems to occur as families try to seek protection in the West, but cannot afford to do it at the same time (Carder 2003; Fink-Nielsen and Kleist 2000). In this strategy of ‘scattered family re-unification’, one family member goes, especially the mother it seems, maybe accompanied by some or all of her children. It is then hoped to attain family reunification with the rest of the family later. This plan, however, does not always work out, especially since asylum and family reunification legislation have tightened throughout Europe – and not least in Denmark. A third implication of European asylum legislation is that the arrival destination of asylum seekers might be contingent – or rather, governed and determined by other actors and factors than the wishes and hopes of the asylum seekers themselves, as in the case of Omar, who wanted to go to Stockholm, but ended up in Denmark. Apart from luck and coincidence, the destination reached seems to depend on which airline and destination the human smuggler asserts is the safest, and whether the route implies transit in a so-called safe country, meaning that asylum seekers in principle have to ask for asylum there. Finally, the price paid to the smuggler plays a role as well: different routes and destinations have different prices (Carder 2003; Lisborg and Lisborg 2003; Farah

43 Another way of securing asylum to the West is through resettlement. Resettle- ment, however, is limited in numbers and only included 1.575 persons in 2002 (UNHCR 2003b), whereas the number of Somali citizens granted asylum in West- ern countries the same year amounted to 8.380 (UNHCR 2003a).

108 2000). The scattered and contingent destinations mean that many Somali families have been dispersed throughout the world, leading some scholars to characterize the dispersed Somali population as a diaspora (e.g. Horst 2003; Montclos 2003; Farah 2000; McGown 1999), as we saw it in chapter two.

Transnational connections As a number of studies show, many Somalis in the West and elsewhere support their family members in Somalia or the neighbouring countries with regular or ad hoc remittances, making crucial contributions to the economies of the households (Maimbo 2006; UNOCHA 2005; Hansen 2004; Horst 2004; 2003; Sørensen 2004; UNDP Somalia and World Bank 2003; Gundel 2002; UNDP 2001). In a recent UN report (UNOCHA 2005), it is estimated that a quarter to one third of the population relies mainly on remittances sent from the West or Arab countries through xawilaad agencies, but that it is mostly urban households – families with better education – that are favoured. Horst, however, notes that money was both sent to and from the Dadaab refugee camp in Kenya, and estimates that at least ten to fifteen per cent of the camp population benefited from remittances (2003, 159). In October 2003 I interviewed Mr. Mohamed Said Duale, the founder and chairman of Dahabshiil Pvt. Ltd. Co. Dahashiil, meaning ‘frying gold, the biggest’ money transfer company in East Africa; it has its headquarter in Hargeisa and has branches in 33 countries, including two offices in Denmark (Dahabshiil 2002). Apart from Mr. Duale, I interviewed four other Dahabshiil agents in Somaliland and Copenhagen, respectively. Mr. Duale estimated that about 800.000 USD was remitted to Somalia (including Somaliland) each day to about 2.000 daily receivers. On an annual basis, these remittances amount to almost 300 million USD remitted through Dahabshiil alone. The total amount of remittances to Somalia in 2004 was estimated as between 750 million and one billion USD – in contrast to the 125 million USD which Somalia received in aid in the same year – emphasizing the role of remittances as “a life line to survival” (UNOCHA 2005). Remittances are both sent to relatives and for business related purposes. According to other agents in Somaliland and Denmark, the most usual transfer is between 50 and 200 USD on a monthly basis, but it can be much larger in cases of business or investment, for instance in land or property.

109 The Somali-Danish informants remitted money in the same volume. About half of the informants sent money to their or their spouse’s relatives on a regular basis, and all of them donated to collections or remitted money on an ad hoc basis in times of special or urgent need. Likewise, all of them had relatives living in at least two and three or more continents – including Europe and Africa, but also North America, Asia and Australia. Furthermore, for almost all of them44, some or all of their siblings as well as one or both of their parents were scattered in different countries. Mohamed Abshir, Aisha Ahmed and Omar Said all had close relatives living in other countries and continents: children, ex- spouses, siblings, cousins, aunts, uncles or parents. But while Mohamed and Aisha had reached the age of retirement and were no longer regular remitters themselves (though they still had care responsibilities), Omar found himself in a situation where he provided for both his own family and his mother. When I first interviewed Omar Said, he was divorced and lived in a rented flat with his children staying over from time to time. Furthermore, Omar supported his mother in Nairobi with 50 to 100 USD every month. He had also given an elderly sister a substantial sum of money a few years previously, in order for her to start a small shop and become self-providing. Two siblings were living in a refugee camp, a younger brother lived with the mother, a brother was living in Somalia, and a sister in Canada. Omar also had cousins in Norway, the UK, Canada, and the US as well as uncles, aunts and cousins in Saudi Arabia, Pakistan, Ethiopia, and Somalia. Most of the relatives had migrated after 1989. Omar was speaking to his mother once or twice a month; the contact with his other relatives varied. While his mother was the only relative whom Omar supported on a regular basis, he and the other relatives in the West would help the others if they were in need. They had also collected a larger amount of money to finance an educational stay for the younger brother in a European country, but Omar emphasized that he wanted the brother to return to Kenya to help their mother. Omar explained:

44 Among the informants included in this study, the only ones with all their siblings and both parents in Denmark were in families where one of the parents had obtained family re-unification for the entire nuclear family.

110 My mother lives in Nairobi and as I said, she is completely reliant on me. I send her 50 or 70 dollars every month, or 100 dollars if I am able to. So … it’s not only her, there are more people living and eating with her. It is the only way people can survive; there is no work, there no nothing. I have told my siblings that I cannot send 100 dollars to each of them every month, but sometimes we have fundraising for my family who live in different places […] I have to support them if I can and since I have work, I am remitting a bit more money than I used to do.

Omar thus lived in a transnational family where the relatives living in western countries helped out the ones in need in Africa. As the eldest son and one of the only children living in a western, high-currency country, his sense of responsibility towards his relatives was marked. Halima Abdi, a woman in her mid 20s and the last informant I present in this chapter, also had a transnational family and transnational responsibilities, though her story was very different.

‘Some of us were born here, so we think it’s pretty normal’: Halima My father came first … he was seventeen when he left Somalia and worked in different places in Yemen. His elder brother worked there, so he was like, ‘I want to go see the world; I don’t want to be in Somalia any longer’. They were many young people going to Yemen or other places and then they heard about different places in Europe and that was how he heard about Denmark. He was totally international as a young man; he has been to China, everywhere! Then he came to Denmark and five years later he married my mother. He was visiting his hometown in Somalia and saw this very tall, very beautiful girl, who tossed her hair … and then they married and had kids [in Denmark].

When I was ten years old, we had to learn Somali and it was a completely foreign language. Suddenly my mother thought that it was about time we learnt it, it was just something we had to do, as a kind of pre- caution, like it was also my language. To be quite honest, in some ways I am not Somali. On the outside, maybe, but not inside. There are some things I will never understand and there are some things I try to understand. You cannot say that I have a Danish background, but an in- between background. I am more like in-between. You know, some of us were born here, so we think it is all pretty normal. It is like we can see Denmark with some other eyes and function like a connecting link between our own people and … Danish society that we know really well.

I was in Somalia for the first time in 1999 and I didn’t say anything for two weeks. There were so many things I had to chew on and that I did not know about. The next time was better. So I think the third time will be even better, because then you know it in a completely different way. I have been thinking about doing some mountain hiking with a backpack and some decent hiking shoes and then just wandering around. I just have to check out that there is no war.

111 My parents moved to the UK a couple of years ago. Their family lives there. Me and my siblings didn’t really have time to … everybody min- ded their own business, high school and so on, and then we thought, that they are getting older … and that they will need their family. Everybody had left home, so we thought that they should stay somewhere where they felt a bit more at home, like Somalia: they actually missed it. What made the whole thing roll is that my younger sister said ‘I don’t want to go to school here, I want to study in the UK, I want an international education’. And then all of a sudden, she moved and then they all moved. It is only one of my brothers and I who are still here. You know, all the Somalis go to the UK now, but I am not interested. I will go on vacation and see my family, but not emigrate.

Halima Abdi is the only informant who was born and has lived her whole life in Denmark; she was not a migrant, but thought of herself as something ‘in- between’ and ‘not really’ Somali. In contrast to most of the other informants, Halima had not established her own family at the time of the interview and neither her parents nor siblings were in need of economic support. Still, Halima had a transnational family and responsibilities with relatives scattered in no less than fourteen countries. Halima’s parents and most of her siblings were among the many Somali-Danes who had moved to the UK where their relatives had lived for two generations. Halima also had a sister in Saudi Arabia with whom she e-mailed every day, and cousins or uncles and aunts in Canada, the US, France, Norway, Sweden, Dubai, Russia, Australia, Somalia, Kenya and Eri- trea. The relatives in Dubai were wealthy, while the ones in Europe, Kenya and Eritrea were doing okay, Halima explained. The cousin in Russia, however, was in economic trouble and had called two years ago with the message saying, ‘I have married, but I have not found any work yet … you have to help me until I have found a solution’. Halima and two of her sisters decided to help out and started to remit about 100 USD a month to the cousin. Furthermore, in times of crisis in Somaliland, Halima Abdi would get plenty of phone calls appealing for economic help, giving her, she said, a terrible headache. As Halima’s example shows, the transnational flows of money, attention, and care were not only directed towards Europe, North America and Africa, but also to other regions. Furthermore, while Omar’s responsibilities seemed to be connected to the fact that only he and a sister were living in the West, Halima’s case suggests that transnational responsibilities can also be invoked (cf. Horst 2003). They were both engaged in transnational social fields of “multiple interlocking networks of social relations” (Levitt and Schiller 2004, 1009), consti-

112 tuted by obligations and care. Furthermore Halima’s example points to the fact that many Somalis continue to be on the move.

Continued migrations As time has passed, many Somalis have obtained citizenship in their western countries of asylum or residence. Contrary to the belief that naturalization mir- rors – or should mirror – a unitary orientation towards the new country of citizenship, a EU or North American passport eases the possibility for continued mobility. When Somalis return to mainly Somaliland for shorter or longer periods to buy land, build a house and maybe try to make a living, the great majority seem to have acquired western passports. Travelling with a western passport not only makes mobility easier – and legal – it also guarantees the opportunity to return to a western country if civil war breaks out again (Fink-Nielsen, Han- sen, and Kleist 2004). During July and August, thousands of visiting ‘western’ Somalilanders loaded with cash, presents, and video cameras go on holiday to Somaliland. While some of these people might end up staying for a while, the majority go back to their western countries of residence, or maybe back and forth (Hansen 2006). It seems like it is especially the men who (wish to) settle in Somaliland, often backed by financial support from their wives or kids in the West (Fink-Nielsen, Hansen, and Kleist 2004). Similarly, there is an ongoing movement of Somali-Europeans to Britain, especially from the Netherlands, Sweden and Denmark – like Halima’s parents and siblings. These Somali-Dutchmen, Somali-Swedes and Somali-Danes have obtained EU citizenship and now move on to London, Birmingham, Leicester, or Cardiff (Nielsen 2004). In case of inter-European migration, women are said to be at least as eager as the men to go. This ‘second family reunification’, as one Somali-British man characterized the movement, adds one more wave to the history of Somali migration where the mobility of those with western citizenship is eased, while the mobility of those without the desired passports has become even more circumscribed, dangerous and more costly. Furthermore, it can be seen as a move where Somali families, who might have been separated not only by the civil war, but also by the western asylum regimes, try to unite with their close or more distant families. Likewise these migrants might be moving on with friends to obtain international education, to find work, or to try

113 their luck in an environment that is perceived to be more accommodating than Scandinavia or the Netherlands. In the words of many Somalis and some researchers, the ongoing movements are a result of ‘the nomadic spirit’, a wish to keep on looking for ‘greener pas- tures’ (Nielsen 2004; Horst 2003). To refer to the ‘master story’ of Somali nomadism underlines the continuity in the history of Somali migration as well as in individual lives where mobility might have been an ongoing life condition – whether voluntary or not. Furthermore, identifying oneself or others with ‘the nomads’ can be seen as a way of emphasizing cultural continuity and of mobilizing images of being adventurous, tough and independent, rather than marginalized, displaced and helpless, as Somali refugees are often portrayed in their new countries of residence.

Chapter resume: Scattered dreams and changing mobility Roughly speaking, the history of the Somali-speaking region can be told as a history of changing geopolitical positions during colonialism, the Cold War, and the post Cold War ‘new world-order’, where the strategic location of the Somali speaking areas grew in importance and then declined. It can be told as a history of nationalist dreams of unity followed by disintegration, of manipulation of the Somali genealogical lineage clan system, and of massive displace- ments. The West still dominates and frames Somali migration, despite the dissolution of the colonial empires, the end of the Cold War, and the appearance of the Schengen agreement in the EU. Civil war and continuing unrest continue to separate people, making mobile livelihoods an option for many people, whether such moves and livelihoods are welcomed or not. Maps have been drawn and redrawn; borders have been created, opened, closed, or moved. Somali history can also be told as a history of different kinds of migration and mobility. The stories of Mohamed, Aisha, Omar, and Halima make it clear that there are both continuities and changes in the migration history of Somali nationals – including those who later became western citizens. Their narratives show that the categories of political and economic migrants, or ‘refugees’ and ‘immigrants’, might be more fluid than is often thought. Mohamed Abshir lived in different European countries before coming to Denmark as a labour migrant, but has, in spite of living here for more than 30 years, continued to have close

114 relations to Somaliland. Aisha Ahmed left Somalia as a professional, a labour migrant, but entered Denmark as an asylum seeker. Omar Said both fled Soma- lia and entered Denmark because of the civil war, while Halima Abdi was born in Denmark and considered Somalia a place to go on vacation. Finally, I have shown how the global dispersion of Somalis has been followed by transnational connections, ongoing migrations and, not least, remittances. At least a quarter of the population in Somalia is dependent on remittances, and almost all the Somali-Danish informants sent money on a regular or ad hoc basis. I discussed how the ‘master story’ and practices of clan in particular are often invoked to understand the tragedies of the civil war, and how the image of nomads is called upon to understand the phenomenon of transnationalism and migration, thereby accentuating the continuity of Somali mobility and life situations. During the chapter, I have thus provided the international historical and political context for Somali migration and transnational practices. I now turn to the Danish context, where I introduce the reception and settlement process and go into detail about how the informants articulated their lives in Denmark.

115 116 Chapter 5 Ambivalent encounters. Life in Denmark

I have two visions of Denmark: one that knows me and who isn’t at all afraid of me – and then the one that doesn’t know me (Saphia Mansoor, 2003).

Somalis started to come to Denmark in larger numbers from the middle of 1990s to search refuge from the Somali civil war or to be united with their families. In this chapter, I present the context of reception surrounding their arrival and settlement. As I presented the Somali migration history, embedded in both local and global developments, so do I start this chapter by applying a specific historical gaze, retelling (parts of) Danish history as a history of migration. I end this account with a specific focus on Somali migration to Denmark. I then turn to how Somalis in Denmark articulate their life in Denmark and encounters with Danish society. As Saphia Mansoor’s words show us, these encounters are often of an ambivalent nature. Finally, I conclude the chapter by suggesting that this ambivalence can be understood in terms of the figure of ‘the stranger’ and recognition of difference.

From multi-national state to small nation-state Europe has always been a region of migration and Denmark has been no excep- tion45. Situated at the mouth of the Baltic and at the very interface between western continental Europe and the Scandinavian peninsula, Denmark has been a country en route of traders, businessmen, craftsmen, missionaries and travel- lers. Indeed, according to historians Scocozza and Jensen, historians and ar- chaeologists used to agree that most of Denmark’s cultural development was due to migration; more recently, however, the idea that Danish culture developed independently of immigrants and conquerors has evolved (1994, 393). In a global perspective, the formation of global empires by European powers and the

45 When nothing else is specified, I have based the following section on historians Bent Østergaard (1983), and Benito Scocozza and Grethe Jensen (1994), while the international perspective especially refers to Wimmer and Schiller (2002).

117 overseas trade between Europe, Africa, the Americas and Asia provided much of the wealth which unleashed the industrial revolution in the 18th and 19th centuries (Castles and Miller 2003, 56ff). Migration – in the shape of slavery and European conquerors and settlers – was at the heart of this development. Fur- thermore, the many wars throughout European history have resulted in displaced and persecuted populations, of which some came to Denmark to find refuge. Foreigners have also been invited to Denmark with the purpose of supporting the Danish economy, such as Dutch and German farmers and craftsmen, who were summoned to share their knowledge of farming and construction between the 16th and 18th centuries. During this period, the Danish nobility, elite and royal family were part of a cosmopolitan network and many were emigrants or the descendants of emigrants. In the era of patriotism, it was loyalty towards the state – not the nation or the nationality – that was considered important. The history of Denmark can be described as a movement from a medium- sized European ‘multi-national’ state to a small nation-state. Over different periods of time, Denmark has had possessions in Scandinavia, the Baltic, the duchies of Schleswig and Holstein in Northern Germany, the North Atlantic islands of Greenland, Iceland and the Faeroe Islands, and a few small overseas colonies in Africa, the West Indies and Asia. The country has, however, subsequently lost, sold or been forced to cede these possessions over the centuries46 – especially in 1815 and 1864 where Denmark suffered ill-fated fortunes of war. With these defeats, Denmark definitively became a small state and turned its attention to local nation-building, paraphrased in the famous saying by Enrico Dalgas, the son of a Danish-Italian businessman: ‘What is outwardly lost, must be inwardly won’. The consolidation of the Danish nation-state with a parliamentary system, granted by the Danish constitution in 1849, the development of Danish nationalism, the growth of associations and the cooperative movement thus collided with the political defeat and the ensuing relative homogeneity of the population of ‘mainland’ Denmark (as the country still possessed the colonies of Iceland, Greenland and the Faeroe Islands). Denmark’s development into a small nation-state took place in a time of intensive globalization and nation state building. La belle époque, the period from the 1870s to World War One, was generally one of widespread colonial and la-

46 Greenland and the Faeroe Islands have self-governments, but are still part of the Danish Kingdom.

118 bour migration, with open European borders and few demands for passports or visas (Schiller and Wimmer 2002, 312-312; cf. Østergaard 1983, 199). Den- mark, however, had just lost land and concentrated on domestic politics. Still, this was a period of migration and industrialization in Denmark as well. Swed- ish seasonal workers constituted a cheap workforce at the end of the 19th century, and many settled. They were succeeded by female Polish workers at the beginning of the 20th century, the so-called ‘beet girls’, who laboured in the sugar beet fields on the islands of Lolland and Falster. Jews also arrived in Denmark, escaping the pogroms of the Russian Tsar’s regime at the beginning of the 20th century. At the same time, there was massive emigration from Europe to (especially) the Americas. Due to poverty, hunger and dreams of ad- venture or ownership of one’s own land, made possible with the 1862 Home- stead Act, millions of people emigrated to the US. Again, Denmark was no exception. It is estimated that about ten per cent of the Danish population emigrated between 1866 and 1900, mostly in the 1880s (Scocozza and Jensen 1994, 254). Denmark was indeed a country of both emigration and immigration. After World War One, labour migration tended to decrease and borders to close. This was a period characterized by disrupted economies and political crises. Denmark, neighbour to the ever more threatening German Reich, controlled the borders strictly, and many of the refugees who arrived during the 1930s were only permitted temporary leave to remain, living under poor conditions, or were sent back. Still, as is well known, a number of the refugees did save their own lives (Østergaard 1983, 199ff). The horrors of World War Two led to the creation of the UNHCR and the 1951 Geneva Convention, which aimed at pro- tecting refugees from persecution and securing their rights. The Convention, which only applied to events which had taken place before 195147, was later extended with the 1967 Protocol. Denmark was among the first countries to sign the Convention. Both the Geneva Convention and the Protocol secure the rights of individuals fleeing from individual political persecution, thereby reflecting the nature and perceptions of conflicts in the Cold War. There were thought to be three lasting solutions to refugee situations: exile in the country of first asylum, resettlement in a third country, and voluntary repatriation. While repatria-

47 The Convention had two formulations, making it possible to make it apply to events occurring in Europe only, or in Europe and elsewhere (UNHCR 1951, Arti- cle 1, B).

119 tion was considered the ideal solution, resettlement was most practised during the Cold War (Chimni 1999). During this period, Denmark mainly received refugees from the communist countries, and the arrival of about 1200 refugees from Hungary in 1956 led to the creation of the Danish Refugee Council (Dansk Flygtningehjælp 1996). Though refugees and refugees’ rights were emphasized in the post-war period, the bipolarity of the Cold War also further threw suspicion on migration (Schiller and Wimmer 2002, 318). Borders were strictly policed and only refugees from communism were permitted to resettle in the West. The Cold War, however, was also a period of labour migration within and to Europe. The economic boom of the 1950s and 1960s resulted in a lack of labour in western Europe. The demand for workers led to the UK, the Netherlands and France turning to their (former) colonial populations, while West Germany started to recruit guest workers. In Denmark, the economic boom started in the late 1950s and by the early 1960s the unemployment level had dropped to such a low level that the export-oriented industries started to worry about the lack of manpower – in spite of more and more women joining the labour market and the so-called ‘Greenlandic manpower reserve’. The Danish Aliens’ Consolidation Act of 1952 made it relatively easy to enter and obtain permission to work in Den- mark. The mid-1960s were also the time when the level of immigration slowly began to exceed emigration (Matthiessen 2000, 41), and new groups of migrants started to come to Denmark.

Guest workers and refugees The high performance of the economy in the 1950s and 1960s resulted in a further development of the Danish welfare state, based on high taxes, solidarity and universal social welfare. As Wimmer and Schiller point out, the development of the European welfare state48 extended the idea of ‘the people’ as not only comprising nation and citizenry, but also as a group of solidarity, circumscribed by “state boundaries [marking] the limitation of access to these privileges” (2002, 318). This was the overall context of the country in which the guest workers arrived.

48 I do not touch upon the discussion about different kinds of welfare states here. See for instance Larsen (2003) and Abrahamson (1990).

120 The majority of guest workers were young male workers from Southern Europe, Yugoslavia, Turkey and, from the beginning of the 1970s, Pakistan49. Many labour migrants came to Denmark to look for work, especially after the decline in the German economy in 1967 and the tightening of the immigration legislation in Germany and Sweden the year after. Their purpose was not necessarily to stay in Denmark. They were seen as temporary workers, expected to move on when their manpower was no longer needed: hence the term gæstear- bejdere (guest workers) or fremmedarbejdere (foreign workers). It was during this period that Mohamed arrived, coming from Germany and looking for work. With his Somali background, he came from significantly further away than his Turkish and Yugoslav counterparts; still his arrival was typical of the late 1960s: young, male work migrant, looking for employment and, perhaps, ad- venture. Mohamed Abshir, however, was not the first Somali to settle. A few Somali sailors had arrived from 1965 onwards, and some of them later married Somali women – as did Mohamed Abshir and Halima Abdi’s father. As Danish historian Bent Jensen (1999b) has shown, the contemporary political and media debate for or against immigration started during this period, focusing on the possible expenses of social welfare for the guest workers, the cost of their possible repatriation, the competition between them and the Danish workers, and worries over racism and exploitation. While employers were in- clined to be in favour of more immigration, unskilled workers, the unions and left-wing parties tended to oppose it. In the fall of 1973, Europe was hit by the oil crisis, which was followed by economic recession and rising unemployment. In Denmark, as elsewhere in Europe, this led to an immediate and permanent freeze on immigration. The freeze did not apply to Nordic citizens or to citizens of the member states of the European Community which Denmark had joined in 1972. Neither did it bring a halt to immigration from outside these countries. The large majority of the guest workers did not, however, go back, but settled and gradually established their own families, often through family re- unification. As work-related immigration was banned – and the circulation of migrant workers thereby stopped – family re-unification and asylum seeking became the two major ways of gaining access to Denmark. The number of for-

49 This section is based on Danish historians Bent Jensen (1999b) and Søren Peder- sen (1999).

121 eign citizens in Denmark thus continued to rise in spite of the immigration freeze. In the late 1960s, refugees came from Czechoslovakia and Poland, and in the 1970s mostly from Chile and Vietnam. In the 1980s, Tamils, Iranians, Leba- nese, Iraqis, and Vietnamese and Polish refugees were recognized (Pedersen 1999, 250-251). During the 1960s, the so-called de facto refugee status was established to offer protection to those who had not fled from personal persecution, according to the 1951 Refugee Convention and 1967 Protocol definition, but who still needed protection, such as defectors from the communist countries. In 1983 a new Alien Consolidation Act was passed, which granted a number of rights to refugees and asylum seekers and institutionalized de facto refugee status. At the same time, visas became obligatory for countries outside the Nordic Countries and the European Community. The Alien Consolidation Act was subsequently tightened during the following years. In the 1970s, the concept of ‘integration’ started to enter the Danish political and public debates about immigration. Especially from the late 1970s onwards, the debates started to become more polarized when ‘immigrant sceptics’ such as the anti-Muslim and anti-immigrant Progress Party claimed that immigrants threatened Danish identity and culture, while their critics emphasized the Dan- ish responsibility for the guest workers and their families (Jensen 1999b). Ac- cording to Jonathan Schwarz (1990) and Mehmet Ümit Necef (2000), who have both written extensively on immigration and integration in Denmark (and who both entered Denmark as immigrants with an American and Turkish background, respectively), the Danish debate changed in character after the immigration freeze in 1973, from a focus on immigration as a labour market and economic problematic to a cultural problematic. While their conclusions concerning what ought to be studied differ (Schwartz points to power relations and social conflicts, whereas Necef calls for a debate about whether and how ethnic minorities challenge the Danish welfare and nation-state), they both agree that there has been a culturalization of the public debates and research concerning migrants in and migration to Denmark. According to Necef, this tendency was strengthened by the influx of refugees from the middle of the 1980s, when the number of recognized refugees – especially from the Middle East – grew twenty-fold between 1983 and 1985. This development, he claims, changed the idea of foreigners. From being primarily related to the labour market, migrants

122 became associated with the institutions of the welfare state, as persons in need of help. At the same time, the political wings changed position on the question of immigration. The right wing, which had previously been pro-immigration, was sceptical about refugees and family re-unification, while the left wing took a more accommodating stance regarding the ‘refugee question’50. There was, according to Necef, almost national consensus that it was very difficult to rein- tegrate migrants and their family re-unified spouses in the labour market. The dominating issue therefore became if, and how, the ‘immigrants’ culture’ should be supported and preserved, and whether this should be a matter of public responsibility51. The political centrality of immigration and refugees became even more apparent when the so-called Tamil Gate52 caused the resignation of the Conserva- tive and Liberal government in 1993; this was succeeded by a four-party government with the Social Democratic Party and the Social Liberals as the two main parties. As we shall see, the political climate also changed significantly in the last part of the 1990s, when a larger number of Somalis had arrived in Denmark – evolving into an even more outspoken emphasis on notions of culture and integration.

Somalis in Denmark Until the end of the 1980s, there were only very few people of Somali background in Denmark. Several of the informants, however, arrived at this time as students, family re-unified or asylum seekers. Musse Sheikh, now a bio-medical scientist and a volunteer social worker, was one of them. In Somalia he had studied law at university and obtained a scholarship to Denmark in the early 1970s. Sheikh, who was very critical of the regime, applied for asylum in Den- mark, but the political situation in Somalia was unknown to the Danish police. He recalled:

50 As Necef also notes, parts of the Social Democratic Party, the so-called ‘mayor wing’ were rather sceptical. See also Jensen (1999a). 51 In many ways, the debate – in Necef’s reading – has similarities to Charles Tay- lor’s (1994) discussion of recognition of cultural identities and ‘cultures’, though Necef and Taylor probably would not agree about the consequences of and ways of dealing with immigration. 52 A political scandal concerning the deliberate delay of family re-unification with Tamil refugees.

123 The policeman knocked on the table and told me that, ‘you guys are always lying’. I only weighed 68 kilos at the time and I was as thin as a finger and the policeman was two metres tall like me, but double the size. I can remember him sitting there on the other side of the table, reading these four or five typewritten pages about the Geneva Convention and human rights, and that I hoped the Danish government would help me. He knocked on the table and said that I was lying because how could I write such an application myself?

Musse Sheikh appealed to the Danish authorities, but was told that he could go to (the former colonizer) Italy, where Somali citizens did not require a visa. In the meantime, however, he had fallen in love with a Danish girl, whom he married and he was allowed to stay. Though the experience with the police was frustrating, Musse Sheikh remembers his first year in Denmark with pleasure. The other informants who had arrived in Denmark during the 1970s and 1980s also emphasized their initial positive experiences, often with some nostalgia. Hawo Yusuf, now a childcare teacher, came in the middle of the 1980s. As she had family in the UK, she was heading towards London, but the flight had a transit stop in Copenhagen, and Hawo Yusuf was told to ask for asylum in Denmark. “There were not so many refugees when I came to Denmark and only a very few women, who asked for asylum”, Hawo explained, continuing that it was not difficult to come to Denmark at that time; people were very friendly and treated her nicely. Both Hawo Yusuf and Musse Sheikh pursued further education in Denmark and both have been very engaged in Somali-Danish associations for a number of years. Their time of arrival and reception were, as Hawo later noted, very different from the Somalis who arrived in the 1990s. In the 1990s, following the civil war, a larger number of Somali asylum seekers and later family re-unified started to come to Denmark. Most asylum seekers were admitted as de facto refugees, granted by their lineage affiliation (Fink-Nielsen and Kleist 2000, 68), as we saw it in Aisha Ahmed’s case. To re- cite one’s lineage (or alternatively, to learn another clan lineage by heart) thus became crucial to obtain protection – not only in war-ridden Somalia, but also in Denmark. Denmark was not necessarily the final destination, but often a coincidence if the plane was in transit – like in Hawo’s and Omar’s cases. Still, whether asylum in Denmark was planned or not, the country became a major country of residence for Somalis in the West (see also Table 2 in chapter 4 and Table 4 below), both for asylum seekers and family re-unified. From 1996, however, the Danish Immigration Service started to carry out DNA profiling in

124 Somali family re-unification cases in order to tighten control and lower the number of family re-unifications (Fink-Nielsen and Kleist 2000, 68-69). As Ta- ble 4 shows, DNA-profiling did indeed lower the number of family re- unifications. The table also shows that the number of Somali citizens in Den- mark peaked in the late 1990s and the early 2000s and then started to drop again. This decline is at least partly due to a rise in naturalizations and, to a much lesser degree, repatriation. The table also shows a dramatic drop in the number of recognized refugees, due to the abolishment of the de facto refugee status in 2002. This meant that the recognition rate of Somali asylum seekers, which had been running at 91-92 per cent from the middle of the 1990s, dropped to 40 per cent in 2003 (Udlændingestyrelsen 2004c; 1999)53.

Table 4 Number of Somalis in Denmark, 1990-2005 1990 1992 1994 1996 1998 2000 2002 2004 2005 Year 393 1.249 3.638 6.925 11.890 14.265 14.585 13.099 11.275 By citizenship - - 3.789 7.091 12.113 14.856 17.299 17.363 16.952 With descendants 138 641 897 1.702 946 538 646 49 10 Refugee status - - 711 1.560 707 - 624 192 202 Family reunified 2 4 7 32 159 1.189 2.263 2.022 1.709 Naturalized - - - 275 619 436 814 698 709 Emigration 2 0 2 1 23 45 43 10 25 Repatriation*

Source: (Dansk Flygtningehjælp 2006; Udlændingeservice 2006; www.statistikbanken.dk 2006; Udlændingestyrelsen 2005; 2002; 1999; Danmarks Statistik 1996). - No statistical specification. In 2000, due to a technical change in the compilation of the positive decisions on family re-unifications, this number could not be broken down by nationality (Udlændingestyrelsen 2005, 44). Where nothing else is implied, all numbers refer to Somali citizens. Naturalized persons are thus not included in the number of emigrants.

* The number of repatriations does not specify the number of repatriated Somali citizens who have used their right to return to Denmark.

53 Furthermore, there was a general decline in asylum applications submitted to Western countries in 2005, also with regard to Somali applicants (UNHCR 2006, 4).

125 As we shall see, a large number of Somalis arrived in a period when the development of a multicultural, multiethnic society evolved into a contested and politically debated issue, and was followed by a range of new policies. The So- mali group was, in many ways, in the eye of the hurricane. This becomes especially apparent if the reception of the Somalis is compared with that of the Bos- nians54.

In the eye of the hurricane In the early 1990s, the fall of the Iron Curtain had led to fear of a big influx of migrants from the East European countries and the former USSR. It turned out, however, that it was the civil war in the former Yugoslavia which caused the displacement of altogether 3,7 million people, of whom about 700,000 fled to western Europe; about 18,000 Bosnians came to Denmark (Dansk Flygtningehjælp 1996). The Bosnian refugees were first offered temporary leave to remain and placed in specially-constructed ‘refugee villages’; they were not to become a part of Danish society, but rather maintain an orientation towards ‘home’. However, as the war did not end as quickly as expected, they were allowed to apply for asylum and repatriation and emigration support arrangements were passed in 1994. Likewise, the Bosnians were a special target group for Danish repatriation efforts, administered and practised by the Danish Refugee Council (Grünenberg 2006; Fink-Nielsen and Kleist 2000, 64-68). In this sense, Denmark followed an international policy development in which repatriation was prioritized and idealized55. The fact that many Bosnians would most probably stay in Denmark and not go back stirred up the debate about refugees and ‘limits of integration’ again (Jensen 1999a, 376ff). Still, while the presence of the Bosnians thus was debated in the first half of the 1990s, they were also seen as ‘genuine refugees’ or ‘war refugees’ – in contrast to bekvemmelighedsflygtninge – economic refugees, literally ‘convenience refugees’. As several observers have noted, the Bosnian

54 See Grünenberg (2006) for an interesting analysis of the senses of home and belonging among Bosnian refugees in Denmark. 55 As I have shown elsewhere (Fink-Nielsen and Kleist 2000), the UN was a central motor in this development, which Denmark followed. During the 1990s, voluntary repatriation was increasingly emphasized and state-supported, and in 1999, the different repatriation support arrangements were united in a Repatriation Act with effect from 2000.

126 refugees were perceived as ‘much like us’ (Grünenberg 2006; Fadel, Hervik, and Vestergaard 1999, 176ff), and often contrasted with Somalis positioned at the opposite end of a presumed immigrant hierarchy and perceived as ‘very difficult to integrate’. This situation seems to be somewhat parallel to that seen in other Scandinavian countries56. In an article about the media coverage of ‘the difficult Somalis’, three Danish anthropologists suggest that the Somalis took over the position as the most visible group in the Danish refugee discourse from the Bosnians, but that in the Somali case media and politicians chose to focus on otherness and difference, rather than sameness (Fadel, Hervik, and Vestergaard 1999, 176). They list a range of articles and headlines in both local and national newspapers in 1997 which focus on cultural differences and conflicts between Somalis and Danes, on Somali problems concerning health, housing, clothing, family relations and lack of education, and accusations of Somalis cheating the social welfare system. The articles were larded with quotes from discontented mayors, social workers, doctors and politicians (including the then Minister of Social Affairs, Karen Jespersen, and the future Prime Minister, Anders Fogh Rasmussen). A series of letters to the editors was published, where Somalis were criticized for their ‘medieval culture’ or accused of being bogus. Furthermore, Somalis were also at the centre of a campaign instigated by the Danish tabloid newspaper Ek- strabladet in 1997 about de fremmede – ‘the strangers’57. The campaign was apparently launched to provoke a debate about the presence of the refugees and immigrants in Denmark. One of the central stories was about the Somali refugee Ali and his ten children and two wives (though one of these was arguably an ex-wife) and ‘his income’ (the total amount of social welfare received by Ali, his wife and his ex-wife) of more than 600.000 DKR (Jørgensen and Bü- low 1999), the equivalent of almost 100.000 USD. This stirred up a political debate comparing the rights of non-Danish and Danish citizens to social welfare. Both Fadel et al. and Jørgensen and Bülow conclude that the media’s representation of Somalis went uncriticized, and it developed into its own genre which was copied all over the country. Furthermore, they claim that the cover-

56 For the situation in Norway, see Assal (2004) and Fangen (2006a, 2006b), for Sweden, see Halane (2004). 57 In Danish, the words for stranger and foreigner can both be translated to fremmed. While fremmedarbejder best is translated to foreign workers, the term fremmede can both be translated to foreigners and strangers.

127 age demonstrated a construction of absolute otherness and backwardness of Somalis in relation to the Danes (cf. Fink-Nielsen and Kleist 2000). That many Somali-Danes felt this to be an uncomfortable and degrading situation is reflected in the fact that the Council of Somali Associations in Denmark complained to the UNHCR in March 1997 about their treatment in Denmark (cf. Møller and Togeby 1999, 5). According to Danish sociologist Morten Ejrnæs (2001), the media campaign had a direct influence on Danish politics. Ekstrabladet presented the story about Ali to the Danish government, which responded with outspoken indignation. Later that year – and after the continued media campaign – a bill was presented, lowering the available economic support and forming a part of the Integration Act of 1999 (L 474). Apart from the reduction of social welfare benefits, the Act also introduced a policy of forced dispersal of refugees in the Danish municipalities and a three-year integration programme58 (Kjær 2003, 217). Ejrnæs argues that the ‘Ali story’ and the media campaign were used to substantiate this tightening of integration policies. The integration debate had entered a new stage, directly focusing on the challenges and survival of the Danish welfare state (cf. Necef 2001; 2000), and often combined with the anxiety of ‘the immigration sceptics’ towards the preservation of ‘Danish culture’, not least in relation to Islam. Somalis also came on the political agenda in the late 1990s in relation to repatriation. In 1997 the Danish Immigration Service claimed that parts of Soma- lia were safe areas. This caused a political debate about repatriation and the return of Somalis in Denmark, culminating in an agreement between the Danish and Somaliland governments concerning cooperation in relation to the return of rejected Somali asylum seekers. At the same time, the Danish government also considered the donation of 44½ million DKR (about 740.000 USD) to the reconstruction of two Somali provinces, including Somaliland. Though the two arrangements were not officially connected, the implicit relationship between the two agreements was much criticized: while the one concerning the return of rejected asylum seeker was passed, the other, related to development aid, was turned down (Fink-Nielsen and Kleist 2000, 69-70). Following this debate, the

58 Though the law has been changed several times, the overall principles of forced dispersal was still the same in 2006.

128 Progress Party and the Danish People’s Party59 demanded that all Somalis citizens should be deported from the country – if necessary, Tom Behnke, MP for the Progress Party (and later MP and spokesman on law for the Conservative Party) proposed, with parachutes over Somalia. While Behnke had to withdraw his statement, his suggestion became much debated in the media, and several of the Somali-Danish informants I talked to in 1999 mentioned it as evidence of the hostility towards the Somali-Danish group (Fink-Nielsen and Kleist 2000, 102-103). In 1998 the Danish Refugee Council (DRC) obtained another special programme appropriation to finance activities concerning voluntary repatriation, with Somalis as a special target group. The DRC held a range of seminars, established a newsletter in Danish and Somali, a hotline, counselling activities and supported a number of so-called scouting trips to Somaliland and Somalia. Likewise, the DRC employed Mohamed Gelle – a former sailor and chief officer – to further network and information activities (Fink-Nielsen and Kleist 2000). Repatriation was high on the Danish agenda60, though the means of attaining this goal differed. The rightwing parties, and as we have just seen, the Progress Party and Danish People’s Party, wanted repatriation to be the end goal of refugee policies, while the government and leftwing parties emphasized voluntary repatriation as the ideal solution. Given the political and media debates of these years, it is not surprising that some Somali-Danes did not feel that they were particularly welcome and sometimes had rather ambiguous fields of belonging towards Denmark. No matter which political orientation, the promotion of repatriation involved a notion of refugees’ homes as elsewhere, a politics of belonging which was to be realized through return (Fink-Nielsen and Kleist 2000). While the media campaign of 1997 might have been the most outspoken example of public exposure of Somalis as minority group in Denmark, the elec-

59 The Danish People’s Party is a rightwing, nationalistic and populist political party, established by a breakaway group from the Progress Party in 1995. Like the Progress Party, the Danish People’s Party has branded itself as an anti-immigrant and anti-Muslim political party. 60 It should be noted that the DRC explained their support of Somali repatriation as based on a range of queries from Somali citizens about return and, according to one of the employees, because the DRC wanted to do something nice for the Somalis (Fink-Nielsen and Kleist 2000, 70).

129 tion campaigns in 2001 and 2002 were characterized by a previously unseen focus on migrants and ethnic minorities in Denmark, again often termed de fremmede – ‘the strangers’ or ‘the foreigners’. According to Danish human rights’ lawyer Kim. U. Kjær, the subject of foreigners turned into a domestic issue during the 1990s when it was taken hostage – often, Kjær notes, with ‘populist un- dertones’ (2003, 14-15). Furthermore, the Danish political debate and the continuing tightening of asylum, immigration and integration rules and policies at- tracted international attention and criticism from various UNHCR and EU agencies, which remarked that ‘the Danish tone’ was disturbing and sensational (ibid.). In 2001 the Social Democrats and Social Liberals lost government power and were replaced by the Conservative and Liberal parties, granted parliamentary majority by the Danish People’s Party. The new government promised to reduce the number of foreigners arriving in Denmark by tightening the asylum and immigration legislation as well as making integration of foreigners already in the country more effective and less costly. These pledges were realized with the so-called ‘foreigner packages’ in 2002 and 2003 (Kjær 2003; Beskæfti- gelsesministeriet 2002). Legislation concerning family re-unification was se- verely tightened, as were the rules for obtaining and keeping permanent residence permit, which were changed from three to seven years and, at the same time, divided into three separate extensions – thereby also making it possible to redraw the permit each time (Kjær 2003, 222ff; L 365). As mentioned earlier, de facto refugee status was abolished and replaced with a protection status, the so-called b-status. This new status restricts the possibility of obtaining protection in Denmark to people who risk the death penalty or torture if they return, but does not include people who ‘only’ risk imprisonment or civil war related insecurity (Kjær 2003). Most Somali asylum seekers thus became excluded from obtaining asylum and protection in Denmark. It was also emphasized in the government’s ‘foreigner package’ that “refugees shall not become immigrants” and that refugees shall be sent back if the situation in their home country changes before they obtain a permanent residence permit (Beskæftigelses- ministeriet 2002, my translation). In 2002, 2003 and 2004, Somalis re-entered the media and political debate, coinciding with the period when I had started to prepare and carry out my fieldwork. Once again, Somalis were depicted as an especially difficult group,

130 singled out for their ‘medieval culture’ and ‘barbaric practices’, this time concerning girl circumcision and so-called correctional education. Newspaper headlines, such as ‘Somalis send their children home to Africa: Not allowed to be too Danish’ (B.T. 2006), and ‘Too Danish for Denmark’ (Information 2003) appeared. Once again, bills and arrangements were tightened with reference to Somalis, making it punishable to carry out girl circumcision abroad (L 386 2003). The maximum age of family re-unified children was lowered from 18 to 15 to hinder the alleged problem of correctional education as well as there were other restrictions in the Aliens’ Consolidation Act (L 171 2004)61. Again, So- maliness and Danishness were constructed as two opposite poles and, often undocumented, ideas of correctional education were used once again to restrict access to Denmark. Several Somali-Danish activists and spokespersons were active in the debate, particularly in relation to the question of girl circumcision. And again, Somali-Danes protested to the UN about their treatment by politicians and the media62. In December 2005 the Danish Nationality Law was changed, making it considerably more difficult for non-western foreign citizens to obtain Danish citizenship, and requiring, among a range of other conditions, the passing of a test in the Danish language, as well as in Danish culture, society and history63. And in March 2006 the importance of knowledge of Danish culture and society was further emphasized with the ‘Declaration on integration and active citizenship in Danish society’ (Ministry of Integration 2006) which refugees and family re-unified persons have to sign. While these two last changes took place after the conclusion of my fieldwork and thus were not part of the overall contexts for the interviews, they do illustrate the political develop-

61 The existence of girl circumcision in Denmark remains debated. According to activists engaged in information and prevention work, there are cases, while other spokespersons claim that girl circumcision does not exist in Denmark. 62 Mohamed Gelle complained to the UN Committee on the Elimination of Racial Discrimination via the Centre for Documentation and Counselling against Racial Discrimination. The Committee came to a decision in 2006, criticizing Denmark for racial discrimination (Grund and Pinholt 2006). 63 The emphasis on ‘Danish culture’ was further accentuated by the introduction of kulturkampen – the ‘cultural battle’ – which included the establishment of a range of canons of Danish literature, art etc. At the Conservative Party congress in 2005, Danish Minister of Cultural Affairs, Brian Mikkelsen, stated that the purpose of the canon was (among other things) to be used against ‘Muslims with medieval norms’’. Mikkelsen later claimed that he was misunderstood (www.dr.dk, 29.09.05).

131 ment and direction in the first years of the 2000s. The development shows us that the discussions from the 1970s over whether the newcomers should take on ‘Danish culture’ or preserve their own, as Necef phrased it (2000), had tilted towards the first option. We can say that migrants and minorities now have to become Danish and profess to Danish norms and culture. This raises the questions as to how Somali-Danes articulated their situations in Denmark. There is no unified answer to that.

Ambivalent encounters While one way of framing the Somali reception story in Denmark is in terms of continuing discrimination and hostility, another side of the coin is that a significant number of Somali refugees and family re-unified have obtained protection and residence in Denmark – especially until the tightening of the Aliens Act. Still, it remains the case that the reception of Somalis in Denmark has been ambiguous and sometimes even hostile. Pushed to its extreme, the stereotypes of Somali-Danes were constructed in terms of absolute otherness and insurmount- able cultural difference in the media and political campaigns, creating a new ‘master story’ of Somaliness. The term de fremmede – the strangers – indicates that Somalis and other migrant groups were positioned as outsiders in Danish society, as people who do not belong in the country. The emphasis on repatriation, restricted access and return can be said to accentuate this non-belonging. If we construct a hierarchy of minority groups in Denmark – or rather of the stereotypes of migrants and minorities – ‘the difficult, backwards Somali refugee’ would be situated at the lower end, at least since the middle of the 1990s. I want to emphasize that although such a hierarchy may be constructed at the discursive level, it has real consequences in everyday life – and, as we have seen, in politics. We should remember, however, that hierarchies are negotiated and are situated within time and space; they are not set in stone and they do not necessarily apply to other settings. What they do, however, is to constitute a shared context of identification and belonging for individuals with Somali backgrounds. Not in the sense that everybody has had the same experiences and reacted in the same ways, as I have already cautioned against believing, but rather in shaping the ways Somaliness and Danishness could be made into meaningful categories.

132 The informants expressed two overall strategies for experiencing and coping with the public exposure of Somali-Danes and the debate about the strangers. One approach was to view it as an overall, general, impersonal political debate where personal racist or discriminatory incidents were unpleasant exceptions, if they ever occurred. I term this the exception approach. Another was to perceive it as part of as an institutionalized, hostile atmosphere, which was reflected in everyday encounters, as an everyday discrimination approach. Both approaches share the overall experience of being in the midst of a sometimes rather tire- some public debate, and encountering the stereotype of ‘difficult, backward Somali’. What the two approaches do not share is if such encounters are to be perceived as general or personal, institutionalized or exceptions. Thus, while some of the informants told me about personal experiences of discrimination, many also emphasized that they had not been subjected to discrimination on a personal level. Both groups of informants emphasized their gratitude for the protection and help they had received in Denmark, and both also worried about the political development. The informants thus all placed themselves in moderate positions. None of them expressed the view that Danish society as a whole was hostile towards the Somali-Danish group, though several explained that this perception could be found amongst more marginalized Somali-Danes. Likewise, none articulated the view that the Danish debate and policies were fair or ‘healthy’. Below, I analyze different ways of coping with Danish society and ‘Danishness’ through the examples of two women and two men. I start with Saphia Mansoor.

‘It’s like Denmark has two faces’: Saphia and Fatuma Saphia Mansoor grew up in a relatively wealthy family in a major town in the southern part of Somalia. When the war broke out, she fled to Kenya and later to Denmark, while several of her siblings went to Canada and the US. Aged nineteen at the time, she was placed in a Danish asylum centre at the beginning of the 1990s, following the normal procedures. “It was such a terrible time”, Saphia told me. To be assigned a number, instead of her name. To stand in a long line to get a meal. To share a small room with several other women who sometimes had male visitors at night. “Things you never thought would happen to you”, as Saphia said. After three weeks, she was moved to another centre with a very bad reputation amongst the other asylum seekers. “If you don’t turn

133 into a prostitute, you’ll surely become a drug addict”, they told her. In the new centre, however, Saphia accidentally ran into a Danish woman she knew from Somalia and later moved in with her family by special permission. She started studying and met her future husband, a Danish man. At the time of fieldwork, Saphia Mansoor had finished her studies and was on maternity leave. In many ways, her story was thus one of success and achievement. Still, she described her relationship to Denmark as of a double nature. It is, Saphia said, like Den- mark has two faces. I have two images of Denmark. The Danes that I know and mix with, my friends. They are like one life, one world; and then the world outside them that doesn’t know me, that generalizes me as … as a group, as Somali or a general group … Somalia this and that … and which means that all of us are generalized into one group, instead of appreciating individuality. That said, it is only the media and the official Denmark that holds that view of us. When I am with my friends, the people who know me, it is something completely different … In a way, you can say that … I have two visions of Denmark, one that knows me and who isn’t at all afraid of me – and then the ones who don’t know me. And sadly, those who don’t know me become more and more numerous. This statement, though phrased in different ways, was recurrent among many of the Somali-Danish informants. The public collective image, the stereotype of ‘the Somali refugee’, defined the agenda for how individual Somalis were encountered in Danish society, and this stereotype often overshadowed the possibilities for being met as an individual. Saphia’s solution was to distinguish between ‘friends’ and ‘the rest’, thereby establishing a zone of security and avoiding the gross generalization of ‘all Danes’. Fatuma Mohamed, another young woman, came to Denmark through family re-unification. Living in Mogadishu, her family was middle class – “not rich, not poor, like a normal family with our own house”, she said. The family had to flee at the outbreak of the civil war and was separated in the course of events. For three months they received no news of Fatuma’s father, but finally learned that he had reached Denmark, and that they should make their way to Ethiopia and then to Denmark by means of family re-unification. Unlike Saphia, Fatuma described her first experience of Den- mark as “just fun and interesting”. She went to school and started further education to be a schoolteacher. At the time of the interview, Fatuma Mohamed had married a Somali-Danish man with whom she had two small children. While it had been easy and fun to arrive in Denmark, everyday life was proving more difficult. One reason was the difficulty which Fatuma’s husband had experienced in getting a job matching his Danish education. Another was the recur-

134 rent encounter with the stereotype of the ‘backward and difficult Somali’. Fatuma explained:

Well, the first comments you get … by Danes who realize you can speak Danish …’hey, where are from?’, ‘I am from Somalia’, ‘oh, but you speak Danish!’, It’s like we cannot do anything at all … So you have to explain all the time, like ‘we are not like this, we are human beings as well, we also went to school’.

For Fatuma and many others of the informants, such experiences were irritat- ing, tiring and possibly perceived as degrading. Some reacted by presenting themselves as good examples, trying to ‘de-mystify the Somalis’, as another young woman phrased it. Other informants, however, emphasized that they only rarely or never encountered such prejudices, articulating the exception approach. Bashir Jama, for instance, a Somali-Danish man in his early 30s whom I met in Hargeisa, was of the opinion that such incidents were self-inflicted and largely a result of being lazy and hanging out instead of pulling oneself together and starting to work. Referring to his own life in Denmark, he claimed:

Such people are on social welfare, but I was not, I had my work. I woke up at eight in the morning, went to work and came back, so I only had time off during the weekend, you know, so I did not have time to blame the social counsellor or something like that. The people who complain have time for it because they are unemployed or on social security. I did not have the time to walk around the streets with ten or twenty friends – I didn’t even have so many friends. I was very busy back then.

Bashir Jama, who had established his own business in Somaliland and had worked hard to realize it, thus articulated the idea that everyone is the architect of his or her own fortune. He also emphasized, however, that this was easier in Somaliland than in Denmark, where everything is already established and where there was no ‘free space to do business’. He thereby established himself as an exception in contrast to other Somali-Danes, and was one of the few informants who did not mention the Danish debate. As Saphia and Fatuma’s examples show us, it may not be as simple as that. Still, while only few of the other informants shared the view of Bashir, many agreed that a life on social welfare was difficult. So did Mahmoud Sheikh and Abdirizak Ali.

135 ‘You have to manage your life’: Mahmoud and Abdirizak I first met Mahmoud Sheikh at a big international Somali conference held in a Copenhagen suburb in 2003, an event I come back to in much more detail in chapter seven. Mahmoud, a man in his mid 30s, volunteered to be my translator and later invited me home to meet him and his wife to talk more about my project. I visited them twice, both times while Mahmoud was taking care of one of his children, who were home sick from school or day care. He and his family lived in a big apartment block, and both he and his wife were unemployed at the time of fieldwork. Mahmoud Sheikh had, however, ‘a famous name’. Not only was his father a respected and well-known public official, but Mahmoud himself used to be famous sports star when he was a young man. After civil service, he had taken up advanced studies in Mogadishu, and had a promising future ahead of him. The war changed everything. Mahmoud had to flee, first to Kenya and then to Denmark via several East European countries. In Denmark, he began in his own words “to reconstruct his destructed life”, starting in the 8th grade to learn Danish. After a few years, he took up further education, but could not continue with this when he moved from Jutland to Copenhagen. At the job centre, they told Mahmoud that he was wasting his time. “We have seen a lot of people like you who want to stay home and have a good time” the case officer proclaimed, and continued, “this is why so many Danes don’t like refugees”. Instead, she proposed that he should pursue unskilled work rather than continue studying64. Mahmoud, seeing himself as a man of resources and a certain status, was very upset by this. At the time of interview, he was unemployed and, as the father of five small children, struggling on a tight budget. “All the Danes know about me is that I am a Somali refugee”, Mahmoud told me, and continued to explain how tired he was of being generalized as problematic because of his Somali nationality. “We are all human beings”, he repeated several times, “we are all human beings, but we have lost everything, give us a chance”. Still, when I asked Mahmoud to present himself, he started by thanking the Danish society, but quickly turned to the difficult situation he was facing:

I thank the Danish society for how they treated us … until the middle of the 1990s. Then it became a bit … difficult. We have not got the help we

64 I have presented Mahmoud’s story to several social counsellors, who have pointed out that Danish law prescribes that immigrants become self-providing as quickly as possible, rather than pursuing further education.

136 need. Some people have lost their parents, their children, their closest family and they have many problems, both psychological and physical problems. […] We don’t have a life. Life is to taste some things, for instance to get a job and integrate oneself. We are people with our own resources and Denmark should learn to exploit them – we have our own education, we speak many different languages, we are adult and we are ready to go to work. Now we just hope that our children get a chance.

Like Saphia and Fatuma, Mahmoud found it difficult and frustrating that the same yardstick seemed to be applied to all Somalis. “We are different, some are nomads, some are illiterate, some are very well-educated”, he said; referring to the story about Ali in Ekstrabladet, he emphasized that “not all Somalis make the same stupid mistake, it was just Ali”. What bothered Mahmoud Sheikh was how the stereotype of Somali refugees, created by the media and politicians, shaped how ordinary people perceived individuals with a Somali background. “The Danes are media people”, he said, explaining that many of them get their knowledge about refugees from the media, making them think of Somalis as second-rate people. Still, Mahmoud maintained that “most Danes are really okay”, and that he had never experienced any ‘real’ racist incidents, though misunderstandings did occur.

I was at a parents’ evening at my daughter’s school, talking to a teacher, and I was just sitting there with all the parents and I felt like I was a part of Danish society. Then the teacher said, ‘we are Danish, you, you, and you, but you are not…. and we are tired of your child’. I was so shock- ed! I was looking around, but I was the only one sitting there. Then I said, ‘Why? Just forget where I come from, I accept that I am not Dan- ish, okay. But my children are also a part of this school. If you want to discuss them, please don’t do it in front of everybody else’.

For Mahmoud and many others – men as well as women – the situation and future of the kids was one of the most worrying issues. As Mahmoud implied, it is one thing to be misunderstood as an adult, but another if one’s children remain positioned as strangers and non-Danes. Still, as Mahmoud Sheikh’s example also bears witness to, adults worried about their own situation as well. Abdirizak Ali was another example of a man who had had a hard time coping with life in Denmark. Abdirizak, a man in his forties, lost his parents at a young age and grew up as an orphan. Thanks to the economic support of an older brother working in Saudi Arabia, he, as the only one among his siblings, managed to get an education in Mogadishu and to flee when the civil war broke out. Abdirizak Ali, however, had to leave his pregnant wife behind and received

137 no news from her for a long time. He first went to Asia before arriving in Den- mark in the beginning of the 1990s. At that time, he finally got in touch with his wife and after a few more years, she came to Denmark through family re- unification. This was the first time Abdirizak had seen their child. The first years in Denmark were very difficult, Abdirizak recalled. Like Mahmoud, he wanted to pursue further education and like him, he was told by a social worker not to be so ambitious. And like Saphia, he never thought he would be a refugee and be dependent on other people’s charity.

I was 28 years old, when I came to Denmark; I was educated, and I had worked in Somalia, and I had my pride … But I lost it when I came to Denmark … after I went to the social security office and got social security. I was so … No! I would much rather work, but I couldn’t find a job. It was so difficult because of language barriers, because there was no work at that time. It was a very tough and difficult experience for me to receive social security. And I got a bit aggressive, because when the social worker said I should do this or that, I thought, ‘no, they are just saying that because they provide for me’. My social worker didn’t understand me. She said, ‘why is he such a hard man?’ […] It is very tough to live in Denmark: you have to manage your life, otherwise you lose yourself and you don’t get any respect, neither from the society nor your family, if you don’t work. But if you work, pay taxes and provide for your family and yourself, there is more independence, there is more pride … and satisfaction with your life.

During his first years in Denmark, Abdirizak Ali started to chew khat, a stimu- lant leaf widely consumed in the Horn of Africa. He did, with his own words, waste his time and after a while, he decided that he had to change his situation. Abdirizak started doing different odd jobs and some translation work. At the time of fieldwork, he had a regular job and was very active in various Somali associations. For Abdirizak, and so many others, the experience of flight and exile was very upsetting; so were the lack of respect and personal opportunities he first encountered. Like Mahmoud, whose future had seemed so bright and promising, his life took an unexpected turn: he became a refugee and was unable to realize his dreams and ambitions of higher education. Abdirizak – like Aisha – arrived in Denmark with education and work experience but, like many others, could not exploit it. As already mentioned, none of the informants with education from Somalia had been able to transfer their competences, and the few who actually found work within their field had gained at least some of their education from another country to Somalia, or had

138 gone through further education in Denmark. For them and many others, the civil war had resulted in great losses in terms of friends and families who had been killed or separated from each other, in terms of loss of earthly belongings and finally, in terms of loss of personal opportunities, safety and social esteem. As we just saw, the latter losses were sometimes aggravated by their situations in Denmark and by changes in family and gender relations.

Gender change and gender nostalgia Changes in gender and family relations due to migration were one of the most hotly–debated issues at the public and semi-public seminars I attended during fieldwork. Similarly, this issue was discussed with great interest in many of the interviews and conversations. We should remember, however, that changes in gender and family relations are not a new phenomenon, but have occurred throughout history. In Somalia, the introduction of Islam, colonization, independence and civil war have all resulted in significant changes in gender relations (see for instance Gardner and Bushra 2004; Warsame 2002; Kapteijns 1994). Migration and exile in the West is thus not a break with a stable ‘traditional past’, but rather another moment of change – albeit a significant one. Changes in gender and family relations seem to take place and are debated throughout the West65 as well as in the Somali-speaking region (e.g. Horst 2003; Warsame 2002). As already noted, the civil war split up many families through loss of lives or physical separation because of the chaos of sudden flight. Many women became the heads of household, many married couples divorced, and many families were further separated because of the asylum system in the West (though many have also since reunited through family re- unification). The main contours of the debate concern whether life in the West threatens ‘traditional’ Somali and Muslim family and gender relations, where men are expected to provide for their families in contrast to ‘the liberation of women’. While only a few people criticized ‘the liberation of women’, the difficulties of many Somali men in the West to be breadwinners for their families were generally seen as a big social problem. Again, we should remember that

65 For country specific accounts of gender relations in the West, see Affi (1997) and Mohamed (1997) for the situation in Canada; Halane (2004) for Sweden; Utteh for Germany (1997); Griffiths (2002) for London; and Mcgown (1999) for London and Toronto in relation to Islam, as well as Gardner & Bushra (2004) for a general analysis.

139 the novelty of this development should not be overstated: women have been and are indeed working in Somalia, as Aisha’s case showed. Rather, I will characterize the debate as a kind of nostalgic discourse towards ‘traditional’ gender ideals mingling with ideas of Somaliness, Islam and Westernness, making it a site for negotiations of proper masculinity and femininity. In some articulations of Somali culture, conservative gender ideals would be defended (or explained) with reference to the patriarchal clan system, culture and Islam, as based on an ideology where adult and elderly men are regarded as political subjects, while women are responsible for the home and family (cf. Hansen 2005; Kallehave 2003; 2001; Lewis 1994). Mohamed Omar Arte, a former Somali politician living in Hargeisa, gave an example of such gender ideals. Being a religious man, Arte emphasized that both women and men should be God-fearing, but otherwise articulated complementary gender ideals. Men achieve respect, Arte stated, by “giving, rather than receiving”, by good self conduct, by being trustworthy, by the credibility of the clan, and by taking responsibility. Women, on the other hand, Arte continued, should take good care of the family and be good mothers, keep the household and themselves clean, and act with a reserved attitude. Politics should be men’s business. Arte’s gender ideals reflect a notion of gendered space and conduct in which men are related to public space and public decisions, while women should be more inwardly oriented towards the affairs of the family and the home. This is a classic patriarchal, complementary, and di- chotomous gender ideal which can be found in many different societies. Now, let’s be very clear. These gender ideals were challenged and contested in Somaliland as well as in Denmark. While they might not be widespread in Somaliland, many, if not most, of the Somali-Danish informants did not agree or would not like to adhere to them. Certainly, one of the most powerful politicians in Somaliland at the time of fieldwork was a woman – Foreign Minister Edna Aden – reminding us not to confuse ideals and practice. Arte was articulating gender ideals from a position in Hargeisa, not analyzing how gender relations were actually practised. Furthermore, Arte was also highly critical of the general performance of Somali men in the West. They are, he said, having major problems, being a burden to the family and feeling alienated because the system gives more priority to the women, indicating a transfer of authority from husbands to the system. Furthermore, Arte continued, women and the children

140 are more adaptable to the situation. The stability of the conservative gender ideals was thus disturbed in exile, where men became fragile, refusing to give up their previous privileges, while and women were stable and adaptable66. Many of the informants shared the view that women were faring better in the West, exploiting and enjoying the possibilities and rights of the western welfare states, while men were having a harder time, unable to fulfil the gender and family ideals of the male provider, witnessing their families falling apart. I will characterize this discourse as ‘masculinity at risk’.

Masculinity at risk The high unemployment and divorce rate for Somalis in Denmark and the rest of the West was a shared worry among most of the informants. At seminars and in interviews, it was explained to me that families are breaking up at an un- precedented rate, causing what has become known as the phenomenon of getting the black plastic sack: women, tired of their unemployed, good-for-nothing husbands chewing khat instead of helping out with the daily chores, kick them out with their belongings in a big black rubbish bag. This phenomenon also seems to be well known in other parts of the West. At the 9th International Con- gress for Somali Studies (ICSS) held in the Danish town of Aalborg in Septem- ber 2004, Somali-American sociologist Abdi Kusow told the participants about his fieldwork in Dixon, Toronto. He was studying five huge residential blocks where the large majority were Somali residents, wondering why the Somali women watched the video surveillance of the main door as if it was a sitcom show. Why, he asked them. Because, the women replied, they were watching the men getting the black plastic sack. This anecdote almost killed the audience – of whom many were Somalis – with laughter. Likewise, some Somali-Danish women found it a scream when I asked them about this expression. Still, it remained true that gender and family relations caused a lot of worries to the in-

66 Tina Kallehave (2003; 2001; 2000) has analyzed how different types of Somali families are recognized by the Danish welfare state. One of her conclusions is that there are two main types of Somali families, defined in relation to what she terms ‘the republican state’ (i.e. modern nuclear families) and ‘the clan state’ (i.e. traditional and collectively based types of social organization). While the first type can be recognized as such by the Danish authorities, the second cannot. Kallehave’s work thus serves as a reminder to distinguish between types of social organization and families.

141 formants – men as well as women. A Somali-British woman and activist in London summed up the situation in the following way, which seemed to apply in the Danish context as well:

It can be very hard for the men. In the welfare offices, the men are on an equal footing with the women. Many men had very high positions before the war, and it is difficult for them to accept that they have lost them. They keep on talking about what they used to have and what they used to be. They don’t want to start again from scratch in school with their kids and be equal with teenagers. It is easier for the women after the war. They still have the responsibility of the house, the kids etc. In Somali- land, the men pay and do nothing else and they don’t want to do anything else. Here, the men think that the system is pushing them, replac- ing their authority and that the women are doing the same. Here, there is power sharing.

This general analysis was mainly shared among the Somali-Danish informants, as the following conversation between three women shows. The women were members of a woman’s association and had been asked by the chairperson to participate in a focus group interview with me. They were aged between 20 and 40. One had been in Denmark for 20 years, and the others for about ten years. When I asked the women about the divorce rate in Denmark, one of them explained to me that “the men change”; she was quickly reproved by one of the others who pointed out that women change as well, making the first women comment that “the women get integrated, but the men behave as though they are in Somalia”. Why? Because, these women said, “the men want to make all the decisions; they want to work, but they are on social welfare and have problems because they would like to provide for their families”. These women thus agreed with the analysis above, concluding that, “the men cannot do anything here. They had power in Somalia and now they say that ‘Denmark is a ladies’ country’”. To characterize Denmark as a ‘ladies’ country’ was not a recurrent statement in the interviews, though the idea of women faring better was. Still, one Somali-Danish man blamed ‘women’s liberation’ and independence as the root of the problems in the families. With inspiration from masculinity researcher Robert Connell (2005), I suggest that the discourse of ‘problematic men’, ‘stronger women’ and ‘failed family relations’ provides an overall gender order – or according to one’s perspective, a gender disorder – in relation to which many individuals and groups were associating their own and others’ experiences. This discourse was evident at seminars and not least at the 9th Inter-

142 national Congress of Somali Studies. In almost all the plenum speeches, the resources and efforts of women were emphasized, while men were disapproved of being too focused on nitty-gritty ‘homeland politics’ and status. Likewise, in the workshops about gender relations, gender and family relations were much debated and worried about – and Somali men much criticized. While statements such as ‘women become stronger in Denmark and other foreign countries, they divorce the men and criticize them’, and ‘women have higher status in the West than men’ were recurrent, several women also emphasized family-related troubles such as social control, violence, worries over the future and education of children and young people. However, it remained a fact that only very few So- mali men – and, indeed, also only few western men – participated in the gender panels, while high-status Somali men and women criticized ‘the problematic men’ during keynote presentations. These observations tell us that the gender order is positional, articulated from some positions (many women and some high-status men), and concerning other positions (low status men). It thus not only relates to ideas of masculinity and femininity, but also to social class. Furthermore, there is a certain paradox: While the discourse claims that ‘women are doing better’ and ‘men are failing’, the employment rate and associational and political engagement have both remained higher for men than women in Denmark. This reminds us that the gender ideal of men as providers and political subjects does not stand in opposition to the discourse of ‘problematic men’. On the contrary, it is rather the reverse side of this ideal; it is a discourse which mainly concerns un- and underem- ployed men, but has relevance in an overall context for the performance of appropriate masculinity – as a standard of failure. Another aspect of this discourse, however, is that it was not only in circulation among Somalis in the West and in the Somali-speaking region, but also among Danish social counsellors and public officials, as Mahmoud’s case showed us, inscribing the discourse in relation to the idea that gender equality has been obtained in Den- mark. In this sense, the discourse of problematic men and women becoming liberated supported notions of ‘Somali culture’ and Islam as patriarchal and Den- mark as a country of gender equality. Let us now return to the conversation with the three women. When I pro- ceeded to ask if women are faring better, all three of them agreed and continued talking about the problems of men chewing khat, explaining that they are using

143 it because they are worried about unemployment and harassment ‘because they are black’. Still, they emphasized that khat abuse might cause huge problems in the families, especially if one or both spouses also remit money to relatives. However, though these three women were critical towards Somali-Danish men, they also sympathized with the men’s frustrations of not being able to provide for their families. Some women emphasized their own development towards becoming independent and achieving education and self-confidence in the West, while others – such as Aisha Ahmed – had the experience of being viewed as an unskilled, problematic single mother and social security claimant without any resources, rather than the active, self-providing woman with a long professional career she used to be. There was, in other words, no unified ‘female experience’ or female discourse. There were of course also differences in how the men talked about gender and family relations. While one man claimed that “the men and women are op- posites; there is no respect, but a lot of fighting”, others talked with love and af- fection about their spouses and emphasized the importance of education for both their female and male children. Likewise, I met a couple of Somali-Danish men on paternity leave with their babies, as well as several part-time or full- time single fathers. Still, both men and women agreed that the majority of childcare and housework tends to be done by women, and that it is very hard work to have children in Denmark without the extensive family networks that many families used to have in Somalia. Also, both men and women worried about unemployment and many men and some women missed the ‘male provider’. No women, however, seemed to lament the possibility for both men and women of having an income, though many women shared the view that it was particularly difficult for the men, as we just saw above. Similarly, many men and women articulated a kind of ‘gender nostalgia’ to the days before civil war, flight and exile, where gender and family relations were stable and secure (cf. Kimmel 2005). Or so it seemed, perhaps, in the distant memory of past times and places which can never be re-established. Abdirizak Ali also shared the opinion of difficult times in Denmark. While he maintained that the challenges in the country were not harder as such for men than for women, he still emphasized the problem of unemployment and the ensuing lack of respect that many men seem to encounter. According to Abdiri- zak, this caused considerable frustration, among other things, because the men

144 cannot carry out the duty of providing for their families that Islam and widely held cultural gender ideals prescribe.

The men are not providing for their families; they just receive social security like their wives, and it’s not a lot of money […]. They are trying to find employment, but they are not able to because of language barriers and lack of education. It is very hard for them […] Sometimes they try to communicate with the Danes, but they are turned down, or when they apply for a job, they are turned down and then they start to withdraw, or they feel discouraged and they get so upset […] After a while, they think ‘I don’t have any opportunities here, what shall I do? Denmark doesn’t need me, so I have to find another place to live’. Because you cannot be on social security your whole life, you just cannot, it is a very … tough life, a lonely life, an unpleasant life.

For Abdirizak, the solution was to find work, thereby gaining respect and ‘managing his life’. The feeling of not being able to provide for himself and his family did not, however, make him complain about the ‘liberation of women’; on the contrary, it made him emphasize the importance of education for both his female and male children. “My girls need to be able to fend for themselves”, he said and continued, “I don’t want my girls to think that the men should provide for them; they need to provide for themselves and not be dependent on a man”. Abdirizak, then, though also a religious man, was thus far from the gender ideals articulated by Mohamed Omar Arte, a point which emphasizes the need to be cautious towards general claims about the nature of men and women.

Identifications and categorizations For both Abdirizak Ali and Mahmoud Sheikh, the humiliation of going to the social office, of encountering the stereotypes of failed and problematic Somali- Danish men, and of not being able to pursue further education or find work matching their qualifications, resulted in an ambivalent relationship to Denmark and to considerations of maybe leaving the country. While asylum in Denmark did provide highly valued security from the civil war as well as health, education and peace, there remained a feeling of insecurity about the future of the children, as Mahmoud stated. So was a feeling of not really belonging – or rather, not really being accepted in Danish society. Mahmoud explained to me that he was “between two countries”.

It is actually a tough question. I will say that I am between two countries. When my children watch the news, they ask me, ‘Dad, is there al-

145 ways war where you came from?’. One part of me is Danish. And another part … I still hope to return to my home country one day. So I am like two parts. But my children are 100 per cent Danish and they have a Danish mentality.

When I asked the informants about how they thought of themselves – as Dan- ish, Somali or Somali-Danes, most of them expressed the same dilemma as Mahmoud: being between Denmark and Somalia/Somaliland, while emphasizing that their kids are completely Danish. Some of those who arrived in Den- mark in the 1970s or 1980s explained that they were mostly Danish because they have lived in Denmark for most of their lives. Another set of articulations, however, pointed to the experience of not being accepted as (even partly) Dan- ish, but being positioned as non-Danish or outside the boundaries of Danish- ness, as the following interview excerpt between Saphia and Fatuma shows.

Nauja: Do you think of yourselves as Somali-Danish or Somali or Dan- ish or something completely different?

Fatuma: You can say that I am from Somalia, but anyway … I have picked up many, many things from the Danes where I can say that I am a part of the Danish society […]

Saphia: I … because of my education, my Danish education, my teachers, my friends, they are all Danish. So in a way I feel a part … but I would never say that I am Danish. I would say that I am a Somali woman who is a part of Danish society. I don’t know why … I really appreciate my education, you know, I am very grateful for what I learned and that Denmark received me when I needed it. I will always do my best for Denmark. But even if I am a Danish citizen, deep down I feel Somali. I tell my husband that if I die in Denmark, he will have to send my body to Somalia. That is what I think.

Fatuma: But then we cannot request that the Danes … accept us as Dan- ish, can we?

Saphia: But they need to accept us as are we are, as a part … otherwise Denmark will never be multiethnic, if everyone has to become Danish. We will never get blue eyes or blond hair. The Danes have to accept us. […] You don’t throw away your background and your identity – your former identity – just because you have lived here for ten years […] He- re there is nothing like ‘I am an African-Dane’ or … Latin-Dane or Arab-Dane or Pakistani-Dane or anything like that. It’s either–or, and that’s what I don’t like.

For Saphia, one of the barriers of inclusion in Denmark was the exclusivity of Danishness as reserved for white-skinned bodies with blue eyes and blond hair.

146 For her, the absence of hyphenated identities such as African-Dane from the dominating public discourse on new and old citizens’ belonging in Denmark, excluded her and other non-white persons from being considered Danish. Saphia, then, considered herself a part of Danish society, but not Danish, wanting to expand the exclusive category of Danishness and make it reflect multiethnic reality. As we have already seen, this tension relates to the overall debate of the definitions and limits of ‘Danish culture’ and Danishness. The relationship between being a part of society and part of a nation – between civic and national belonging – turns into a Catch 22: migrants are expected to become Danish (for instance by dressing and behaving in certain ways), but still cannot be accepted as such, because the dominant understandings of Danishness remain too exclusive. For Fatuma, the lack of inclusion created a feeling of ‘insecurity in the heart’ and a pressure to give up her ‘roots’, while feeling that she will remain a stranger anyway. Not everybody, however, articulated this dilemma. Halima, as we saw in chapter four, described herself as ‘in-between’, but with the important correction that Denmark constituted home and society, while Somalia was a place to go on holiday. Likewise, another young woman pointed to the irrele- vance of the question, explaining to me that her friends like her as she is, and that it was her personality and friendships that mattered. For her, the question of singular or hyphenated national identities did not seem to be an interesting or relevant one, in contrast to the more concrete question about social relations. Many others, however, articulated their feelings that the ‘Somali stereotype’ set the limits for their access to ‘Danishness’. However, it should be mentioned that I also heard of Somalis who were accused of being ‘too Danish’ in the sense of being too busy or ‘too liberated’, not behaving in ‘proper’ Somali ways, as we just saw it in connection with gender relations. The boundaries between ‘Soma- liness’ and ‘Danishness’ were going in two directions. Still, it remains true that for many the opportunity of being accepted as an equal Danish co-citizen seemed to be almost non-existent. Likewise, many shared the idea that life would be easier in the UK, Canada, and other multicultural societies, which were seen as offering better opportunities for a dignified and equal life: places where the categories of andengenerationsindvandrer (second generation immigrant), pæredansker (Dane through and through) or de fremmede (‘the strangers’ or ‘the foreigners’) were not constantly in play and where it would be pos-

147 sible to be a part of society without the skin colour or religion ‘getting in the way’67.

Between strangeness, Danishness and Somaliness One way to explain the stereotype and the absence of complicated, transnational identifications is through sociological work on ‘strangers’. Inspired by Georg Simmel’s famous essay on ‘the stranger’ (1950), sociologists Zygmunt Bauman (1998a; 1991) and Bülent Diken (1998) have written about how strangers are fundamentally embedded in ambivalence. Strangers blur, in Bauman’s words, “the master-opposition between the inside and the outside” (1991, 53), being a permanent other, causing uneasiness and anxiety. He proposes that the stigmatization of strangers is a response to the complexity and insecurity of late-modern life, marked by the uncertainties of economic globalization and societal changes. Bauman (1998a) suggests that the anxiety of an uncertain present and future are transformed into policies of controlling and policing immigration. In a similar vein, Diken proposes that the stigmatization of certain migrant groups is a way for receiving societies to handle migrants and their multi-stranded fields of belonging. Inspired by Simmel (1950), he claims that migrants’ ambiguous identities and fields of belonging are reduced and stabilized into singular, unambiguous categories – belonging only there and thus not here. The categories of ‘us and them’ are, in other words, stabilized, ordered and (re)- established, as is the belonging to national and cultural categories, i.e. ‘Danish’ versus ‘Somali’ and so on. This process works, in other words, counter to complex fields of belonging and supports an idea that incorporation in Danish society requires singular, unambiguous and non-complex identifications and engagements which ‘fit’ with quite specific ideas of ‘Danishness’, sometimes explicitly articulated as non-Muslim. Bauman and Diken propose, in other words, that strangeness can be seen as part of an ordering process, sparked by ongoing global changes where migrants become the symbol of (what in the eyes of some is threatening and upsetting) change. Furthermore, the anxiety concerning some migrants can be seen as a re-

67 The fact that many Somalis have family members in the UK, that language problems might be fewer, and the idea that jobs are easier to find obviously also contributed to the attraction of these places. Whether London and elsewhere actually offer a better life is indeed another story (Bloch and Atfield 2002; see also Harris 2004; Griffiths 2002).

148 flection of what Liisa Malkki (1992) has termed the national order. In this perspective, culture, identity, and nations are seen as overlapping and discrete entities contained within the territories of nation states, while migrants are viewed as a deviation, a threat towards cultural and national coherence. Or rather some migrants. Because as Bauman (1998b) has suggested, there is a mobility hierarchy based on the ease and choice of where to move and settle and if and how people are welcomed, tolerated, or shut out. We can say that while recognized Somali refugees and family re-unified were let in, they still were not welcome, but rather reluctantly tolerated. They are the ambivalent strangers, not really outside and not really inside either. The descriptions of Saphia, Mahmoud and many others of being met as a stereotype and not as individuals can be said to reflect such experiences of strangeness; of being tolerated (rather than being physically assaulted or expelled from the country), but not generally welcomed. However, as we have seen, there were different ways of handling such experiences. Saphia, whose situation may be thrown into even more relief by her life situation with Danish friends, husband and education, distinguished between ‘the two faces of Den- mark’: those who knew her versus those who did not. Abdirizak reacted by turning into ‘a hard man’ until he found employment, but still wanted to leave the country.

Recognizing difference We should also remember that the ‘stranger argument’ is not the only explanation for the ambivalent encounters with Danish society or for the absence of hyphenated identities, as Saphia reflected upon above. There are more versions of ‘Danishness’ in circulation than the nationalist and reductive ones. As Saphia said, Denmark has two faces, and while one is based on ignorance and sometimes hostility, the other is friendly and accommodating. Rather, the stereotypes and the version of unitary, exclusive ‘Danishness’ form a context for identification. Just like the discourse of ‘problematic men’ is not an analysis of the situation of individual men, but a context for the performance of masculinity. Oth- erwise, the analysis becomes part of the overall generalization and stereotyping processes, fixing Somali-Danes (or other groups) in a particular position in Danish society as ‘the strangers’ – or as ‘problematic men’. Furthermore, there is more to the categorization of strangers than ambivalence. Sara Ahmed points

149 out that “recognition of strangers bring into play relations of social and political antagonism that mark some others as stranger than other others” (2000, 25, emphasis in original)68. Ahmed thus extends the discussion of ‘strangers’ through the question of cultural intelligibility and a visual economy, as I discussed in chapter two. Some bodies are, Ahmed writes, “already recognized as not belonging, as being out of place” (ibid., 21, emphasis in original). This turns our attention both towards gazes and signs of the bodies (Søndergaard 1996) – how some bodies are recognized as out of place, and “how the stranger is an effect of processes of inclusion and exclusion” (Ahmed 2000, 6). In a Somali-Danish context, the intersection of raciality and Islam seemed to be established as the opposite pole to the dominant ideas of ‘Danishness’. As Rikke Andreassen (2005) and Peter Hervik (2002) show in their analyses of the media coverage of, respectively, visible minorities in Denmark and religion, Is- lam has become a central marker for difference in Denmark, intersecting with categories of gender, ‘race’, nationality and sexuality. While Islam thus is and has been discussed in public and political debates in relation to ideas of ‘Dan- ishness’, notions of ‘race’ are much more controversial in a Danish context. There is no official language of ‘race relations’, as for instance in the UK, US or Canada, and the concept is usually understood as referring to Nazism or apartheid. This does not mean, however, that racism does not exist in Denmark, rather that it is seen as a (highly regrettable) exception from the otherwise tolerant and inclusive Danish society (cf. Castles 2000). In this sense, Gilroy is accurate when he states that racism often becomes a non-issue, because ““race” ought to be nothing” (2004, xv, emphasis in original).The publication Factsheet Denmark (Ministry of Foreign Affairs of Denmark 2006) can be seen as example of this attitude, listing official efforts and political statements against discrimination and for diversity. In Denmark, racism is thus usually understood as biological racism, which is highly unjustifiable, but almost non-existent. Another way of conceptualizing

68 Ahmed criticizes Baumand and Diken for stranger fetishmism. She further criticizes Diken for concealing “differences between ways of being displaced from home” (2000, 5). I strongly agree in the critique that the strangers should not be used to include all kinds of displaced persons as a kind of unifying concept (like diaspora should not). Still, I find that Diken’s thoughts are inspiring in an analysis of processes of stigmatization and ‘the strangers’ as a specific category.

150 ‘inferiorization’ of certain groups, can be as so-called new racism or culturalism (Gilroy 2004; Røgilds 2002; Castles 2000). In new racism, biology has been replaced and supplemented by ideas of culture and Islam as emblems of overwhelming and threatening difference. Often, however, new racism is not seen as a branch of racism, but is at the centre of political and public debates about immigration and Islam, as mentioned above. According to some politicians, researchers and debaters, this is necessary and a sign of open-mindedness; no debates should be closed with reference to racism or political correctness (cf. Cas- tles 2000, 185-186). While I agree that research should be critical and that it would be flaw to analyze the Danish debate from just one perspective, I strongly disagree, however, when and if public, political and academic debates lead to gross generalizations and ideas of culture and religion determining behaviour. While biological racism thus is officially banned, new racism is not. The discussion about biological and new racism turns our attention to how different modes of differentiation are invoked, recognized and intersecting in sometimes very derogatory ways. According to Abdi Kusow (2004), Somalis become black in western societies, in contrast to in Somalia, where clan rather than skin colour is the primary social stratification. Likewise, some of the participants in the 9th International Congress for Somali Studies expressed the view that ‘Somalis are learning to be persons of colour’, i.e. learning to be positioned and categorized by the colour of their skin. Most of the informants emphasized that they had not experienced downright racism – in the sense of being physically assaulted69. Still, several informants gave examples of unwanted racialization – from being characterized as ‘the black flower’ to their coffee being called ‘negro sweat’. One man explained that such incidents are expressions of Danish humour and continued to tell me how he deals with Danes who call him names when he passes:

They said, ‘hey, negroes are not allowed here!’. But I said, ‘listen, I have already showered three times today, I cannot get it off’. Then they said,

69 Though violent biological racist incidents seem to be rare in Denmark, they do occur. An example happened in 2005 when neo-Nazis attacked a Somali family living in a village on the island of Funen, and swastikas were painted on their house. In response, the local football club arranged a demonstration in support of the family. Likewise, there was massive media and political disassociation and denuncia- tion.

151 ‘wow, he understands our humour, come on, sit down’. So you need certain resources and an understanding of Danish humour and culture.

None of the other informants, however, took comfort in ‘Danish humour’, but rather got angry and sometimes frustrated at not being able to answer back in eloquent Danish, or they tried to ignore such incidents. And while some of the male informants recalled (with some nostalgia) how it was considered exotic and attractive to be a young black man in Copenhagen in the 1970s, most of the informants seemed to experience racialization as an uneasy mix of ideas of Is- lam and cultural difference. As with Saphia who, when heavily pregnant with her first child, was passed by an elderly woman who hissed that ‘they are always so pregnant’. In this incident, notions of skin colour and gender seemed to intersect with the idea of African or immigrant women having many children. Other times, ideas of culture and religion intersected with gender and, perhaps, skin colour. For instance, when Mahmoud was told that he was just trying to stay at home and do nothing or when Halima, covering her hair (and actually speaking with a heavy Copenhagen accent) met people who thought that she did not speak a word of Danish. Or when Aisha’s social worker could not imagine that the black woman with a scarf, refugee and mother of five, used to work as a professional secretary. The exact reasons behind such encounters cannot be established here – derogatory ideas about skin colour, Islam, women, single mothers – and this is precisely one of the points: that these modes of differentiation intersected and reinforced each other in various ways.

Chapter resume: Ambivalence and (mis)recognition I started out by outlining the modern history of Danish migration history to provide the context of Somali immigration and settlement. I showed how the larger number of Somalis arrived at a historical moment when the public and political debates about migrants and minorities were changing, and that Somalis were given a specific and exposed position in the debates about ‘the strangers’ from 1997, adding to gendered stereotypes of ‘difficult Somali refugees’. I pro- ceeded to discuss how Somali-Danes cope with the stereotypes and their fields of belonging and identifications, arguing that individuals are often met as stereotypes and not in their individuality. This stigmatizing stereotyping happens both at the general level, by the constant exposure of Somalis and other Muslim, non-western migrants as ‘the strangers’, and in everyday encounters,

152 for instance in the social office or at parents’ evenings. As a result, the stereotypes constitute a context for identification and belonging, often indicating a relation of non-belonging to Denmark. We might also say that they create a context of misrecognition. It was not only the Danish media and politicians, however, who circulated discourses of Somali gender relations. I then went on to discuss gender ideals and showed how the civil war, flight and exile in Denmark had resulted in a big loss of social position and social esteem for many Somali-Danes. Loss of social position and social esteem seem to be especially difficult for some men to deal with, in that they are unable to realize their ambitions in or transfer their competences to Danish society, causing a kind of ‘gender nostalgia’. The lack of recognition of competences – or in the terminology of Axel Honneth, the absence of recognition of achievement and social esteem – had a gendered perspective, contributing to a discourse of ‘failed masculinity’ and ‘liberated women’. Like the stereotype of ‘difficult Somali refugees’, the discourse of failed masculinity constitutes a context for the measure and performance of gender relations, a gender (dis)order. Both women and men risk being perceived as culturally unintelligible as ‘proper’ men and women, but this seems to be especially apparent for the men. Finally, inspired by Bauman and Diken’s reflections about the sociological figure of the stranger, I suggested that the stereotyping of Somalis and other migrant groups could be seen as a way of stabilizing ambivalence. In some cases, the stereotypes result in a reduction of cultures and identities that make hyphenated identities seem improbable and the categories of Somali and Danish mutually exclusive. Many of the Somali-Danish informants articulated ambivalent identifications and fields of belonging where the intersections of raciality, gender and Islam turn into a site for overriding difference – whether temporarily or more permanently. We can say that the stereotype of Somali-Danes was recognized – in the sense of identification – as culturally intelligible as strangers. The collective image of Somali refugees in Denmark can thus be described as one of general misrecognition and social subordination with public notions of inferiority, rather than reciprocal recognition or status equality in Nancy Fra- ser’s sense or, in Honneth’s words, recognition of social esteem. However, my analysis also shows that the individual experiences of stereotyping varied and

153 were articulated very differently – from notions of hostile media and politicians to insignificant exceptions, captured in the two terms of everyday discrimination approach and exception approach. This analysis raises several questions: what did the informants do to cope with their situations in Denmark? To which places and situations did they direct their engagements and loyalties? And which social relations and arrangements made it possible to be recognized as a proper and competent Somali-Danish man or woman? In the next chapter, I explore these questions in relation to associational engagement.

154 Chapter 6 ‘Making a difference’. Somali-Danish associational engagement

Associations are something particularly Danish and, internationally, you call Denmark the country of associations (Musse Sheikh, 2003).

It was in the 1990s that most Somalis came to Denmark … and there was a civil war in Somalia, there was hunger in Somalia, and I remember these pictures on the television about Somalia, the misery and the many dead people. There is always this obligation among most Somalis, that you help other Somalis, you help one another, and these thoughts also appeared among some of the refugees who had arrived earlier and who had become Danish citizens, and who were educated and had started to work […] We were a group of people who got together and said, ‘what should we do, we have to do something’. And then we started to organize a collection of money on the street […] and we said, ‘we will collect money, we will make an exhibition, we will stage a big concert to collect money to help out hunger victims’ […] We actually collected a lot of money. This is where it all started. It was like the starting signal that made us say, ‘well, we can organize some things, we can do something’ (Mohamed Gelle, 2003).

In June 2003 I interviewed the coordinator of the Somali Network in Denmark, Mohamed Gelle, about the history of Somali-Danish associations. Gelle explained the proliferation of associations with four reasons: firstly, the large number of Somalis in Denmark; secondly the need for humanitarian help in Somalia, caused by the civil war; thirdly, the cultural obligation to help each other; and, fourthly, the feeling that it actually mattered – that ‘we can do something’. In this chapter, I explore Somali-Danish associations. I start by introducing the Somali-Danish associational scene and associational engagement, analyzing the history and the social principles of organization which guide the associations. I thereby seek to establish the conjectures of a possible Somali- Danish associational field (cf. Bourdieu and Wacquant 1996). I then turn to a case study of a women’s association to examine these issues in more depth. Fi- nally, I discuss and analyze associational engagement in relation to learning

155 processes and political culture, arguing that associations form platforms for action where political agency, responsibility and ‘culture’ can be recognized and enacted. But first, let me present the associations involved.

The Somali-Danish associational field I have included altogether fifteen Somali-Danish associations in the study, though I came across and heard about a much larger number of associations and clubs. Nine of the associations were oriented towards social or cultural issues in Denmark. I include the Somali Network in Denmark, coordinated by Mohamed Gelle and connected to the Danish Refugee Council (DRC). Strictly speaking, the Somali Network is not an association and, accordingly, is not formally organized with an executive committee and annual meetings. The network has supported and initiated different initiatives, ranging from debates about youth education and repatriation through to employment-related issues. Likewise, I include the Council of Somali Associations in Denmark, an umbrella organization consisting of many Somali associations. The remaining seven associations offer different kinds of social and cultural activities for their members as well as providing information and course activities about life in Denmark. They include two women’s associations, a youth association, a local neighbourhood association in Copenhagen, two associations based on regional affiliation in So- malia and, finally, a reproductive health association with Danish, Somali and other African members. Six of the associations were mainly transnationally oriented. Five supported reconstruction projects in Somalia and Somaliland and out of these, two were umbrella organizations with members in more countries than Denmark. One was the Somscan & UK Cooperative Associations, as mentioned in the intro- ductory vignette: a Scandinavian-British-Somali organization with the purpose of building a new neighbourhood in Somaliland to further collective return. The other umbrella organization was the Development Organization of East Africa (DOEA), which had as its purpose the coordination of transnational support to the south of Somalia. Both umbrella organizations had a broad transnational orientation in terms of their members and mobilization, while their visions of reconstruction and development were directed towards specific regions in So- malia. This was also true of two other associations, which had members based in many countries. Only one association had an exclusive Somali-Danish basis.

156 The last association included in the study was a Danish branch of a Somaliland political party, also part of a transnational structure oriented towards Somali- land. With the exception of the Somali Network in Denmark, the Council of Somali Associations in Denmark, and Somscan & UK, all names of associations have been changed. In spite of the relatively high number of Somali-Danish associations, only a rather limited literature exists (Mikkelsen 2007; 2003a; 2003c; Hjære 2003; Hjære and Balslev 2001). Mikkelsen suggests that Somali-Danes might be one of the most active groups in ethno-national and religious associations (Mikkel- sen 2003b, 124; 2003c, 57-64). The Danish Centre for Voluntary Social Work reaches a similar conclusion in a survey of immigrant associations in Denmark, where they estimate that the 55 Somali-Danish associations marked them as having the highest number among ethno-national associations in Denmark in 2001 (Hjære and Balslev 2001, 15)70. Hjære and Balslev, however, do not present a distinct profile of Somali migrant associations in their survey. This high number of Somali associations is also found in other western countries, such as Norway (Assal 2004, 108-114), Finland (Saksela 2003, 254), Sweden (Johnsdottor 2002, 39), and the UK (Hopkins 2006; Harris 2004; Grif- fiths 2002), and Toronto (Hopkins 2006). In the UK, where most research has been carried out, Somali associations were reported to be high in numbers, with more 100 in London alone (Harris 2004, 66). Both David Griffiths (2002) and Gail Hopkins (2006) have examined Somali associations in London as part of larger fieldworks, and both point to difficulties of fragmentation along the lines of clan affiliation (Hopkins 2006, 371; Griffiths 2002, 173-175), resulting in a general problem of invisibility and low impact. In the words of Hopkins:

Although the high number of Somali organizations in London might be argued to be the Somali strategy of gaining wider representation for themselves, wider representation has not been the outcome. Instead, the presence of numerous small organization reflects clan divisions and has resulted in a lack of united direction and objectives amongst Somalis which acts against wider representation” (2006, 370-371).

70 In 2001 the centre carried out a survey of ethnic minority associations. Out of the 55 Somali-Danish associations the centre approached, only 11 replied, whereas the centre received a reply from 17 Bosnian and 14 Turkish associations (Hjære and Balslev 2001, 57). It seems to be extremely difficult to get an overview of the exact number of existing associations.

157 Griffiths, however, cautions against believing that clan tensions are the only cause of division, as they may “disguise other, more narrowly economic or political considerations” (2002, 175) or, as Hopkins also suggests, the many small associations may reflect a wish to create employment opportunities (2006, 371). She further compares the situation in London to Toronto, where Somali associations have been more organized and coordinated since the establishment of an umbrella organization in 1995 (ibid., 373), thereby furthering their political visibility and success in obtaining funding (ibid., 376). As we shall see below, Somali-Danish associations have similarities to the situation in both London and Toronto.

Fiery souls: A history of Somali-Danish associations In the following section, I retell the development of Somali-Danish associations, based on an interview with the coordinator of the network, Mohamed Gelle. Up until the middle of the 1990s, when a growing number of Somali asylum seekers, refugees, and family re-unified started to come to Denmark, there were only very few official associations in Denmark71. In the middle of the 1990s, then, Gelle and a group of other educated men and women from all over Somalia – “fiery souls with the heart in the right place”, as he characterized them – formed the Somali Resource Group as a response to the humanitarian catastrophe in Somalia. The Somali Resource Group cooperated with Save the Children, and later with the Danish Refugee Council (DRC) as well. Coopera- tion between these NGOs and Somali-Danish associations has continued, as has the associational engagement of several of the members of the Somali Resource Group. In the middle of the 1990s, as more and more Somalis arrived and settled in Denmark, the number of Somali-Danish associations started to grow as well. Associations were founded in the largest cities and soon “in all provincial towns where Somalis arrived, they started to organize themselves”, Gelle explained to me. These associations were organized around culture and religion in order to

71 In 1987 ‘The Somali Association’ (translated from Danish, Den Somalisk Foren- ing) was established as a cultural association ‘against loneliness’ and ‘for justice’ (Mikkelsen 2007, chapter 6). Two years later, there were two registered associations which had applied for economic support from the Ministry of Social Affairs. In 1999 the number had grown to 12. The latter applications were directed to the Ministry of Interior (Mikkelsen 2003a, 116).

158 create a mosque, a place of prayer, and a place to educate children and youth. Later, new kinds of associations appeared, focusing on support for various kinds of development projects in Somalia and on remittances. These transnational associations tended to be based around one particular locality in Somalia, and shared kinship affiliation. In the late 1990s, Gelle recalled, this developed into a situation where “there were so many associations that they started to oppose and compete with each other”. The introduction of forced dispersal of immigrants and refugees in the Integration Act of 1998 probably also made the number of associations proliferate, as Somalis in integration programmes were dispersed outside the larger cities. In order to further cooperation, the Somali Resource Group and the DRC arranged a big conference where the Council of Somali Associations in Denmark was established with the aim of coordinating the associations and to ‘speak with one voice’. As many as 18 Somali-Danish associations became members because, as Gelle explained, “at that time, there were a lot of writings about the Somalis in this country, so the will was there. It was necessary to have one voice”. Gelle was elected as the chairperson of the Council. This new organization, however, could not control the still growing number of associations, and the situation seemed to repeat itself.

We discovered that every time we came to an assembly or that we should take a stand on something like integration or that women need to educate themselves and work, then there might be one of these very religious associations in the Somali Council, which would oppose to it. That means we discovered that there were hidden agendas, and we started to suspect each other.

In 1998, while this process was going on, Gelle was employed as a seminar coordinator by the DRC. The DRC had obtained a grant from the Ministry of Inte- rior to support a “pilot project of dialogue- and network creating activities for Somali refugees in Denmark” (DRC 2000a, 1). In 1999 a range of seminars tar- geting different groups and issues within the Somali-Danish group was arranged by the DRC (DRC 2000a; 2000b), and the idea of a Somali network was introduced and established with Gelle as a coordinator. When the pilot project ended, the DRC received further funding for the network activities, first from the Ministry of Interior and later the Ministry of Refugee, Immigration and In- tegration Affairs (or, in short, the Ministry of Integration). Gelle, however, emphasized the continuity with and embeddedness in the hitherto Somali-Danish

159 development, where the network had been established as part of an ongoing democratic process and the work of the Somali Council. At the time of my fieldwork in 2003, there were thus a large number of So- mali-Danish associations all over Denmark. The following list was found at the then homepage of the Somali Network in Denmark in June 2003:

The Council of Somali Associations in Denmark72, Somali-Danish Friendship Committee, Forum for Toleration, Unity and Reconstruction of Somalia, Somali Women’s Associations in Copenhagen, the Somali Culture Association, the Somali Friendship Association, Union of the Somali Nation in Denmark, the Somali Association in Denmark, Somali- Danish Friendship Association in Nykøbing Falster, Multi-ethnic Friendship Association in Maribo, the Somali Association in Hobro, the Somali Women Association in Mundelstrup, Somali Family Association, SORA, Nolay’s Relief Organization, Somali Community, Somali Women’s Association in Århus, Somali Association in Holbæk, the So- mali Association in Esbjerg, the Somali Association in Kolding, the So- mali Radio and Media Association in Kolding, Somali Media Associa- tion, Somali Radio in Copenhagen, Radio Dalo, Somali Radio in Hern- ing, DanSom TV (www.somalnet.dk/dansk/forening.htm 2003)

The 31 associations were located all over Denmark and were of different scope and size: from the nation-wide umbrella organization of the Somali Council through to very small provincial town-based clubs. Indeed, the number of associations on the list is quite impressive. Still, it was far from complete, Gelle explained to me. On the other hand, many of the associations might not exist any longer or only ‘on paper’, while others have been established in the meantime (cf. Hjære and Balslev 2001, 15). The Somali-Danish associations formed, in other words, an associational landscape in motion. There are at least two points to be made here. Firstly, that both Somali and Danish actors – including associations, individuals, Danish NGOs and Minis- tries – were involved in this development; and it must be seen in relation to the Danish opportunity structure and within the institutional context of funding and support. Secondly, that the account of the development from the Resource Group to the Somali Council and finally the Somali Network, as I present it here, is a positioned account. In this account, the associational development is articulated as a continuous struggle for establishing a pan-Somali body that includes Somalis from all over the Somali Republic, no matter what their regional, political or clan-related affiliations might be. Indeed, the high number of

72 I have translated most of the names on the list from Danish.

160 associations in Denmark and London has similarities to the clan-based fragmentation in Somalia in times of political crisis73. According to Gelle – and many others – the attempt for unity must be seen in the light of the proliferation of associations and the increasing tensions and divisions. Furthermore, it was an attempt to improve the exposed and contested situation for Somali-Danes, who therefore needed, according to Gelle, to act and speak with ‘one voice’. Follow- ing Taylor and Honneth, we might say that one aspect of the proliferation of associations and the attempt to unify them could be seen as a struggle for recognition, rights and respect for Somali-Danes in response to harm and misrecognition in the public, media and political spheres in Denmark. However, as Gelle also expressed, the struggle for unity was accompanied by diversity, fragmentation, and sometimes antagonism. To speak with ‘one voice’ might, in other words, have been in contrast with another desire, namely to be heard for one’s specific needs and wishes. Still, in spite of troubles and divisions, associations seemed to remain an important way of organizing in Danish society.

‘In Denmark, you need to organize yourself to be heard’: Gelle Other Somali-Danish key persons also shared the positive assessment of the integrative potential of associations. “Associations are something particularly Danish and, internationally, you call Denmark the country of associations”, I was told by Musse Sheikh, who continued that he and other key organizers were outspoken proponents of organization and always told people to organize themselves. Likewise, Gelle explained to me that one of the reasons for the proliferation of Somali-Danish associations was that “people hoped they would be heard, that they would be included in the processes of decision-making, that they, at least, could be a part of a dialogue”. In their perspective, then, associational engagement constitutes a way towards influence, political participation

73 Clan-based fragmentation has not only appeared due to the civil war. In 1969 – the year when Siyad Barre seized power in a military coup – 64 political parties ran for election in Somalia, at that time a country with four million inhabitants (Samatar 1991, 17).

161 and attaining a voice as a particularly Danish strategy74, which had to be learnt to get on in Denmark. In spite of the apparent enthusiasm for organization, Musse Sheikh and Mo- hamed Gelle both expressed that a feeling of ‘it does not matter anyway’ had started to manifest itself and that it had become more difficult to organize people over the last few years. The reasons, they said, both related to practical and economic obstacles, but also the realization that associational engagements as a strategy of participation and inclusion did not have the intended effects. In the words of Gelle:

The spirit of organization has started to fall a bit […] The spirit to develop something and do something big is no longer there, people are tired. We have always told people that in Denmark, you need to organize yourself if you want to be heard. Denmark is a country of associations and in that way you can get a voice and defend yourself. But I think people realized that it does not work anyway, nobody will listen anyway, and the smearing campaign does not stop anyway, so it does not matter.

Both Gelle and Musse Sheikh emphasized that the feelings of ‘it does not matter anyway’ and that ‘it has become more difficult’ were widely shared in Dan- ish society and were not, as such, particular to the Somali-Danish associations. Still, in spite of the possible frustrations of not being heard, the Somali-Danish associational engagement was impressive. This raises the question that if it ‘does not matter anyway’, why did Somali-Danes keep on organizing themselves? I suggest that associational engagement was not only a strategy to participate in Danish society, but was also related to negotiations of positions among other Somalis – in Denmark, Somalia and Somaliland as well as in other western countries. In order to explore this aspect further, I now turn to an analysis of the social principles of organization75 guiding associations; i.e. the modes of differentiation that were mobilized and articulated as central to associational involvement. Together with the history of the development of associations, this

74 As indicated above, to organize oneself in an association is not a particularly Danish strategy, as the high number of Somali associations in other countries also bears witness to. Still, the idea that associations are a particularly Danish phenomenon is often articulated in Denmark (see for instance Ministry of Integration 2003). 75 The term social principles of organization stands in contrast to the formal principles of organization, such as statutes, executive committee, and the economic principles in associations like membership fees etc.

162 dimension is another central component in my outline of the conjectures of a Somali-Danish associational field.

Social principles of organization An analysis of the social principles of organization sheds light on the understanding of processes of inclusion and exclusion in relation both to Danish society and transnational engagements. This is important, not only to avoid generalizations regarding the nature of Somali-Danish associations, but also to challenge methodological nationalism and move beyond the logic of the container state society, where transnational relations tend to be ignored (cf. Wim- mer and Schiller 2002; Beck 2000). Furthermore, I want to highlight the contextual and intersectional nature of the social principles of organization: how different modes of differentiation are mobilized and articulated in a number of ways, sometimes reinforcing, sometimes overshadowing each other. Still, I will posit that the social principles of organization cluster around two overall themes: culture and the body as well as kinship and politics. While the former especially took its departure in notions of gender, the latter related to questions of lineage and state formations.

Gendered and religious spaces To the Danish observer, one of the most striking features of the Somali-Danish associational scene was gender division. Though both women and men established associations or participated as ordinary members, the majority of them were dominated and frequented by men in their thirties, forties and fifties. This gender and age feature, however, is not a particularly Somali-Danish phenomenon as such, but a characteristic shared with other ethno-national associations (Mikkelsen 2003b) as well as with Danish voluntary associations (Koch- Nielsen and Clausen Dalsgaard 2002)76 – especially in relation to political work. Likewise, many Somali associations had gender-divided activities, where men had access to club premises one day a week and women another day. Furthermore, while many Somali-Danish associations were frequented either by men or women, I only heard of women’s associations and a fathers’ group

76 In Denmark, however, an equalizing development can be found when the 1990 gender distribution is compared with 1999 (Koch-Nielsen and Clausen Dalsgaard 2002).

163 which explicitly characterized themselves as gender-divided. In this sense, women (and fathers in relation to parenting) seemed to be the exception from the norm, specifying their gendered particularity. We cannot, in other words, go from an observation of gender division to claims of ‘dominant men and oppressed women’. Certainly, claims of male dominance as natural, normal and inevitable would not remain uncontested in Somali-Danish associations. Fur- thermore, gender division worked both ways – placing both men and women out of place in specific contexts and privileging them in others. Such mecha- nisms are well known in Denmark and other western countries. Rather, we may say that there are gendered ways of organization and expectations of gendered behaviour in both Somali as well as in Danish society which may reinforce each other, not necessarily as a conscious strategy, but as more or less explicit discourses of the nature of women and men, thereby circumscribing or affecting notions of gendered space (cf. Massey 1994). Indeed, there were numerous examples of implicit gendered social space. In a Somali-Danish associational context, many social clubs are not termed as ‘men’s clubs’, but in practice, they seemed to be frequented exclusively by men. A Copenhagen branch of a Somali money transfer company, for instance, that I visited quite a few times, functioned as an informal social club – or an extra living room, as the manager said – for Somali-Danish men living in the neighbourhood. There were often several men in the place, drinking tea, talking and praying, and while women used the money transfer facilities, no women (but the curious researcher) were hanging out. After a while, my main contact in the place suggested that we relocate our talks to a local cafeteria where it was ‘calmer’; perhaps to get me out of the place, perhaps to talk in a more private space where no other Somali-Danes would be listening. Likewise, many more formal associations had all-male executive committees and all-male representatives, even if they were not considered to be ‘men’s organizations’. With the obvious exception of women’s associations (where only women were members), most of the associations I came across were similar. According to the (male) informants, this was not an explicit strategy. In the words of Ali Hersi, a university-trained associational organizer, “it is not like that there is no admit- tance for women […] It is because it is always the men who take up the big issues, because they pay and they organize it”. In other words, the gender division might not be deliberate, in the sense of purposefully excluding women, but

164 rather be a result of cultural and practical impediments which reinforce each other. Likewise, there is a high degree of gender division in Somali society (cf. Kusow 2003). One aspect of gender as a social principle of organization thus concerned gendered space, social networks and time schedules. As Ali Hersi further explained to me, the decision-making and coordination of associations had an “invisible, mobile organization”, requiring a flexible time schedule and personal mobility. This kind of organization makes it more difficult for women to participate, Ali Hersi explained, continuing:

It is the men who go the neighbourhood with all the different clubs, the men who have the time to hang out in the clubs; the women do not have this option. They can meet and chat of course, but not as much as the men.

In Ali Hersi’s analysis, the associational organization thus had a gendered time and space economy. Because much of the organization’s work and planning was carried out in informal, sometimes spontaneous meetings, taking place in clubs that women did not and were not expected to frequent, women’s participation was more difficult, relegating them to chatting at home. It was, Ali Hersi said, a question about networking; networks that women may not have access to – qua their gender and qua their often rather extensive childcare and housework responsibilities and thus their less flexible everyday life. Let me emphasize, however, that while both men and women invoked gender division in different contexts and for different purposes, other men and women opposed what they thought was coercive gender division. The point is that ideas of Somali culture might be invoked and mobilized to legitimize, prescribe or explain women’s and men’s behaviour, but did not determine it (cf. Benhabib 2002). Likewise, Is- lam77 was often invoked as a reason for gendered behaviour to explain or impede women’s or men’s participation in certain activities as (in)appropriate or (im)proper. Like ‘Somali culture’, religious prescriptions acted as a repository of (potential) organizational principles, mobilized to define and regulate proper gendered behaviour relating to who could and should participate in which activities, how they should behave and dress, and what activities were suitable to

77 Let me emphasize that my reflections about Islam do not take dimensions of spirituality into perspective, but remain limited to how Islam was mobilized in associational involvement.

165 be debated or practised by whom. According to some key activists, the degree of religious devotion resulted in outspoken power struggles between the religious liberals and the religious conservatives where gender was a key theme. These struggles concerned which issues were appropriate to discuss with whom, and whether meetings and seminars ought to be gender-divided as ‘the conservatives’ insisted, but ‘the liberals’ would reject. Seminars about youth and education, for instance, where both men and women discussed issues like falling in love, contraception and sexual relationships, were reported to have caused an outcry from ‘the religious’. One man argued that the religious dimension had become central for many people because of the big losses suffered in the civil war (cf. McGown 1999, 208). However, he also suggested that the influence of the Imams was diminishing, as Somali-Danes are getting more and more used to life in Denmark. Still, this man emphasized the huge influence of the religious leaders:

They exercise the religious dimension because they say that they communicate the words of God and all what they say is always stressed with a verse from the Koran or the hadiths that gives it a certain credibility and a certain weight and then it is easier to control people’s behaviour, to control them socially, their opinions, how they dress, how they behave, all these things. On the other hand, they can also quickly raise money to support projects in the homeland. At the end of the day, it is easier for them to mobilize people when they have a religious message.

Several other people mentioned overt social control as a part of the religious community. Other informants, however, emphasized the spiritual guidance of Islam and support of fellow Muslims in their everyday lives. Almost all informants included in the study were religious to varying degrees, from ‘liberal’ to devout. Likewise, several stressed their respectable family backgrounds, religious knowledge and practices, and decent lifestyle. Still, none of them were among the most conservative or traditionalist. On the contrary, several of them contested conservative gender and religious ideals, and associations were sometimes employed as a means of challenging such notions, while others used Is- lam as an active frame of reference – sometimes in the struggle for women’s rights, claiming that it is not Islam, but cultural interpretations of Islam, which subordinate women. There was, in other words, no all-compassing, fixed and decisive interpretation of Islam, Somali culture or gender. Religious and cultural prescriptions and

166 understandings varied according to context and to who interpreted them. Cer- tainly, there were powerful discourses, practices, and authorities which were claimed to be correct and true but, to be sure, others disputed many of these as- sertions. Still, the centrality of gender as a principle of organization was striking and relates, I think, to a perceived double nature of gender as both natural and cultural. This is not a theoretical statement in favour of the ‘sex-gender’ division in some gender studies (e.g. Scott 1986). Rather, I want to propose that gender holds a double potential for mobilization, referring both to embodiment and culture, which can be invoked to sustain or challenge social hierarchies and order. Or in other words, to practice, articulate, prescribe or invoke ‘Somali’ or ‘Muslim’ gender relations, ideals can be seen as (an aspiration for) cultural intelligibility through the exclusion of some positions and activities and the inclusion of others.

Clan affiliation In addition to gender and (degrees of) religiosity, lineage or clan affiliation was the third mode of differentiation that almost all of the informants articulated as a constitutive part of Somali-Danish associational engagement. Like gender, lineage was also perceived to be both of an embodied and cultural nature. On the one hand, Somalis are born into their father’s lineage and both women and men follow it (in principle) throughout their lives. In this sense, clan affiliation is a congenital mode of differentiation, which is so to speak located in ‘the body’, holding powerful potential for mobilization. In addition, the lineage system has both gender and generation dimensions. As a patrilineal kinship system, only fathers pass on their lineage to their children. Furthermore, traditionally, adult and especially elderly men possess a special position as negotiators and decision-makers in what Tina Kallehave has termed the ‘clan state’ (2003), i.e. social and political arrangements determined by the Somali lineage system. Women’s lineages, however, are also considered highly important, for instance in relation to alliances between different clans through marriage (see Lewis 1994). According to one Somali man, women are “the social glue keeping So- mali society together, while the men are diffusing it”. Furthermore, he continued, women’s clans are invisible, but crucial to know in order to understand Somali politics and alliances.

167 On the other hand, lineage is also widely articulated as ‘part of the culture’ and as a political issue: a flexible system of belonging and political allegiances, as we have just seen, where relatedness, distance and closeness change according to context (cf. Luling 2006; Lewis 1994, ch. II). Clan is often related to issues of loyalty towards specific places and regions and, therefore, specifically after the civil war, with political orientation. As mentioned in chapter four, the civil war has caused a reterritorialization of fields of belonging where the place of upbringing and clan affiliation are sometimes in tension (cf. Barnes 2006), and where clan affiliation tends to be more powerful for reasons of security. There was widespread agreement among the Somali-Danish informants that both clan and regional affiliation structured and defined many associations although, in practice, regional affiliation would often be defined by the logic of clan. Other associations, however, were or aimed to be pan-Somali. Clan thus constituted another central social principle of organization, which worked along and across gender and religious devotion in different ways and at different levels of cooperation. The organizational continuum ranged from the inter-clan to the sub-clan level. Inter-clan organization refers to associations which include members from different clan families, for instance both Isaaq and Darod; this level of organization might be regional or pan-Somali, depending on the level of inclusion of different clan families. Sub-clan organization, however, is based on one sub-lineage branch to one of the six major lineage families, for instance the Habar Yonis or Habar Awal sub-lineage branches of the Isaaq clan family. Due to the flexible nature of the lineage system, however, sub-lineages can also be divided into further sub-sub-lineages (causing many a joke about different sub-sub-sub-sub-sub-sub-clans). Thus, while sub-clan associations might be organized around just one delimited lineage branch, they might also include a wider branch of kinship related families, thereby attaining a regional basis as well. Some associations were explicitly based on lineage affiliation, whether on the inter- or sub-clan level, while the organization was less explicit or unarticu- lated in other ones. In the first case, different positions in the executive committee or other organizational boards were rotated or divided between different clans. This principle might be used both to overcome accusations of clan dominance, but also – according to some – to cover up such dominance. A member of a Danish sub-branch of a Somali political association, for instance, reported

168 that while committee positions were explicitly divided between different clans, members of the clan that had initiated the association always preserved a majority of the committee positions anyway. Most often, however, clan was not explicitly articulated as a deliberate principle of representation in Somali-Danish inter-clan or pan-Somali associations78, which included – or aimed to include – many or all lineages. In these and other cases, clan was rather articulated as a non-issue that did not or should not need discussion, at least not with outsiders (such as a curious Danish researcher). Many associations seemed to have an explicit or unspoken ‘politics of silence’ where political questions related to the civil war and clan affiliation were prohibited to discuss “to avoid fanning the flames”, as one man expressed it. Finally, many associations claimed that lineage did not play any role in the organizational structure at all. This was how many informants initially described their method of organization (cf. Hopkins 2006, 371). I do think that the principle of no-clan organization was an ideal for a number of associations, and it is entirely possible that many Somali-Danish activists never gave lineage a second thought. Indeed, several young women explained that for them, clan affiliation means very little. In their opinion, young people care less about lineage affiliation: it is mostly the concern of elders and the more politically interested. Still, as practically all informants claimed that clan was important, either for themselves or for other Somali- Danes, I will argue that lineage does matter in associational life. Not necessarily for everybody and not in the same way, but as a social principle of organization which could be mobilized, and which indeed often has been mobilized. As the different principles of clan organization indicate, there were very different assessments of the potential and nature of clan. While some perceived lineage as a constructive tool (captured in the notion of heer as a social contract or treaty (Lewis 1994, 84; Samatar 1994a, 109)), others thought of it as related to civil war and distrust – as clannism, an impediment for peace and cooperation (for a further discussion of these aspects, see Luling 2006; Samatar 1994a). In this view, lineage orientation fixed and delimited loyalties, restricted cooperation, and thus more ambitious and effective organizations. For others, it was

78 Lineage representation seemed to be more accepted in Somali contexts. The transitional federal government (TGF), elected in Nairobi in October 2004 as well as the Somaliland guurti, the House of Elders, are both examples of institutionalized lineage representation.

169 an efficient cultural tool, which could be of great help in coordinating associational work. Ali Hersi, for example, was very engaged in several associations and supported, among other things, a sub-clan reconstruction association in the area of Somalia where his sub-clan dominates. The sub-lineage organization, he explained to me, was simply the easiest way, because “you know who is who, all clans know their own people, so you don’t have to be formal to each other”. For Ali Hersi, then, the lineage structure was a kind of tool that can be mobilized and that can build on existing relations of trust.

It is built-in trust, it is so easy. I know it is my family; it is my clan, my sub-clan. If somebody wants to cheat, he will have to cheat all of us and then he is finished, his reputation will be destroyed. But, on the other hand, there are always excuses like, ‘you argue against me because you are from another clan’, and things like that […] But I don’t support a certain clan or the clan system as such. Clan for me is culture, identity; it is something we are embedded in. We cannot reject that. It is wrong to say that clanship destroyed Somalia. It is the abuse of clan that destroyed Somalia. That means clan is a tool, and it became a political tool which was abused. It might as well have been used positively to build the country.

In this understanding, clan is a culturally available tool which is easy – and necessary – to employ in organizational work because of the in-built trust and social control of sub-lineage relations. This aspect might become even more articulated in the absence of state authority. The quality of immediate trust is perhaps one of the most potent – and dangerous – features of lineage as an associational principle; it has an ambiguous potential and can be both a destructive and constructive organizational tool. The trust of one’s own lineage can be – and, according to many informants, indeed was – coupled with distrust of other lineages. In this sense, the lineage system had elements of both bridging and bonding social relations, to use Putnam’s (2000) concept. Given the manipulation and exploitation during the years of Siyad Barre, and the atrocities committed in the name of clan during the civil war, this ambiguity is not surprising. As Ali Hersi indicated, clan works as a generative principle of loyalty and trust, sometimes turning into a principle of exclusive and predetermined loyalty – that “‘you argue against me because you are another clan’”, as he said. In this logic, lineage organization was seen to produce antagonistic relations, trust as well as distrust, loyalty as well disloyalty. One man went so far as to say that the civil war had hit Copenhagen and that the Somali-Danish group was deeply divided

170 between the clans. “It is an oral war where people fight to get their share”. In the same way, a woman who had lived in Mogadishu during the civil war referred to the war when she explained that many people do not trust each other, and this is why, she further claimed, there were so many different associations.

We still have this anger within, at least most of us who have experienced the war. We have difficulties forgetting and we have difficulties trusting one another. When you leave, you think, ‘where did she come from’ and so on. I think it is things like that which mean that it takes a long time to make us a group. That is why there are so many associations in this country, perhaps hundreds of Somali associations. I think they are based on clans; that is my opinion.

In this woman’s opinion, the main reason for the proliferation of associations was thus due to fragmentation and unresolved hostility, pointing to continued conflicts and distrust in the aftermath of the civil war. This point also reminds us that Somalia – in contrast to many other post conflict societies – has not gone through a successful national reconciliation process, supported by the international community79.

Territorialization of loyalties Another aspect of lineage as a social principle of organization concerned the territorialization of loyalties. This problematic was often both connected with clan affiliation and the question of ‘origin’ – the tension of u dhashay – ku dhashay (Barnes 2006) – of being born to a lineage and in a place. For instance, while most many people of the big Isaaq lineage family (the dominating lineage family in Somaliland) appeared to be in favour of Somaliland, I did talk to a few Isaaqs, who did not share this opinion. Likewise, although many of the informants from other clans appeared to favour a united Somalia, there were still exceptions. Territorialized loyalties were often organized around lineage logic, whether a region or even a neighbourhood, Still, as we shall see in chapter seven, the relationship between regional and clan identification and loyalty was not unequivocal, but subject to negotiation and contestation. The division between proponents of a united Somalia and political recognition of Somaliland as a nation-state was one of the most central political dis-

79 As mentioned in chapter four, Somaliland went through several peace conferences in the 1990s, organized and supported by local authorities as well as by So- malis abroad – but without international support.

171 tinctions for the informants. It concerned the very territorial and political exten- sion and definition of the Somali Republic, and most of the informants had strong opinions regarding it. Depending on political observation, this conflict either dated back to the time of colonization or to 1991, when Somaliland se- ceded from Somalia. No matter which way, the conflict had been exacerbated in relation to associational development during the later years. This became clear when I returned to some of the informants in 2003 that I had also interviewed in 1999: Karim Suleiman, for example. In 1999 Karim was very engaged in a pan- Somali organization and active in a regional association as well. In 1999 he explained to me:

In the daytime, we are all Somalis, we hold meetings and talk together and we have the Somali Council, and we have the associations and the chairmen and all that … but at night then people meet in small groups – not just in their clan, but in the sub-, sub- sub-clan, right. They discuss their internal or private problems. So it does mean something, but we are good at hiding it and we are also good in creating an understanding for our rules … we have a paragraph stating that we cannot mention the political situation of Somalia in our discussions: It is forbidden to discuss politics to avoid fanning the flames so to speak, like if we should help here or there. So we aim to help our members locally here in Denmark, and then the individual associations take care of the local [Somali] interests. And I am very involved in a specific region of Somalia, but still I am the spokesman for everybody and not just for some.

In the 1999 interview, Karim Suleiman, strongly emphasizing that he was speaking on his own behalf in the interview and not for the pan-Somali association, articulated his position as one able to represent all Somali-Danes as well as being in favour of Somaliland and engaged in regional politics. While the pan- Somali association employed a ‘politics of silence’ towards political questions in Somalia in order to act with one voice and improve the situation for Somali- Danes ‘here’ in Denmark, local associations focused more explicitly on specific parts of Somalia or Somaliland. In 2003 this situation had changed and Karim Suleiman was no longer active in pan-Somali associations, but explicit in his political opinion in favour of the region. Furthermore, as the fragmentation of the Somali associational landscape indicated, not everybody believed in the idea of a pan-Somali body even in the late 1990s. Abdullahi Hussein, a former activist in the Somali National Movement, SNM, and an outspoken proponent of Somaliland, told me how he had first believed that all Somalis should help each other after Siyad Barre’s downfall in 1991. Abdullahi did, however,

172 quickly lose faith in pan-Somali cooperation – whether in Somalia or in Den- mark. “I agree that Somalis should come together with their problems in Den- mark”, he explained to me in 2003, “but to make an umbrella organization that can run the interests of all the Somalis – it cannot work. It couldn’t survive”. For Abdullahi Hussein, the idea of pan-Somali associations was unviable from the beginning and he contended that there were no longer any ‘real’ pan-Somali associations in Denmark. In his words, ”it is not all Somalis; it is only a group of Somalis from a certain region or a certain clan. That is the reality. But it is hidden, it is not official”. Certainly, not all Somali-Danish activists would agree with this characterization, but many shared the disillusion of the possibilities of ‘speaking with one voice’. For another man, the development of the Somali- Danish associations had gone from unity to politicization in a way that reflected the political situation in the Somali Republic. He said:

In the beginning, there were no divisions at all. I don’t think that people were conscious if they were from Somaliland or the South […] It was after the creation of the Somali Council […] that it became critical, where the consciousness of Somaliland …entered the picture, not before. It was only then that I realized that some people from Somaliland wanted to … detach themselves, also from the associational work in Denmark, in the diaspora. But I don’t feel like opposing or cooperating, I don’t care, not at all […] Somalia was one Somalia for me, always one Somalia. Such a statement is an explicit political stance in favour of a united Somalia. To articulate Somalia as one Somalia in contrast to detachment and fragmentation was one of the central political distinctions dividing the informants. This meant that there were ongoing struggles defining what pan-Somali and Somali meant: if such notions encompassed all present or former citizens of the Somali Repub- lic, including people from self-declared Somaliland, or if they were excluded or chose not to participate for political reasons. It also made the criteria of success more difficult: insofar that there was a tendency for Somaliland proponents to be less involved in activities defined as pan-Somali80. A wide range of the informants – from different regions – supported the opinion that there was no

80 As I have not incorporated religiously defined Somali associations, I cannot assess if this division included religious association as well. However, several informants emphasized that religion was one of the issues uniting Somalis across political orientation (though, as we have seen, the degree of religious devotion and con- servatism was up for debate).

173 pan-Somali body that had the support of all Somali-Danes, if this category was to include both proponents of the Somali Republic and Somaliland. This does not mean that there were no Somaliland proponents who engaged themselves in activities and associations which were not explicitly in favour of Somaliland – far from it. But it did indicate that the position of representing all Somali-Danes regardless of their political orientation towards Somaliland and Somalia had become more difficult and contested than in 1999. Some, though far from all, Somaliland activists had stopped engaging themselves in the Somali Network or other pan-Somali associations. Sometimes they claimed their activities were not supported because proponents of a united Somalia prevented them with the reason that officially ‘Somaliland does not exist’. Proponents of the Somali Re- public, on the other hand, sometimes accused Somaliland activists of contributing to the disintegration of Somalia. On some occasions, both groups used rather harsh words to describe the ‘other side’. Most people, however, maintained that politics is one thing and everyday life another. Still, it remained that the Somali-Danish associational scene was politicized. To sum up, the informants came from all over Somalia and from most lineage families, and their political loyalties, lineage affiliation and regional orientation varied. While I do not have detailed knowledge of lineage affiliation for all informants and thus cannot offer any coherent picture of it, I do claim that no single clan family dominated the entire Somali-Danish associational scene (though some clan families may have dominated certain associations). Politi- cally, key activists were divided according to the question of Somalia and Somaliland and this is perhaps where the biggest differences between the main activists appeared. Most key activists shared an outspoken interest in the political development of Somalia and Somaliland, but it was mostly Somaliland proponents who articulated an explicit political engagement.

Continuums of identification Gender, Islam, lineage and political orientation were thus the four most outspoken social principles of organization in the Somali-Danish associational field. Together, these principles constituted a social space where meanings of ‘Soma- liness’, Islam, ‘proper’ gender and relations, integration, and opinions concerning the future of Somalia and Somaliland were negotiated as well as contested. This implies that Somali associations made up a field in Bourdieu’s sense in

174 that certain sets of conflicts and power dynamics were at play between the associations and associational key activists, structured around these social principles of organization. However, rather than a network of objective relations between positions, as Bourdieu (1992, 16) formulated it, these principles constituted continuums of identification. Thus, while some associations and actors had clear-cut positions, others were more contextual and relational, meaning that it would be difficult to situate them in an unambiguous way. Likewise, as we have seen, some key persons had changed political position during the last few years, further emphasizing the changeable nature of the Somali-Danish associational scene. The analysis of the social principles of organization thus serves as an overall framework for the further examination of struggles for recognition and negotiations of positions through associational engagement, not of the individual struggles and positionings. In the following case study, I turn to how the social principles of organization were enacted and negotiated in one Somali-Danish association that I here call the Women’s Association.

The Women’s Association During the late summer and autumn of 2003, I spent an afternoon once or twice a week at the premises of the Women’s Association, helping with Danish classes and other activities. I knew several of the organizers and key persons in the association from my earlier fieldwork in 1999 where they were already engaged in various kinds of activities and associations. I had met most of them at a seminar for Somali-Danish women, to which they had been invited, arguably because of their positions in the Somali community. At the seminar, several of the women emphasized the need to have ‘their own place’, some premises where they could meet, discuss and do things together. Some of the women established the Women’s Association the same year, focusing on women with an affiliation to their home region, i.e. the region where some of them had grown up or to which others were related to by clan affiliation. I met several of the women again at another seminar in 2001, this time arranged by the Women’s Association itself with Hawo Yusuf, the initiator and chairperson of the Women’s Association, as the main organizer. The focus of the seminar was on the resources of Somali women and the challenges they face in Danish society. However, when Hawo welcomed the participants, she also emphasized that the date of the seminar was a celebration of an important ‘national day’ in the home

175 region. The seminar thus both comprised discussions regarding integration and served as a long-distance nationalist marker. While most of the Danish seminar participants probably did not notice this, the celebration had a clear political message for the Somali-Danish participants. At the time of fieldwork in 2003, the Women’s Association had become an established association with its own premises in a community house and a range of activities. Not only was it oriented towards everyday challenges in Denmark, it also offered Danish classes, exercise activities, different kinds of practical courses such as baking and sewing, and advice on more general issues such as integration, prevention of girl circumcision and HIV. The Women’s Associa- tion had also succeeded in obtaining economic support for a local project in the home region and several of the key organizers were very concerned about its political development, appealing for political and economic support. It thus had both a Danish and transnational orientation. Being open to all women related to the home region, the Women’s Association was a gender-divided, inter-lineage association with a regional focus and a liberal religious position. It was organized by an executive committee, with statutes, had about 120 members and, apart from courses, it arranged social gatherings and public seminars, and celebrated important events in the home region. As in 1999 several of the organizers and executive committee members were still very active in other associations and activities. Knowing several of the organizers, my fieldwork in the Women’s Associa- tion was carried out along two tracks. One the one hand, I was active in the Danish classes in the association during summer and autumn 2003, helping out as a volunteer and observing the everyday practices. This gave me an opportunity to talk with the ‘ordinary members’ following Danish classes, and offered an insight into the challenges they faced in relation to housing, health, employment, the Danish social system, and education of children and youth. Chal- lenges concerning everyday life and inclusion in Danish society which were, in short, shared by many of the other informants – women as well as men. On the other hand, I interviewed and met with Hawo Yusuf as well as several of the other organizers and key persons in the Women’s Association throughout my whole fieldwork. These encounters were not necessarily related to the Women’s Association as such, but more often to other activities of the organizers, who were engaged in an impressive web of activities and networks – both in the

176 Women’s Association and in other associations. My fieldwork among the organizers was thus of a quite different nature to my observations in the Danish classes among the so-called ‘ordinary members’. In the following vignette81, I share some of my fieldwork observations during Danish classes in the Women’s Association.

Three afternoons in August and September Copenhagen, August 2003.

On my way to the Women’s Association, Awa passes me on her bicycle. Awa is one of the women who attends the Danish classes. She is hardworking, dili- gently studying for a Danish test needed to apply for Danish citizenship. She goes to language classes in the mornings and then supplements these with extra sessions in the Women’s Association a couple of afternoons every week. Awa wants to learn Danish to become naturalized, and to better communicate with the teachers in her children’s school and the parents of their friends. Therefore she immediately starts practising Danish as soon as she arrives at the premises of the Women’s Association and continues until she stops for prayer together with a few of the other women. While they pray, the other women drink sweet tea with milk and chat in Somali, taking a break from the Danish class. I talk with Amina. She is a divorced single mother with four kids. It is very difficult, she tells me, very lonely. Like Awa, she has joined the Women’s Association to improve her Danish, but also for company and to talk with other women from her home region. Many of the them face a range of practical problems in their everyday life in Denmark, problems, which may be reinforced due to their lack of knowledge of the Danish system, misunderstandings, misinformation and, sometimes, poor Danish skills. Amina for instance, has a problem with her flat. She has sublet it and now the contract has run out. She urgently needs a new place and does not know what to do. Maria, one of the Danish volunteers, promises to see what she can do.

81 The vignette is a reconstruction based on my field notes: translated, anonymous excerpts from my observations and the formal research conversations which I carried out. While I retell some of the discussions and things going on during Danish classes, I do this in general terms, changing personal details and all names in order to respect confidentiality and wishes for anonymity.

177 When the Danish class is over, I ask Awa if she wants to practise some Somali with me. Okay, she says. The Danish volunteers leave, while most of the So- mali-Danish women stay. We practise a few basic sentences, much to the amusement of the other women. They chitchat, drink more tea, use the phone and surf on the Internet. One of the women holds up some new clothes in front of her, a yellow petticoat, and mimics a catwalk, making the other women gig- gle. We leave after an hour or so. On my way home, I pass a group of the women waiting at a bus stop, carrying heavy bags filled with groceries.

Copenhagen, August 2003.

I talk to Amran Nazir, who used to run a shop in Somalia. She has only had little schooling and fights with the difficult Danish pronunciation. Furthermore, she, like some of the other women, suffers from physical disabilities, which make the chores of everyday life as well as learning Danish difficult. I ask Am- ran to translate some words for me, but she has a hard time concentrating. She is thinking, she says. ‘Of what?’, I ask. ‘My hometown …and things like that’, she replies.

After class, I join Noora Hassan. Noora is a skilled Internet surfer, often using the computer in the corner to browse the many Somali homepages for news and interesting stuff. Enthusiastically, she shows me a range of useful homepages, including a Somali-English translation service and several Internet photo ar- chives from Somalia. She points out places she used to hang out in when she studied in Mogadishu many years ago. Noora readily discusses political, social and cultural affairs in Somalia and Denmark. In spite of a lifelong career and fluent English and Arab language skills, she has not been able to find employment in Denmark and now uses her skills in the association. She tells me which books to read and instructs me not to leave out the more thorny issues which I may encounter.

Copenhagen, September 2003.

It is the last session before I go for two weeks to Hargeisa in the self-declared Republic of Somaliland. There is a lot of excitement about my trip and one of

178 the women arranges a scarf around my hair. ‘Oh, now you look beautiful!’, she exclaims with a loud laugh. Amran is visibly nostalgic, talking about camel milk and other Somali delicacies. One of the women has asked me to take a pair of Ecco shoes, a comfortable Danish shoe brand, to her brother living in Har- geisa. Another is worried about some siblings, from whom she has not had any news for a long time. Mulkki, a board member and student nurse, arrives. She is dressed in a beautiful outfit: a red scarf with golden flowers arranged like a turban and a black shawl with fine prints wrapped around her shoulders. She is wearing perfect make-up and smells of sweet perfume. Mulkki bought the clothes during a holiday in Somalia last year and I think to myself that I will have to do some shopping in Hargeisa. When I leave, I get warm hugs from everybody and I promise to take lots of photos to show when I come back.

Dealing with everyday challenges In October, then, after spending a couple of afternoons each week at the Women’s Association, I went on a two-week field trip to Somaliland. My visit created a lot of interest and excitement among the women and I expected that it would be a turning point in relation to my fieldwork in the association. It some ways it was. After my return, I intensified my conversations with Noora, now including discussions about clan relations and, given my new insights into the state of affairs in Somaliland, more detailed talks about the political situation. Still, only few of the other women were interested in participating in my project. While they were exited about my trip, looking at photos and compliment- ing me on my appearance in Somali clothes in Hargeisa, most of them were not interested in participating in any interviews. When I asked, some simply said no, others agreed, but due to practical obstacles our encounters were either can- celled or postponed. From one perspective, the refusal and hesitation of the members might seem to be a reflection of traditional and conservative gender ideals that some of the ‘ordinary members’ in the Women’s Association were subscribing to. Indeed, one of the Somali-Danish organizers characterized some of the women as conservative and reserved. But as I have earlier argued, we cannot isolate one single factor such as gender, but must also include other modes of differentiation as well. This analytical demand was underlined by the fact that it was relatively easy for me to gain rapport and conduct lengthy interviews with both male and

179 female organizational key persons in other associations. Another explanation of the hesitation might be that the women chose to protect their privacy in a context where it was actually possible to do so – in contrast to, for instance, the social security office, the Immigration Service and other authorities. The massive negative political and media exposure may further contribute to insecurity about what it means and what personal consequences it might have to participate in a research project. In this sense, the insistence on not discussing private affairs can be seen as a way of establishing a place to breathe freely, so to speak. Again, such a place might be more urgently needed for women and men in more disadvantaged situations, dependent on social welfare and thus subjected to close contact with and control of public authorities. The protection of their privacy and the uncertainty of what it may imply to share personal details with other people can be seen as an aspect of everyday challenges in Denmark (cf. Møller and Togeby 1999)82. Furthermore, the pace of their lives was breathtaking: School or activity programmes; meetings with the social office; children to be taken to day care, school, spare time activities, the doctor, looked after when sick; parents’ evenings to be attended; groceries to be bought; relatives to be taken care of; social parties and events not to be missed. The fact that quite a number of the women in the Women’s Association were single mothers, had absent husbands or otherwise had total or the major responsibility for the children and housework also points to a gendered aspect where everyday life in Denmark seemed to demand even more time and attention from many women. The Women’s Association might thus not only be a place to breathe freely in relation to public authorities, but may also function as a centre for social gatherings with parties and social events where the women could get together. This social aspect, which was furthered with the access to their own premises, was an important function of the

82 In their report about discrimination among Somalis and three other ethnic minority groups in Denmark, Møller and Togeby conclude that there is a group of socially isolated Somalis with limited knowledge of Danish society and with poor Danish language skills, who feel highly insecure about their lives in Denmark (Møller and Togeby 1999, 78-79). While I do not think the women joining the Dan- ish classes in the Women’s Association can be classified as part of this group (as these women actually joined an association), elements of the overall insecurity might be shared. Obviously, the refusals to participate can also be seen as – simply – a refusal, not grounded in any social conditions. Still, when compared with other Somali-Danes, the difference in willingness to participate was thought provoking.

180 association and its everyday orientation. In a theoretical perspective, one aspect of the association can be seen as an arrangement of the production and reproduction of social capital – in Bourdieu’s sense of group membership (1986). In contrast to Putnam, Bourdieu localizes social capital on the individual and group level as social connections, which can be converted to other kinds of capital. He defines the concept as:

the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance and recognition -- or in other words, to membership in a group -- which provides each of its members with the backing of the collectivity-owned capital, a “credential” which entitles them to credit, in the various senses of the word (1986, 248-249).

Social capital as membership of a group or a network is thus a way of gaining access to other people’s knowledge and resources at the individual level; the sources of social capital are the network itself, whereas the effects are the array of material and institutional benefits provided by the networks (cf. Portes 2000, 4). This raises the question of group membership and, inspired by Putnam, the distinction between bonding and bridging social relations. According to Hawo Yusuf, the organizers were old friends, the members all knew each other and they were all related to the same region in some way or the other. In this sense, we may say that the social networks established in the Women’s Association had a bonding character. Still, as Putnam reminds us, the bridging-bonding distinction is rarely a question of either-or, but is a conceptual differentiation (2000, 23-24). Thus, while the ‘ordinary members’ joining the association might find new friends from the same region, the organizers, qua their activities and cooperation with Danish and other Somali-Danish associations as well as with several women and minority umbrella organizations, had even greater opportunities to enlarge their networks in a bridging sense. Furthermore, the Danish language training classes offered a venue where Danish volunteers and the members had an opportunity to meet and talk. There were also counselling activities during Danish classes, an opportunity which was both used by the Somali women practising their Danish as well as by other members joining specifically for receiving good advice or practical help. The counselling took place on an informal basis: sometimes the volunteers translated papers, or in more rare cases, they talked to the social security offices and

181 other authorities. In a few cases, the volunteers also helped out with practical problems83, sharing their knowledge and networks with the women – for instance when volunteer Maria tried to find Amina a flat. However, much to the frustration of the Danish volunteers (the researcher included), a number of the ‘ordinary members’ were not particularly interested in extended social interaction; whether they were too busy, too shy, or craving for a place to breathe without being asked personal questions. Still, it would be wrong to conclude that the Women’s Association did not further processes of social inclusion. Be- cause, as Putnam points out, both bridging and bonding networks are necessary. In the words of Putnam, “strong ties with intimate friends may ensure chicken soup when you’re sick, but weak ties with distant acquaintances are more likely to produce leads for a new job” (2000, 362-363). For people in marginalized situations, bonding networks might be crucial for managing the challenges of an extremely busy everyday life in the first place. In this sense, the bonding character of the Women’s Association may play a significant role in creating “specific reciprocity and mobilizing solidarity” (ibid., 22), which makes everyday life easier. As many of the members were unemployed single mothers, specific reciprocity might be necessary to cope with the challenges of everyday life and to avoid isolation, which indeed is an important aspect of processes of inclusion.

‘I was always a nationalist’: Noora The Women’s Association, however, was not only oriented towards inclusion in Danish society and social and cultural Somali-Danish issues, but also towards the Somali home region. The transnational orientation was reflected in the As- sociation’s development project in the region, the long-distance nationalist orientation as well as, on a more everyday basis, the eagerness of many of the women to get news from Somalia and their home region. While Noora Hassan definitely was the keenest Internet surfer among the women, others listened to the Internet version of the BBC Somali, a Somali-speaking BBC news programme. Generally, however, the women joining the Danish language training classes did not express any real interest in politics. Amran Nazir, for instance, explained that: “I come from there, but I don’t go into politics, just the societal

83 The volunteers emphasized that they only carried out informal counselling activities which did not require a professional counselling background.

182 issues. I am interested in the people, the society. I hear about politics, but I do not involve myself.” Still, when I asked her why she was a member of the Women’s Association, Amran wondered with surprise why I could ask such an obvious question. “I am from the region!”, she said, “I have many girlfriends in the Association”. When I further asked if she belonged to any other associations, Amran answered that she was a member of the Somali public. This is an interesting answer. Not only did Amran include both the social and regional orientation of the Women Association in her reply (and considered it obvious that she would join it for these reasons), she also related associational engagement to participation in a public sphere – even if she defined herself as uninter- ested in politics. While Amran Nazir and many other members only seemed to have a vague interest in politics, Hawo Yusuf and Noora Hassan were openly political. “I was always a nationalist”, Noora told me and continued, “My father told me that we should serve our country. I always think of helping my region and what I can do. I don’t think a lot of other women think like that, though there are exceptions”. For Hawo Yusuf, the home region and inclusion in Denmark were equally central issues in the Women’s Association. Therefore the association ran courses about Danish society for the members as well as aiming to prepare them for repatriation in the (very) long term. As repatriation is high on the Dan- ish migration policy agenda, this objective might not only be a reflection of a long-distance nationalist political cause, but also a strategic priority in relation to the possibility for funding in a Danish context. Still, unlike Noora, Hawo did not grow up in what she now termed her home region and only became interested in its political destiny later. Thus, when I asked Hawo if the ordinary members cared as much about the political struggle of the home region as she did, she explained to me that the idea and need to establish a regional association only appeared when she was positioned as an infidel.

Hawo: It’s funny, in the beginning we were from all over Somalia, but then they [from other parts of Somalia] started to gossip that we are infi- dels and all kinds of stuff and I realized that they said … that I am from this region … and then they started to say, ‘she does not go to the mosque’. But I told them that ‘I am very devout, I don’t have to go to the mosque’.

Nauja: Who said that?

183 Hawo: Almost always the women [from other regions], so then I started the Women’s Association. But there were not many people who believed in that idea, even one of my good friends did not believe in it, she went to another association. So another friend and I ran the Women’s Asso- ciation, and there was also an umbrella organization which helped me a lot. And when they [other women from the region] saw that our association was working, they all started to come.

Hawo Yusuf – a devout Muslim and an assertive, outspoken woman – thus articulated the regional orientation as an alternative to gossip and social pressure. She thereby pointed to a linkage between the religious and political principles of organization. Other informants, however, argued that the opposite was the case: political orientation and lineage did not matter in religious associations. For Hawo, however, this did not seem to be the case. Still, she did not reckon with the logic of regional affiliation but, on the contrary, subscribed to it, as only women from the home region were admitted to be members of the Women’s Association. The orientation towards everyday challenges in Den- mark, Somali social and cultural gatherings, and the transnational orientation were thus simultaneous, enabling the Women’s Association to position itself as both a potential actor in the home region as well as in Danish society. Firstly, forming a women’s association may create a space for action, articulation and recognition both in relation to the Somali-Danish associational scene as well as to Danish society. Given the already existing structure of gender division, organizing as women made it possible to be recognized as cultural intelligible actors – as decent women, not collaborating too closely with men – as well as potentially creating positions for speaking out and working in partner- ship with other associations and actors, which might have been difficult in those dominated by men. Furthermore, women’s associations might also be seen as a strategic response to collaborating with men who may (strive to) dominate the associations, not leaving much space and initiative for women. In this sense, women’s associations both respected cultural ideals of gender division as well as moving beyond the idea of politics as an exclusively ‘male sphere’. Further- more, in a Danish context, where ideas of oppressed refugee women are widely circulated and gender equality and women’s participation are articulated as highly important for integration, women’s associations can easily be recognized as holding an integrative potential, which is worth support. Thus, also in the context of Danish society, women’s associations might appear more culturally

184 intelligible. Secondly, forming a regional association could be seen as the establishment of a political actor as described above – as collective spokespersons with a capacity to speak for their ‘community’ and thereby potentially influenc- ing various important discussions and decisions as, for instance, when Hawo Yusuf and the other organizers collaborated with NGOs and went to international conferences, representing the Women’s Association. In this perspective, forming or joining an association can be seen as creating a platform for action to realize activities and negotiate positions in different social spaces.

Learning processes One of the things that struck me most when I returned to some of the informants I knew from 1999, was exactly how many of them had realized at least some of the visions they had shared with me back then through the means of forming or joining an association. Certainly, Hawo Yusuf and her collaborators in the Women’s Association were not the only ones. Almost half of the informants in this study had initiated their own association or played an important part in organizing and running one or more associations, and some of the most evident learning processes were related to such experiences. In the interviews, several of the informants emphasized their personal efforts towards realizing their ideas and the hard work it had required and often still did. Some of them also underlined the difficulties they had encountered – both due to lack of knowledge about establishing an association, but also due to misunderstandings from the members. An activist explained to me that “we are not used to dealing with associations. So it is like a learning process for us; it is like we are learning a new culture”. This ‘new culture’, then, could be said to be the political culture of voluntary associations and organizations. This relates to dealing with the re- quirements of regulations, statutes, annual general assemblies, executive committees, dues, fund raising etc. Furthermore, it can be said to relate to the practices of doing voluntary, unpaid work and organizing for a purpose, which may be more common and abstract, than contributions from the founders and organizers. To learn such practices and skills was widely articulated by Somali- Danish activists as a valuable and useful experience. Such learning processes were, however, both challenging and demanding. Take Fatima Omar’s example.

185 Fatima Omar had lived in Denmark for a number of years and knew a lot of people. Like Hawo, Fatima founded a Women’s Association in the late 1990s with the purpose of creating a forum for Somali-Danish women to meet and enhance their knowledge of Danish society. She therefore contacted a lot of women and invited them to her home, telling them that she had something to discuss. About 25 women showed up and Fatima told them about her own problems of dealing with the Danish school system, and suggested that they created a group to talk to teachers, to get information, and to inform the system about their views. The women told her that they thought it was a good idea, Fatima said and continued, “but they did not know where to start, so I came up with the idea and they wanted me to go on and they supported me. So I started the association, but at that time I did not know what to do or how to get funding”. Fatima Omar therefore decided to follow a course, started to collaborate with a Danish social worker, and finally sought advice from a large Danish NGO. They told her that she needed the addresses of the members.

So I went to about 50 women and I sat down with everybody, explaining why I needed their addresses. But I did not write down their date of birth. When I came back to deliver the addresses, they [the NGO] said, ‘you need their dates of birth’. So then I went back again and got 50 dates of birth and we made the association. Then somebody told me that the members must pay dues so that we can apply for funding. […] I told the women that they must pay, but they did not understand why. ‘Did you make the association because you want money?’, they asked me. So I had to explain and explain and explain. It was very difficult. And then we got money, 100.000 Danish krones. When I was reading the letter regarding the funding I got really, really scared. I just wanted to send the money back.

The act of forming and running the association turned out to be learning process – and a rather demanding one. Not only because Fatima did not know how to do it in advance (and seemed to get almost absurdly insufficient advice from the NGO), but also because she had to carry out a lot of explanation and negotiation with the (future) members. That the process of establishing an association, obtaining funding for it, and making it work can be a challenging one was widely shared among the organizers I interviewed. Fatima presented her story as getting a good idea, striving to realize it, facing many challenging difficulties, but finally succeeding! Other organizers told similar stories. And indeed, there were many success stories. Quite a number of the associations included in

186 this dissertation had successfully applied for funding from different ministries, municipalities, private funds, and even the European Commission. In a very few cases, organizers were temporarily employed as coordinators or counsellors, but this was definitely an exception. First and foremost, getting funding meant that ideas and projects could be realized, premises could be rented, courses and seminars could be given, and sometimes a Danish project worker could be hired for a short or longer period of time. Obviously, this should not lead us to believe that all attempts to form new associations succeeded or that all associations were successfully doing what they wanted to, getting funding and attracting members. Rather, the fact that I was told several success stories reveals at least two things. Firstly, that there actually were a number of well-functioning associations with skilled organizers who did a good job of fundraising. Secondly, and in a methodological perspective, this tendency points to a bias in my empirical material. Just as it was easier for me to get access to associational organizers and key persons, it was also easier to get through to well-functioning associations with some degree of success. But to return to Fatima: why was she scared when she got news about the funding?

New political culture and opportunities Like several of the other associations, Fatima Omar’s succeeded in obtaining considerable funding. And, in both hers and other cases, such achievements both provided opportunities and caused misunderstandings. According to Fatima – as well as a number of key persons in other associations – many ‘ordinary members’ did not fully understand how associations work, especially in relation to financial issues. This is one of the reasons why Fatima was initially scared of receiving money, fearing that the members would expect to get some of it for themselves. She and her Danish collaborator, however, insisted that the women should learn to understand the purpose of obtaining funds, thereby turning misunderstandings into a learning process. This, however, was not an easy task. Fatima continued:

Then the members started to come to me, telling me that ‘I have a big bill, I would like to pay it’, ‘I would like a baby carriage’ or ‘I have problems, I would like to borrow 10.000 [Danish krones]’. Then I told them that ‘this money is for activities, it is association money, you cannot just distribute it’, and I explained and explained […] Some of them started to call me at home and told me, ‘you used our name and now you are rich and don’t want to share with us, but you keep the money to yourself’.

187 After about four or five months they started to understand, but by that time some of them had left the association, because they did not understand what we did with the money. But all the time, we tried to educate them, tell them how we got money, how we used them. And some of them thought that if you make an association, you get money. So altogether six new associations were established, but they could not continue, because they did not know how to fundraise and how to establish an association. When they quit this association, I heard them say, ‘we have started a new association’, but they could not go on.

In this account, there are several learning processes going on. Some of the members apparently ended up understanding what funding meant and how it should be used, thereby gaining an insight into the practices of voluntary associations. This democratic lesson, however, did not involve the members who had already left to create their own association in order to ‘get money’. Their reactions can be said to reflect a logic of specific reciprocity (cf. Faist 2000; Putnam 2000) where resources are expected to be shared within the group in contrast to being used for (what may seem like) more abstract common activities. We may also say that they lacked the necessary social capital to obtain access to the right information about fundraising and mobilizing potential members. In addition, it may be argued that the Danish authorities and NGOs also seemed to lack the appropriate social capital to reach these women and inform them about how to run and fund an association. Fatima was not alone in her frustrations or misunderstandings relating to how funding should and should not be used. Another woman blamed it on ignorance about how associations work in Denmark. In her words, some people believed that, “when we say that the association has been awarded some funding, they think that it is money in our pockets”. The idea that there would be a personal economic profit in doing associational work sometimes seem to lay the foundations for jealousy and misunderstandings, even if many associations had chosen to have their finances administrated by a larger NGO or use external auditing. Thus, when I asked key persons whether they had experienced misunderstandings and jealousy, most of them affirmed it. “Of course!”, one man said, but continued that it does not matter, because he had his cause, and a lot of people had never heard about voluntary work before coming to Denmark. “Now they know and many have apologized to me”, he went on. Other informants, however, claimed that they had never experienced any jealousy, emphasizing that they always maintained a modest and honest attitude, avoiding troubles and

188 politics, respecting everybody. One aspect of jealousy, a man contended, could be that some key persons were showing off, thereby exposing themselves to the law of Jante84 – a ‘who-do-you-think-you-are-attitude’, often thought to be particularly Danish, but which, according to these men, was also at work among Somali-Danes as well. For the associational key persons I interviewed, associational engagement denoted possibilities for doing something they were enthusiastic about doing – as well as a lot of voluntary, unpaid work. It meant joy, respect, challenges and valuable experiences as well as worries, troubles and huge phone bills. Still, activists also told me about broad support and recognition for their commitment. Inspired by Honneth (2006; 2003), we might say that they achieved recognition for social esteem and merit for their achievements. Still, many of the associational key persons downplayed the aspects of prestige and social esteem, emphasizing their personal satisfaction gained from making a difference as their most important fuel for their actions. An activist against girl circumcision summed up her feelings like this: “I have encountered resistance … and it has been tough, but I thought, ‘I am not giving up’, because I know what was at stake”. She thereby pointed to her contribution to a larger social struggle for change as well as to the ambiguous nature of much voluntary work. As she and other informants said, misunderstandings can be seen as an expression of limited or no experience of voluntary associational work. Given the history of colonization, dictatorship and civil war, it should be little surprise that many people living in the Somali Republic had only experienced relatively little exposure to voluntary associations, if any. In this perspective, associational engagement involved ‘democratic education’ and learning processes of associational political culture. These learning processes involved not only organizers of associations, but also ‘ordinary members’ and observers in the longer term. Still, the Somali political-historical perspective alone does not tell us the whole story, as it ignores the different life experiences and social positions – and thus also the varied exposure to political practices. We also need to see the learning processes of forming and joining an association in a Danish societal context and

84 The law of Jante was made famous in a book by the Danish-Norwegian writer, Aksel Sandemose. The law of Jante refers to ten commandants about not feeling special. It is often regarded as something particularly Danish.

189 the Danish opportunity structure where associations are regarded and supported as cornerstones in civil society and political culture. One way of obtaining resources in Denmark is through an association, and quite substantial funds are channelled from the state to various kinds of associations. As many Somali-Danes are not very well established in the labour market or in the Danish political sphere, to form or join an association may seem to be one of the most accessible ways of participating in Danish society (cf. Fraser and Honneth 2003; Fraser 2000) – as well as a way of obtaining social esteem (cf. Honneth 2006; 2003). Associational engagement may make it possible to be heard, to actually realize projects and perhaps be funded; possibilities which otherwise could seem to be difficult or even impossible to achieve. Funds, however, are not unlimited and according to Somali-Danish key persons, it was easier to obtain funding in the 1990s than in 2003. This meant that many associations did not have the economic means to rent premises with facilities and thus only carried out very few activities. It also obviously meant that associations were competing for funding with each other as well as with a lot of other groups and organizations. Furthermore, it meant that associations were competing over attracting and keeping members, as possibilities for funding often depend on the number of members in a given association. Thus, what some observers and associational organizers termed ‘associational shopping’ – referring to members going from one association to another – might result in problems for funding. Seen in this light, we can say that learning processes and misunderstandings related to associational engagements may both be connected to changes in the political culture (some may have) inherited from Somalia as well as to the situation in Denmark, where the overall socio-economic profile combined with the possibility of (limited) funding may make associations appear as a relatively attractive and open option.

Establishing and negotiating positions So far, I have contended that associations can be seen as platforms from which it might be possible to speak and ‘be heard’, to get funding for different activities and realize ambitions because of the collective dimension which they represent – and because of their role in Danish society. Being the chairperson, initiator or organizer of an association is thus to speak and act for more people than oneself with the enhanced legitimacy that follows. Following Bourdieu, such

190 positions imply a concentration of social capital that mandates actors “to speak and act in its name and so, with the aid of this collectively owned capital, to exercise a power incommensurate with the agent’s personal contribution” (1986, 251). That some people represent others through institutionalized social and political arrangements is of course a well-known phenomenon all over the world – both in the Danish representative, constitutional democracy and in the traditional Somali clan system. What is interesting here is that Somali-Danish associations can be seen as a way of establishing and negotiating positions at several levels, which sometimes include the aspect of representing the association as a ‘constituency’ in relation to Danish authorities, and is sometimes connected to regional or clan affiliation logic in relation to the Somali political scene in Denmark, Somalia and elsewhere (cf. Kallehave 2003). Associational key persons thus had a ‘double potential’ for positioning themselves – in Denmark as well as in transnational social space.

‘Ordinary members’ and spokespersons In the narratives of forming associations, of having a cause and fighting for it, of encountering resistance and misunderstandings, but also of gratifying experiences, associational key activists articulated themselves as capable, successful and enterprising people. As we have seen, they accentuated their positions by referring to ‘ordinary members’ and ‘ordinary people’ as less enterprising and without the necessary exposure to and knowledge about democratic ways of organization. This positioning between organizers, on the one hand, and ‘ordinary members’ and ‘ordinary people’ on the other, was significant and shared by many of the informants, across gender, regional affiliation and other divisions. The majority of associational key persons emphasized their educated background, whether their education was obtained in Somalia, Denmark or elsewhere. What most of them thus shared – in spite of other differences – was a middleclass and urban background. Several of them were the sons or daughters – or grandsons or granddaughters – of former high-ranking politicians, government officials, freedom fighters against the colonial powers, or traditional leaders. They were, in other words, of good families and some of them highlighted their prestigious family background by using a family or first name which

191 would remind other Somali-Danes of their lineage affiliation85. The distinction between ‘key persons’ and ‘ordinary members’ was thus often accentuated by socio-economic and lineage positioning. Somali-Danish associational spokespersons were thus often already in relatively privileged positions in regard to their background in Somalia, a position which might be further strengthened through their associational engagement in Denmark, where they might be given authority to influence decisions which may not only concern their own association, but also other Somali-Danes. Fur- thermore, like many other voluntary associations in Denmark, Somali-Danish associations often seemed to be run and initiated by a few people (cf. Gun- delach and Torpe 1999, 84). While some of the associational key persons were engaged in just one association, others were veritable organizational nodal points with committee positions in several. Several of the informants emphasized their position as Somali-Danish or Somaliland-Danish spokespersons. However, while these spokespersons often enjoyed strong support from certain quarters of the Somali-Danish group, they were often contested by others – especially in relation to the Somalia–Somaliland question and other sensitive issues. Nevertheless, when some informants did claim such positions, they did so with reference to the extended and intensive voluntary work they arguably had undertaken over a number of years, endowing them with experience, political credit and confidence and, as Bourdieu reminds us, with extended social capital. However, we should also remember the fluidity and mobility of the Somali- Danish associational scene. The fact that anybody can set up an association in Denmark makes it easier to establish a position (of claiming) to represent collective standpoints. It also means, however, that it may be difficult to maintain a position as a spokesperson, especially because of the differing interests in relation to religious devotion and regional orientation, which have been aggravated by the civil war. This makes the point of the proliferation of associations as related to both vivid commitment as well as fragmentation even more important. Still, several of the key activists had been ‘in the game’ for quite a number of years, though their positions and associations had changed over time. In this

85 Somalis usually have three names: Their first name, father’s name and father’s father’s name. Some of the informants used their grandfather’s or great- grandfather’s name instead of their father’s or, in the case of the men, chose to use their father’s name as their own first name.

192 perspective, while the associational landscape changed – associations appear- ing, flourishing, and disappearing again – many of the key people remained central. They had, as Bourdieu might say, a ‘feel for the game’ (1996a, 103). The position as an ethno-national spokesperson, however, was also made possible because of a tendency in Danish society to perceive migrant and minority groups as homogeneous communities and ignore or downplay internal differences. As Somali-Danish associational spokespersons, some key persons functioned as informal counsellors for NGOs, public authorities and ministries, thereby having an actual influence on how the Somali-Danish group was perceived and an at least potential influence on policies and the distribution of funds within the Somali-Danish group. In this sense, they were in a position to actually participate in the symbolic struggles of defining what it means to be integrated, to be Somali, to be Danish, to practise Islam, democracy and gender equity – or, in other words, to participate in the struggles for defining the matrixes of cultural intelligibility. Still, while Somali-Danes supported or challenged the authority of associational key persons, the Danish media and authorities sometimes seemed to be more eager to identify a few people to ‘speak for the community’ – that is as ethno-national spokespersons – perceived to speak and act with legitimacy as well as to possess authentic insights into the lives of other Somali-Danes. For instance, in a report about the channels of influence of ethnic minorities in Denmark, Inge Hammar and Ole Bruun write that Danish society assumes that minority associations are representative of their compatriots and criticizes associational spokespersons for not understanding or living up to these expectations (2000, 43-44)86. I do not want to assert whether or not Somali-Danish associational key persons live up to such demands or if they possess these insights. My aim here is, firstly, to point out that the underlying logic of representation at the ethno-national level might reinforce certain versions of ‘Somaliness’ and ignore

86 According to Ulrich Schmidt-Hansen, head of section in the Danish central administration and PhD candidate at the University of Copenhagen, the Ministry of Refugee, Immigration and Integration Affairs prefers to restrict their cooperation with migrant and minority associational actors “to those who are well connected to their hinterlands, in the sense that they can be trusted as stable representatives of the group they claim to represent. However, during the last five years the cooperation between immigrant organizations and the Ministry has been further restricted and the representatives are primarily included in temporary projects” (personal communication, June 14, 2006).

193 others. Secondly, that ‘the problem’ of associational spokespersons is related to conditions in both Danish society as well as in the Somali-Danish associational scene and thus cannot be reduced to a question of ‘better spokespersons’ or ‘better representation’. But thirdly, that several associational key persons were very critical towards ‘one-man associations’, claiming to speak on behalf of all Somali-Danes. The legitimacy of spokespersons was thus a shared concern among both Danish authorities and Somali-Danes. The potential ‘slip’ between associational and ethno-national spokespersons thus remains debatable. In my view, it is both necessary and desirable that the participants in the democratic and political processes come from all spheres of Danish society. It does, however, place a huge responsibility on the spokespersons if they are perceived as representing all or at least many Somali-Danes to safeguard a lot of different interests, which they do not necessarily all share. As feminist and cultural study voices have noted, pluralist multiculturalism tends to convey power to already established authority and overemphasizes cultural difference, while vesting particular interests (cf. Benhabib 2002; Hall 2000; Appiah 1994). Now, the Danish ‘integration regime’ can hardly be characterized as pluralist multiculturalism, as we have seen. Still, the idea of cultural wholes or entities is shared to some degree (though without much entitlement to specific ‘cultural rights’) – especially in some of the public, political and media debates. In this perspective, both well-meaning proponents of recognition as well as nationalist or populist voices can be said to underemphasize the conflicts involved in defining and recognizing collective identities or ‘a culture’. While the former may believe in a perfect democracy where all voices and interests are heard and all antagonistic relations can easily be solved – as Honneth and Taylor have faced criticism for (Fraser and Honneth 2003; Benhabib 2002) – the latter might have an interest in defining homogeneous categories of ‘us and them’, for instance in terms of ‘the strangers’. Both perspectives can be criticized for, respectively, ignoring or exploiting the antagonistic aspects of politics. These considerations thus support a more conflict-oriented approach to recognition and political participation, as Bourdieu (2005; 1996a; 1994), Mouffe (1997) and Fraser (2003; 2000) propose. It thereby complicates the question of processes of inclusion in relation to struggles for recognition and negotiations of positions – in Denmark as well as in transnational social space. In the next chapter, I elaborate more on the transnational dimension of these as-

194 pects, turning to two case studies of involvement in the reconstruction and development of Somalia and Somaliland.

Chapter resume: Platforms for action In this chapter, I have analyzed the development of the Somali associations in Denmark and the social principles of organization guiding them. Somali-Danish associations started to proliferate at the end of the 1990s, articulated as a specific strategy of participation in Danish society. This, however, might have been almost too effective, as the number of Somali-Danish associations is very high, bearing witness not only to huge commitment, diversity, and the dispersion of Somalis all over Denmark, but also to fragmentation and lack of collaboration. I argued that Somali-Danish associations are organized around dynamics of positionings and sets of conflict and collaboration, constituting a political field in Bourdieu’s sense of the concept. The social principles of organization include lineage and regional affiliation, political orientation concerning Somaliland versus the Republic of Somalia, gender division, and degree of religious devotion, dividing as well as uniting Somali-Danish associations. I then showed how these principles were enacted in the Women’s Associa- tion where gender, religious practice, and regional affiliation were central dimensions. In this case, I also discussed issues of social capital, arguing that the associations form spaces for accumulating social capital in Bourdieu’s sense of group membership, which had both bridging and bonding dimensions. For the so-called ‘ordinary members’, the Women’s Association seemed to offer a venue for social gatherings as well as a place to breathe freely in an otherwise very busy daily life, while organizers and key persons were engaged in a number of activities and collaborations. Likewise, I discussed how the organizations in the Women’s Association – as well as in many other associations – positioned themselves vis-à-vis the so-called ‘ordinary members’ in terms of knowing the Danish system and understanding the political culture of voluntary work in associations. Learning how to establish, organize, obtain and manage funding was seen as a valuable and sometimes tough learning process, and associational key activists often experienced jealousy and suspicion. Still, associational key persons might use associations to establish positions as ethno-national spokespersons in Denmark as well as positions in transnational space: as platforms for action. While it might be difficult to gain recognition for achievement and so-

195 cial esteem in Danish society – let alone to get a job matching one’s competence and ambitions – associations offer a possibility for obtaining funding to realize projects and make a difference in Denmark, as well as in Somalia and Somaliland, as we shall see in the next chapter, where I turn to assessing how recognition and positioning were negotiated in transnational reconstruction projects, mobilizing ‘the diaspora’.

196 Chapter 7 Mobilizing ’the diaspora’. Transnational involvement

Dear friend. It is a pleasure for the above-mentioned organizations to in- vite you to the Somali International Conference, which will be held in Copenhagen, Denmark. The goal of the conference is to: Challenge war and instability with micro-development projects initiated by the Somali Diaspora; explore in what way the Diaspora can stabilise and improve living conditions in Somalia; make it easier for Somalis living in exile to return to their country (DOEA Conference invitation, 2003).

Their experiences living in Western Europe and the fact that most of their children are born and raised there, makes the Somscan & UK community appear different when back in Somaliland. A difference that the members are aware of and wish to make use of. A stated objective of Somscan & UK is to strengthen the connections between Somaliland and Scandinavia/UK, and they perceive themselves as promoters of democracy and collectiveness, being a living example themselves (Jama and Jensen 2004, 2).

***

In April 2003 I was invited to participate in a big international Somali conference in Copenhagen, arranged by three Somali-Danish associations. As the invitation quoted above states, the conference had as its particular aim to further and collaborate the transnational support offered by the ‘Somali diaspora’, and at the conference the Development Organization of East Africa (DOEA) was constituted. Later that year, I visited the Somscan & UK project in Somaliland, and I interviewed Somscan & UK members in Copenhagen, Somaliland and London. Somscan & UK also articulated their visions in diasporic terms, as “Diaspora joining forces to better the chances of return to the homeland”, as the paper above, prepared for the 9th Somali Studies International Conference in Aalborg, was titled. DOEA and Somscan & UK were not the only Somali actors taking up the vocabulary of diaspora, as I have already shown in chapter two with the examples of research, policy reports, and seminars addressing the so-called ‘Somali diaspora’. Indeed, this development does not only apply to the Somali case. Initiatives by dispersed migrant populations have recently at-

197 tracted increasing interest among policy makers, scholars, NGOs, and migrants themselves. Migrants and ‘diasporas’ are seen as possible agents of development and change, which not only remit money to their countries of origin, but also transfer ideas of political, social and cultural change (e.g. Global Commis- sion on International Migration 2005; Schiff and Ozden 2005; Levitt and Sørensen 2004). In this chapter, I explore Somali transnational reconstruction and development involvement, focusing on two case studies of the DOEA conference and the Somscan & UK, respectively. I present and analyze the event and the projects, paying specific attention to how these actors mobilize in the name of diaspora, how they negotiate loyalties, positions and social esteem, especially in relation to the tensions between clan and regional identification, and finally how the different situations in Somaliland and southern Somalia shape visions of development and, perhaps, return. But let us start with an afternoon in Co- penhagen, on the premises of a community house in the suburbs, now transformed into a conference hall, hosting the DOEA conference.

The Development Organization of East Africa I was invited to participate in the DOEA conference by one of the organizers and we agreed that I should write the conference report, thereby providing both them and me with a good reason for my presence. Omar was one of the main organizers of the three-day conference and he and Mahmoud took part in trans- lating for me. After the conference, I discussed the proceedings with the organizers and undertook several follow-up interviews with Omar and Mahmoud. I also e-mailed several of the other conference delegates and met with three Brit- ish participants in London in March 2004, as I come back to in the next chapter. The tensions between the official version of the conference report and the more unofficial comments offered an opportunity to examine the negotiations of loyalties and the visions of development and governance at play. Furthermore, participating in the conference demonstrated the importance of seeing and being seen. This became clear to me when a video was shown.

Copenhagen, April 2003 I am sitting next to Mahmoud together with the other participants in the DOEA conference, about 40 Somali men. They are an international crowd, from Scan- dinavia, the UK, Switzerland, Italy, and even as far as the US and Australia.

198 For two days now, they have been discussing various development projects, presented by associational chairpersons and revered elders. Now they are carefully watching a video, commenting on it in low voices. The video is a recording of a trip to the Southern part of Somalia made by a Somali-Swedish doctor, Dr. Adam Ibrahim. It shows a run-down school and an equally neglected hospital; it shows women selling camel milk in the market and a local soccer team pos- ing in worn-out T-shirts. Mahmoud Sheikh giggles when he tells me that they only have nine T-shirts for the whole team and have to share if there are any substitutions. One man suddenly sees a relative appear on the screen. He has had no news from her in more than four years and is visibly moved. After the video, tea and vanilla flavoured biscuits are served and I am told jokes and an- ecdotes about the video and the memories it provoked.

The next day, an umbrella organization is constituted which is to coordinate contributions to the two regions donated by Somalis in the west. While there is agreement to focus solely on these regions, the conference participants choose to name their organization ‘The Development Organization of East Africa’. ‘Maybe other parts of East Africa will need us one day; it is important to keep an eye on the future’, Mahmoud explains to me. Denmark is then chosen to house the secretariat of the organization, and co-ordinators of water, resettlement, education, as well as communication and community issues are appointed. Everybody seems optimistic and enthusiastic with the conference serving as a ‘meeting between the intellectuals in diaspora’, as one of the chairmen phrases it. Finally, the preliminary date of a new conference is set and a group picture of all participants is taken outside in the spring sun.

A meeting between important men At first the video intrigued me. It was one of the highlights of the conference, causing emotions, memories and enthusiasm: definitely more than just an enter- taining break from the presentations of reconstruction projects. Still, I could not figure out exactly why the video seemed to be so important and why I kept thinking about it. After a while, I realized that one way of understanding it was as a means of proof. Proof that economic donations reach the schools and hos- pitals they are meant for, proof of the importance of transnational support. Dr. Ibrahim’s video, however, was not the only visual documentation taking place

199 at the conference. The conference participants also documented the event and their participation, videoing, photographing, and recording the proceedings for local Somali radio and TV stations, as well as for circulation in their respective organizations. The many ambitious development projects – including plans for a brand new hospital, economic support to existing schools, the construction of new water bore holes, and an extensive re-settlement project – could thereby be shown to many more people. The overall atmosphere of the conference was definitely that of a meeting between important men. Many of them were dressed in dark suits, and the exchanging of business cards and checking of mobile phones emphasized the professional atmosphere of the event. The trip made by Adam Ibrahim was, he told me during a lunch break, only possible because his father was a general major in the former government and knew the figureheads in the Transitional National Government (TNG). He himself could, the doctor continued, be at the very top of political life in Somalia himself, but he did not want to. Like Ibrahim, many others of the participants and the majority of the initiators were part of the former elite – or the sons of it. Well educated and well articulated; the son of a former minister, former regional administrators and commissioners, former government top employees, a former colonel in the army, and several traditional clan leaders, ugaas. While titles such as Doctor and Engineer were mentioned with pride, the positions held in the former regime, however, were not to be written in the conference report. And while some of the conference participants – like the doctor – readily told me about his father, others were more reserved about their backgrounds.

Homeland politics When I talked to participants, everybody agreed that the conference had been a great success and, as one of the chairmen had phrased it, “a meeting between the intellectuals in diaspora”. Just as importantly, it was an important step towards further cooperation and coordination, as the invitation stated. Instead of five different associations doing the same things, they could now work together and coordinate their plans and initiatives. The participants were invited because of their activities and core positions within their respective associations and communities, I was told. They were committee members in one or more organization, as well as involved in Somali media outlets such as local radio, local TV and web pages, thereby constituting nodal points of communication and organi-

200 zation – as the various recordings and documentations also bear witness to. I do not know the number of associations represented at the conference, but a rea- sonable guess would be that each participant was active in at least one to two associations; thus between fifty and one hundred different Somali associations were probably represented. Greater cooperation between all these associations and the creation of a common platform was a central aim of the establishment of the umbrella organization. Omar Said explained to me:

We need a common organization to strengthen each other, to cooperate, to reduce competition and to have more cogency. So even if people work in their own local area, they can still say that they are members of DOEA and prove that they co-operate with people – not only in the diaspora, but also in Somalia, and in the local town. Then you get more credibility and cogency than if you are on your own. If you can say that you work with people in the States, Italy and in the local town, and if you can say that we have different projects, but we are still one organization with one board, that will show donors that there is a certain coherence, rather than it is all divided. […] Another thing is that you don’t have to overlap each other. If some people already support a school and some people in Finland, for instance, want to support a school in the area too, they don’t have to start all over again. You can learn from the people, who are already there. […] So you can do more and inspire each other.

Mobilizing and coordinating the efforts of ‘the diaspora’ would thus strengthen and improve the various reconstruction projects in the region. Furthermore, according to Omar, DOEA would also prevent the so-called ‘shopping culture’ in the region, when a local association appeals to a Somali association in the West for help, but then when problems and disagreements arise, it backs out and contacts yet another western-based association. DOEA would, in short, not only enhance cooperation, effectiveness and cogency between Somali associations in the West, but also impress donors and be able to exercise a greater control over the local associations and, in the longer run, in the region. At the same time, the creation of a common platform also concerned the So- mali-Danish associational landscape. The DOEA conference was set up by three Somali-Danish organizations, whose names and objectives referred to different territorial principles: The Homestead, The Nation, and finally The Horn. The Homestead was involved in reconstruction projects in a specific part of the region. The Nation was established in the middle of 1990s as an association for people from the whole region to provide a forum for discussing homeland poli-

201 tics as well as dealing with challenges in Denmark. It was thus founded on a regional basis, but split into two in 2000, when The Horn was established as a sister organization. The reason, I was told, was that a faction in The Nation supported armed struggle. The majority in The Nation, however, was opposed to this, so they expelled the faction and decided that The Nation should focus exclusively on integration issues in Denmark. However, according to one man, active in both associations, many of the members were

still interested in working a little with politics and the homeland [...] So we in The Horn said that we are not going to work with politics, but we will work with … humanitarian projects, relief […] And we said that we will concentrate on this specific area in the beginning, but then later, perhaps, do other things in the Horn of Africa. And that is why our organization is called The Horn.

The difficulties of ‘homeland politics’ were thus resolved by dividing The Na- tion into two: a transnational humanitarian relief and an integration-oriented organization. At the same time, the supporters of armed struggles were ruled out, while several of the executive committee members remained active in both associations, thereby maintaining their involvement in both reconstruction and integration. This incident shows the ambivalence of ‘homeland politics’ in different ways. From one perspective, ‘Somali politics’ was equated with armed struggles or what Benedict Anderson (1994; 1998; 1992) has termed long- distance nationalism. In Anderson’s understanding, long-distance nationalism is “a menacing portent for the future”, as the long-distance nationalist, Anderson posits, is “well and safely positioned in the First World” and does not have to live with the consequences of his or her deeds (1994, 137). Apparently, the majority of members in The Horn and The Nation shared Anderson’s critique of long-distance nationalism as well, expelling the supporters of armed struggles. Still, members in The Horn were “interested in working a little with politics” on the one hand, while stating, on the other, that they “were not going to work with politics”, but focus on relief and humanitarian projects. This double aspect of the concept of politics, simultaneously rejected and embraced, must be seen in relation to the specific situation in Somalia: the years of violent and corrupt dictatorship followed by the total collapse of the state and civil war. No wonder then, that politics is mostly associated with violence and corruption in such a

202 context. Indeed, Chantal Mouffe’s emphasis on the necessity of democratic political struggles is even very relevant to such a situation. She writes

When there is lack of democratic political struggles with which to identify, their place is taken by other forms of identification, of ethnic, nationalist or religious nature, and the opponent is defined in those terms too. In such conditions, the opponent cannot be perceived as an adver- sary to contend with, but only as an enemy to be destroyed (1997, 6, my emphasis).

In other words, the dictatorial regime and the civil war where armed factions and warlords were loitering and killing in the name of clan affiliation can be said to be the negations of legitimate political positions. Warlords became politicians in the sense of being the ones defining the principles and distinctions between ‘us and them’, as Mouffe defines politics, often on the basis of clan affiliation. In consequence, politics turned into the acceptance of the lineage logic as the overall principle of collective identity and as warlords’ affairs: dirty, corrupt, violent business. Many Somalis engaged in collective reconstruction projects adhered to the distinction between relief and politics, emphasizing that they were not involved in politics. The relations of the so-called ‘diaspora’ to the homeland were, in other words, not that simple.

‘Clan is a bridge we need to cross’: Omar A consequence of the Somali civil war is thus a situation in which kinship and clan affiliation play a central and contested role. We have already seen how lineage has a double aspect, signifying safety and security in one’s own lineage versus distrust of and potential danger from other ones. Likewise, we have seen how the area dominating a person’s lineage became his or her ‘area of origin’ where (relative) safety could be obtained and to where the field of belonging was now supposed to be directed. This was also the case at the DOEA conference. According to Omar, the participants in the conference – who were all said to originate from the same provinces – might never have set foot in them.

Maybe they come from Mogadishu, maybe from Hargeisa. But they cannot turn back … they cannot have confidence in Hargeisa because things are dealt with according to clan affiliation. They might have grown up with an Isaaq in the south, but they do not have this belonging to the south, because the clan is not from there, so even if they have never seen Hargeisa, they would feel that they belong there.

203 The tension between lineage and regional affiliation is not a new phenomenon in Somali history, but has remained central – especially in the south of Somalia, where “the competing claims between ‘u dhashay – ku dhashay’ […] are most urgent and divisive due to the more heterogeneous history of expansion, settlement and territorial occupation” (Barnes 2006, 489, emphasis in original). The massive displacement of people to their ‘areas of origin’ has led some to talk about ethnic expulsion. Whether this term is appropriate or not, the entangle- ment of origin, territory, lineage affiliation and security has created new dynamics of belonging. One of the more indirect objectives of the conference was exactly to deal with lineage affiliation as a central and contested social principle of organization. This, however, was not an easy task. As Omar later phrased it, “clan is a bridge we need to cross”. Bridges, we all know, connect people, places; more metaphorically, bridges connect concepts and ways of thinking. Cros- sing a bridge might get you from one place or one state of mind to another – from local loyalty to regional identity, perhaps. But crossing a bridge might also be a stormy experience, where you need to keep your balance not to stumble. As we have seen, Putnam (2000) also uses the bridge metaphor in his distinction between bridging and bonding social capital. If we apply his optic, we can characterize the conference as an example of both aspects, depending on the perspective taken. We may say that the conference primarily furthered bonding social capital, in that the participants were all related to the same area. Yet the conference also had a bridging function in its attempt to establish trust between different clans and generations in that region. This shows us that the discussion about bridging and bonding is closely related to how communities and groups are constructed and delimited. In order to distinguish between bridging and bonding in the first place, it is necessary to establish a stable social inside or core as well as a surrounding outside – whether this outside is perceived as the rest of Somalia or Danish society. Indeed, if Somalis from the region are established and seen as the well known inside, then the DOEA conference could be characterized as an event reproducing bonding networks. This view, however, ignores internal differentiation, as well as the fragmentation and dispersion following civil war and the need for reconciliation. In other words, the distinction between bridging and bonding social networks only makes sense in a perspective in which the inside and outside is analyzed as relative positions. The lineage system is, as we have seen, a highly flexible and contextual system, with

204 shifting positions of inside and outside, proximity and distance, with both bridging and bonding dimensions. Still, following the years of dictatorship and civil war, the bridging aspects of lineage affiliation seem to have been overshadowed by its bonding functions. Thus, to regard clan as a bridge that needs to be crossed may represent both a wish to emphasize trust and cooperation and to overcome clan affiliation altogether.

Principles of governance and identification Clan affiliation is, as we have seen earlier, of a double nature. It is a vital source of solidarity, security and division in Somalia as well as in migrant associations, constituting a generative principle of trust and distrust. Trust is a central issue in processes of reconciliation and development, not least in the Somali case. At the conference, the establishment of trust had a relatively straightforward aspect, namely to be able to meet and discuss – in spite of disagreements. When I asked Mahmoud Sheikh what he thought about the conference, he ex- claimed:

It was so great, so great! It was the same mentality […] The whole of Somalia should know that the young generation and elderly people can make their own choices and set up a big conference and accept each other and discuss how we can return and reconcile […] The most important thing was that Somalis came from many different countries; they travelled and spent time together and they accepted each others’ opinions and thought of the general interest of the country, of the region …that is the most important […] Before, we didn’t know what each other stood for, but now we trust each other, we have shown each other an open mind and we have discussed all sorts of things.

Getting to know each other across countries and across generations, creating trust and showing each other an open and tolerant mind was thus of utmost importance to participants. Mahmoud’s statement can be read as a commitment to the possibility of democratic politics – to meet, discuss and reconcile without being or becoming enemies87. We can also read it as a statement of diasporic formation. Not as an essential feature, defining all aspects of life, as Thomas Blom Hansen (2005) has warned us against believing, but rather in the sense of Werbner’s (2002) idea of a moral community of shared solidarity and commit-

87 I want to add that the DOEA conference was certainly not the only event where Somalis met in peace; the point here is how the DOEA participants emphasized this aspect.

205 ment. Mahmoud’s statement, however, also turns our attention to the process aspects of this: the diasporic ‘community’ is not given, not already there, but it is mobilized in spite of distance between countries and generations. The conference objective that ‘the diaspora’ should ‘stabilize and improve living conditions in Somalia’ was thus not as straightforward as it might seem at a first glance. For who is included in ‘the diaspora’ (cf. Anthias 1998; Brah 1996)? The origin of the participants was, as we have seen, ‘from the region’, either through upbringing or clan affiliation. The ‘Somali diaspora’ in this case, did not refer to all Somalis dispersed around the world, but to those related to the two provinces. This raises the question of which principle of belonging should guide the mobilization of ‘the diaspora’, or in other words, the tensions of con- flicting and competing loyalties. One of the main messages from the speakers at the conference was precisely this: to stop clan identification and promote a regional identity, which would lend itself more easily to modern state- and nation-building than the segmented and mobile kinship lineages. Both kinds of identity, however, relate to different kinds of abstractions, which take us back to the discussion of (long-distance) nationalism. The regional identity model bears a resemblance to Benedict Anderson’s (1991) claim that the nation is imagined as an abstract horizontal and territorially defined brotherhood. Only here, it was less about embracing nationalism and more about reckoning with the idea of ‘areas of origin’ defined by lineage dominance. One might argue that this reckoning might be a first step towards a possible ‘re-nationalization’ of Somalia by means of overcoming segmentary lineage loyalties. Indeed, for the conference participants and many others, clan loyalty and modern state formation (even at a regional level) were in oppositional relationships. We can formulate this as a tension between different principles and visions of governance: between a segmentary vision of kinship solidarity and a modernist vision of territorial loyalty, seen as a mark of modernity. In the words of Mahmoud, it was also about re-introducing a democratic way of doing politics. He said:

The only thing we think about is when will there be peace and stability in our homeland. We have seen one hopeless peace conference after another … So we thought that this is the chance for our generation … who live in the West and who have seen their way of dealing with the problems, their ways of doing politics.

206 Still, it is worthwhile remembering that tensions between pastoral, decentral- ized principles of governance versus modern and centralized state formation have been debated since the formation of the Republic of Somalia in 1960 (Ahmed 1995b; Samatar 1991). They have, however, been dramatically reinforced by the civil war and the arguable disintegration of the republic. The struggle for regional identification was thus related to long-standing efforts of state modernization as well as to the inspiration and exposure of democratic visions of governance, which ‘the diaspora’ had experienced in exile. If we remember that many of the participants at the conference were part of the former elite or their sons, it becomes clear that the former modernizers and administrators of the Somali Republic constituted at least part of ‘the diaspora’. Their transnational reconstruction projects may, in other words, constitute a continua- tion of the nation-building and modernization process, though now at long- distance. For the conference participants, however, it was important to position themselves vis-à-vis the civil war and politics, as we have already seen, accentuating their peaceful, uncorrupted and non-violent attitude in contrast to the warlords in Somalia. As several participants noted, warlords only know war and are not interested in peace, as this would mean the end of their livelihoods, where they, on the one hand, attend international peace and reconciliation conferences with all bills paid, while on the other, make money on war and terror in Somalia. People ‘in diaspora’ in contrast, the participants emphasized, have been exposed to new experiences, education and democracy. Thus when one of the organizers proudly concluded the conference by characterizing the event as “a meeting between the intellectuals in diaspora”, this statement can be seen as a political positioning of the participants: as people of peace, education, and resources, as people who contribute to the development of Somalia via their peaceful engagement in reconstruction and, who might contribute to the future of Somalia as well. Participating in the conference, documenting it on video and audio recordings, might thus be a way of positioning oneself, one’s association and one’s qualifications with regard to the political map of Somalia. To identify oneself and one’s collaborators as part of ‘the diaspora’ can be seen as a way of establishing a position on this map: to be recognized as a political actor.

207

Returning to the desert? The Somscan & UK Cooperative Associations The Somscan & UK Cooperate Association also wanted to promote return and reconstruction, but were in many ways in a different situation to that of the DOEA. Firstly, while the DOEA focused on the south of Somalia, haunted by insecurity and political instability, Somscan & UK turned its attention to the more stable Somaliland. Secondly, while the DOEA conference could be a starting point in a long process, Somscan & UK was a well-established organization. Still, as we shall see, these two umbrella organizations also shared some of the problems and expectations they had to deal with. Somscan & UK – in its initial stage – was established in May 1999 in Swe- den, when a small group of Somali-Swedes founded an association with the purpose of rebuilding and returning to Somaliland. At this time, Somaliland had been through two major peace conferences and had proved itself to be relatively stable, and many Somalis living in the West had started to visit the country during summer vacations. Likewise, some had actually started to return for shorter or longer periods of time – or were considering doing so (cf. Fink-Nielsen, Hansen, and Kleist 2004). The Somali-Swedish group, however, had greater ambitions: they wanted to return collectively through the joint acquisition of land and the establishment of a new neighbourhood just outside the town of Bu- rao (Ibrahim and Farah 2001). The idea of collective return quickly spread to Denmark, Norway and later the UK, where similar organizations were soon established. In 2000 these organizations decided to work together and formed the umbrella organization, Somscan Cooperative Associations, which was further extended to Somscan & UK Cooperative Associations when three British organizations joined in. The two Danish branches played a leading role in the organization of Som- scan & UK at the time of my fieldwork, and its main secretariat was in Copen- hagen, where office space was borrowed from the Danish Refugee Council. One of the key persons in the Danish branches was Elmi Samir, who had arrived in 1989 as a refugee and had become one of the members of the Somali Resource Group and later the Somali Council. When I first met Elmi Samir in 1999, he was already talking about return, but was – like many others – con-

208 cerned about the economic, educational and health related aspects, especially in relation to the return of his family. The organized return of Somscan & UK to a Scandinavian-style neighbourhood thus seemed like an ideal solution, and Elmi used a lot of his spare time to work voluntarily for Somscan & UK.

Organized return In 2003 Somscan & UK consisted of eight organizations that each owned a piece of land located next to each other. There were one Swedish, two Danish, two Norwegian and three British groups. The pieces of land were bought for prices that ranged between 15.000 and 20.000 USD. Each Somscan & UK member family had paid 1.000 USD to cover the acquisition of land as well as other costs. Since then, the price of land in Burao had risen considerably. Alto- gether, 330 families were members of Somscan & UK at the time of fieldwork, and the project was closed to new members. In October 2003 about 15 families were said to have returned to Somaliland and settled in Hargeisa, Burao or other Somaliland towns. Likewise, the Somscan & UK project had caused interest among local Burao families, who longed to get out of the crowded town and own a bigger house. The year before, in 2002, the Danish Refugee Council, already active in Somaliland, submitted a proposal to the European Commission, prepared in cooperation with Somscan & UK. The objectives of the project were to avoid a situation in which the expected return of Somscan & UK members would drain scarce local resources, and to address the issue of the insufficient infrastructure of Burao. The project kicked off in the summer 2003, when the DRC received a positive response and a large grant of 550.000 Euros for the ‘Community Based Assistance to Repatriation Project’ by the EC High Level Working Group on Asylum and Migration (Jensen 2004). The EC money was donated to upgrade the town’s water supply and renovate a primary and secondary school. As the two schools were located close to the Somscan & UK area and a water pipeline would be led through this area as well, the Somscan & UK residential area would also certainly benefit from this project. Furthermore, the organization was represented in the steering group of the EC project as stakeholders and the DRC had employed two Somscan & UK members to coordinate the renovation in cooperation with other local and national partners. In addition, the Somscan & UK was establishing electricity in the area, funded by its own means.

209 In this way, the Somscan & UK constituted a rather unusual hometown association, if it can be considered one at all. The Somscan & UK members were not only compatriots from Burao raising funds to improve their hometown (cf. Caglar 2006; Landholt, Autler, and Baires 1999), but originated from many different localities. Still, as pointed out by Landholt et al., hometown engagement is not only about reconstruction projects, but also about creating social networks among fellow compatriots abroad. While the latter seemed to apply to Somscan & UK, the organization still differed in many aspects. The efforts to rebuild Burao were directed towards the establishment of a new neighbourhood with the ultimate aim of return, and they received indirect EC-funding to do that. Rather, we might say that the ‘hometown’ was in the making, creating a new kind of transnational social formation.

Buying land, negotiating lineage In 2000 and 2001 two delegations from Sweden and Denmark went to Somali- land to explore the possibilities of buying land and building a new neighbourhood on a collective basis. The delegations chose to buy land outside Burao for several reasons, I was told. First of all, Burao is situated in the middle of Soma- liland: it is the heart, the very centre. If not all roads lead to Burao, then a good deal of them do, making it easy to access other Somaliland cities, not least Mogadishu, the Somali capital and commercial centre. Secondly, Burao was completely destroyed during the civil war and internal conflicts in 1995 and 1998, when heavy inter-clan fighting took place. It has only been modestly rebuilt, compared to other Somaliland cities. To rebuild Burao is crucial in making the whole Somaliland idea work, Somscan & UK members told me. Finally, the third reason was the availability of land in Burao, which was much cheaper than that surrounding Hargeisa. These three reasons: centrality, nation building and the availability of cheap land might, however, be supplemented by a fourth: clan affiliation. As in the south of Somalia, the question of origin also plays a major role in Somaliland. The clan aspect had a double nature in Somscan & UK. On the one hand, a specific objective was to overcome the clan logic in creating a mixed neighbourhood, where people of different lineages would actually live together. As several of the initiators of Somscan & UK were not from the dominating lineages in Burao, this was not an empty promise. On the other hand, a large percentage

210 of Somscan & UK members were said to originate from Burao and to be of one specific sub-clan. There was thus a certain tension within Somscan & UK concerning the question of lineage. Or rather, the lineage question was more or less abstract or concrete for the different members. At the time of fieldwork, Burao was divided into two areas, mainly inhabited by two different clans. A third clan lived in between the two others. All three were Isaaq, the dominating clan family in Somaliland. As just mentioned, some Somscan & UK members were not Isaag and had never lived in Burao, and they thus had to establish a local connection at the beginning of the project. Therefore, when the Swedish association wanted to buy land, they contacted Dahabshiil, not only the largest Somali money transfer company in East Africa, but also a transnational business company itself. However, Dahabshiil was not only global in this context, it was also local, as the founder of Dahabshiil is from Burao and his brother runs the local branch there. Salman Jamal, one of the Somscan & UK employees, explained:

The first group contacted Dahabshiil. Because with the first group, we said, we have to contact someone we trust. Because we don’t know what is happening there […] We contacted Dahabshiil and he [the boss] was the one who arranged the contract about the land issue and ensured that the owner is the real one […]. But the rest [of the Somscan & UK groups], they found their relatives and some other guys in the city, so the process became easy afterwards.

Through the assistance of Dahabshiil, a piece of land was bought for a favour- able price. The other Somscan & UK organizations were more locally connected and thus did not have to establish their position in the same way. Indeed, buying land in Somaliland is not a simple issue, but requires local knowledge to ensure the seller is the actual and only owner of the land and that the deal is recognized and respected. People often just seized land, I was told, and local government officials might even be selling government property. In other words: to buy land for a new neighbourhood is not something you just do if you do not have the local connections, networks or contacts.

Creating a Somali-Scandinavian neighbourhood

211 When we88 visited Burao in October 2003, the construction of three houses in the Swedish area had just begun, but the plans were ambitious. Changing the desert into a thriving neighbourhood worth living in for families returning from the West required access to water, schools, electricity and new, spacious houses with certain comforts, we were told. Indeed, this was the beginning of a longer process, as the following vignette bears witness to:

Burao, Somaliland, October 2003 Even in the early morning, it is very hot. The landscape is flat, with yellow- reddish sand and small bushes, a few trees. There are many goats, some camels far away, and a few people walking around in the distance. Nothing else. We drive on a dirt road outside Burao. The heart of Somaliland, as people say here. Salman tells the driver to stop. ‘This is the Danish land’, he says. It is silent and peaceful. And empty.

Inside Burao, the atmosphere is different. Unlike Hargeisa, where many houses have been rebuilt, the scars from the civil war are visible here. A lot of houses are in ruins, and old tanks are scattered here and there. Nobody knows what will happen, Salman says, but continues that it is quiet at the moment. There haven’t been any shootings at night for three days now. Apparently, it is a good sign. People from Burao are tough and strong, I am told, and they certainly shake my hand with strength. Salman tells us that you need clan affiliation in order to do things around here, but that he is from another clan, not from Bu- rao. Now, however, he can use Somscan & UK as a reference when he negoti- ates with the local authorities. We joke that Somscan & UK will be like a clan family with different sub-branches – Danish, Swedish, Norwegian, and British.

Later we visit the areas where squatters and returnees from Ethiopia live. First we see the squatters. They have recently been placed there, living next to the Somscan & UK area, but without any roads going there or any facilities. Then we drive further on to the returnee settlement. Both camps consist of huts and tents made out of old tarpaulins and sacks. There is a police station, a health clinic, and a renovated primary school. But otherwise, nothing is there, it

88 I visited the Somscan & UK together with Peter Hansen and a Somali-Danish Somscan & UKmember.

212 seems. It is a far cry from the visions of the Somscan & UK neighbourhood, with spacious, comfortable houses in an area with a playground and a park.

The contrast between the planned Somscan & UK residential area and the squatter and returnee camps was indeed striking. Take a look at the drawing below, showing the imagined Somscan & UK dream house: the big house, the car (a Volvo?), the man leaving the house in a suit, and the two children accompanied by an adult next to the house. It is a dream house. Another drawing of the furniture layout of the house depicts a combined dining and living room with a dining table, a couch and a TV, a modern kitchen, a pantry, three toilets, a bath, three bedrooms, one master bedroom, and two verandas. Certainly, this was a dream house compared to the huts of the returnees and squatters. Compared to more traditional Somali houses, where people sit on the floor and bathrooms and kitchen are built separately, it was definitely also a comfortable and western-style residence. And just as important, it was also a nice house compared to the often rather small apartments in social housing complexes that many Soma- lis in the West live in.

The Somscan & UK Dream House

Drawing made by Ibrahim Yassin and reproduced with his kind permission.

The drawing of the house also had symbolic or representational aspects. The drawing was not only a model of the Somscan & UK house, it could also be said to be a model for a certain kind of lifestyle, a dream house in more than one sense. Not only was it a dream house compared to the usual housing in the

213 West and more traditional homes in Burao – though it was not more luxurious than other new houses built by returnees to Somaliland; it was also a dream house in the sense that it visualized dreams of a middle class life that many might hope for, but fewer have realized. As Salman Jamal said, the drawing shows what you could do, what you might dream to do – not what Somscan & UK members actually would do. The design and comfort of the Somscan & UK dream house, as well as the availability of unlimited electricity and water, also had a gendered dimension, or at least that was what the men told me. As in other cases of considered or actual return to Somaliland, men seemed to be more eager to go than women (Fink-Nielsen, Hansen, and Kleist 2002; see also Goldring 2001 for parallels in the Mexican case). One of the biggest problems, one man confided, was how to persuade the ‘ministers of interior’, i.e. the women, to settle in Burao. Elmi expressed it like this, when I asked if women have any special considerations:

Yes, they require the education and health system to be working. Also, when you build a house, you will have to make it easier for them to work in it. For instance, you will have to have a washing machine, a stove and a fridge and all what you are used to getting here. You cannot go back to working with an old fashioned stove. […] This is why it is so important that we get our own electricity to cover our needs. People use so many electric appliances.

This view was echoed in Somaliland. Salman, however, not only related the question of gender to the comforts of western welfare states in terms of good kitchens or a proper health service, but also to issues of local power relations and notions of gender in general. Asked why the men seem to return first, Sal- man said:

I think the men return first because men tolerate much more, and the women and the children don’t have the patience… The women’s first question is ‘where is the midwife?’ and ‘the children need go to school’. So they send the men out to the hard times and the harsh areas, and they try to figure something out.

Some women are different, Salman explained, but he still doubted that they could resist the severe pressures of local government as well as men did. Paral- lel to the expression of women as the ministers of interior, women were thereby related to the domestic sphere, while men were articulated as the active negotia-

214 tors in the public and political world. It reflects a gendered political culture, in which men dominate the public and political life89. At the same time, it was also an expression of a gendered social order, where men were seen to be the crea- tors of an environment to which women and children can return and feel safe and comfortable. The socio-spatial dimensions of this social order related to how space was perceived and lived – the appropriate places and spaces of gendered bodies. The Somscan & UK neighbourhood was thus not a total transfer of Scandinavian life, then, as Scandinavian ideals of gender equality were not part of the envisioned life in Burao, which rather seemed to be a restoration of more conservative gender and family relations. Like women, teenagers were another group said to require special needs and spaces. Elmi Samir explained:

This is one of the things we consider very seriously to make sure that we do not ruin their lives and their future. We need to be prepared and therefore we are starting with the school; if they get home, they need a real school at a western level. This wealth they have there, they go to day care, they go to school … they have lots of things, lots of activities. If we just move them home without a school, without day care, that will be boring. Now most of our children are born and raised here in this country, in Europe, in the Scandinavian countries, right, so if we just move, our children will have a hard time in Burao … That is why we hope to do some things before we go home. We have to do something for them, like a football or baseball stadium or a playground or some li- braries – places they can use in their spare time […] If we don’t, we can be 100% sure that our children and teenagers will never be satisfied, because they grew up here.

To satisfy women, children and youth, special arrangements were needed. This was reflected in the overall arrangement of the Somscan & UK residential area, where about 15% of all space was reserved for public facilities such as local shops, a market, an elementary school, playgrounds, a mosque, parking lots, a medical clinic, a police station, and a park. If realized, the area would be much like a kind of ‘mini-Scandinavia’, a small self-sufficient neighbourhood that would not need much contact with the rest of Burao. ‘Going home’, then, would be less a question of returning to the local Somali lifestyle and more about realizing a western middleclass lifestyle situated in Somaliland surroundings. Still,

89 Such a political culture is not, of course, unique to Somaliland. Though the context is different, there are parallels to the gendered engagement in Mexican hometown associations, described by Goldring, where “men [are cast] as active and prominent in the “public” realm of politics and women in supporting, passive roles” (2001, 503).

215 as we have seen, the Scandinavian-like public facilities were to be embedded in a social order with more ‘traditional’ or ‘conservative’ gender and family ideals. The visions of the Somscan & UK residential neighbourhood, however, were the long-term goal: the actual realization of it remained vulnerable. Sal- man Jamal expressed it like this:

We might not build all this stuff in the next ten years, it is open. We will see now. Reality now is that we are building houses and fixing water and rehabilitating the two schools in the area and this is where our children will go with the local children. When a lot of families come back, then we must come up with the next plan.

However, this more realistic vision of the future might clash with the expectations of other Somscan & UK members, who required more facilities before they would consider return. The tense clan composition in Burao was another issue, which might make life in Burao risky. Because, as Salman continued, “we don’t know what the future holds here. We have some plans, but we don’t know if they are sustainable”.

The squatter incident The sensitivity of constructing a new neighbourhood in Burao became very visible when the ‘squatter incident’ happened in September 2003. Close to the Somscan & UK area was a big secondary school, the Sheikh Bashir Secondary School. As in so many other public areas, squatters, about 700-800 people, had lived in the school grounds since 1996. The squatters seemed to consist primarily of people originating from Burao, either former IDPs (internally displaced persons) or poor townspeople with no other place to live. They were, in other words, not returning from the west to a comfortable house, but living in their aqal, huts made of old tarpaulins, clothes, sacks and other available materials. Together with returnees from Ethiopia, IDPs constitute one of the most vulnerable groups in Somaliland. In 2003 there were a an estimated 40.000 IDPs in Somaliland (Global IDP Project 2004) and about 215.000 refugees returned from Ethiopia from 1997 to 2003 through voluntary UNHCR-supported repatriation. Returnees and IDPs have been characterized as “among the poorest of the poor” and returnee and IDP camps in Burao as having a desperate need for assistance (UNOCHA 2004). The difference in living standards and conditions

216 of returnees from the west or the Gulf and returnees from Ethiopia or internally displaced people was thus immense. That such differences resulted in differing interests became clear when the Sheikh Bashir Secondary School was to be renovated and the squatters therefore had to move out. First they were told to move to a resettlement area for returnees from Ethiopia, located a few kilometres outside Burao, which we visited. The squatters turned down this suggestion. Then a piece of land was bought for them with the EC project money, located next to the Somscan & UK area, but still further away from the town. This, Somscan & UK members told us, had been expected to satisfy the squatters, as they would be the owners of the new land. However, on the big move-out day the squatters resisted, a fight broke out, one squatter was shot dead and two policemen were hurt. At night some of the squatters burnt down parts of the school, and the next day, students from the school were on the street ready for revenge, rumours said. Luckily this did not happen. After some negotiation, the squatters agreed to move, the students were temporarily relocated in another school, and the rebuilding of the school began, along with the construction of three houses in the Swedish section, “to demonstrate that we are serious about it”, as Salman explained. The squatter incident was interpreted in several ways. The official explanation was that the squatters had received the wrong information and protested against being ejected from the school to the bush. This was the explanation of Elmi as well as other Somscan & UK members in Somaliland, a squatter leader, and government officials in Burao. If only the squatters had been given the correct information, there would have been no problems, people said. Elmi was quick to reassure me (at that time on my way to Somaliland and pretty worried) that everything was under control. Even if the incident was tragic, it would not influence decisions over whether or not to return. Conflicts were to be expected, even if this one was unusual. It was all just a matter of communication. Another interpretation was based on clan. The majority of squatters were said to be from the clan living in between the two larger clan areas in Burao. The area of the school, as well as the land that the Somscan & UK bought, were also said to be largely dominated by this clan. The hypothesis was that the squatters – or rather a warlord related to the squatters – wanted to maintain part of the Sheikh Bashir Secondary School as clan land. A clan territorialization process in other words that the EC renovation interrupted. The refusal of the

217 squatters to move to the resettlement area first proposed could also be seen in this light, as the returnees from Ethiopia were mainly from the two other clans. The new squatter resettlement was named the ‘Ali Hussein Resettlement’ after a hero in the Somaliland resistance movement, SNM, with the same clan lineage as the majority of the squatters. The clan interpretation thus points at the conflict between two kinds of legacies: claiming the right to land in the name of previous settlement, backed by clan power, and claiming the right to land in the name of private property, backed by state power. Furthermore, the name of the squatter resettlement made it possible for the (former) squatters to draw on both legacies: private property and clan. A third interpretation related to the issue of social class. The Sheikh Bashir School was closer to town than the resettlement area the squatters were offered, the school has brick buildings and shade, and the squatters had settled there for several years. Being poor, the squatters did not have the abilities to submit NGO-assisted proposals to the EC, let alone buy their own land and build nice houses. In this way, the squatter incident could be said to mirror the overall hierarchy of social class, mobility and destinations, where the majority of people who have gone to the West have been among the wealthier ones. Due to the re- strictive visa and asylum policies, getting to the West as an asylum seeker is a very expensive process, meaning that poor people or people without connections to relatives in the West or the Gulf states have more limited destinations (Sørensen 2004; Castles and Miller 2003; Lisborg and Lisborg 2003; van Hear 2002a). Getting asylum in the West in contrast to staying in the region circum- scribes future opportunities as well, such as the conditions under which an eventual return takes place. In this perspective, Somscan & UK enlarged already existing class divisions in Burao. Still, the squatters also had their say in the development of the course of action: they were not deprived of agency, even if they were in the hands of a ‘warlord with money and guns’. However, while the squatters succeeded in getting a better deal than they were offered at first, the local and national government of Somaliland acted in favour of Somscan & UK to promote the return of ‘the diaspora’. The squatter incident thus exemplified several of the challenges and conflicts that Somscan & UK – and more generally return and reconstruction projects – face and have to deal with. Like the participants in the DOEA conference, Som- scan & UK sought to replace or supplement the clan logic to promote principles

218 of territorialization – in this case, resulting in the emphasis on private property as well as democratic principles of governance, showing “a collective point of view”, as Elmi and the Somscan & UK paper expressed it. Other members explained that some of them might become future politicians, maybe even ministers, though Elmi Samir strongly disassociated himself from politics. Though a multiparty system had been introduced in Somaliland in 2003 and some of the Somscan & UK were politically active, Somaliland politics was often related to corruption, though not necessarily to violence as in the south. Still, like DOEA, Somscan & UK could be characterized as potential political actors with visions of governance, reconstruction, and principles of collective identification (cf. Mouffe 1997) – and furthermore, as an actor that had the support of the local and national governments. Elmi, however, saw his role within the social organization of people and in relation to associations, passing on his experiences from Denmark. As he, the DOEA organizers and many others phrased it, ‘the diaspora’ should transfer skills and competences to their homeland. Like we also saw in the analysis of the DOEA conference, ‘the diaspora’ seemed to be endowed with special moral cause and responsibility.

‘The diaspora’ at work The expectations of ‘the diaspora’ were thus widely shared. I now discuss the DOEA conference and Somscan & UK together, to analyze how different institutions and actors in Somalia, Somaliland and elsewhere mobilized the category of ‘the diaspora’. As I did not undertake fieldwork in the southern provinces of Somalia, it is impossible for me to analyze how the notion of ‘the diaspora’ was employed there. But it seemed to be used in the same way. For instance, in a letter that Dr. Adam Ibrahim from the DOEA conference passed on to me, the Council of Elders in a town in the south of Somalia appealed for support for their health facilities, mentioning the help from the “Somali Diaspora living in USA, Finland, UK, and Denmark”. In Somaliland, ‘the diaspora’ was widely used as a category to indicate the Somali populations abroad. Both the mayor and governor in Burao had clear expectations of ‘the diaspora’. The governor explained:

It is good if the diaspora returns with money and help with their accommodation. We thank the countries that they are living in. If they cannot take care of themselves, they should not come back, but send money. As the government we cannot afford their accommodation, as they demand

219 more than the local people. They must come with their own materials and money. When we find oil and diamonds, then they can come back. We’re happy that some come back, but as a government we cannot afford it…

Returning from the West with empty pockets was not welcome. ‘The diaspora’ was expected to take their responsibility in rebuilding Somaliland by supporting people economically and initiating projects that would benefit the local people as well. The economic support of family members was not expected to cease, even in the case of return. “If you have left without money, you cannot come back without money, resources or land”, as the mayor said. The people who did return without these resources were scorned, called nafo, handicapped. The logic seemed to be that unsuccessful returnees from the West should not strain the fragile and very poor Somaliland economy; if people could not contribute, they had better stay where they are. The reality of such expectations was also emphasized by returned Somalis from the West that I talked to. To return not only required the ability to take care of oneself, but also to support other family members, even quite remote ones. Thus, while relatives of Somscan & UK members in Burao were happy about the prospect of their sons’ or brothers’ possible return, they were also worried about the prospects for their livelihoods and, thereby, their own livelihoods, which often consisted of remittances sent from the West. Such contributions were not expected to stop in the case of return. Salman, for instance, supported his parents with 200 USD every month and spent about 100 USD on people who came by asking him for money. The economic support from the West was a crucial livelihood strategy, as remittances are the main resource of hard currency in Somaliland and the sole economic income for many households (Horst 2004; 2003; Gundel 2002; UNDP 2001), as we have seen. This ‘economy of sharing’, in which the wealthier provide for the less fortunate, was explained both as an accepted cultural expectation and as a burden, which makes it difficult to return (see also Horst 2004; 2003). The expectations of continuing economic support of relatives, and of being in the vanguard of the rebuilding of the country indicate that people from the West were expected to maintain their social position as (often) more privileged than the local people. Or in other words, the moral economy of repatriation as ‘the ideal solution to refugee situations’, where refugees return from exile to rebuild their homeland and live like the locals without exacerbating class divi-

220 sions (cf. Fink-Nielsen and Kleist 2000) was a far cry from the realities of people from ‘the diaspora’. Rather, they were expected to make a difference through sending remittances or through initiating development projects. At least on the official level, return was only welcome, if they could continue these efforts – otherwise it was not.

Social remittances One way to analyze the expected role of ‘the diaspora’ is through reference to social remittances, a term suggested by sociologist Peggy Levitt (1998). As the ‘social’ in the term indicates, the concept extends the notion of remittances from economic support to broader visions and transfers that migrants and their counterparts engage in. Levitt defines social remittances as “the ideas, behav- iours, and social capital that flow from receiving to sending country communities” (1998, 962), which may result in different kinds of social change90. Ac- cording to her, there are three main kinds of patterns of interactions: recipient observers (passively taking in new ideas); instrumental adapters (altering rou- tines for pragmatic reasons); and finally purposeful innovators (searching, se- lecting, and absorbing new things). Likewise, Levitt distinguishes between three types of social remittances: firstly, normative structures (ideas, values, be- liefs); secondly, systems of practice (“actions shaped by normative structures” (ibid., 934) and organizational strategies); and thirdly, social capital understood as the transfer of prestige and social esteem. Both the DOEA and especially Somscan & UK could be said to be examples of purposeful innovators, seeking to change normative structures and systems of practices – especially in relation to the challenge of the clan logic. Regional identification (and its consequences for governance), the right of private property (and the consequences for the regulation of settlement) as well the planned mix of different clans in the Somscan & UK residential neighbourhood, comprised elements of normative structures and systems of practice. As we have also seen, both the DOEA conference and Somscan & UK led to the transfer

90 At a seminar about social remittances, arranged by the Danish Institute for Inter- national Studies, November 9, 2005, Levitt commented that she has extended her notion of social remittances from a one-way transfer to multi-directional, multi- layered and circular flows. While this latter approach no doubt is more theoretically adequate, I think the earlier approach is useful in analyzing how many actors articulated their expectations of ‘the diaspora’ as a one-way transfer.

221 and establishment of social capital in Bourdieu’s sense of group membership, as when Somscan & UK became a new collective actor through which Salman, without lineage affiliation to Burao, was able to negotiate with the local authorities. Likewise, the DOEA conference provided an opportunity for people to meet, get to know each other and – through the recording, photographing and videoing of the proceedings – to disseminate the message of the conference as well as their engagement and contributions. While economic and material support was indeed considered important and necessary, it was definitely not the only focus and purpose for transnational reconstruction actors such as DOEA and Somscan & UK, which saw themselves as agents of development and modernity, contributing to new scenarios for the future. They were indeed engaging in purposeful communication (Levitt 1998, 936). This was also apparent in a smaller project that Omar had initiated, the Water Project. The Water Project was very different from both Somscan & UK and the other transnational reconstruction projects which I heard about. It was a small Somali-Danish organization, with both Somali and Danish members, who were primarily living in Omar Said’s neighbourhood in Copenhagen. The project was not related to repatriation or return, but had as its aim the provision of clean water through the establishment of a pump in a town in southern Somalia. In 2002, a big Danish company had donated a water pump to the Water Project and money was collected from both local Somali-Danish members and various Dan- ish sponsors. After more fundraising, the Water Project succeeded in collecting enough money and Omar travelled to Somalia, where he monitored the installation of the pump. The Water Project, though perhaps the smallest and least ambitious presented at the DOEA conference, was still related to bigger processes of change and development. As Omar explained, the plan was not simply to establish and donate the pump, but to create an association, inspired by the Dan- ish tradition for co-operative organizations, andelsbevægelsen. Before he left, he explained to me:

I think we should discuss how our experiences from the Danish associational life can contribute to the local community. When I go there, I will establish an association with an executive committee, a chairman, and all that, and I will translate ordinary Danish statutes. So in that way, they can learn a lot of things from us, right.

222 For Omar, then, the water project was necessary in itself, as access to clean water was badly needed in the area, but it was also part of the overall purpose to contribute to new methods of organization and governance in the local community, or in other words, to initiate learning processes of democratic and responsible organization. He hoped that the establishment of a co-operative association would make the local people feel responsible for the pump. But he also hoped that it would contribute to a new vision of society and counteract tendencies of dependency. He said:

Our purpose is not only to provide clean water, but to create another society, and to share our experiences with them and tell them the rules of the game. […] They have to learn that nobody just donates things all the time. They must contribute something to their community as well. We can try to motivate people, to change the way they think, so there is a feeling of peace and stability, a feeling of society and community. In that way, we can share our experiences.

The water project, then, was part of a larger vision of development, where new ways of organizing and of perceiving community and society were emphasized in terms of local initiative and responsibility. Like Elmi in Somscan & UK, Omar wanted to share his experiences of associational life in Denmark, not only installing a water pump, but also hoping to instil initiative, responsibility and democratic practices – or, in other words, democratic norms as well as practices. In Elmi’s words, Somscan & UK members should share their experiences with the local people, because “the people in the country don’t have the resources, they don’t have the mentality we inherited from Europe”. They were seen as different, as the Somscan & UK paper also stated (Jama and Jensen 2004, 2). For Mahmoud Sheikh, exile in Denmark had also given him a new perspective on politics. He explained to me:

The only thing we think about, is when will there be peace and stability in our homeland. We have seen one hopeless peace conference after the other … So we thought that this is the chance for our generation … who live in the West and who have seen their way of dealing with the problems, their ways of doing politics.

Both Omar, Elmi and Mahmoud thus explicitly related these ambitions with their experiences in Denmark – or more broadly in the West – because of the exposure to democratic systems, efficiency, and associational engagement.

223 They thereby accepted the moral responsibility to try to share their experiences and make a difference. In the DOEA case, the participants also envisioned an alternative way of doing politics from the warlordism in the south of Somalia.

‘Walking with two legs’: Salman Both Somscan & UK and the DOEA conference had reconstruction and return as their central aims. Likewise, Somscan & UK members and DOEA conference participants situated their engagements as ‘homeland’ and ‘diaspora’ commitment, emphasizing their moral responsibility for sharing experiences and skills. As we have seen, the clan tensions were also a shared concern and challenge. There were, in other words, a lot of similarities – in spite of the obvious different political context between the southern provinces of Somalia and Burao in Somaliland. It was, however, still very different to return to the southern provinces of Somalia to ‘check things out’, as Adam Ibrahim had done. Though a Somaliland trip did require gifts and surplus money, it certainly did not demand the same security considerations as a visit to the southern provinces. To return to Somaliland for a longer period of time, however, was another story: a long-term process, where different conditions should be fulfilled. As we have seen, these included peace, security, a livelihood, a sufficient educational and health system, and a house. And, in most cases, western citizenship to secure continued mobility and access to the West. Salman referred to this as a strategy of ‘walking with two legs’: keeping relations to the Scandinavian country of residence as well as to Somaliland. We might also say that Salman and others were situated in a transnational social field of “interlocking networks of social relationships” (Levitt and Schiller 2004, 1009), which both provided obligations and opportunities – for instance through holding a western passport. Indeed, as I and others have pointed out, voluntary repatriation to Somaliland is more likely to happen with the acquisition of a western passport than through the repatriation support offered by European governments91 (Fink-Nielsen, Hansen, and Kleist 2004; 2001). Furthermore, return often seemed to be of a more circular and temporary character (Hansen 2006) than the permanent

91 During my stay in Somaliland, I heard about just one family who had used the Danish repatriation support and stayed so long that they had lost their right to return to Denmark. All other Somali-Danish returnees, who were staying for a longer time, held Danish citizenship.

224 movement envisioned in mainstream ideas of repatriation. A western passport as a security mechanism was considered important because of the unstable situation in Somaliland, which still has not been recognized by any country. The Somscan & UK members shared this anxiety – and the strategy to overcome it: about 90% of the Danish members were estimated to hold Danish passports. Also in this way, the Somscan & UK was a ‘dream project’: it required an optimistic belief in the future and stability of Somaliland to invest in – not to mention return to – Somaliland. The focus on repatriation no doubt stimulated the policy interest and most probably also the economic support from the EC. Still, it would be a mistake to believe that all the 330 Somscan & UK member families would actually return immediately and permanently if only water was supplied and schools renovated. These surroundings were part of the preconditions, but not decisive. As expressed in the Somscan & UK paper where the flexible time horizon was emphasized:

The vision of a “fully equipped” neighbourhood with playgrounds, schools, police station, health clinic etc. remains far in the distance, but the members are equipped with patience and willing to progress step by step, not following a set time frame. Some are ready to repatriate shortly, whereas others will await more facilities to be in place before they will give up their lives in Scandinavia/UK and return their families. Even, some might end up not returning at all (Jama and Jensen 2004).

When I asked Somscan & UK members about their plans to return, there was a tension between hopes, ambitions and preconditions. One man explained that he needed more education and that his children first needed to grow up and get an education. Elmi, who had wanted to return to Somaliland since I first met him in 1999, still wanted to go, but did not have money to build a house. Another man had spent much time in Burao, but had not built a house at the time of fieldwork. Neither had Salman. All of them thus continued to be connected to their Scandinavian home countries. The ‘two legs’ strategy could also be seen to reflect this fact. For Omar and Mahmoud, the question of return was less complicated in certain ways: due to the political situation in the south of Somalia, it was simply not an option. Though Omar did return to monitor the installation of the water pump, his life was in Denmark. “I have children in Denmark and they have already started school, so it seems like I will stay in Denmark”, he said and continued, “I think that so long as I stay in Denmark, I will try to feel a part of so-

225 ciety and contribute to it”. Omar, then, wanted to make a difference in more places – both in Denmark and in Somalia. Likewise, Somscan & UK members in the West mentioned ‘making a difference’ as one of the prime reasons for returning to Somaliland. Elmi expressed it like this:

It is mostly that we should participate in the rebuilding and the only way to do that is to contribute yourself and go and do something about it. […] In a war, you know, all the educated people, all the people who are important to the country are the first ones to move. And when they move here, even if you are a doctor or an engineer or a professional secretary, they don’t use you. But if you go back, you are a person who is respected, you are a doctor, but here you are nothing. So the people living in Somaliland or the people living abroad, they say, okay, we want our trained people. […] So when they come back, they can do something, they can participate in the reconstruction.

Returning to Somaliland was thus also a possibility for upward social mobility and respect through the recognition of contributions, social esteem and qualifications – both in the sense of recognition of achievement and social esteem, as Honneth suggests, as well as in terms of participation (cf. Fraser and Honneth 2003). Coming to Denmark or anywhere else as an asylum seeker, having to start one’s education all over again or being on public welfare was sometimes experienced as an immense decline in social esteem and life opportunities, as we have seen. Given these circumstances, a straightforward analysis of the question of return to the Somscan & UK neighbourhood could be that return – once sufficient facilities are established – would simply offer the best of both worlds: Scandinavian comfort in a Somaliland environment. This might be the case for some of the Somscan & UK members I talked to in Somaliland and Denmark, who were not part of the executive committee, but who had returned or planned to do so because they felt unwanted in Denmark, where they were or had often been unemployed: where they had gone through experiences of social subordination and misrecognition, in other words. For this group, the engagement in Somscan & UK seemed to be less framed in the rather abstract development and nation building discourse; it was more about emphasizing the opportunity to buy cheap land, probably one of the best investments in Somali- land. For the Somscan & UK organizers, however, the reasons for return were, as we have seen, meshed with the discourse of rebuilding Somaliland and making a difference – rather than getting out of Scandinavia or the UK or making a

226 good investment. None of the leading members of Somscan & UK expressed any personal encounters of racism, though all the interviewees agreed that the situation was difficult for Somalis in Scandinavia. The positions of ‘being a successful man’ and ‘victimized Somali refugee in a hostile host society’ did not seem to be compatible. We might say that the Somscan & UK organizers were taking in terms of the Somali stereotype through their engagement, making it a non-issue, a non-component of their identification. Another aspect of this positioning was that the organizers actually seemed to ‘have made it’. Though they might not all have realized the educational ambitions they had dreamed of, they had practical and organizational skills, they held jobs, but worked extra hours for Somscan & UK, and might even be engaged in other kinds of community work as well. They were, in sum, organizational nodal points with lots of social capital by virtue of their engagement in a range of activities and relations. While they – with the exception of the two employees in Bu- rao – did not earn a salary, they did gain experience, respect and social esteem in the Somali communities in the West and in Somaliland. In this way, their social esteem seemed to be related to positions within the ‘Somaliland diaspora’ and in Somaliland (cf. Goldring 2001; 1998; Schiller and Fouron 2001), rather than in relation to the western host societies, where transnational engagement is much less appreciated. While the question of return was much less acute for the DOEA participants, the same dynamics seemed to be at work: recognition of social esteem was established and accumulated within the transnational fields themselves, but could not necessarily be transferred to the Danish context.

Chapter resume: Agents of modernity and development Through the case studies of DOEA and Somscan & UK, I have showed how articulations of ‘the diaspora’ as agents of development and modernity were shared. Both Somscan & UK members and DOEA participants framed their engagements between ‘homeland’ and ‘diaspora’ commitment, where the latter was articulated as a moral cause to share and transfer their skills, competences and economic sources that a life in western democracies were supposed to ensure. The diaspora discourse, as we saw, was not only articulated by western- based activists, but just as much by authorities in Somalia and Somaliland. Identifying as ‘the diaspora’ (cf. Axel 2004) offered individual migrants and

227 migrant groups to opportunity to position themselves as part of a moral community and be recognized as responsible and potential political actors. ‘The diaspora’, however, was not only mobilized by migrant groups themselves, but also by their counterparts ‘at home’, for whom the contributions of ‘the diaspora’ were urgently needed. The mobilization of ‘the diaspora’ was thus a process that more players were engaged in and had interests in. Likewise, members and participants in Somscan & UK and DOEA shared the challenges of dealing with the clan logic and the ambitions to reckon with it. While they both could be described as inter-lineage organizations, they clearly had regional delimitations as well. Also, while both actors wanted to displace the lineage logic with modern principles of governance in terms of private property rights and regional identification, they were also embedded in the clan logic, having to deal with clan realities on the ground (as Somscan & UK did) or in the organization itself (as the DOEA participants did at the conference). There were differences between the two actors, though. The security situation was unresolved in the south of Somalia, making return much more difficult. In Somaliland, however, Somalis from the West were returning in larger numbers – even if mostly going back and forth – and expectations to their contributions were high. Should Somscan & UK members return, they were expected to keep on contributing to their families and the development of Somaliland. Return with empty pockets is not welcomed in Somaliland and most (potential) returnees seemed to engage in a transnational ‘two-legs’ strategy, maintaining relations, belonging and rights to both the West and Somaliland. They were, in other words, situated in a transnational social field (cf. Wimmer and Schiller 2002), consisting of both obligations and opportunities. Finally, both the Somscan & UK project and the DOEA conference were opportunities to establish and accumulate social capital and recognition: to make contacts and be recognized for one’s competences, achievements and social esteem. As we saw, these engagements had gendered aspects, both in terms of the visions of the Somscan & UK residential area, but just as striking in relation to the gendered organizational space: both were exclusively male. In the following discussion, I elaborate on the gender aspect, linking it to questions of recognition, social esteem and participation. I thereby attempt to gather the theoretical and empirical threads of the dissertation, returning to the questions I posed in

228 the introduction and on the way in this analytical journey. I turn, in other words, to the theoretical and analytical horizons. PART III

HORIZONS

229 230 Chapter 8 Spaces of recognition

I don’t think there is anything preventing us from integrating in our society, on the one hand, and supporting the reconstruction in our homeland, on the other (Omar Said, 2003).

How do Somali-Danes negotiate positions and struggle for recognition? How do their involvements in local and transnational associations relate to processes of inclusion in Denmark? And how can we understand the proliferation of claims and identification made in the name of diaspora? In this chapter, I revisit these questions through a further elaboration of my analysis and the theoretical discussions. I claim that associational engagements form social spaces in which achievement, agency, and cultural intelligibility can be performed and recognized. I term this phenomenon for spaces of recognition. Spaces of recognition refer to the ensemble of social relations, generated in relation to associational engagement, which enable, establish, or negotiate recognition. It is thus not the associations as such which are the spaces of recognition, but the effects of associational engagement. My concept of spaces of recognition is thus close to Bourdieu’s notion of field and his emphasis on struggles for symbolic capital. Like Bourdieu, I delimit the ‘borders’ of spaces of recognition as where the social relations cease to have effects, and I also claim that games of positioning are negotiated and struggled for within this space (cf. Bourdieu and Wacquant 1996). However, by proposing the concept of spaces of recognition, I want to highlight the double analytical perspective of recognition as relating to, on the one hand, social struggles and, on the other, cultural intelligibility. This double focus thus directs the research attention towards both what social actors struggle for – how they do it – and the negotiations of ideals of ‘proper’ comport- ment, gender ideals, and social positions. Furthermore, in an analysis concerned with transnational involvement, the spatial dimensions of recognition are important, raising questions of where and in which social spaces recognition is struggled for. I start this exploration by taking the readers on a walk around Southall in London.

231 Struggles for recognition

Walking around Southall During the DOEA conference, I had talked quite a bit with some of the Somali- British participants living in London and told them about my plans to study in the UK. One of them, Mr. Ismail, a pleasant middle-aged man, had told me to call him when I was in London and so I did. About a year later, then, I met with Ismail and two other DOEA participants at a Somali restaurant in Southall, a multiethnic neighbourhood in west London. After a delicious meal and a chat about the economic and political prospects in the south of Somalia, we went for a walk around Southall. That evening I wrote the following field notes:

While we are walking the streets of Southall and visiting various Somali shops, we encounter a number of men that my companions know. We meet a medical Doctor (a cardiac surgeon, I am told), a former top politician, and several other important people. The men greet each other, exchange a few words, and introduce me. […] After our walk, Ismail ac- companies me back to Victoria Station. During the long train ride, he explains to me that the men appreciate it when they are addressed with their title, such as Doctor, President, Admiral, or Director. Here in Lon- don they are often greeted just by their name, like a teenager, he sighs, and continues that it is very difficult to have been in a high position and not be recognized with respect (Excerpt from field notes, London, March 2004).

The surgeon, the politician and, not least, Ismail himself, once a successful banker, now earning his living as a bus driver, were all examples of men who had experienced a huge loss of status when they came to the West, unable to continue their professional lives and uphold their positions as men of status, competence and qualifications in British society. The fact that they – all part of the (former) Somali elite – were running into each other in the run-down streets of working-class, multiethnic Southall illustrates an enormous change of life circumstances. According to Mr. Ismail, life in Britain may be even further aggravated by the absence of respect, when men of status are ‘greeted like teenagers’ – with their first name and without the appropriate titles. Names, however, are not ‘just names’ in a Somali context, but genealogy; they connote history, identity, and status, reminding other people of ‘who you and your family are’, “validating membership in a group” (Luling 2006, 471). Ismail, for instance, was the son of sultan, a traditional Somali clan leader, and his name conveyed

232 social esteem and respect, positioning him – among other Somalis and other people who knew him – in a radically different position from his job-related position as a (black) bus driver. Acknowledging social position and competences, showing respect, and greeting in appropriate ways were all ways of recognizing social esteem – of ‘repairing’ and re-enacting prestigious social positions, respect, and masculinity. Now, how far can we take Mr. Ismail’s example in a Danish context? Quite a long way, I will argue. Firstly, while the UK has a reputation for being more accommodating towards Somalis than Denmark, and the British employment market for being more open, un- and underemployment is a big problem among Somalis in the UK as well (Bloch and Atfield 2002; Bloch 1999). Still, due to fewer language problems and business regulations in the UK, employment issues might be more difficult to address in Denmark than in the UK. Secondly, several informants emphasized that, as a former colonial power, the histories of the UK and Somaliland are interwoven. Thirdly, the UK has explicit multicultural policies, whereas Danish political debates are often specifically against multiculturalism92. According to David Griffiths (2002), Somalis in London are invisible and generally ignored, while the situation in Denmark might be characterized as one of exaggerated media and political exposure. However, stigmatization and social problems seem to take place both in Denmark and the UK (cf. Harris 2004). Likewise, the break-up of family and gender relations appears to be shared in the UK, Denmark and in many other western countries (e.g. Gardner and Bushra 2004; Griffiths 2002; Affi 1997; Utteh 1997). In short, comparing Denmark and the UK at this general level leads to the conclusion that the UK certainly has a better reputation than Denmark, which is increasingly notorious for its tough asylum and immigration policies: the overall situation in both countries, however, appears to be much the same in many ways.

Recognizing achievement One of the questions that the walk around Southall with Mr. Ismail and his associates brings to fore is the degree to which positions from Somalia can be used in the new context in the country of residence to establish authority and respect. This question relates, in other words, to if and how positions and recogni-

92 Like Denmark, the UK is also tightening its asylum and immigration legislation.

233 tion can bridge, or are interrupted by, different contexts in transnational social fields. Let us have a closer look at this question. In the dissertation so far, I have argued that the stereotype of ‘difficult Soma- lis’ creates a context of negative expectations that individual Somali-Danes may have to deal with in their encounters with Danish society. Following Nancy Fraser’s suggestion that the normative goal of recognition is to participate as peers, I will claim that the general standing of Somali-Danes is often connected to experiences of misrecognition and social subordination, hindering their full participation and inclusion. As I showed in chapter five, many of the informants experienced the circulation of the public stereotype as extremely difficult, lumping them all together as subordinated women or lazy men, uneducated, probably illiterate and, in all cases, as not belonging to Danish society. Seen through the light of the public stereotype, Somali-Danes are culturally intelligible as strangers (cf. Ahmed 2000), not as individuals and, following Fraser, not as peers. This experience of misrecognition might be felt as a split between one’s self-perception and general societal identification. To recall, Honneth and Taylor insist that identity is constituted through social interaction. If there is no social interaction, no dialogue, identities cannot be negotiated – and the stereotypes cannot be challenged. However, I am not going to claim that the context of stereotyping distorts identity formation in Taylor and Honneth’s sense. I am claiming that it definitely makes the negotiations of positions, achievement, and social esteem more difficult vis-à-vis Danish society – and it makes it harder to participate fully as a peer. This last aspect is emphasized by the fact that it is very difficult to transfer educational and employment-related qualifications to Denmark, which again makes it even tougher to counter the stereotype and to obtain recognition for merit and achievement in the country. In some cases, this might result in a vicious circle of (experiences of) misrecognition. Let me emphasize once again that there were major differences between the informants’ exposure and reactions to stereotypes. Some informants, like Saphia, distinguished between the public stereotypes and her friends, family and colleagues as two different faces of Danish society. This division reminds us that though stereotyping could be seen as very frustrating, many of the informants also had friends and colleagues who appreciated and recognized them as friends and as colleagues without paying attention to ethnicity and nationality. Other informants had established positions as experts on issues concerning

234 Somalis in Denmark or, as we saw it in chapter six, as ethno-national spokespersons of Somali associations. Indeed, some of the associational key persons received public (and well-deserved) praise for their efforts, as well as economic support for their associations – especially, it appeared, if they addressed the Danish political priorities of employment, gender equality, or repatriation. Still, these cases were relatively few. And yet the key activists did not usually escape from the stereotype as context; rather, they were positioned differentially to it, as exceptions from it and/or experts on it. Finally, some were in a situation characterized by the seemingly impossible or very difficult transferability of achievement from Somalia in terms of recognition in Danish society, as I argued above. I suggest that Somali associations form social spaces where recognition of achievement and social esteem could be established and individuality ‘re- stored’. Obtaining recognition in associations or through associational engagement, however, is not an unproblematic strategy. As Honneth (2006; 2003) argues, in contemporary, liberal, capitalist societies, recognition of achievement is primarily about being recognized as a productive citizen – to have gainful employment. In a Danish context, migrants’ employment and self-provision is publicly debated as one of the biggest challenges to the survival of the Danish welfare state. While employment is determining the rights to social benefits for all citizens in Denmark, it also forms part of the conditions for obtaining a permanent residence permit and naturalization93 for third country citizens, such as Somali nationals. The rights and social importance related to employment are thus immense in Danish society. However, as Honneth reminds us, what is evaluated as work is a result of the historic development of capitalism and the ‘meritocratization’ of social esteem “according to his or her achievement as a “productive citizen”” (2003, 141). Merit is thus not a neutral category. Some activities are evaluated ‘as gainful employment’ while others fall outside the “result of a group-specific determination of value” (ibid., 141), for instance housework, much care work, and a lot of the voluntary work undertaken in associations.

93 With the tightening of the Danish nationality law in December 2005, applicants for Danish citizenship must have been self-providing for at least four years out of five (Ministry of Integration 2005a, § 23, 2).

235 The suggestion that recognition for achievement and social esteem can be obtained through associational engagement is thus not that straightforward. Firstly, while associations are valued as a cornerstone in Danish civil society, involving oneself in such an organization does not count as gainful employment but, as stated in the Manual for new members of Danish Society, as leisure time (Ministry of Integration 2003). Associational engagement might thus be a way of achieving social esteem and using qualifications in a sphere of society other than work; it does not attract societal recognition, or ensure rights related to employment. This hierarchy of values is demonstrated in the departmental order concerning permanent residence permits, where engagement in associations carried out alongside one’s employment and promoting integration is emphasized as a sign of significant attachment to Denmark (Ministry of Integration 2005c, § 3, 2). It is thus a specific kind of associational engagement that counts, but still only as a kind of ‘top-up’ on employment, not in itself. Furthermore, in contrast to the US and UK where migrant organizations often have employees (cf. Hopkins 2006; Cordero-Guzman 2005), Somali-Danish associations generally do not offer such prospects for employment. Thus, while associational life is respected and praised in Denmark, it rarely contains any economic prospects, and it thereby neither ensures economic nor political rights as such. My point here is not to make a judgement as to whether this is fair or not. Indeed, this is a much larger discussion relating, among other things, to the economic composition of the welfare state and the nature of capitalism. Still, as Honneth emphasizes, the question of what counts as ‘gainful employment’ is a socially and culturally constructed phenomenon, a debate which may well be the focus of future struggles for recognition (Fraser and Honneth 2003, 188; cf. Willig and Petersen 2004). For now, my point is to emphasize that in Denmark and other capitalist societies, paid work is more valued as a contribution to society than associational engagement. Or, in other words, involvement in Somali associations and reconstruction projects does not necessarily result in general societal and public recognition but, I will claim, rather in group-specific recognition.

Repairing gender (dis)order If associational engagement does not generally lead to recognition as productive citizens or peers in society, then what were associational key persons and or-

236 ganizers recognized for and recognized as, and by whom? I will argue that gender and social esteem were two central aspects. As I showed in chapter five, Somali gender relations in the West are intensely debated, creating a gender (dis)order (Connell 2005). In a transnational perspective, however, there are not only local gender orders, but also others from more places (cf. Levitt and Schiller 2004, 1015). The gendered stereotype of Somalis in Denmark, the UK and elsewhere, along with more traditionalist and conservative gender ideals are part of local gender orders which have transnational dimensions as well, reminding us of Massey’s (1994) insistence on the simultaneity of spatial scales in social spaces. One aspect of gender orders is in the context of relational positioning (Brah 1996, 183), creating a social space where discourses and ideals of masculinity and femininity are identified, negotiated or contested. What is more, gender, social class and (loss of) social esteem are closely related. This intersection of positions can be seen as part of the gender (dis)order where the social, economic and political changes in the wake of globalization have caused a global ‘masculinity crisis’ (cf. Kimmel 2005; Turner 2004). For researcher Michael Kimmel, the masculinity crisis is related to situations where men “believe themselves to be entitled to power – by a combination of historical legacy, religious fiat, biological destiny, and moral legitimacy – but they believe they do not have a power” (2004, 418, emphasis in original). The social mobility of these men, often of working class and lower middle class background, is either blocked or on a downward path; they are losing their positions as breadwinners, losing their desired livelihoods, losing esteem and authority and, perhaps, losing a feeling of control over their lives – and reacting against it. In the cases Kimmel analyzes, the responses are militarist masculinity such as white su- premacists and terrorism. His analysis thereby emphasizes Honneth’s point: that the discrepancy between expectations of or entitlement to recognition and experiences of the absence of recognition result in frustrations and in fuel for social and political struggles. Kimmel’s examples also remind us that struggles caused by experiences of misrecognition are not always liberating, but might very well be violent and abusive. However, Kimmel’s analysis is very different to mine. First and foremost, I am not analyzing radicalism and militarism. Secondly, most of the informants were not of working class background, but had been part of the Somali middle-

237 class or, in a few cases, the elite. Still, in many ways, the situation of some So- mali men in the West can be seen as a parallel to the loss of expected upward social mobility – or indeed the experience of downward social mobility – and the ensuing loss of entitlement to power that Kimmel describes. Furthermore, due to the consequences of civil war, the situation might be even more extreme, as the loss of status, power, and rights is accompanied by the loss of close family members, loss of wealth and property, loss of home, loss of well-known surroundings – and often further aggravated by difficult and low-status positions in the western countries of residence. Experiences of inequality in societies where equal rights are highly praised might cause frustrations (cf. Fraser and Honneth 2003). Obviously, this situation also applied to women. Still, it was striking that men’s loss of social position and social esteem was much more articulated than women’s; that men were seen as having lost more than women, who often were viewed as gaining more power in the western countries. The idea of (more) powerful women was supported by the stories of ‘failed’ men not contributing to the daily chores, and certainly not fulfilling the gender ideal of being breadwinner and the head of family. Indeed (some) men felt that women were favoured. “Here the men cannot even control the television”, a Somali-British man said as an illustration of the loss of male authority in the family that, according to him, many Somali men face. Likewise, Somali women also experienced change in social position and loss of social esteem in the West, and found that they could not realize their ambitions or use their qualifications. Remember Aisha Ahmed, who came to Den- mark at the beginning of the 1990s and could not continue her professional career. Let’s repeat the last words in the life story:

It was horrible to come to Denmark. I was always a very active person and here I became a passive person, another person. I often miss a lot of my inner person, my old self. My friends and family have noticed it too. ‘What has happened to you? they ask. I worry so much. When I told my social worker about my career, she did not believe me. She did not even bother to read my file.

For Aisha, coming to Denmark meant a loss of social and economic opportunities as well as misrecognition of her work-related achievements. It resulted in a much more difficult economic situation for her and her kids, who ‘used to have more than just enough’, as she said. Other female informants also told me about

238 their relatively privileged backgrounds in Somalia, contrasting them to their present situations in Denmark. Still, it remains that men’s loss of social position and social esteem and possibilities was more articulated and debated among both male and female informants than women’s situations. Now, how can we understand this situation? Below, I will suggest two analyses: one focusing on empowerment and one informed by theories of positioning and performativity.

Empowerment or cultural intelligibility? One way of understanding the emphasis on men’s loss is as a mirror of the fact that while many women have gained access to their own income (or, alternatively, social security), many men have lost privileges and respect, especially if they are unemployed in the new country of residence. Furthermore, men have lost the ‘right’ – morally and legally – to control their families, if necessary by means of violence. There are numerous stories circulating about the ‘999 option’: women, children and youth calling the police to report their violent or threatening husbands or fathers (cf. Harris 2004, 63). I did not come across a Danish equivalent of a ‘112 option’, which is the emergency phone number in Denmark. Still, the theme of women rejecting and taking advantage of their husbands was recurrent. One Somali-Danish man put it like this:

The women are between two cultures, and that is a problem. There are many problems in the families and many divorces. As soon as the women get freedom and their own money, they get westernized and ad- dicted to having their own money. […] In Denmark, the woman wants the man to stay and share the work. They say, ‘you are not supporting, you are not a real man’. Some men are leaving or are being thrown out. It is very embarrassing to be thrown out of your own house, maybe even by the wife that you brought here.

This man thus connects money, ‘cultures’, expectations about gender ideals and the division of domestic labour with masculine pride and shame, arguing that the women cause problems because they are between cultures and are expecting men to adapt. Such ‘cultural impurity’ or, rather perhaps, ‘Western cultural overload’ was often framed in terms of too much liberation and too much independence, resulting in rejection of ‘traditional’ family values, making women

239 disrespect, divorce and throw out their husbands94. In this sense, yes, women were empowered and men had lost privileges. Losing social positions and gendered social esteem has aspects both of changing socio-economic and socio-cultural relations. Nancy Fraser’s point of bivalent collectivities (1995), referring to how gender, ‘race’, and other modes of differentiation are embedded in both socio-cultural (mis)recognition and economic (mal)distribution, illustrates the ‘masculinity crisis’ as one of simultaneous loss of respect and authority, as well as of economic opportunities and control. However, there are also certain problems with this explanation. To claim that Somali men’s loss of social position is more articulated than Somali women’s because men have lost more – i.e. a corresponding relationship between practice and discourse – ignores the differences between men and between women as well as presupposing a certain set of pre-civil war gender relations in Somalia, where men were breadwinners and women were housewives. Certainly, while this might have been the case for some families, maybe many, it would be wrong to conclude that it applied to each and everybody. Just think of Aisha. As Fraser argues, we need to “eschew the “stable pyramid” picture of subordination, which assigns every individual to a single “status group”” (2003, 59) and we need to give up the idea that women and men are homogeneous categories, neatly ‘fitting’ into already established positions. Inspired by the focus on performativity and positioning (Butler 2004; 1999; Ahmed 2000; Davies and Harré 1990; Hall 1990), I will thus suggest that another aspect of the discourses surrounding the difficult situations of men and the ‘too liberated women’ is as an act of gendering in itself. The repeated articulations of men’s change in social position and loss of social esteem and men’s problems are also ways of establishing ideals of how gender relations ought to be, of ‘proper’ and ‘failed’ men (cf. Kimmel 2005), establishing a continuum of masculinity ideals. I will suggest that one dominant gender ideal of ‘proper Somali masculinity’ in the West is about acting with dignity, providing for – and controlling – ones’ family, taking advantage of the opportunities in the West in terms of education and employment, involving oneself in ‘the commu-

94 The divorce rate in Somaliland has also grown since the civil war (Warsame 2002, 54).

240 nity’, and concentrating on Islam – in spite of loss and a difficult situation95. This ideal gains meaning through its contrast to the notions of ‘failed men’ and culturally impure, overly liberated women who cannot be controlled, but on the contrary, might end up taking control over their family and husbands. ‘Failed masculinity’, I will argue, is not the result of loss of social esteem in itself, but of failing to deal with it in an appropriate way – for instance by chewing too much khat, by losing control over and respect from ones family, by not being able to make a difference in anybody’s life. Talking about other men’s problems can be seen as part of the efforts to uphold cultural intelligibility in a context of loss and changed life circumstances, establishing a position of proper masculinity. The same mechanism of dealing inappropriately with life in the West might obviously be at work for women in relation to the discourse of ‘too liberated women’ and women’s (claimed) empowerment. Notions of ‘proper femininity’ were especially related to how women handled their families and, in the eyes of more conservative minds, their independence. However, in contrast to how many Somali men talked about other men’s failure, the Somali-Danish female informants seemed to be less hard on each other. Certainly, women were worried about family relations and the possible moral corruption of children and youth. Likewise, women were engaged in defining and struggling about the correct way of practising Islam, thereby also correcting other women’s behaviour. Women gossiped and criticized each other for methods of dealing with associations and politics, sometimes quite harshly. Still, I rarely came across women positioning themselves vis-à-vis other women in terms of ‘proper’ or ‘improper’ femininity. I am not suggesting that this was because Somali-Danish women showed more solidarity with each other than men did. Rather, I suggest that women often remained culturally intelligible as mothers and caretakers with responsibility for the family, in contrast to many men, who could not realize the ideal of the male breadwinner and/or deal with their losses in ‘appropriate’ ways. In this sense, the two perspectives might also reinforce each other: while some women were relatively empowered in terms of rights and economic support, still remaining culturally intelligible as mothers, some men lost privileges and could not realize the masculinity ideal as providers.

95 Other ideals include being a skilled speaker and virtuous poet, two virtues that many Somali men and women praise.

241 Symbolic capital and symbolic struggles It is important to emphasize the relativity and contextuality of the claims above. As I wrote, some women were empowered and some men lost privileges. Rec- ognition and negotiations of social esteem and cultural intelligibility also have a dimension of social class. The higher the social position before flight or exit from Somalia, the bigger the distance between the old and the new life – between the former and present social positions. Think of Mr. Ismail, the son of a sultan and once a banker; think of Aisha Ahmed, a former professional secretary, and think of Mahmoud Sheikh, who used to be a sports star and who had expected to become a politician. A former Somali diplomat explained to me that he tried not to think of his life trajectory from the top of society with an elite lifestyle, expensive cars, and the best private school for the kids, to a modest life in a suburb to avoid falling into ‘a black hole’. Still, not only the former elite experienced loss of social positions and social esteem. Abdirizak, an orphan and the only of his siblings to get an education and gain asylum in the West, did not come from a privileged background, but had still managed to change his life situation and social position in Somalia. In Denmark, however, he remembered his time as a social security claimant as degrading and humiliating – as misrecognition of his capabilities and status. Though Abdirizak Ali eventually became a teacher in Denmark, he was never capable of using his educational background. One of his responses was to engage himself in associations and voluntary work, thereby establishing an alternative space for using his qualifications – and being recognized as a man of competences by other Somalis. To return to Mr. Ismail: he was using his banking background as a cashier in a Somali-British association as well as serving as an informal counsellor to several British politicians who came to his home to discuss immigration affairs and Somali politics. He and Abdirizak Ali were thus examples of how recognition and social esteem for achievement was not only obtained through work, but also through associational engagement and in other domains of interaction in spite of the lack of recognition for their professional qualifications. Inspired by Honneth, we might say that they did not have the opportunity to fully express their professional self-realization, but that they used their qualifications in a sphere other than work, i.e. in associations and voluntary work, thereby gaining recognition from other Somalis. This mechanism was also at work among the female associational key persons. It shows us

242 that recognition of achievement cannot be reduced to mere employment, but must also offer the opportunity to use qualifications and competences in order to participate as peers, as Fraser expresses it. It also shows us that achievement and social esteem is embedded in gender ideals and cultural norms. In this way, associational engagement can be analyzed as ‘repair’ and ‘restoration’ of gender and social esteem, as we saw above – being recognized as a proper and respected Somali man through group-specific recognition in the Somali transnational social field. The level of recognition of achievement thus depended on ‘the recognizers’, so to speak, on the social space within which recognition took place and was evaluated, whether transnationally or among compatriots in Denmark. When I presented my preliminary analysis of associational engagement as related to social esteem and recognition to the informants, however, they all emphasized that quests for status and prestige were not part of their or others’ reasons for engagement, even if some of them accepted that they enjoyed the social esteem and respect they obtained through their activities. Rather, several informants explained to me, associational key persons already had prestigious positions and a ‘good name’, this being one of the reasons why they were encouraged to join executive committees or talk at seminars in the first place. This was emphasized by the fact that most of the associational key activists were educated and generally knew Danish society well, again mirroring more privileged socio-economic backgrounds in Somalia, as higher levels of education has been a privilege of the few. In this way, the question about the transferability of positions that I raised in the beginning of this chapter thus seems to be confirmed; associations did work as a vehicle for transferring and negotiating positions, but mainly within Somali social fields. However, to reduce associational engagement in this way would be to fail to appreciate the associational activities and achievements as well as the huge commitment of key persons, their many hours of work and, sometimes, personal costs. Rather, inspired by Bourdieu, I will suggest that negotiations of positions worked as a kind of practical sense of the game (cf. Bourdieu and Wacquant 1996, 101ff) or a practical sense of honour (Bourdieu 2005, 57). Associational key persons did not engage themselves as a result of premeditated calculations to obtain prestige and social standing; rather they mastered doing what people in their social position should do in culturally intelligible ways – not least in re-

243 lation to ideals of Somali masculinity. Certainly, to claim that associational involvement is a means of obtaining prestige and personal benefits would be seen as misrecognition of their commitments. Indeed, accusations of self-interested speculations in ‘the community’ sometimes circulated as rumours concerning how key organizers ‘used the name’ – or money – of the members, or more generally of Somalis in Denmark, for personal gain, benefits or prestige. I will suggest that Somali associational engagement connotes strong norms of collectivity, reciprocity, and selflessness, and that obtaining social esteem and prestige only was possible if associational engagement was seen as living up to these norms. In many ways, my analysis of group specific recognition of status has parallels to Bourdieu’s reflections of symbolic capital (1996a; 1994). In Bourdieu’s terms, symbolic capital can be understood as ‘honour’, reputation and prestige in specific fields (2005, 57), though not obtained through deliberate attempts, but as a kind of practical recognition. He writes:

It must be asserted at the same time that a capital (or power) becomes symbolic capital, that is capital, endowed with a specifically symbolic efficacy, only when it is misrecognized in its arbitrary truth as capital and recognized as legitimate and, on the other hand, that this act of (false) knowledge and recognition is an act of practical knowledge which in no way implies that the object known and recognized be pos- ited as object (1990, 112, emphasis in original).

Symbolic capital is thus not valid if it is recognized as capital, or in other words: to be engaged in an association or ‘the community’ does not convey social esteem if the engagement is seen as a means of obtaining social esteem and personal benefits. Furthermore, symbolic capital produces and reproduces visions of the legitimate social world (1994, 63). We can see the efforts to repair social esteem and gender (dis)order as attempts to re-establish social order within specific fields and social spaces. These are symbolic struggles in Bourdieu’s terms. However, in the contemporary, late modern, globalized and porous world (cf. Taylor 1994), people are living and dealing with more sets of social positions and more than one status order (Fraser and Honneth 2003; Fra- ser 2000) – not least transnationally engaged migrants (cf. Levitt and Schiller 2004). In my view, the transnational dimension of recognition theory is seri-

244 ously underdeveloped and needs much more attention96 because, as I show in this study, struggles for recognition and negotiations of positions in transnational social space are not limited to just one locality. While the recognition of social esteem and achievement might, on the one hand, be seen as an act of pre- serving social order within ‘the Somali community’, it may also, on the other, be seen as a desire – or the slightest hope – to change the social categories of perception in the surrounding society. It is thus related to social positions in both the Somali group as well as in Danish society. We might also say that not only was there a struggle for recognition of achievement and social esteem taking place, but also a struggle for cultural intelligibility as equal co-citizens and peers. The question then is, if they succeeded; if associational engagement furthered inclusion in Danish society. I turn to this question in the next section of the discussion, focusing on the relationship between processes of inclusion and involvement in transnational social space.

Processes of inclusion and transnational involvement

Between fragmentation and collaboration One of the striking characteristics of Somali-Danish associational involvement was the impressive commitment and the many initiatives. The high number of associations might seem to be a reflection of diversity as well as a proof of So- mali-Danish willingness to organize in ‘Danish ways’ through associations. Furthermore, the many associations also offered a plurality of ‘scenes’ for using competences (whether or not these could be employed in Danish society) and for struggles for recognition of achievement and social esteem – albeit not in deliberate ways, as I have just argued. There was, however, a downside to the many associations, in terms of fragmentation and, sometimes, lack of collaboration between associations. Ironically, even if the many associations reflect the success of the Somali-Danish strategy of associations as the means of organiza-

96 With some parallel to the critique of methodological nationalism, Fraser points to decentring of the Westphalian territorial-state frame in the contemporary stage of globalization (see also Dahl, Stoltz, and Willig 2004, 380). Nancy Fraser does criticize the tendency to force transnational social struggles into a national frame, what she terms ‘the problem of misframing’ (2003, 92; see also Dahl, Stoltz, and Willig 2004, 380). Still, in my view, Fraser does not really develop this perspective sufficiently.

245 tion in Denmark, the fragmentation of associational life might, in the end, di- minish the public recognition of Somali associations in Danish society at a general level (cf. Hopkins 2006). It would be mistaken, however, to explain this situation as an unequivocal reflection of Somali disintegration. Rather, I will argue, it reflects both the Danish and Somali contexts. On the one hand, the fragmented nature of associations relates to struggles over accessible but still relatively scarce economic resources available to migrant associations in Den- mark, as well as to the Danish policies of forced dispersion where Somalis have been placed all over Denmark, thereby possibly furthering the need for local associations. It thus partly reflects the Danish political opportunity structure. On the other hand, the associational fragmentation has clear parallels to the political situation in Somalia and Somaliland where lineage-based coalitions dominate the social and political life, resulting in political fragmentation, sometimes into very small units. As we have seen, this situation ‘spilled over’ into some Somali-Danish associations, which organize themselves according to clan, regional affiliation or political orientation. Still, the Somali-Danish associational landscape was not only characterized by fragmentation, but also by diversity and collaboration. The Somali clan system is inherently flexible and characterized by ongoing tensions between fragmentation and collaboration (cf. Luling 2006; Lewis 1994). However, the diverse associational landscape was not only divided in terms of clan or regional affiliation but also according to gender and degrees of religious devotion. It thus reflected transnational and ‘ethno-national’ conflicts, divisions and loyalties as well as cultural and religious norms. Furthermore, while (some) Somali-Danish associations were organized around exclusive principles of membership such as regional or clan affiliation and/or gender division, others were pan-Somali and based on inclusive membership. Rather, we may say that there was a symbolic struggle over principles of organization and the constitution of and negotiations for political authority: of whose words and visions should be considered legitimate. The struggles for recognition were thus both struggles for rights and public acknowledgement, fuelled by harm and misrecognition; they were struggles for obtaining and exploiting opportunities in Denmark, Somalia or elsewhere, and they were struggles for cultural intelligibility – of being recognized as a person of competences, esteem, enacting ‘proper’ gendered behaviour. They thereby both re-

246 lated to inclusion in Danish society as well as to transnational social spaces, and they were both related to issues of self-realization and to participation.

At the centre Struggles for recognition and negotiations of positions thus cannot be fixed in just one place, but have both local and transnational dimensions, relating to a broader ensemble of dynamics and conflicts. As the small vignette from South- all bears witness to, recognition and positioning took place on the streets as well as within associations and seminars. Likewise, many of the informants were also locally and transnationally involved in many other ways, taking care of their families, supporting relatives, doing business, and perhaps being politically engaged. They were thus negotiating loyalties, obligations and commitments to different people and places through associations as well as by other means. They were, in other words, positioned in an overall transnational social field, “through which ideas, practices, and resources are unequally exchanged, organized and transformed” (Levitt and Schiller 2004, 1009). Take Hawo Yusuf, Abdirizak Ali, and Elmi Samir, for instance. All three were holding full-time jobs, remitting money to relatives and providing for their own families in Denmark. Furthermore, all three of them were very engaged in associational activities. At the time of fieldwork, Elmi was concentrating on Somscan & UK, but he used to be very involved in Somali associational activities in Denmark. Abdirizak was involved in a veritable multitude of different efforts and activities in both Denmark and Somaliland, including associations and several smaller business initiatives. Hawo, finally, was engaged in both the transnational and local activities of the Woman’s Association. In addition, as they were all considering relocating to Somaliland or Somalia, all three were also contemplating how to secure not only their own, but also their dependants’ livelihoods. In short, they were immensely busy, managing and dealing with obligations and ambitions in a transnational social field, offering both opportunities and demanding responsibilities. And take Omar Said and Mahmoud Sheikh. They were also supporting relatives in the Horn of Africa and were also involved in transnational reconstruction projects. However, their projects were not directed towards the region where their dependants lived, as their families had fled Somalia and had not returned. Still, according to Mahmoud, it was about keeping the future open, securing one’s reputation by using one’s good

247 name. “You exploit your name, your father’s and your father’s father’s name”, Mahmoud explained to me, continuing how making ‘a sacrifice’ – doing something for other people – is a way of ensuring respect for one’s name in the future, which can be important if, at some point, one wishes to return. He did, in other words, establish social and symbolic capital as a kind of possible future ‘credentials’ (cf. Bourdieu 1986, 249). Mahmoud thereby not only contributed to local development and reconstruction, acting as a responsible man; he also invested in his future and reputation. Mahmoud, however, was unusual in the sense that he articulated his involvement in these terms97. Omar, for instance, emphasized his engagement as obligations arising out of his relatively privileged situation in Denmark. When I asked him about his local and transnational involvement, he stated:

I don’t think there is anything preventing us in integrating in our society, on the one hand, and supporting the reconstruction in our homeland, on the other. Because all the educated people left Somalia, so it is necessary that we do something. If we don’t, there are only warlords left … I am a member of the local integration council and do different things, I mean local things, right, and on the other hand, I think, we should spend some time on supporting our homeland …I think that if you are okay yourself, you should help other people, right.

Omar thus articulated his simultaneous responsibilities and engagement in Dan- ish society and Somalia as arising out of a sense of obligation. As I argued above, his engagements can be seen as arising out of a practical sense of mas- tering ‘community’ and associational involvement, intimately connected to senses of obligation and moral commitment. Indeed, due to his impressive engagement in both Danish society and reconstruction projects, Omar might obtain recognition from both Danish and Somali actors as a good Danish and transnational citizen. As we have seen, Omar was not the only informant involved in more places at once. Indeed, most of the transnationally involved key persons, organizing and initiating reconstruction projects, were holding full- time jobs, had spent many years in Denmark, had Danish citizenship, and were usually also involved in associations and initiatives to support processes of inclusion for other Somali-Danes. They were at the centre of a transnational so-

97 We might also say that Mahmoud was unusually open in his attempts to explain to me what he did and what was going on. Still, Mahmoud was not in the same (associational) executive position as Omar or Elmi, for instance.

248 cial field. Interestingly Mahmoud, who was unemployed and in a less established position in Danish society, was also more explicit in his articulation of obtaining social esteem from associational involvement than the others. Another (interrelated) position in the transnational social field was as a connecting link. Halima Abdi, whose story I told in chapter four, was born in Den- mark, and through occasionally going to Somalia on holiday, was also transnationally engaged; yet she articulated her situation differently. Not wanting to live anywhere else than Copenhagen and not having siblings, children, or parents in Somalia, Halima did not position herself as a possible future actor or politician in Somalia; she was not engaged in any transnational reconstruction projects, but focused on her life here. At first glance, she might appear not to have any transnational responsibilities. However, Halima owned land in her parents’ town of origin and she was a part of a large extended family network, spanning localities in Africa, Europe, North America, Asia and Australia. Re- mitting money to a cousin in Russia as well as the extended family in times of need, Halima had obligations across borders, embedded in family expectations and a sense of obligation. Halima, then, was also part of a transnational network of obligations and social capital, even though she emphasized that she would not, herself, take advantage of it. Likewise, Halima Abdi was very engaged in different activities for Somali children and youth in Denmark. At the time of fieldwork, she was planning to create a pan-Somali association, independent of regional and clan affiliation, focused on information about Danish society. Halima, emphasizing her particular position as somebody who was both Somali and Danish, explained that she had special obligations and qualifications to me- diate between Danish society and Somalis in Denmark, to work as a connecting link. Likewise, other informants described their positions as links, intermediaries and brokers because of their knowledge of Danish society. Some described themselves as practical leaders or representatives, as we saw it in chapter six; others viewed their association as a kind of embassy, helping out and advising Somalis in Denmark, due to their privileged understanding of Danish society. Indeed, some of these associational key persons were very transnationally engaged: the position as connecting link does not stand in any opposition to transnational involvement. Rather, it presumed knowledge of and involvement in

249 Danish society as well as knowledge of and involvement in the situation of Somalis in Denmark.

In the margins Not all of the informants, however, were engaged in associations or wished to function as connecting links. Khalid Osman, a Somali-Danish man in his thirties was outspoken in his disengagement from Somali-Danish associations. Remitting money to his mother in Mogadishu, Khalid was transnationally involved, but did not wish to be a part of what he saw as too much political talk of clubs and associations. He explained to me:

I sometimes go to a club, which a lot of elderly men attend. They listen to the BBC, read the Internet and photocopy news from the Internet. They talk about politics all the time. They talk about politics there, but they are living here and want to stay here. I tell them to think about their social counsellor instead. I tell them to concentrate on their lives here. You should not live here and bother about politics there. They just sit and talk.

Khalid Osman was thus explicit in his distance from Somali clubs, associations and politics, indicating that in his view orientation towards Somalia impedes inclusion in Danish society. His disengagement might be reinforced by the fact that, at the time of interview, his father was a member of the transitional parliament. According to Khalid, his father was just talking and talking, not doing anything to change the situation, exactly like the men in the club in Denmark. And Khalid was fed up with men talking about Somali politics: he wanted peace, and he wanted people to focus on their lives here. Still, should there be permanent peace in Somalia, Khalid would be the first to go back, he claimed, not least to exploit the economic opportunities in the country. Thus, although Khalid Osman argued against an orientation towards Somalia, his objections mainly seemed to be against what he saw as the absence of focus on life in Denmark. Somalia was not completely out of focus, so to speak, but it was not his main orientation. Rather, Khalid articulated a position of selective distance in the transnational social field. Khalid was not the only informant who was not engaged in transnational reconstruction projects. A few informants expressed points of view similar to his, being explicit in their criticism, while others simply said that they were not interested and preferred just to remit money to their relatives. And some infor-

250 mants could not afford to send any kind of money at all, or only very rarely. This was the situation for many of the women that I talked to in the Danish language classes in the Women’s Association. They did not support any reconstruction projects, mostly explicitly emphasized that they were not interested in politics, and hardly ever remitted money to their relatives. In contrast to some informants, whose relatives were doing well and did not need any contributions, the reason was that they did not have the economic surplus themselves. “I would like to”, one woman said, “but I don’t have the money”. She thereby pointed to the fact that transnational engagement also requires resources: To struggle for recognition through involvement in transnational reconstruction projects, to establish social positions and obtain social esteem, to accumulate symbolic and social capital do not come for free, so to speak. These women might thus also be positioned in the margins of Somali transnational social spaces. Not necessarily because they did not want to participate, but because they could not afford to. The women joining Danish classes mostly did not speak Danish very well, were unemployed, some did not have much education from Somalia and several were single mothers. They were, in many ways, among the more vulnerable, positioned in the margins of Danish society as well as, it seemed, of the transnational social spaces. Not necessarily because they had chosen to disengage themselves, as Khalid Osman had done to a certain degree. Rather, they seemed to lack resources and were immensely busy with making ends meet in their daily lives in Denmark.

Migrant associations and processes of inclusion Above, I have outlined four positions in the broader transnational social field, employing metaphors of distance and closeness: at the centre, as connecting links, at a selective distance and in the margins. They could also be conceptualized in terms of degrees of simultaneity of connection in the transnational social field and involvement in Danish society. As such they were temporary locations, “rotating back and forth and changing direction over time” (Levitt and Schiller 2004, 1011), rather than definite positions and unequivocal directions of orientations and loyalties. My analysis shows that the positions in broader transnational social field (of which associational engagement is but a part) are structured by the interrelationship between obligations and opportunities and by the degree of resources, as well as by the subscription to or distance from social

251 and cultural principles of organization – such as gender relations, clan and regional affiliation. Negotiations of positions and struggles for recognition through associational engagement were thus often reinforced by other transnational involvements. My analysis hence supports the theoretical claim that there is no necessary contradiction between transnational involvement and processes of inclusion and integration in the receiving country (cf. Snel, Engbersen, and Leerkes 2006; Mezzetti and Piperno 2005; Levitt and Schiller 2004; Portes 2003; Itzigsohn and Saucedo 2002). It shows that it is not the most disadvantaged who are transnationally engaged, especially not at the organizational level, as transnational involvement requires resources. On the contrary, the most transnationally involved were also well educated, held jobs, were Danish citizens, and most had spent many years in Denmark98. Furthermore, most of them were also engaged in activities aimed at supporting the inclusion of fellow Somalis in Dan- ish society. Still, there were a few associational key people who were not transnationally engaged on the associational level, but no associational key activists were exclusively transnationally involved. Thus, while involvement in Danish society did not necessarily imply transnational involvement, organizing and initiating transnational involvement seemed to go hand in hand with involvement in Danish society. My material indicates, however, that the tendency of simultaneous involvement is less clear if we turn from the institutionalized level of organizing and initiating reconstruction projects to the donations of ‘ordinary members’ and individual remittances to relatives. Or, in other words, while the Somali-Danish transnational associational ‘entrepreneurs’ and key persons were very engaged in both Danish society and Somalia, those not involved at this organizational level might still remit money (and thereby be transnationally involved), but might not necessarily be as well established in Denmark. How, then, does associational engagement relate to the question of processes of inclusion and transnational involvement in relation to migrant associations? My analysis supports Layton-Henry’s (1990) claim that involvement in ethno- national migrant associations (including transnational involvement) enhances

98 Studies of associational involvement in Danish voluntary sports associations suggests a similar tendency: That the associational key persons are to be found among well-educated and well-established men, often with employment related management experience (Ibsen and Habermann 2005, 25; Ibsen 1993).

252 processes of inclusion in the country of residence, especially at the organizational level. To organize and initiate associational activities implies democratic learning processes, as I argued in chapter six, not least in relation to the political culture of voluntary work. Associational engagement might thus further political inclusion via “the educational aspect of participation in an association”, as Kaspersen and Ottesen argued (2001, 128). They furthermore claim that this political aspect is the most important virtue of associations, rather than processes of assembling people across culture and differences (Gundelach and Torpe 1999). The case of Somali-Danish associations suggests that aspects of (political) participation were indeed the most important ones. According to key persons such as Mohamed Gelle and Musse Sheikh, associations were initially seen as a strategy of participation in Danish society, as a particularly Danish way of getting a voice in the Danish democratic processes. This relates to aspects of participating as a peer, as Nancy Fraser formulated it, in relation to politics. In- deed, several of the male and female associational key organizers were members of the local integration councils, and some of them were active in a Danish political party. One had (unsuccessfully) run for Parliament, and several others had been encouraged to run for political office and were considering such moves. One man achieved election as a member of the local municipal council in 2005. In all of these cases, the informants had extensive experience with associational work, as well as in collaboration with Danish authorities such as local municipalities, integration councils or government agencies. Their engagement can be seen as classic examples of political participation and inclusion in the democratic system in Denmark (cf. Mikkelsen 2003a). In this sense, associations offered opportunities for becoming politically active in the democratic system in Denmark, demonstrating a potential for political integration in Habermas’ sense, as “loyalty towards the political culture […] rooted in an interpretation of constitutional principles” (1994, 134). Still, I will claim that organizing and initiating associational activities also had the potential for bridging social networks or cross-cultural encounters. As- sociational key ativists simply had to communicate a lot with other major players and stakeholders – whether Somali, Danish etc. – in order to organize activities and, not least, to apply for funding. Their involvement both presumed and contributed to the density of social capital in Bourdieu’s sense of “networks of

253 connection” (1986, 249). ‘Ordinary members’ on the other hand, who joined associations for social purposes and/or to get practical help, might not necessary be involved in such activities. This is not to say that associations failed to further processes of inclusion for ‘ordinary members’. Rather, such processes might take place on another level. Joining an association, establishing new networks (even if they were not among Danish NGO representatives), receiving practical help, following courses and, perhaps, gaining an insight into organizational work might indeed support inclusion in Danish society. Still, more intensive processes of inclusion seemed to be related to actual involvement in organizing associations and activities – though joining in with activities might be a step towards getting involved oneself (and, of course, take place simultaneously with organizing activities). Thus, associational engagement furthered processes of inclusion, but the kind of inclusion depends on the type of engagement and the persons involved. To return to Layton-Henry’s theory, I also have a correction to make. His ideal typical phase model of migrant associations conceptualized as linear development from orientation towards the country of origin – through a middle phase – and then towards the country of residence is not supported by my study. Rather, as we have seen, associational key persons did not turn away from transnational involvement; on the contrary, some of them became engaged after having established themselves in Denmark. We need a more open-ended scenario (cf. Cordero-Guzman 2005) and we need new analytical concepts to analyze these processes. The notion of simultaneity of connection (Levitt and Schiller 2004) is useful here, as we have already seen, as well as Mezzetti and Piperno’s suggestion of transnational integration. This concept relates to the “positive practices of transnational individuals, communities and political subjects who – thanks to a good level of integration at the local level” contribute to the “promotion of human, social and economic development” (2005, 8). Indeed, the associational key persons in this study are specifically examples of transnational integration. As I have argued above, their involvement in associations and transnational integration seemed to be guided by the combination of obligations, expectations, and opportunities as well as available resources and positions. Furthermore, these involvements often had transnational perspectives, as we have seen, directed towards Somalia or Somaliland, sometimes articulated

254 as part of a broader diasporic engagement. In the last part of the discussion, I explore this dimension.

In the name of diaspora As I have shown in chapters two and seven, the concept of diaspora is in the air; it has travelled from theology to social sciences and migration studies as well as further on to policy makers, organizations and, not least, migrants’ own concep- tualizations of their situations. Indeed, the conceptual journey of diaspora can be described with Giddens’ concept of a double hermeneutic (1987), referring to the dialectical relationship between researchers’ and other social actors’ uses and appropriations of concepts. I find that the mobilization and identification aspects of diaspora are of particular interest in my exploration of fights for recognition as relating to both social struggles and cultural intelligibility. I thereby connect the recognition and diaspora perspectives. In the following analysis, I continue this theoretical exploration, starting by addressing the question of belonging and homeland, considered so central in many accounts of diaspora, followed by an analysis of how ‘diasporas’ are mobilized. Rather than taking ‘diasporas’ for granted as social and political entities, I analyze the mobilization processes (cf. Axel 2004), focusing on what – and who – is claimed in the name of diaspora. I thereby end on a more general, theoretical level, linking recognition and diaspora theories.

Between homeland and society What and where is home? This is a central and contested concern in diaspora studies. At first glance, the answer to this question seemed quite simple, as the big majority of the Somali-Danish informants stated that Somalia or Somaliland was their homeland. Still, the question of origin was a complex and sometimes contested one, as illustrated with the notion of u dhashay – ku dhashay – the tension between clan and regional affiliation (Barnes 2006). In some cases, Somaliland had become the homeland (given the historical development where Somaliland was (re-)declared an independent state in 1991). In other cases, Somalia was the claimed homeland and national unity defended. Due to the civil war, continued strife and conflict, and the de facto disintegration of the Somali Republic, Somalia as homeland was complicated, offering no polity to identify with, no nation-state. Some of the informants articulated their grief,

255 frustration and sometimes embarrassment towards the political development in Somalia, especially when things went just too wrong. For instance in March 2005, when parliamentarians in the transitory parliament in Nairobi got into a fight with chairs and sticks. “I was so embarrassed”, one woman said, continuing that the whole world was watching the fracas on CNN. Feeling embarrassed and humiliated – and not just disengaged – the woman identified as a part of a Somali nation which, once again, exposed its conflicts to the whole world. She and many others expressed an ambivalent relationship to Somalia, grieving and/or being disgusted by the civil war, while still remaining attached to the country. They were hoping for better times and, sometimes sentimentally, remembering the time before the war and before Barre’s dictatorship as delightful and easy. Some recalled hanging out with friends and family, some remembered playing football on the beach, while others evoked the silence and gran- deur of the desert, or the beauty of the sun setting in the Indian Ocean. How- ever, given the aftermath of war and disintegration, these memories were often followed by a silent sigh, situated in the nostalgia of past events and past life. Somalia as homeland was thus articulated as a tension between (sometimes nostalgic) memories and the realities of today. Likewise, it was constituted as a tension between an abstract national level and regional and local affiliation, towards which actual loyalty, involvement and support to reconstruction projects tended to be directed. Very often in the so-called ‘areas of origin’, the area dominated by one’s clan family had become the original home region, as we have seen, whether people had actually grown up there or not. In this sense, there were clear parallels between the development in Somaliland and Somalia. However, as we also saw it in chapter seven, the tension between regional and clan affiliation was much debated and some informants were explicitly in favour of a regional identity, rather than clan affiliation. Likewise, other informants expressed a rather pragmatic stance, explaining to me that while, say, Somaliland was their homeland now, should there be peace in Mogadishu, then this would be the one and only place they would want to live – in “the city of my dreams”, as one Somaliland man said to me. Somali-Danish ideas of a homeland, home region, hometown or homestead were, in other words, flexible and contextual, directed towards simultaneous, multiple spatial scales. Similarly, we have also seen that the idea of a (far-away) homeland in many instances was coupled with ideas of Denmark as the society in which daily life

256 takes place. “I don’t think there is anything preventing us in integrating in our society, on the one hand, and supporting the reconstruction in our homeland, on the other”, as Omar Said said, while Saphia Mansoor described herself as “a Somali woman living in Danish society”. Finally, some of the informants, especially the ones who had visited Somaliland or Somalia, emphasized that they were both Somali and Danish. When I asked Elmi Samir, for instance, he laughed and explained to me that he realized his ‘Danishness’ when he visited Somaliland a few years ago, putting his identification as Somali into perspective. At the same time, Elmi emphasized that integration does not mean an end to Somali identity, adding that those who claimed that they are only Somali would change their minds should they spend some time in Somaliland or Soma- lia. A woman I interviewed both in Somaliland and Copenhagen simply referred to herself as ‘Danish’, when I visited her in Somaliland. Living abroad seemed, in other words, to intensify the feeling of being ‘Somali’ (Kleist 2002; cf. Anderson 1994), but visiting the ‘homeland’ equally emphasized the sense of belonging to the country of exile. While such senses of belonging might be about a ‘double strangeness’ or ‘double distance’, it might also be about having two homes. Or rather, perhaps, as we just saw: to both belong to and engage in one’s society and one’s homeland. Somali or Danish identification was, in other words, not set in stone. Rather, these reflections show that fields of belonging are ambivalent and contextual phenomena, dependent on time, place and personal biographies, embedded in dreams and memories; sometimes guided by a certain pragmatic sense of the possible, and sometimes shaded in nostalgia. The informants thus articulated a homeland which could be analyzed in diasporic terms: ambivalent, longed for, sometimes distant. The question is, however, did the informants articulate themselves as part of a diaspora?

Diasporic identification According to Axel, diaspora has become a potential “globally mobile category of identification” (2004, 27). The case of Somali migration supports this asser- tion. As we have seen, notions of diaspora have become part of the vocabulary of Somali migration, employed by Somali migrants and stayers, policy-makers and researchers alike. Though the term was sometimes used as a neutral one with which to refer to the spatially dispersed Somali population, it quite often had implicit or rather explicit moral dimensions, invoking or appealing to po-

257 litical or economic support. In chapter seven, I showed the expectations of ‘the diaspora’ to transfer remittances and act as agents of modernity and development – as a moral community, as Werbner (2002) expressed it. As I mentioned in the introduction, I was struck by how much the notion of diaspora was employed by Somali actors in comparison to my first fieldwork: on Somali homepages, in reports, by policy-makers and researchers, Somali- Danish migrants, and not least among the informants in Somaliland. At first, this made me think that there had been a ‘diasporic turn’ among the Somali- Danes. However, a closer look at the empirical material showed that while some – mainly the most active and transnationally oriented Somali-Danish associational key persons – did use the term, they almost exclusively did so in relation to collective reconstruction and development projects. Furthermore, a lot of the Somali-Danish informants did not use it, or only starting talking about ‘the diaspora’ when I asked them about it. More than once, I was asked to explain exactly what diaspora means by informants who had heard or read about the term, in one case in a seminar report that I had written. This was a rather sobering question in relation to my initial diaspora excitement. Compared to the quite extended references to ‘the diaspora’ articulated by returnees, politicians and project workers in Somaliland, it seemed like transnationally engaged So- mali-Danes were ‘learning’ to use the term, going through a process of diasporic identification, inscribing themselves as part of a larger moral community of agency and commitment. In this way, diasporic identification can be said to supplement and accomplish the identity and recognition work, offering a framework of interpretation, as Thomas Blom Hansen suggested (2003), uniting the idea of a dispersed Somali population with moral and economic obligations towards Somalia or Somaliland.

The re-birth of a nation It is thus a central point that identification in the name of diaspora was mobilized, rather than emerging as a reflection of transnational dispersion (cf. Söke- feld 2006). Furthermore, if we follow Axel, the analysis of the context of diaspora should be moved from spatial dispersal and presumed origin to the production of temporalities (2004, 31). This move suggests that we turn our attention to histories of pasts, presents, and futures, to narrations of ‘before’ and ‘thereaf-

258 ter’99. The events surrounding the visit of Somaliland President Riyaale Kahin and his delegation to London in March 2004100 serves as an illustration of this dimension, showing us how evocations of the past and present can be used to envision or legitimize certain political visions and goals.

The big demonstration London, March 17, 2003

I am standing in Downing Street in front of the Foreign Office, surrounded by Somalis from all over Britain. They are demonstrating for the political recognition of Somaliland, waving banners and flags, shouting and singing, carrying pro-Somaliland posters taped on their bodies. ‘Somaliland Reborn’, the posters state. Some women and children are dressed in the Somaliland flag or have painted it on their faces, embodying the nation – or the dreams of it. I notice a man in a wheel chair with no legs and a younger man walking with crutches, carrying a big Somaliland flag. They are SNM heroes, somebody tells me, in- jured when fighting Barre’s soldiers during the civil war.

All of a sudden a big roar of excitement goes through the crowd. Bristol has arrived! Passengers from six coaches from Bristol enter the scene, walking in from Whitehall Street. They are greeted with applause and great enthusiasm. It’s big now. I can see banners from Manchester, London, Bristol, Cardiff, Bir- mingham, as well as a few Danish, Dutch and Norwegian flags. And Union Jacks, lots of Union Jacks. Why?, I ask one of the guys from Bristol, who has wrapped himself in the Somaliland flag and Union Jack. Because Great Britain has a past in Somaliland, he explains to me. To remind them of their responsibility. And because he has lived in the UK for many years and also belongs here, he adds.

Later the demonstration is relocated to Parliament Square, at the footsteps of Big Ben. At five o’clock, a smaller part of the demonstration moves into Port-

99 This temporality is also captured in the title of Nuruddin Farah’s book, Yesterday, Tomorrow. Voices from the Somali Diaspora (Farah, 2000). 100 For a more detailed description and analysis of this event, see (Kleist and Han- sen 2007; Kleist and Hansen 2005).

259 cullis House, where the all party Parliamentary Group on Overseas Develop- ment features ‘President Riyaale on The Challenge of Somaliland’, as the official invitation states it. During their speeches, the President, the Foreign Min- ister, and other pro-Somaliland speakers remind the audience of the British colonial past and the oppression during the regime of Barre. “Eritrea, Serbia, Bosnia – why not Somaliland?” Riyaale rhetorically asks, while Foreign Minis- ter Edna Adan forcefully proclaims that, ”the diaspora has brought Somaliland to where it is today” and calls for further political diasporic mobilization for the recognition of Somaliland.

After the parliamentary session, the day ends with a fancy reception at the posh Landmark Hotel, certainly worthy of a presidential visit. The luxurious ban- queting hall with marble floors and golden chandeliers is full of beautifully dressed Somalilanders in tuxedos, evening gowns, high heels and shimmering jewellery. There are speeches and an abundant buffet with French, Italian and Somali food. When I leave a few hours later, spirits are high and the party is going full swing. The President, his delegation, representatives of organizations, politicians, models, and reporters are all there, mingling with each other, taking photos, having a good time. The Somali-British elite, in short. They party, celebrating the presidential visit as well as, perhaps, themselves as a community, a nation, a diaspora – in spite of everyday divisions and troubles.

Throughout the events of 17th of March 2004, the demonstrators and the delegation made efforts to legitimize Somaliland as a recognized nation-state. Refer- ring to the British colonial past, they emphasized that recognition of Somaliland will not violate colonial borders; they also reminded the British of the oppression during Barre’s regime. Likewise, they compared the present situation of Somalia as a failed state with the development of Somaliland as a de-facto independent nation-state with a government, police, its own flag etc. The production of temporalities, i.e. the references to the past, the comparison of present- day Somalia and Somaliland, and the visions of possible futures, served to articulate the Somaliland nation and diaspora as a cultural, political and moral entity with a place in the ‘family of nations’, as Somaliland reborn, as the demonstration posters stated. Likewise, Foreign Minister Edna Adan, herself a trained

260 midwife, invoked the metaphor of political kinship and birth during her speech at the Landmark Hotel. She said:

I never expected, when I came to UK to study nursing and later mid- wifery, that I would be a member of a team trying to deliver a nation. Somaliland grew into a nation in spite of the nation which tried to kill them, but not fighting back, respecting human rights. We separated because of basic differences with people from Somalia. We are Somali- landers and they are Somalis. It is not a question of North and South. We are from Somaliland; it has a name and a place on the map. Somaliland is here to stay. You should inform the world of who you are; you should coordinate it.

Edna Aden got standing applause in an enthusiastic and emotional atmosphere. The metaphor of giving birth to a nation is indeed classic nationalist imaginary. It may refer both to the Somaliland constitution, in which it is stated that “the Somaliland nation is a family that has everything in common, such as religion, culture, customs and language” (Republic of Somaliland 2001, preamble), and it may be said to supplement and challenge the rather male-dominated Somali- land political system with a feminization and familiarization of the nation (cf. Eriksen 2002). Going on to address and encourage the audience to “inform the world of who you are”, Edna Aden delivered an outspoken example of (long- distance) nationalism, linking local and global political responsibility, as well as the past, present and future. The Somaliland nation and ‘diaspora’ were articulated as a people opposing oppression and civil war, constituting a moral and political category, a population that identifies with and works for a political project: Somaliland101. ‘The diaspora’ and nation were thus imbued with a collective dimension; a ‘we’ who can be mobilized, who are politically responsible, who are expected to show solidarity and struggle for a common cause: Somaliland. We might say that the nation and ‘the diaspora’ formed – or were mobilized to form – a Somaliland transborder citizenry (Schiller and Fouron 2001, 20), including migrants, their descendants and those who stayed in one single category, united by (long-distance) nationalism. The nation and ‘the di-

101 Also Mohamed Silanyo, leader of the Somaliland political party Kulmieye, whom I interviewed when he was in Copenhagen in October 2004, directly addressed ‘the diaspora’ and emphasized their role in the reconstruction of Somali- land.

261 aspora’ become one, connected by ties of moral commitment and responsibility across borders. Schiller and Fouron distinguish, however, between long-distance nationalism and diaspora. They define the former as “identification with a particular, existing state or the desire to construct a new state”, while ‘diasporas’ are organized around “other forms of transborder ideas about membership, such as based on religion or a notion of a shared history and dispersal” (ibid., 23), but making no claims to nation-state building. I take issue with this analytical distinction. Or rather, in my attempt to analyze what is claimed in the name of diaspora, it is not helpful to leave aside long-distance nationalism as a separate domain of belonging and mobilization. Instead, I content that the concept of diaspora gains part of its strength and intelligibility because it in so many ways has nation-like connotations. The flexible, contextual and ambivalent nature of Somali notions of homeland means that while identification and solidarity are sometimes rather exclusive, delimited and local, diaspora can be mobilized at a broad and inclusive level at other times, organized around ideas of a national homeland, offering an effective framework of interpretation and identification. It can be recognized, in other words.

Recognizing ‘the diaspora’ In the last part of the discussion, I am turning to an analysis of how the concept of diaspora is gaining currency in social struggles as well as its legitimacy and intelligibility in such struggles. I thus engage in a broader and more abstract argument. I have already discussed how diaspora can be understood as a framework of expressions and practices, drawing on the connotations of diaspora as loss, suffering, harm, and dispersion, following the classic perceptions of the concept (cf. Safran 1991). In this regard, diaspora certainly has effective and moral dimensions. This perception of the ‘diasporic experience’ resonates very well with the discursive frameworks of recognition, emphasizing the moral im- peratives of avoiding harm and furthering social justice and equality. In both perspectives, there are elements of longing, unfulfilled dreams and unfulfilled identities. According to Honneth, “the purpose of social equality is to enable the personal identity-formation of all members of society” (2003, 177). The potential connotations of agony and injustice associated with diaspora may reinforce the moral weight of claims made in its name, as the backdrop of the diasporic

262 mobilization and identification. Stuart Hall, writing about cultural identities and diaspora, formulated it as “unstable points of identification or suture, which are made, within the discourses of history and culture” (1990, 226). It is my propo- sition that the theoretical and political ‘recognition turn’ fuels the proliferation of studies and claims of diaspora: that such claims and studies become possible, legitimate and intelligible, perhaps even attractive. Claims of ‘diasporas’ with the connotations of suffering, loss, displacement, and return are imbued with mobilizing power, as well as with potential moral legitimacy and sympathy. The moral dimension and potential for mobilization, however, might be used both by ‘diasporas’ in social or political struggles, but also by governments or other actors, striving to mobilize ‘their diaspora’ to remit money or support the homeland in other ways. The pro-Somaliland demonstration and following events were a case in point in relation to both aspects. Another aspect of the proliferation of the vocabulary of diaspora may be that it is often mobilized and invoked as having a nation-like nature, as we just saw, inscribing itself in the nation state order rather than challenging it, as Clifford (1997) suggests. Indeed, ‘diasporas’ seem to be increasingly understood as political actors, making legitimate and collective claims on behalf of established communities. Or, in other words, the nation-like character of the usages and claims of the Somali or any other ‘diaspora’ might make diasporic spokespersons more intelligible as legitimate political actors with whom negotiations can be made. The diaspora perspective thereby includes both a dimension of victimization and suffering as well as of political agency. We might say that the diasporic identification was related to recognition both in the sense of fuel for social struggles as well as in the sense of cultural intelligibility as responsible transnational citizen and political actor. In the Somali-Danish case, it was striking that it was almost exclusively men articulating diasporic identification, indicating a link between citizen-like aspects of diaspora, nation-building and proper masculinity. That diaspora is not only a masculine affair, however, was shown by Hawo, Noora, and Edna Aden. In this sense, being a part of ‘a diaspora’ is to be part of a larger moral community of commitment and agency, which is recognized and reached for by a number of political actors. It is a position from which struggles for recognition can be formulated and carried out, and from which cultural intelligibility can be obtained.

263 Chapter resume: Spaces of recognition In this chapter, I have argued that Somali associational engagement forms spaces of recognition relating to social struggles as well as to cultural intelligibility. I claimed that associational engagement is used to negotiate and re- establish social positions and social esteem, recognition for achievement, and ‘proper’ masculinity and femininity. Involving oneself in associational work can be one way of using one’s qualifications and being recognized – though most often by other Somalis. Indeed, recognition for associational engagement is primarily taking place at a group-specific level. In this way, associations often serve as vehicles for (re)establishing positions in Denmark, guided by a practical sense of the game, rather than by calculated and premeditated strategies. I further claimed that this situation reflects the Danish opportunity structure, the situation of general social subordination and misrecognition in Danish society, and the political development in the Somali region. We might say that the Somali-Danish associational field is part of an unequal overall transnational social field with different positions, depending on the density of practices and relations. I outlined four overall positions: at the centre, connecting links, at a distance, and in the margins. The two poles of centre and margin are constituted by the degree of density of (many kinds of) transnational practices as well as of the connectedness to Danish society and transnationally. I showed that transnational associational involvement on the organizational level most often goes hand in hand with involvement in Danish society. Indeed, the most active So- mali-Danish transnational organizational entrepreneurs are among the most established in Danish society, indicating that there is no necessary contradiction between transnational engagement and inclusion in the country of residence. Finally, I showed how the claims and mobilization made in the name of ‘the diaspora’ refer to a moral, transnational, and nation-like community of resources and agency, which is claimed, mobilized and recognized at different levels and by different actors. Diasporic identification and mobilization is thus also a way of struggling for recognition and for negotiating positions.

264 Chapter 9 Conclusion

“Many Somalis in the diaspora worry about the same questions of how to manage their lives in a new society as well as the reconstruction of Somalia”, I was told by the organizer of the Somali Network seminar in February 2003, when I was just starting my fieldwork. This overall problem is the crux of this dissertation. I have analyzed and explored it through the specific optic of involvement in Somali ethno-national migrant associations, focusing on involvement in Dan- ish society as well as on reconstruction and development in Somalia or Somali- land. Associational engagement is thus just one particular aspect of the multi- layered relationship between processes of inclusion in Danish society and transnational involvement.

Struggles for recognition and negotiations of positions More specifically, I asked the overall research question: How do Somalis in Denmark negotiate positions and struggle for recognition in Danish society and transnationally through associational engagement? I have endeavoured to analyze this question in a theoretically informed and empirically grounded way. Theoretically, I have combined theories of transnationalism and migrant organizations with diaspora and recognition theories; the latter including both a focus on social struggles and on cultural intelligibility and positioning. This ‘double gaze’ brings to the fore discussions of how social positions are negotiated, established and enacted, as well as how such identity work relates to social struggles and processes of inclusion. I developed the analytical concept of spaces of recognition to capture these dynamics. Empirically, I embarked on a research journey, which was centred in Copenhagen, but which also led me to Somaliland and London. Taking departure in the methodological concepts of location-work and situated knowledges, I set out to analyze how recognition and positioning have local as well as transnational dimensions and how they are enacted and articulated in different contexts. Recognition and positioning in a transnational perspective is a research field still in embryo. The present dissertation constitutes a small step in the develop-

265 ment of this framework, shedding light on some aspects of associational engagement and migrant lives, which otherwise have not received much research attention. It should be emphasized that the study has a particular focus on So- mali-Danish associational key persons, who are quite well established in Danish society. The dissertation is thus not representative in a statistical sense. Yet, I will argue, that this focus is a both a relevant and interesting one, as the associational key persons play central roles in initiating ‘community’ activities. They may function both as bridge builders and connecting links to Danish society and to the so-called ‘diaspora’; they might also act as agents of development in So- malia and Somaliland. There are three main conclusions of the dissertation. Firstly, I conclude that associational engagement is a way of fulfilling senses of obligation and exploiting opportunities, as well as enacting gendered cultural intelligibility. Associa- tional engagement may thus function as a vehicle for negotiating and re- establishing social positions and social esteem at the group-specific level. The high number of Somali associations can thereby also be seen as social spaces for positioning. Positioning, however, is guided by a practical sense of the game, as Bourdieu terms it, rather than through premeditated and calculated strategies for social prestige. Secondly, in consequence to the claims above, I conclude that associational engagement constitutes both a strategy of participation and a means of self- realization. Associational engagement may offer an opportunity to use competences and qualifications which otherwise can be difficult to exploit in Danish society. In this way, associations can be way of obtaining recognition for achievement and social esteem. Associational engagement, however, was also related to conflicts and misunderstandings, and recognition mainly took place amongst other Somalis. Thirdly, I conclude that there is no contradiction between processes of inclusion and transnational involvement in relation to associational engagement. On the contrary, the most transnationally involved associational key persons were also among the most established in Danish society. Furthermore, struggles for recognition and negotiations of positioning had both local Danish and transnational dimensions. In the remaining part of the conclusion, I elaborate and supplement these claims through my answers to the three supplementary research questions.

266 Destinations, commitments and loyalties of associational involvement The first sub-question concerns why and how Somali-Danes involve themselves in associations and towards which destinations, commitments and loyalties they are oriented. I argue that Somali-Danish associations form a political field in Bourdieu’s sense, united and divided around games of positioning, conflicts, and collaboration. I identified four main social principles of organization: gender, clan affiliation, political orientation (in relation to Somalia), and religious devotion. These principles are often interrelated and there were especially close relationships between gender and Islam, and between clan and political orientation. I further argued that the Somali clan system works as a principle of gener- ating trust and distrust, as flexible and contextual solidarity. While some associations were explicitly organized in terms of clan affiliation, others endeavoured to overcome this logic. Loyalties could thus be mobilized at various levels: from local to pan-Somali. The destinations of the associational engagement included Denmark, Soma- lia and Somaliland, and at a first glance, seemed rather uncomplicated. Somali- Danish everyday-oriented associations were concerned with life in Denmark, often explicitly articulating their aims as integration, while transnational associations supported reconstruction and development in Somalia and Somaliland, usually with a transnational organization with Somali members in different countries. A closer analysis, however, reveals that the distinction between these two overall types of associations is not that clear-cut after all. On the one hand, several everyday-oriented associations were engaged in transnational reconstruction projects, and the emphasis on integration was sometimes supplemented with a focus on long-term repatriation. Transnational Somali associations, on the other hand, did not necessarily have a focus on processes of inclusion and participation in Danish society. Still, the link to such processes was evident in the sense that many informants emphasized that their contributions were related to experiences and learning processes in Denmark, or more generally in the West. Furthermore, when we turn from associations to Somalis engaged in associations, the distinction between everyday orientation in Denmark and transnational involvement becomes even more blurred. All the transnationally engaged associational key persons were – or had been – involved in associational activities focusing on processes of inclusion in Denmark. At the level

267 of associational key persons, there was thus no contrast between supporting reconstruction and inclusion in Danish society. Indeed, it seemed that a well- established position in Danish society might strengthen (though not necessarily result in) transnational involvement at the institutional and collective level, pointing to the importance of human, economic and social capital (in Bourdieu’s sense) in furthering transnational practices. The linkage between transnational involvement and inclusion in Denmark, however, seemed less pronounced for ‘ordinary members’, who did not organize activities, or indeed for Somali-Danes not engaged in associations. Rather, their positions in Danish society vary a lot and it is thus not possible to make the same kind of overall analytical claim in their cases. Indeed, my study indicates that the position in the country of residence is less important for involvement in individual transnational practices, such as remittances to relatives, than it is for associational involvement on the organizational level. More research is needed in relation to this issue. I further argue that commitments in associations were articulated as arising out of senses of obligation and responsibility, whether directed towards other Somalis in Denmark, development in Somalia or Somaliland, or both. In Den- mark, such obligations and responsibilities concerned the difficult overall positions of many Somali-Danes, and associational engagement was focusing on furthering processes of inclusion in Danish society through education, information about Denmark, and strengthening of social networks. Likewise, associations were seen as a particularly Danish way of participating in civil society. In Somalia and Somaliland, obligations and responsibilities were related to the fact that the civil war had caused huge human suffering, and that reconstruction, development, and, in some parts, reconciliation is urgently needed. To live in the West is often perceived at least potentially as to be in a situation allowing access to economic resources, education, and exposure to democratic systems: privileges which are expected to be shared, not least with relatives and dependants in more vulnerable situations. In both Danish and Somali contexts, the Somali clan system and its emphasis on helping out kin in need can be mobilized on different levels. Initiating or supporting an organization might be one way of responding to obligations and responsibilities, though it does not seem to function as a substitute to individual remittances to relatives.

268 Obligations and responsibility, however, go hand in hand with opportunities. Such opportunities relate to the Danish emphasis on associations as a cornerstone in Danish society and means of participation as well as, more broadly, to the (expected) collectivity of associations as representing their constituencies. I argue that associations might function as platforms for action in Danish, Somali and/or transnational contexts, from which political positions and participation in decision-making processes can be established or striven for. Associational spokespersons and executive committee members might especially exploit such opportunities. Furthermore, associational engagement also offers possibilities of receiving financial and practical support for activities and projects and could thus contribute to the realization of ambitions and dreams.

Contexts of exit and reception Secondly, I asked to what extent and how the contexts of exit and reception shape associational engagement. The Somali civil war and its consequences form the context of exit for most of the Somali-Danish informants, who came to Denmark as asylum seekers. The civil war has resulted in the disintegration of the Somali Republic, with continued instability and conflict in large parts of Somalia, as well as political fragmentation, often organized around clan loyalties. To a certain degree, this situation is reflected in Denmark, where the number of Somali associations is high and they are relatively fragmented. Further- more, the civil war has caused loss of loved ones, of livelihoods and property, as well as the break-up and global dispersion of many Somali families. This was also the situation for the Somali-Danish informants, many of whom were entangled in webs of transnational obligations and relations. Still, the form, in- tensity, and scale of transnational relations varied greatly: from contact to family members and occasional remittances, to virtual transnational entrepreneurs, engaged in many different activities and at the centre of larger transnational social fields. The largest number of Somali migrants arrived in Denmark in the middle of the 1990s, in a historical and political moment when the Danish political climate had started to change. Somalis in Denmark have been at the centre of several negative political and media debates, referring to gendered stereotypes of Somali refugees as bogus, ‘difficult to integrate’, and/or exploiting the Danish welfare state, creating an overall context of identification and (non)belonging in

269 Denmark for the individual encounters with Danish society. I argue that Soma- lis became culturally intelligible as strangers in this overall context of reception, through intersections of (fantasies of) raciality, gender, and Islam. There were, however, individual responses to this situation, ranging from frustrations of stereotyping as shaping huge parts of everyday life to articulations of discrimination as an exception. In a theoretical perspective, this situation can be described both as social subordination, as Fraser puts it, and as lack of recognition of social esteem, as Honneth formulates it. The Somali-Danish informants and their reactions to life in Denmark were thus quite differentiated. Still, most (though not all) of the associational key persons came from relatively privileged backgrounds in Somalia and had gone through some degree of vocational training or higher education. This background improved their chances for navigating in Danish society, but often also implied huge losses of social position and social esteem in Denmark, compared to pre-civil war Somalia. It was striking that life in Denmark – and more generally in the West – was articulated as especially difficult for Somali men. Indeed, troubled family relations as well as ‘failed men’ were hotly debated among So- malis in Denmark, London, and Somaliland. Inspired by Robert Connell, I suggest that such debates form part of a diasporic gender (dis)order, as a context for the enactment and articulations of gender relations. I posit that associational engagement is partly shaped by these different contexts, as outlined above: the disintegration of Somalia and the manifold losses due to the civil war, the generally difficult position in Danish society, and the gender (dis)order. Associational engagement is, amongst other things, a means of (re-)establishing gendered social positions and obtaining recognition for achievement and esteem at the group-specific level. Being involved in an association offers an opportunity to use one’s competences and to do something for other Somalis in Denmark, or towards reconstruction in Somalia or Somaliland – showing responsibility and initiative. While this was true for both male and female associational key persons, associational engagement especially seemed to offer a means for ‘enacting’ and ‘restoring’ masculinity: to do ‘the rights thing’ as a ‘proper’ man should do, thereby distancing oneself from the discursive figure of ‘failed men’. Let me emphasize, however, that there were different opinions as to what this implied: gender ideals as well as other social principles of organization were highly contested.

270 Still, associational engagement was not only about re-establishing positions and recognition, but also related to processes of participation. In contrast to the labour market, associations remain an open sphere for Somalis in Denmark. The Danish opportunity structure – especially the possibility for establishing an association and applying for funding – makes associations an attractive and accessible arena for participation and for obtaining resources and funding. Indeed, several of the Somali-Danish key persons emphasized that associations were a particularly Danish strategy of participation and a way of obtaining influence and involvement in Danish civil society: getting a voice in decision-making processes. While several of the associational key persons were members in the so-called integration councils and/or functioned as informal counsellors to NGOs, ministries or other agencies, more research is required to assess to what degree the Somali-Danish associations or other migrant associations actually have a substantial influence in Danish politics. However, just as the positive relationship between transnational involvement and processes of inclusion in Danish society was most outspoken in relation to associational key persons, the same mechanism seemed to be at work in relation to associations as a means of attaining participation and involvement in Danish society: that it mostly furthered inclusion for those who were (already) active in the organization of associations and mostly well established in Danish society. I will also add that, while associational engagement is encouraged, it still does not ‘count’ as much as employment towards gaining rights in Danish society. Still, engaging oneself in an association might be a way to at least partly reckon with the position as a victimized refugee or stranger in Danish society, and to participate in Danish society and be a ‘good Danish citizen’.

Transnational engagement, diasporic mobilisation and inclusion The final research question touches upon how transnational engagement and diasporic mobilization relate to processes of inclusion in Danish society. I have already argued that transnational associational engagement at the organizational level went hand in hand with processes of inclusion and participation in Danish society, but that such processes were most pronounced for associational key persons. Let me therefore turn to the particular transnational aspect of diasporic mobilization.

271 As the history of Somali migration shows us, migration, dispersion and political instability are not new phenomena in the Somali region. However, it is only recently that the term ‘diaspora’ has entered the vocabulary of Somali migration for designating the dispersed Somali population, used by policy makers, researchers, Somali migrants and stayers alike. Inspired by Brian Axel and Av- tar Brah – and in line with my focus on location-work and positioning – I propose an analytical focus on whom and what is included in the claims and identifications made in the name of diaspora. I am thereby in favour of a theoretical approach to diaspora studies, which perceive diaspora as a category of identification and a framework of interpretation, rather than as an ideal typical social organization or actor. Analyzing how the category of diaspora was used among the informants in Copenhagen, Somaliland and in London, I argue that it refers to a transnational and nation-like moral community, which is claimed, mobilized and recognized at different levels and by different actors. The term implic- itly (and sometimes quite explicitly) relates to the dispersion and suffering as a result of the Somali civil war. However, in contrast to much of the theoretical writings about diaspora, in which suffering and victimization are emphasized, the references to the ‘Somali diaspora’ mainly denote agency, solidarity and resources. Indeed, there are huge expectations for ‘the diaspora’ to contribute to reconstruction and development, not least from Somalis and from policy- makers. Due to the current policy emphasis, to identify – or be identified – with ‘the diaspora’ offers an opportunity for individuals and groups claiming to be or to represent ‘the diaspora’ to establish political positions and platforms for action as culturally intelligible agents of development and, thereby, political actors. This means, in other words, that the ‘diaspora position’ and mobilization in the name of diaspora offers a framework of identification and intelligibility from which social struggles for recognition can be carried out. Claims and mobilization in the name of diaspora may thus reinforce the struggles for recognition and (re-)establishment of positions already occurring in relation to associational engagement, situating associational actors as part of large-scale transnational efforts of ‘the diaspora’. Indeed, it was the most active associational So- mali-Danish key persons who mostly referred to their involvement in diasporic terms. Diasporic mobilization and identification thus also questions the distinc-

272 tion between processes of inclusion in Danish society and transnational involvement as two separate spheres.

Moving on: Thinking beyond this study In the last part of this conclusion, I briefly outline the implications for the claims and perspectives made in the dissertation. I have argued that associational engagement forms spaces of recognition for social esteem, for making a difference, and for enacting gendered cultural intelligibility, and that these aspects have both local and transnational dimensions. I have also shown that the transnationally most active and enterprising associational key persons are well established in Danish society, claiming that there is no a priori contradiction between processes of inclusion and transnational involvement. My analysis thus implies that the focus on processes of migrant inclusion as exclusively related to the country of residence is too narrow. When the analytical focus is directed towards migrants’ lives and social relations, a more complex social and political situation can be analyzed. On the one hand, I thus assert that studies of migration require methodological and theoretical frameworks, which take a broader point of departure rather than unequivocal senses of belonging and one- dimensional loyalties. But on the other hand, my dissertation also indicates that a strong focus on different positions and interests in migrant groups – locally and in transnational social space – is necessary for more detailed exploration, not least in relation to claims and identifications made in the name of ‘the diaspora’. Likewise, I propose that the recognition perspective offers important corrections to dominant ideas of integration in Danish society (as migrant inclusion is usually termed in Denmark). The focus on the dialogical and reciprocal foundation for recognition – whether formulated as a question of parity of participation or self-realization – suggests that integration is an ongoing and joint process of mutual change and adaptation. These claims have both theoretical and political implications. Politically, they challenge the rather one-sided Danish focus on integration as employment and adherence to Danish values. My dissertation indicates that the possibilities of participating as a peer – to use one’s competences and be recognised for one’s achievement – are crucial factors for processes of inclusion in Danish society, while social subordination and stereotyping impede it. Recognition might certainly be obtained through employment; there are, however, also other

273 spheres of society, such as voluntary work or political involvement, which are important to consider. More generally, I support the view that associational engagement can further democratic learning processes and inclusion in Danish society. Still, my dissertation suggests that more attention should be paid to collaboration between associations, and that the organizational level of associations is strengthened through courses and mentor arrangements. Likewise, the inclusion of women and youth in associations must be encouraged, as many associations are dominated by men. The conclusion that transnational involvement and orientation do not hinder inclusion in Danish society implies that migrants’ transnational fields of belonging and social practices do not necessarily harm their relationship to Denmark. This insight calls for more inclusive visions of Danishness than the rather exclusive and nationalist notions of Danish identity that are currently in circulation. Finally, my dissertation emphasizes the important contributions of migrants to development, an issue which is gaining increasing policy attention at the moment, not least in relation to so-called ‘diasporas’. However, I find it of utmost importance to pay close attention to who is claiming to be or to represent ‘the diaspora’ – and who does not get a voice in this. Theoretically, I suggest that the combination of transnational and recognition perspectives can make important contributions to many other fields, including ‘integration research’ and the general sociology of organizations. Indeed, the struggle for recognition is by no means a problem specific to Somalis in Den- mark or other migrant groups, but rather a much more common issue. There is more theoretical development to be done to combine recognition and transnational perspectives. Thus, although I have found much inspiration in both Fraser and Honneth’s versions of recognition theory, I claim that both their theoretical positions need to be further developed in relation to transnational and spatial dimensions. Likewise, I find that the policy emphasis on ‘diasporas’ and on migrant mobilization in the name of diaspora calls for more research. In accordance with my focus on recognition and transnationalism, I will end this dissertation with a call for collaborative, multi-sited and comparative research. We, as researchers, could learn many things from migrants’ associational engagements, which are, after all, most successful and efficient when they collaborate with each other. We should not only think of new areas for re-

274 search, but also of new ways of carrying out research and sharing our knowledge.

275 276 Post-script

Almost four years have passed since I started fieldwork for my dissertation. Be- low, I briefly outline what has happened to some of the informants and their associations since then. Several of the Somali-Danish informants talked about leaving Denmark and some of them have done so. Abdirizak Ali, the teacher, moved to the UK in 2005, on the initiative of his wife. Saphia Mansoor, who claimed that Denmark has two faces, has also moved to another western country. Mohamed Abshir, who came to Denmark in 1969, now spends a lot of his retirement in Somali- land. Hawo Yusuf, the founder of the Women’s Association, has visited her home region in Somalia and still contemplates going back. Elmi Samir, one of the Somscan & UK organizers, also thinks about returning, but has still not built a house in Somaliland. Fatuma Mohamed, who considered moving to the UK, has bought a house in Denmark and now sees her future here. So does Mah- moud Sheikh, my translator at the DOEA conference, who is no longer unemployed and whose future in Denmark seems brighter. When it comes to associational activities, Mohamed Gelle is still the coordinator of the Somali Network in Denmark and continues to be engaged in a number of activities. Likewise, he is the director of a local minority TV station in Copenhagen. Musse Sheikh, the biomedical scientist and volunteer social worker, is now the chair of the integration council in Ballerup. The Danish language classes of the Women’s Association are no longer much attended, but the association has taken up new activities, also organized by Noora Hassan and Hawo Yusuf. In Somaliland, the Somscan & UK project has progressed considerably, with altogether 16 houses being built in Burao; water and electricity have been installed, and a new school has been constructed in the area. One of the Danish Somscan & UK members has returned. The Water Project has provided clean water for a number of people through the establishment of the pump, but the present situation is unknown due to the flooding in the south of Somalia. Omar Said is now busy with many other activities. The DOEA did not turn into a coordinating actor of contributions from ‘the diaspora’ to the southern Somali regions after all, but the conference may have constituted a step in collaboration between different projects and associations. As Somalia seems be

277 on the edge of civil war again, nobody knows what the future may bring. One thing seems to be sure: reconstruction and reconciliation are as urgently needed as ever.

278 Acknowledgements

I wish to thank everybody who has helped and supported me in making this research possible and gratifying to undertake. First of all, I am indebted to all the Somali participants in the study in Copenhagen, Somaliland, and London – I remain deeply thankful for the help, confidence, and time that so many people have extended to me. Needless to say, the responsibility for the analysis is mine alone. I thank Mohamed Gelle, Musse Sheikh, Ahmed Dhaqane, and Ibrahim Yassin for their much appreciated help, as well as all of the other participants whose names are changed in the dissertation. I want to offer special thanks to (names changed) Aisha Ahmed, Abdirizak Ali, Elmi Samir, Fatuma Mohamed, and Hawo Yusuf for their precious help, as well as to Saphia Mansoor, who has become a dear friend. Likewise, I thank the Somali Network in Denmark, Som- scan & UK, and the Danish Refugee Council. Mahaad sanid! The research was made possible through a scholarship at the Department of Sociology, University of Copenhagen for which I am grateful. I thank my colleagues at the department, especially Vita Christiansen, for her professional and friendly way and Flemming Røgild for sharing interesting literature and beautiful poems. Similarly, I want to thank all my present and former PhD candidate colleagues as well as my ‘lunchmates’. I am obliged to my reading groups and others, who have offered feedback: thank you so much! Special thanks to Kri- stina Grünenberg, Liv os Stølan, Frederik Thuesen Pedersen, Kathrine Vitus Andersen, Stine Adrian, Simon Turner, Marianne Holm Pedersen, Mikkel Ryt- ter, Cindy Horst, Flemming Mikkelsen, and Mette Fink-Nielsen. A big thank you to Peter Hansen for highly appreciated collaboration in Copenhagen, Lon- don and, not least, in Somaliland! Likewise, I thank both of my supervisors, Lars Bo Kaspersen and Ninna Nyberg Sørensen, who – in spite of job changes – found time to offer competent and inspiring supervision as well as enthusiasm and support throughout the process. The dissertation has been part of the Diaspora, Development and Conflict research network, anchored at the former Centre for Development Research (CDR) and from 2003, at the Danish Institute for International Studies (DIIS). Being a part of the Diaspora programme has been intellectually stimulating as well as good fun and I wish to thank all the programme members and seminar

279 participants. I also thank the former Globalization and Governance research theme at DIIS, where I spent six months in 2005 as well as Sussex Centre for Migration Research, University of Sussex, for a Marie Curie Scholarship which made it possible to stay for three months at the Centre in 2004. Thanks to Rich- ard Black for supervision in this period. I am grateful for feedback at conferences, PhD courses and other events where I have presented the dissertation. Special thanks to the participants in the 9th International Conference for Somali Studies, September 2004, Aalborg. Parts of the dissertation are revised versions of published or forthcoming articles and book chapters: ‘Nomads, Sailors and Refugees. A brief history of So- mali migration’, Sussex Migration Working Papers, nr. 23, 2004; ‘The big demonstration. A study of transborder political mobilisation’, AMID Working Paper Series, nr. 42 (with Peter Hansen), 2005; ‘Ambivalent Encounters. Nego- tiating boundaries of Danishness, Somaliness and belonging’ in Kusow, A.M. & Bjørk, S. (eds.) From Mogadishu to Dixon: The Somali Diaspora in a Global Context. Lawrenceville, NJ: Africa World Press/The Red Sea Press, forthcoming; ‘Somali-Scandinavian Dreaming. When ‘the diaspora’ returns to the desert’ in Sørensen, N.N. (ed.) Living Across Worlds. Diaspora, Development and Transnational Engagement. Geneva: International Organization of Migration, in press; ‘Mobilizing ‘the Diaspora’. Somali Transnational Political Engage- ment’, accepted for publication in Journal of Ethnic and Migration Studies. I have benefited greatly from comments and suggestions from editors and readers, whom I all thank, especially Ralph Grillo, Valentina Mazzucato, Joakim Gundel, Stephanie Bjørk, Abdi Kusow and Maja Povrzanovic Frykman. I also thank Terry Mayer for language revision and Anne-Lise Schulze Andersen for setting up the dissertation. Finally, I thank my friends, parents, and neighbours for patience and help. Most of all, I want to thank my loving family: Justin, and my sweet children Sophia and Jonathan for always being there for me.

280 List of references

Abrahamson, P. (1990) 'The Scandinavian model of welfare' in D. Bouget & B. Palier (Eds.) Comparing Social Welfare Systems in the Nordic Countries and France. Paris: MIRE, 29-58. Adam, S. M. & Ford, R. (1997) Mending the rips in the sky. Options for Somali Communities in the 21st century. Lawrenceville: The Red Sea Press Inc. Adler, P. A. & Adler, P. (2002) 'The reluctant respondent' in J.F. Gubrium & J. A. Holstein (Eds.) Handbook of Interview Research. Context & Method. Thousand Oaks, London & New Deli: SAGE Publications, 515-535. Affi, L. (1997) 'The Somali Crisis in Canada: The Single Mother Phenomenon' in H.M. Adam & R. Ford (Eds.) Mending Rips in the Sky. Options for So- mali Communities in the 21st Century. Lawrenceville, NJ & Asmara, Eri- trea: The Red Sea Press, Inc., 441-448. Ahmed, A. J. (1995a) 'Preface' in A.J. Ahmed (Ed.) The Invention of Somalia. Lawrenceville: The Red Sea Press, ix-xv. Ahmed, A. J. (1995b) The Invention of Somalia. Lawrenceville: The Red Sea Press. Ahmed, S. (2000) Strange Encounters. Embodied Others in Post-Coloniality. London & New York: Routledge. Al-Ali, N., Black, R., & Koser, K. (2001) 'The limits to 'transnationalism': Bos- nian and Eritrean refugees in Europe as emerging transnational communities', Ethnic and Racial Studies, (24)4: 578-600. Al-Ali, N. & Koser, K. (2002a) 'Transnationalism, international migration and home' in N. Al-Ali & K. Koser (Eds.) New Approaches to Migration? Transnational communities and the transformation of home. London & New York: Routledge, 1-14. Al-Ali, N. & Koser, K. (2002b) New Approaches to Migration? Transnational communities and the transformation of home. London & New York: Routledge. Ali, A. Q. (1995) 'The Predicament of the Somali Studies' in A.J. Ahmed (Ed.) The Invention of Somalia. Lawrenceville: The Red Sea Press, Amit-Talai, V. (1997) 'In Precarious Motion: from territorial to transnational cultures', Canadian Review of Sociology and Anthropology, (34)3.

281 Andersen, H. (2003) 'Funktionalisme og sociale systemer' in H. Andersen (Ed.) Sociologi - en grundbog til et fag. København: Hans Reitzels Forlag, 27-47. Andersen, S. C. (2005) 'Smid ikke metoderne ud med badevandet. Epistemolo- gisk konstruktivisme og dens metodiske konsekvenser', Dansk Sociologi, (16)3: 45-60. Anderson, B. (1991) Imagined Communities. Reflections on the Origin and Spread of Nationalism. (2nd ed.) New York & London: Verso. Anderson, B. (1992) 'The last empires', New Left Review, 93: 3-13. Anderson, B. (1994) 'Exodus', Critical Inquiry, (20)2: 314-327. Anderson, B. (1998) 'Long-Distance Nationalism' in The Spectre of Compari- sons. Nationalism, Southeast Asia and the World. London & New York: Verso, 58-76. Andreassen, R. (2005). The Mass Media's Construction of Gender, Race, Sexu- ality and Nationality. An analysis of the Danish news media's communication about visible minorities from 1971 - 2004. PhD dissertation, University of Toronto. Anthias, F. (1998) 'Evaluating 'Diaspora': Beyond Ethnicity?', Sociology, (32)4: 557-580. Appadurai, A. (1996) Modernity at Large. Cultural Dimensions of Globaliza- tion. Minneapolis & London: University of Minnesota Press. Appiah, A. K. (1994) 'Identity, Authenticity, Survival. Multicultural societies and social reproduction' in C. Taylor & A. Gutman (Eds.) Multiculturalism. Examining the politics of recognition. Princeton, NJ: Princeton University Press, 149-163. Assal, M. A. M. (2004) Sticky Labels or Rich Ambiguities? Diaspora and challenges of homemaking for Somalis and Sudanese in Norway. Oslo: BRIC, University of Bergen. Axel, K. B. (2004) 'The Context of Diaspora', Cultural Anthropology, (19)1: 26-60. B.T., (2003) Somaliere sender børnene hjem til Afrika: Må ikke blive for danske. BT. 22.04.03. Back, L. (2005) 'Sociologisk lytten - at tale om bemærkelsesværdige ting: Den globale sociologis skala', Social Kritik, 100: 10-25. Barnes, C. (2006) 'U dhashay - Ku dhashay: Genealogical and Territorial Dis- course in Somali History', Social Identities, (12)4: 487-498.

282 Barre, S. (1971) 'The Evils of Tribalism' in My Country and My People. The collected speeches of Major-General Mohamed Siyad Barre. Mogadishu: Ministry of Information and National Guidance, 99-103. Basch, L., Glick Schiller, N., & Szanton Blanc, C. (1994) Nations Unbound. Transnational Projects, Postcolonial Predicaments, and Deterritorialised Nation-States. Pensylvania: Gordon & Breach Publishers. Baszanger, I. & Dodier, N. (2005) 'Ethnography: relating the part to the whole' in D. Silverman (Ed.) Qualitative Research. Theory, Method and Practice. London, Thousand Oaks & New Delhi: SAGE Publications, 9-34. Bauman, M. (2001) 'What you always wanted to know about the origins and usages of that word 'Diaspora' OR The Theology of Exile: hope and retribu- tion', irishdiaspora.net, 1-14. Bauman, Z. (1991) Modernity and Ambivalence. Itacha, New York: Cornell University Press. Bauman, Z. (1998a) 'Europe of Strangers.' in D. Broen & K. B. Bertelsen (Eds.) "Europe of Strangers". Integration policy in a comparative perspective. Co- penhagen: The Danish Refugee Council & The Danish Cultural Institute, 4- 14. Bauman, Z. (1998b) Globalisering. De menneskelige konsekvenser. Køben- havn: Hans Reitzels Forlag. Bech, H. (2005) Kvinder og mænd. København: Hans Reitzels Forlag. Beck, U. (2000) 'The cosmopolitan perspective: sociology of the second age of modernity', British Journal of Sociology, (51)1: 79-105. Benhabib, S. (2002) The Claims of Culture. Equality and Diversity in the Global Era. Princeton & Oxford: Princeton University Press. Beskæftigelsesministeriet. (2002) Regeringens udlændingepakke. www.bm.dk/flygtninge_indvandrere/udlaendingepakken/udlaendingeudspil. asp. Bjørn, N. H., Pedersen, D. A., & Rasmussen, L. K. (2003). Somaliere og det danske arbejdsmarked - om netværk, kommunikation og integration. København: Ministeriet for Flygtninge, Indvandrere og Integration. Bloch, A. (1999) 'Refugees in the job market: a case of unused skills in the Brit- ish economy' in A. Bloch & C. Levy (Eds.) Refugees, Citizenship and Social Policy in Europe. Houndmills: Macmillan Press Ltd, 187-210.

283 Bloch, A. & Atfield, G. (2002). The Professional Capacity of Nationals from the Somali Regions in Britain. London: Goldsmith College, University of London. Bourdieu, P. (1986) 'The Forms of Capital' in Richardson (Ed.) The Handbook of Theory and Research for the Sociology of Education. New York: Green- wood, 241-258. Bourdieu, P. (1990) In Other Words. Essays Towards a Reflexive Sociology. Cambridge: Polity Press. Bourdieu, P. (1994) 'Socialt rum og symbolsk magt' in M.D. Munk, M. Nørholm, & K. A. Petersen (Eds.) Pierre Bourdieu. Centrale tekster inden- for sociologi og kulturteori. København: Akademisk Forlag, 52-69. Bourdieu, P. (2005) 'Kampen om den symbolske orden. Pierre Bourdieu i samtale med Axel Honneth, Hermann Kocyba og Bernd Schwibs', Dansk Socio- logi, (16)4: 41-70. Bourdieu, P. & Krais, B. (1997) 'I mellemtiden har jeg lært alle den sociologiske forståelsessygdomme at kende' in Pierre Bourdieu. Centrale tekster inden for sociologi og kulturteori. Copenhagen: Akademisk Forlag, 111-127. Bourdieu, P. & Wacquant, L. (1992) An Invitation to Reflexive Sociology. Cambridge: Polity Press. Bourdieu, P. & Wacquant, L. (1996) Refleksiv Sociologi - mål og midler. Co- penhagen: Hans Reitzels Forlag. Bradbury, M. (1997) Somaliland Country Report. London: CIIR. Brah, A. (1996) Cartographies of Diaspora. Contesting Identities. London & New York: Routledge. Brettel, C. B. & Hollifield, J. F. (2000) 'Introduction. Migration Theory: Talk- ing across Disciplines' in C.B. Brettel & J. F. Hollifield (Eds.) Migration Theory. Talking Across Disciplines. New York: Routledge, 1-26. Butler, J. (1999) Gender Trouble. Feminism and the subversion of identity. (2nd ed.) London & New York: Routledge. Butler, J. (2004) 'Introduction: Acting in Concert' in Undoing Gender. New York & London: Routledge, 1-16. Caglar, A. (2006) 'Hometown associations, the rescaling of state spatiality and migrant grassroots transnationalism', global networks, (6)1: 1-22. Carder, S. (2003). Choices and expectations. Somali refugee families now in the UK. MA, University of East London.

284 Castles, S. (2000) 'The Racisms of Globalization' in Ethnicity and Globaliza- tion. From Migrant Worker to Transnational Citizen. London, Thousand Oaks & New Delhi: Sage Publications, 163-186. Castles, S. & Miller, M. J. (2003) The Age of Migration. International Popula- tion Movements in the Modern World. (3rd ed.) Houndmiles & New York: Palgrave. Charmaz, K. (2005) 'Grounded Theory in the 21st Century. Applications for Advancing Social Justice Studies' in N.K. Denzin & Y. S. Lincoln (Eds.) The Sage Handbook of Qualitative Research. Thousand Oaks, London & New Delhi: SAGE Publications, 507-535. Chimni, B. S. (1999) 'From resettlement to involuntary repatriation: towards a critical history of durable solutions to refugee problems', New Issues in Refugee Research, 2. Chimni, B. S. (1998) 'The Geopolitics of Refugee Studies: A view from the South', Journal of Refugee Studies, (11)4: 350-374. Clifford, J. (1994) 'Diasporas', Cultural Anthropology, (9)3: 302-338. Clifford, J. (1997) 'Diasporas' in Routes. Travel and Translation in the Late Twentieth Century. Cambridge, Massachusetts & London: Harvard Univer- sity Press, 244-278. Cohen, R. (1997) Global Diasporas. An Introduction. Seattle: University of Washington Press. Collin, F. (1998) 'Socialkonstruktivisme og den sociale virkelighed' in M. Jär- vinen & T. M. Bertilsson (Eds.) Socialkonstruktivisme. Bidrag til en kritisk diskussion. København: Hans Reitzels Forlag, 41-67. Collins, P. H. (1998) 'It's all in the family: Intersections of gender, race, and nation', Hypatia, (13)3: 62-82. Collinson, S. (1999) 'Globalisation and the dynamics of international migration: implications for the refugee regime', New Issues in Refugee Research, 1: 1- 31. Colonial Office (1931). Memorandum. Migration of Somalis from Italian Somaliland. London: Colonial Office. Colonial Office (1960a). Report of the Somaliland Protectorate Constitutional Conference. London: Colonial Office. Colonial Office (1960b). Somaliland 1958 and 1959. London: Colonial Office.

285 Connell, R. W. (2005) 'Globalization, Imperialism, and Masculinities' in M.S. Kimmel, J. Hearn, & R. W. Connell (Eds.) Handbook on Studies on Men & Masculinities. Thousand Oaks, London & New Delhi: SAGE Publications, 71-89. Cordero-Guzman, H. R. (2005) 'Community-Based Organisations and Migra- tion in New York City', Journal of Ethnic and Migration Studies, (31)5: 889-909. Crenshaw, K. W. (1994) 'Mapping the Margins: Intersectionality, Identity Poli- tics, and Violence Against Women of Color' in M. Fineman & R. Mykitiuk (Eds.) The Pulic Nature of Private Violence. New York: Routledge, 93-120. Crisp, J. (1999a) '"Who has counted the refugees?" UNHCR and the politics of numbers', New Issues in Refugee Research, 12: 1-17. Crisp, J. (1999b) 'Policy challenges of the new diasporas: migrant networks and their impacts on asylum flows and regimes', New Issues in Refugee Re- search, 7: 1-8. Dahabshiil (2002) Dahab Shiil Calendar 2003. Hargeisa: Dahabshiil Pvt. Ltd. Co. Dahinden, J. (2005) 'Contesting transnationalism? Lessons from the study of Albanian migration networks from former Yugoslavia', Global Networks-A Journal of Transnational Affairs, (5)2: 191-208. Dahl, H. M. (2004) 'A View from the Inside. Recognition and Redistribution in the Nordic Welfare State from a Gender Perspective', Acta Sociologica, (47)11: 325-337. Dahl, H. M., Stoltz, P., & Willig, R. (2004) 'Recognition, Redistribution and Representation in Capitalist Global Society: An interview with Nancy Fra- ser', Acta Sociologica, (47)11: 374-382. Dahl, K. M. & Jakobsen, V. (2005). Køn, etnicitet og barrierer for integration. Fokus på arbejde, uddannelse og foreningsliv. København: Socialforsk- ningsinstituttet. Daly, M. & Saraceno, C. (2002) 'Social exclusion and gender relations' in B. Hobson, J. Lewis, & B. Siim (Eds.) Contested Concepts in Gender and So- cial Politics. Cheltenham, UK & Northampton, MA, USA: Edward Elgar, 84-104. Danmarks Statistik (1996). Statistisk Årbog 1996. København: Danmarks statistik.

286 Dansk Flygtningehjælp. (1996) Enkeltskæbner og masseflugt. 1956-1996. Flygtninge 4: 1-23. Dansk Flygtningehjælp (2006). Akkumuleret statistik for repatriering af flygtninge og indvandrere, 01.01-05-31.12.2005. København: Dansk Flygtningehjælp. Davies, B. & Harré, R. (1990) 'Positioning: The Discursive Production of Selves', Journal for the Theory of Social Behaviour, (20)1: 43-63. de los Reyes, P. & Mulinari, D. (2005) Intersektionalitet. Kritiska reflektioner över (o)jämlikhetens landskap. Malmö: Liber. Diken, B. (1998) Strangers, Ambivalence and Social Theory. Aldershot: Ash- gate Publishing Company. DRC (2000a) Ansøgning om økonomisk støtte til det somaliske netværk. Kø- benhavn: Dansk Flygtningehjælp. DRC (2000b) Det somaliske netværksprojekt. Evaluering af Dansk Flygtninge- hjælps netværksprojekt 1999. København: Dansk Flygtningehjælp. Ejrnæs, M. (2001) 'Integrationsloven - en case, der illustrerer etniske minoriteters usikre medborgerskabsstatus', AMID Working Paper Series, 1: 1-20. Ejrnæs, M. (2002) 'Etniske minoriteters tilpasning til livet i Danmark - forhol- det mellem majoritetssamfundet og etniske minoriteter', AMID Working Pa- per Series, 18: 1-27. Emerek, R. (2003) 'Integration - eller inklusion? Den danske diskussion om integration', AMID Working Paper Series, 31: 1-14. Eriksen, T. H. (2002) 'The Sexual Life of Nations. Notes on gender and nation- hood', Kvinder, Køn & Forskning, (11)2: 52-65. Fadel, U. H., Hervik, P., & Vestergaard, G. (1999) 'De "besværlige" somaliere' in P. Hervik (Ed.) Den generende forskellighed. Danske svar på den stigende multikulturalisme. København: Hans Reitzels Forlag, 171-213. Faist, T. (2000) The Volume and Dynamics of International Migration and Transnational Social Spaces. Oxford: Clarendon Press. Faist, T. (2004) 'The transnational turn in migration research: perspectives for the study of politics and polity' in M. Povrzanovic Frykman (Ed.) Transna- tional Spaces: Disciplinary Perspectives. Malmö: IMER, 11-45. Fangen, K. (2004) ‘Identity and symbolic ethnicity among Somali immigrants in Norway’. Paper presented at the 9th International Congress of Somali Studies, September 3-5, Aalborg.

287 Fangen, K. (2006a) 'Humiliation Experienced by Somali Refugees in Norway', Journal of Refugee Studies, (19)1: 69-93. Fangen, K (2006b) Assimilert, hybrid eller inkorporeret i det etniske? Tilpas- ning og identifikasjon blant somaliere i Norge, Sociologisk tidsskrift, (14): 4-33. Farah, N. (1986) Maps. London: Picador. Farah, N. (2000) Yesterday, Tomorrow. Voices from the Somali Diaspora. Lon- don & New York: Cassell. Fink-Nielsen, M., Hansen, P., & Kleist, N. (2002) 'Repatriering - afsluttet eller fortsat mobilitet?', Den Ny Verden, (35)3: 52-65. Fink-Nielsen, M., Hansen, P., & Kleist, N. (2004) 'Roots, Rights and Responsi- bilities. Place-making and Repatriation among Somalis in Denmark and Somaliland', Stichproben. Vienna Journal of Critical African Studies, (7)4: 25-48. Fink-Nielsen, M. & Kleist, N. (2000). Tilhørsforholdets politikker. En analyse af hvorfor så få somaliere repatrierer fra Danmark. MA thesis, Institute of International Development & Institute of History, Roskilde University Cen- tre. Fortier, A.-M. (2000) Migrant Belongings. Memory, Space, Identity. Oxford & New York: Berg. Fraser, N. (1995) 'From Redistribution to Recognition? Dilemmas of Justice in a 'Post-Socialist' Age', New Left Review, 212: 68-93. Fraser, N. (2000) 'Rethinking Recognition', New Left Review, (2nd Series) 3: 107-120. Fraser, N. & Honneth, A. (2003) Redistribution or Recognition? A politico- philosophical exchange. London & New York: Verso. Gardner, J. & Bushra, J. E. (2004) Somalia the Untold Story. The war through the eyes of Somali women. London, Sterling & Virginia: CIIR & Pluto Press. Geshekter, C. (1997) 'The Dead of Somalia in a Historical Perspective' in H.M. Adam & R. Ford (Eds.) Mending the rips in the sky. Options for Somali Communities in the 21st century. Lawrenceville: The Red Sea Press Inc., 65-98. Giddens, A. (1987) 'What do sociologists do?' in Social Theory and Modern Sociology. Cambridge: Polity Press, 1-21.

288 Gilroy, P. (1993) The Black Atlantic. Modernity and Double Consciousness. London & New York: Verso. Gilroy, P. (1994) 'Diaspora', Paragraph, (17)1: 207-212. Gilroy, P. (2004) 'Foreword: Migrancy, Culture and a New Map of Europe' in H. Raphael-Hernandez (Ed.) Blackening Europe. The African American Presence. New York & London: Routledge, xi-xxii. Global Commission on International Migration (2005). Migration in an inter- connected world: New directions for action. Switzerland: Global Commis- sion on International Migration. Goldring, L. (1998) 'The Power of Status in Transnational Social Fields' in M.P. Smith & L. E. Guarnizo (Eds.) Transnationalism from below. New Bruns- wick, New Jersey: Transaction Publishers, 165-195. Goldring, L. (2001) 'The Gender and Geography of Citizenship in Mexico-US Transnational Spaces', Identities: Global Studies in Culture and Power, (7) 4: 501-538. Goldsmith, P. (1997) 'The Somali Impact on Kenya, 1990-1993: The View from Outside the Camps' in H.M. Adam & R. Ford (Eds.) Mending the rips in the sky. Options for Somali Communities in the 21st century. Lawrence- ville: The Red Sea Press Inc., 461-484. Gosh, Flora (2005) 'Lower benefits to the refugees in Denmark: Where is recognition?', Paper presented at the annual Danish sociological conference. Greenfield, R. D (1987). Somalis abroad - a neglected factor. Africa Now, 20- 21. Griffiths, D. (1997) 'Somali refugees in Tower Hamlets: clanship and new identities', New Community. A journal of research and policy on ethnic relations, (23) 1: 5-24. Griffiths, D. (2002) Somali and Kurdish Refugees in London. New Identities in the Diaspora. Aldershut: Ashgate. Grillo, R. (2001) 'Transnational Migration and Multiculturalism in Europe', WPTC-01-08, 1-29. Grund, Jens and Pinholt, Kristoffer (2006). FN: Danmark brød racismekonven- tionen. Jyllandsposten. 14.03.06. Grünenberg, K. (2006). Is Home where the Heart is, or where I Hang my Hat? Constructing Senses of Belonging Among Bosnian Refugees in Denmark. PhD Dissertation, Department of Sociology, University of Copenhagen.

289 Guarnizo, L. E. & Smith, M. P. (1998) 'The Locations of Transnationalism' in M.P. Smith & L. E. Guarnizo (Eds.) Transnationalism From Below. New Jersey: Transaction Publishers, 3-34. Guba, E. G. & Lincoln, Y. S. (2005) 'Paradigmatic Controversies, Contradic- tions, and Emerging Confluences' in N.K. Denzin & Y. S. Lincoln (Eds.) The Sage Handbook of Qualitative Research. Thousand Oaks, London & New Delhi: SAGE Publications, 191-215. Gundel, J. (2002). The Migration-Development Nexus: Somalia Case Study. Centre for Development Research. Gundelach, P. & Torpe, L. (1999) 'Befolkningens fornemmelser for demokrati: foreninger, politisk engagement og demokratisk kultur' in J.G. Andersen, P. M. Christiansen, T. B. Jørgensen, L. Togeby, & S. Vallgårda (Eds.) Den demokratiske udfordring. København: Hans Reitzels Forlag, 70-91. Gupta, A. & Ferguson, J. (1997a) Culture, Power, Place. Explorations in Criti- cal Anthropology. Durham & London: Duke University Press. Gupta, A. & Ferguson, J. (1997b) 'Discipline and Practice. 'The Field' as Site, Method, and Location in Anthropology' in A. Gupta & J. Ferguson (Eds.) Anthropological Locations: Boundaries and Grounds of a Field Science. Berkeley, Los Angeles & London: Cambridge University Press, 1-46. Habermas, J. (1994) 'Struggles for Recognition in the Democratic Constitu- tional State' in C. Taylor & A. Gutman (Eds.) Multiculturalism. Examining the politics of recognition. Princeton, NJ: Princeton University Press, 107- 148. Hagi, A. O. & Hagi, A. O. (1998). Clan, Sub-clan and Regional Representation in the Somali Government Organization 1960-1990: Statistical Data and Findings. USA. Halane, F. W. (2004) "Otherness" - a challenge or an obstacle for Somali- Swedish Women in Sweden. Paper presented at the 9th International Con- gress of Somali Studies, September 3-5, Aalborg. Hall, S. (1990) 'Cultural Identity and Diaspora' in J. Rutherford (Ed.) Identity. Community, Culture, Difference. London: Lawrence & Wishart, 222-237. Hall, S. (2000) 'Conclusion: the Multi-Cultural Question' in B. Hesse (Ed.) Un/Settled Multiculturalisms: Diasporas, Entanglements, 'Transruptions'. London & New York: Zed Books, 209-241.

290 Hammer, O. & Bruun, I. (2000). Etniske minoriteters indflydelseskanaler. År- hus: Århus Universitet. Hannerz, U. (2001) 'Introduktion: när fältet bliver translokalt' in U. Hannerz (Ed.) Flere fält i ett. Socialantropologer om translokala fältstudier. Stock- holm: Carlssons Bokförlag, 7-35. Hansen, L. L. (2005) 'At skabe lighed i deltagelse i praksis. Nye veje til ligestilling', Kvinder, Køn & Forskning, (14)4: 17-29. Hansen, P. (2004) 'Migrant transfers as a development tool. The case of Somali- land', DIIS Working Paper, 15: 1-20. Hansen, P. (2005) 'Circumcising Migration', Paper for the third international workshop in the Diaspora, Development and Conflict Programme on "Forms and means of transnational engagement". Hansen, P. (2006). Revolving Returnees: Transnational Return among Somali- landers. PhD Dissertation, Institute of Anthropology, University of Copen- hagen. Hansen, T. B. (2003) Between Autochthony and Diaspora. Indians in the 'New' South Africa. (pp. 1-41). Paper presented at the Annual Meeting of the American Anthropological Association. Chicago. Hansen, T. B. (2005) 'Between autochthony and diaspora: Indians in the "new" South Africa' in P.N.Abinales, N. Ishikawa, & A. Tanabe (Eds.) Dislocating Nation-States. Globalization in Asia and Africa. Kyoto & Melbourne: Kyoto University Press & Trans Pacific Press, 147-170. Haraway, D. (1991) 'Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective.' in D. Haraway (Ed.) Simians, Cy- borgs, and Women. The Reinvention of Nature. London: Free Association Books, 183-201. Haraway, D. (2004) 'The Promise of Monsters: A Regenerative Politics for In- appropriate/d Others' in The Haraway Reader. New York & London: Routledge, 63-123. Harris, H. (2004) The Somali community in the UK. What we know and how we know it. London: The Information Centre about Asylum and Refugees in the UK (ICAR). Hasse, C. (2000) 'Overvejelser om positioneret deltager-observation. Sexede astronomer og kønnede læreprocesser', Kvinder, Køn & Forskning, (9)4: 39- 51.

291 Heistler, B. S. (2000) 'The Sociology of Immigration. From Assimilation to Segmented Integration, from the American Experience to the Global Arena' in C.B. Brettel & J. F. Hollifield (Eds.) Migration Theory. Talking Across Disciplines. New York: Routledge, 77-96. Hervik, P. (2002) Mediernes muslimer - En antropologisk undersøgelse af mediernes dækning af religioner. København: Nævnet for Etnisk Ligestil- ling. Hess, R. L. (1966) Italian Colonialism in Somalia. Chicago & London: The University of Chicago Press. His Majesty's Stationary Office (1928). Somaliland. Report for 1927. London: His Majesty's Stationary Office. Hjære, M. (2003). NEM-VOL Migrant and Ethnic Minority Volunteering. Summary: Final Report Denmark. Odense: Center for Frivilligt Arbejde. Hjære, M. & Balslev, M. (2001). Frivillighed blandt etniske minoritetsforenin- ger. Odense: Center for frivilligt socialt arbejde. Holst, C. (2005) 'Nancy Fraser i kritikkens landskap', Kvinder, Køn & Forsk- ning, (14)4: 56-67. Holstein, J. A. & Gubrium, J. F. (2005) 'Interpretive Practice and Social Action' in N.K. Denzin & Y. S. Lincoln (Eds.) The Sage Handbook of Qualitative Research. Thousand Oaks, London & New Delhi: SAGE Publications, 483- 505. Honneth, A. (2004) 'Recognition and Justice: Outline of a Plural Theory of Jus- tice', Acta Sociologica, (47)11: 351-364. Honneth, A. (2006) Kamp om anerkendelse. Sociale konflikters moralske grammatik. København: Hans Reitzels Forlag. Hooghe, M. (2005) 'Ethnic Organisations and Social Movement Theory: The Political Opportunity Structure for Ethnic Mobilisation in Flanders', Journal of Ethnic and Migration Studies, (31)5: 975-990. Hopkins, G. (2006) 'Somali Community Organizations in London and Toronto: Collaboration and Effectiveness', Journal of Refugee Studies, (19)3: 361- 380. Horst, C. (2002) Xawilaad: The Importance of Overseas Connections in the Livelihoods of Somali Refugees in the Dadaab Refugee Camps of Kenya. University of Amsterdam.

292 Horst, C. (2003). Transnational Nomads. How Somalis cope with refugee life in the Dadaab camps of Kenya. PhD dissertation,University of Amsterdam. Horst, C. (2004) 'Money and mobility: transnational livelihood strategies of the Somali diaspora', Global Migration Perspectives, 9: 1-19. Ibrahim, J. Y. & Farah, A. J. (2001) Frivilligt återvandringsprojekt. Örebro & Copenhagen: Al-hijra Informationscenter & Somscan Cooperative Associa- tions. Ibsen, B. (1993) 'De frivillige lederes sociale profil og uddannelsesbehov', Fo- cus. Idræt, (17)4: 132-136. Ibsen, B. & Habermann, U. (2005) 'Guf' og 'Gab'. Foreninger og frivilligt arbejde i Københavns Kommune. Odense: Syddansk Universitet. Information (2003) For dansk til Danmark. Information. 30.12.03 International Crisis Group (2004) Biting the Somali Bullet. (Rep. No. 79). Nai- robi & Brussels: International Crisis Group. Itzigsohn, J. & Saucedo, S. G. (2002) 'Immigrant incorporation and sociocul- tural transnationalism', International Migration Review, (36)3: 766-798. Jackson, M. (2002) The Politics of Storytelling. Violence, Transgression and Intersubjectivity. Copenhagen: Museum Tusculanum Press. Jama, A. F. & Jensen, L. H. (2004) Somscan and UK - Diaspora joining forces to better the chances of return to the homeland. Paper presented at the 9th International Congress of Somali Studies. September 3-5, Aalborg. Jensen, B. (1999a) '30 års avisdebat om "de fremmede" i Danmark. Debatten om asylansøgerne' in D. Coleman & E. Wadensjö (Eds.) Indvandringen til Danmark. Internationale og nationale perspektiver. København: Spektrum, 333-406. Jensen, B. (1999b) '30 års avisdebat om "de fremmede" i Danmark. Fremmed- og gæstearbejderne 1963-1980' in D. Coleman & E. Wadensjö (Eds.) Ind- vandringen til Danmark. Internationale og nationale perspektiver. Køben- havn: Spektrum, 285-332. Jensen, L. H. (2004) Debriefing note. Danish Refugee Council: Copenhagen. Johnsdottor, S. (2002) Created By God. How Somalis in Swedish Exile Reas- sess the Practice of Female Circumcision. PhD Dissertation, Lund Univer- sity. Jordan, G. (2004) Somali Elders. Portraits from Wales. Cardiff: Butetown His- tory and Arts Centre.

293 Jørgensen, M. W. & Phillips, L. (1999) Diskursanalyse som teori og metode. Frederiksberg: Roskilde Universitetsforlag / Samfundslitteratur. Jørgensen, R. E. & Bülow, V. S. (1999) 'Ali og de fyrretyve k(r)oner. En analyse af Ekstra Bladets kampagne "fremmede" og om læsernes opfattelse af forskellighed' in P. Hervik (ed): Den generede forskellighed. Danske svar på den stigende multikulturalisme. København: Hans Reitzels Forlag, 81-107. Kallehave, T. (2000) 'Stat og hverdagsliv i migrationsproblematikken - to sider af samme sag', Dansk Pædagogisk Tidsskrift, 3: 38-49. Kallehave, T. (2001) 'Somali Migrants', Ethnologia Scandinavica, (31) 25-44. Kallehave, T. (2003). Somaliske livsformer i velfærdsstaten. Udforskning af begreber til analyse af brydninger og processer i immigrationsproblematikken. PhD Dissertation, University of Copenhagen. Kapteijns, L. (1994) 'Women and the Crisis of Communal Identity: The Cul- tural Construction of Gender in Somali History' in A. Samtar (Ed.) The So- mali Challenge: from catastrophe to renewal? Boulder, Colorado & Lon- don: Lynne Rienner Publishers, 211-332. Kaspersen, L. B. & Ottesen, L. (2001) 'Associationalism for A Hundred and Fifty Years - and still alive and kicking: Some reflections on the Danish civil society', Critical Review of International Social and Political Philoso- phy, (4)1. Kent, R., Hippel, K. v., & Bradbury, M. (2004). Social Facilitation, Develop- ment and the Diaspora: Support for Sustainable Health Services in Somalia. London: The International Policy Institute, King's College. Kibreab, G. (1999) 'Revisiting the Debate on People, Place, Identity and Dis- placement', Journal of Refugee Studies, (12)4: 384-410. Kimmel, M. S. (2005) 'Globalization and Its Mal(e)content: The Gendered Moral and Political Economy of Terrorism' in M.S. Kimmel, J. Hearn, & R. W. Connell (Eds.) Handbook on Studies on Men & Masculinities. Thousand Oaks, London & New Delhi: SAGE Publications, 414-431. Kjær, K. U. (2003) Flygtninge i Danmark. En kommenteret introduktion til det danske asylsystem. København: Akademisk Forlag. Kleist, N. (2002) 'Når hjemme er mere end et sted. Møder med Danmark blandt somaliere i diaspora.', Kvinder, Køn & Forskning, (11)2: 31-40. Kleist, N. (2004a) 'Nomads, Sailors and Refugees. A brief history of Somali migration', Sussex Migration Working Papers, 23: 1-14.

294 Kleist, N. (2004b) 'Situated Transnationalism: Fieldwork and Location-work in Transnational Research Methodology' in M.P. Frykman (Ed.) Transnational spaces: disciplinary perspectives. Malmö: Malmö University Press - Malmö högskola, 138-150. Kleist, N. & Hansen, P. (2005) 'The big demonstration. A study of transborder political mobilisation', AMID Working Paper Series, 42: 1-24. Kleist, N. & Hansen, P. (2007) 'Performing Diaspora. Studying the mobilization of a Somaliland transborder citizenry' in F. Osman, M. Mammo & J. Gundel (eds): Somalia: Diaspora and State Reconstitution in the Horn of Africa. London: Adonis & Abbey Publishers. Koch-Nielsen, I. & Clausen Dalsgaard, J. (2002) 'Værdierne i det frivillige arbejde' in P. Gundelach (Ed.) Danskernes værdier 1981-1999. København: Hans Reitzels Forlag, 232-259. Koptiuch, K. (1997) 'Third-Worlding at Home' in A. Gupta & J. Ferguson (Eds.) Culture, Power, Place. Explorations in Critical Anthropology. Dur- ham & London: Duke University Press, 234-248. Kusow, A. M. (2003) 'Beyond Indigenous Authenticity: Reflections on the In- sider/Outsider Debate in Immigration Research', Symbolic Interaction, (26)4: 591-599. Kusow, A. M. (2004) 'Contesting Stigma: On Goffman's Assumption of Nor- mative Order', Social Interaction, (27) 2: 179. Kvale, S. (1997) Interview. En introduktion til kvalitative forskningsinterview. København: Hans Reitzels Forlag. L 171 (2004) Lov om ændring af udlændingeloven og integrationsloven. L 365 (2002) Lov om ændring af udlændingeloven og ægteskabsloven med flere love. L 386 (2003) Lov om ændring af lov om straffeloven og udlændingeloven. L 474 (1998) Lov om integration af udlændinge i Danmark. Laclau, E. & Mouffe, C. (1985) Hegemony and Socialist Strategy. Towards a radical democratic politics. London & New York: Verso. Landholt, P., Autler, L., & Baires, S. (1999) 'From Hermano Lejano to Her- mano Mayor: the dialectics of Salvadoran transnationalism', Ethnic and Ra- cial Studies, (22)2: 90-315.

295 Larsen, J. E. (2003) 'Velfærdsstat og velfærdssamfund' in H. Andersen (Ed.) Sociologi - en grundbog til et fag. København: Hans Reitzels Forlag, 149- 168. Larsen, J. E. (2004). Fattigdom og social eksklusion. Tendenser i Danmark over et kvart århundrede. København: Socialforskningsinstituttet. Lather, P. (2001) 'Validity as an Incitement to Discourse: Qualitative Research and the Crisis of Legitimation' in V. Richardson (Ed.) Handbook on Re- search on Teaching. Washington D.C.: AERA, 241-250. Layton-Henry, Z. (1990) 'Immigrant Associations' in Z. Layton-Henry (Ed.) The Political Rights of Migrant Workers in Western Europe. London, New- bury Park & New Delhi: SAGE Publications, 94-112. Levitt, P. (1998) 'Social Remittances: Migration Driven Local-Level Forms of Cultural Diffusion', International Migration Review, (32)4: 926-948. Levitt, P. (2001) 'Transnational migration: taking stock and future directions', global networks, (1)3: 195-215. Levitt, P. & Schiller, G. N. (2004) 'Conceptualizing Simultaneity: A Transna- tional Social Field Perspective on Society', International Migration Review, (38)3: 1002-1039. Levitt, P. & Sørensen, N. N. (2004) 'The transnational turn in migration studies', Global Migration Perspectives, 6: 1-13. Lewis, I. M. (1994) Blood and Bone. The Call of Kinship in Somali Society. Lawrenceville, NJ: The Red Sea Press. Lewis, I. M. (2002) A Modern History of the Somali. Nation and State in the Horn of Africa. (4 ed.) Oxford, Hargeisa & Athens, Ohio: James Currey, Btec Books, Ohio University Press. Lilius, M. S. ed. (2001) Variations of the Theme of Somaliness. Åbo: Center for Continuing Education, Åbo Akademi University. Lisborg, A. & Lisborg, M. (2003). Destination Denmark - om asylmigration, menneskesmugling & mobilitetsstrategier. MA thesis, Institute of Interna- tional Development Studies & Geography, Roskilde University Centre. Luling, V. (2006) 'Genealogy as Theory, Genealogy as Tool: Aspects of Somali 'Clanship'', Social Identities, (12)4: 471-485. Lykke, N. (2003) 'Intersektionalitet – ett användbart begrepp för genusforsk- ningen', Kvinnovetenskaplig Tidsskrift, (21) 3.

296 Maanen, J. v. (1988) Tales of the Field. On Writing Ethnography. Chicago & London: The University of Chicago Press. Macdonald, C. L. & Merril, D. A. (2002) '"It Shouldn't Have to Be A Trade": Recognition and Redistribution in Care Work Advocacy', Hypatia, (17)2: 67-83. Mahler, S. J. & Pessar, P. R. (2001) 'Gendered geographies of power: Analyz- ing gender across transnational spaces', Identities - Global Studies in Cul- ture and Power, (7)4: 441-459. Maimbo, S. M. (2006) Remittances and Economic Development in Somalia. An Overview. Social Development Papers (38): 1-64. Malkki, L. (1992) 'National Geographic - the Rooting of Peoples and the Terri- torialization of National Identity Among Scholars and Refugees', Cultural Anthropology, (7)1: 24-44. Mansur, O. A. (1995) 'The Nature of the Somali Clan System' in A.J. Ahmed (Ed.) The Invention of Somalia. Lawrenceville: The Red Sea Press, 117-134. Marchal, R., Mubarak, J. A., Buono, M. D., & Manzolillo, D. L. (2000) 'Glob- alization and its impact on Somalia', UNDP/UNDOS, 1-13. Marcus, G. E. (1995) 'Ethnography in / of the World System: The Emergence of Multi-sited Ethnography', Annual Review of Anthropology, 24: 95-117. Massey, D. (1994) Space, Place and Gender. Oxford: Polity Press. Matthiessen, P. C. (2000) 'Indvandringen til Danmark - historik og demografi' in G.V. Mogensen & P. C. Matthiessen (Eds.) Integration i Danmark om- kring årtusindskiftet. Indvandrernes møde med arbejdsmarkedet og vel- færdssamfundet. Aarhus: Aarhus Universitetsforlag, 40-56. McGown, R. B. (1999) Muslims in the Diaspora: The Somali Communities of London and Toronto. Toronto, Buffalo & London: University of Toronto Press. Mezzetti, P. & Piperno, F. (2005) 'Migrant-development nexus and local authorities in Italy: towards a European Transnational Integration Approach', DIIS Brief, 1-11. Mikkelsen, F. (2001) Integrationens paradoks. Indvandrere og flygtninge i Danmark mellem inklusion og marginalisering. København: Catinét Re- search. Mikkelsen, F. (2002) 'Indvandrere og civilsamfund. En forskningsoversigt ved- rørende etniske minoriteters deltagelse i civilsamfundets samt kulturmødet

297 mellem minoriteter og danskere på arbejdspladsen, i boligområder og i foreninger', AMID Working Paper Series, 19: 1-30. Mikkelsen, F. (2003a) 'Indvandrerorganisationer i Danmark' in F. Mikkelsen (Ed.) København: Nordisk Ministerråd & Akademiet for Migrationsstudier i Danmark, 95-170. Mikkelsen, F. (2003b) Indvandrerorganisationer i Norden. København: Nor- disk Ministerråd & Akademiet for Migrationsstudier i Danmark. Mikkelsen, F. (2003c). IntegrationsStatus 1999-2003. 5 år i et integrationsper- spektiv. København: Catinét Research. Mikkelsen, F. (2007) Transnational solidaritet i forandring. Indvandrernes sociale, religiøse og politiske mobilisering i Danmark. 1965-2005 i et interna- tionalt perspektiv. Aalborg: Akademiet for Migration Studier i Danmark. Minister for ligestilling (2005). Beskæftigelse, deltagelse og lige muligheder til alle - regeringens handlingsplan for at nedbryde kønsbestemte barrierer til uddannelse, arbejde og foreningslivet blandt kvinder og mænd med anden etnisk baggrund end dansk. København: Minister for ligestilling. Ministry of Foreign Affairs of Denmark. (2006) Factsheet Denmark. Copenha- gen. Ministry of Integration (2003) Citizen in Denmark. A manual for new members of Danish society. Copenhagen: Ministry of Refugee, Immigration and Inte- gration Affairs. Ministry of Integration. (2005a) Cirkulæreskrivelse om naturalisation. Copen- hagen: Ministry of Refugee, Immigration and Integration Affairs. Ministry of Integration (2005b). Årbog om udlændinge i Danmark 2005. Kø- benhavn. Ministry of Integration. (2005c) Bekendtgørelse om meddelelse af tidsubegræn- set opholdstilladelse. LBK 826 24/08/2005. Copenhagen: Ministry of Refu- gee, Immigration and Integration Affairs. Ministry of Integration. (2006) Declaration on integration and active citizenship in Danish society. Copenhagen: Ministry of Refugee, Immigration and In- tegration Affairs. Mohamed, H. S. (1997) 'Somali Refugee Women's Experience in Kenyan Refugee Camps and Their Plight in Canada' in H.M. Adam & R. Ford (Eds.) Mending Rips in the Sky. Options for Somali Communities in the 21st Cen-

298 tury. Lawrenceville, NJ & Asmara, Eritrea: The Red Sea Press, Inc., 431- 440. Mohamoud, A. A. (2005). Reversing the Brain Drain in Africa. Harnessing the intellectual capital of the diaspora for knowledge development in Africa. Amsterdam: SAHAN. Montclos, M.-A. P. d. (2003) 'A Refugee Diaspora. When the Somali go West' in K. Koser (Ed.) New African Diasporas. London & New York: Routledge, 37-55. Mortensen, N. (2000) 'Differentieringer og integrationer - eksklusioner og inklusioner' in J.E. Larsen & I. H. Møller (Eds.) Kontinuitet og forandring. Nye differentierings- og integrationsformer i samfundet? København: Sam- fundslitteratur, 99-140. Moser, I. (2004) 'On beoming disabled and articulating alternatives: The multiple modes of ordering disability and their interferences', Cultural Studies, (19)6: 667-700. Mouffe, C. (1997) The Return of the Political. London & New York: Verso. Moya, J. C. (2005) 'Immigrants and Associations: A Global and Historical Per- spective', Journal of Ethnic and Migration Studies, (31)5: 833-864. Møller, B. & Togeby, L. (1999). Oplevet diskrimination. En undersøgelse blandt etniske minoriteter. København: Nævnet for Etnisk Ligestilling. Necef, M. Ü. (2000) 'Den danske debat om indvandrere og flygtninge 1964- 2000 - synspunkter i og på forskningen', Jyske Historiker, 89: 133-159. Necef, M. Ü. (2001) 'Indvandring , den nationale stat og velfærdsstaten' in P. Seeberg (Ed.) Ubekvemme udfordringer. Aktuelle tendenser i den danske og europæiske indvandrerdiskussion. Odense: Odense Universitetsforlag, 31- 64. Nielsen, K. B. (2004) 'Next Stop Britain: The Influence of Transnational Net- works on the Secondary Movement of Danish Somalis', Sussex Migration Working Papers, 22: 1-21. NSHF (2002). Transfer of Competence and Human Resources from the Somali Diaspora to Somalia. Report from the seminar held under the auspices of NSHF - The Nordic Governments' Joint Consultation Group for Refugees. Stockholm: NSHF. Olwig, K. F. (1999) 'Narratives of Belonging: Life Stories in Family Networks of West Indian Background', CDR Working Paper, 99.7: 27-48.

299 Olwig, K. F. & Hastrup, K. (1997) Siting Culture. The shifting anthropological object. London & New York: Routledge. Osman, A. Y. (1996) In the Name of Our Fathers. London: HAAN Publishing. Oxford Advanced Learner's Dictionary of Current English (2000) (6th ed.) Ox- ford & New York: Oxford University Press. Owen, R. (1985). Migrant Workers in the Gulf. London: The Minority Rights Group. Park, R. E. (1928) 'Human Migration and the Marginal Man', The American Journal of Sociology, (XXXIII) 6: 881-893. Pedersen, S. (1999) 'Vandringen til og fra Danmark i perioden 1960-1997' in D. Coleman & E. Wadensjö (Eds.) Indvandringen til Danmark. Internationale og nationale perspektiver. Copenhagen: Spektrum, 233-284. Portes, A. (1998) 'Social Capital: Its origins and applications in modern sociology', Annual Review of Sociology, (24)1-24. Portes, A. (2000) 'The two meanings of social capital', Sociological Forum, (15)1: 1-12. Portes, A. (2001) 'Introduction: the debates and significance of immigrant transnationalism', global networks, (1)3: 181-194. Portes, A. (2003) 'Conclusion: Theoretical convergencies and empirical evidence in the study of immigrant transnationalism', International Migration Review, (37)3: 874-892. Portes, A., Guarnizo, L. E., & Landolt, P. (1999) 'The study of transnationalism: pitfalls and promise of an emergent research field', Ethnic and Racial Studies, (22)2: 217-237. Putnam, R. D. (1995) 'Bowling Alone: America's Declining Social Capital. An interview with Robert Putnam', Journal of Democracy, (6)1: 65-78. Putnam, R. D. (2000) Bowling Alone. The Collapse and Revival of American Community. New York: Simon & Schuster. Ragin, C. S. (1992) '"Casing" and the process of social inquiry' in C.S. Ragin & H. S. Becker (Eds.) What is a Case? Exploring the Foundations of Social Inquiry. Cambridge: Cambridge University Press, 217-226. Rayne, H. (1921) 'Introduction' in Sun, Sand and Somalis. Leaves from the note-book of a district commissioner in British Somaliland. London: H.F. & G. Witherby, 5-7.

300 Republic of Somaliland (1994) Repatriation of Somaliland Refugees in Ethio- pia and Other Neighbouring Countries. Policy and Programme of Somali- land Government. Hargeisa: Republic of Somaliland. Republic of Somaliland (1996) Submission on Statehood and Recognition of Republic of Somaliland. Hargeisa: Republic of Somaliland. Republic of Somaliland (2001) The Revised Constitution of the Republic of Somaliland. Unofficial English Translation. ww.somalilandforum.com/somaliland/constitution/revised_constitution.htm Richardson, L. (1997) Fields of Play. Constructing an Academic Life. New Brunswick, New Jersey: Rutgers University Press. Richardson, L. & St. Pierre, E. A. (2005) 'Writing. A Method of Inquiry' in N.K. Denzin & Y. S. Lincoln (Eds.) The Sage Handbook of Qualitative Re- search. Thousand Oaks, London & New Delhi: SAGE Publications, 959- 978. Røgilds, F. (2002) 'Race, lejrtænkning og nonracial humanisme - en samtale med Paul Gilroy', Social Kritik, (14)84: 44-64. Safran, W. (1991) 'Homelands in Modern Societies: Myths of Homeland and Return', Diaspora, (1)1: 83-99. Saksela, S. (2003) 'Mångkulturella organisationer och invandrerorganizationer i Finland' in F. Mikkelsen (Ed.) København: Nordisk Ministerråd & Akademiet for Migrationsstudier i Danmark, 235-282. Samatar, A. (1994a) 'The Curse of Allah: Civic Disembowelment and the Col- lapse of the State in Somalia' in A. Samtar (Ed.) The Somali Challenge: From Catastrophe to Renewal? Boulder, Colorado & London: 95-146. Samatar, A. (1994b) The Somali Challenge: from catastrophe to renewal? Boulder, Colorado & London: Lynne Rienner Publishers. Samatar, S. S. (1991). Somalia. A Nation in Turmoil. London: Minority Rights Group. Save the Children Fund (1994). The Somali Community in Cardiff. Cardiff: Save the Children Fund, Wales Division. Schiff, M. & Ozden, C. (2005). International migration, remittances and the brain drain. Washington: World Bank. Schiller, G. N. (1999) 'Transmigrants and Nation-States: Something Old and Something New in the U.S. Immigrant Experience' in C. Hirschman & P.

301 Kasinitz (Eds.) The Handbook of International Migration: The American Experience. New York: Russel Sage Foundation, 99-119. Schiller, G. N., Basch, L., & Blanc-Szanton, C. (1992) Towards a Transna- tional Perspective on Migration. Race, Class, Ethnicity, and Nationalism Reconsidered. New York: The New York Academy of Sciences. Schiller, G. N. & Fouron, G. (2001) Georges Woke Up Laughing. Long- distance Nationalism and the Search for Home. Durham & London: Duke University Press. Schiller, G. N. & Wimmer, A. (2002) 'Methodological nationalism and beyond: nation-state building, migration and the social sciences', global networks, (2)4: 301-334. Schrover, M. & Vermeulen, F. (2005) 'Immigrant Organisations', Journal of Ethnic and Migration Studies, (31)5: 823-832. Schwartz, J. (1990) 'On the Representation of Immigrants in Denmark' in F.Røgilds (Ed.) Every Cloud Has A Silver Lining. Copenhagen: Akademisk Forlag, 42-52. Scocozza, B. & Jensen, G. (1994) Danmarkshistoriens Hvem Hvad og Hvornår. Politikens etbinds Danmarkshistorie. København: Politikens Forlag. Scott, J. (1986) 'Gender: A Useful Category in Historical Analysis', American Historical Review, (91)5: 1053-1075. Silanyo, Ahmed Mohamed (1986). Statement of the Somali National Move- ment on the recent peace initiatives between Somalia and Ethiopia. Liberty. Magazine of the Somali National Movement, 2-3. Simmel, G. (1950 [1908]) 'The Stranger' in K.H.Wolff (Ed.) The Sociology of Georg Simmel. New York: The Free Press, 402-408. Skak, T. (1998) 'De er så svære at integrere' in A.-B.P.Steen (Ed.) Kan vi leve sammen? Integration mellem politik og praksis. Copenhagen: Munksgaard, 96-120. Smith, J. K. & Hodkinson, P. (2005) 'Relativism, Criteria, and Politics' in N.K. Denzin & Y. S. Lincoln (Eds.) The Sage Handbook of Qualitative Research. Thousand Oaks, London & New Delhi: SAGE Publications, 915-932. Smith, M. P. & Guarnizo, L. E. (1998) Transnationalism From Below. New Brunswick, New Jersey: Transaction Publishers. Snel, E., Engbersen, G., & Leerkes, A. (2006) 'Transnational involvement and social integration', global networks, (6)3: 285-308.

302 Sökefeld, M. (2006) 'Mobilizing in transnational space: a social movement approach to the formation of diaspora', global networks, (6)3: 265-284. Somali Relief Association (1992). The Somalis: An Invisible Community in Cri- sis. London: Haan Associates. Søndergaard, D. M. (1996) Tegnet på Kroppen: Køn: Konstruktioner og koder blandt unge voksne. København: Museum Tuscalanums Forlag. Søndergaard, D. M. (2005) 'Academic Desire Trajectories. Retooling the Con- cepts of Subject, Desire and Biography', European Journal of Women's Studies, (12)3: 297-313. Sørensen, N. N. (1995). Telling Migrants Apart. The experience of migrancy among Domican locals and transnationals. PhD Dissertation, Institute of Anthropology, University of Copenhagen. Sørensen, N. N. (2004) The Development Dimension of Migrant Transfers. Fi- nal Report. Copenhagen: Danish Institute for Development Studies. Sørensen, N. N. & Olwig, K. F. (2002) 'Mobile Livelihoods: Making a Living in the World' in N.N. Sørensen & K. F. Olwig (Eds.) Work and Migration: Life and livelihoods in a globalising world. London & New York: Routledge, 1-20. Sørensen, N. N., van Hear, N., & Engberg-Pedersen, P. (2002) 'The Migration- Development Nexus. Evidence and Policy Options State-of-the-Art Over- view', International Migration, (40)5: 1-47. Staunæs, D. (2003) 'Where have all the subjects gone? Bringing together the concepts of intersectionality and subjectification', NORA, (11)2: 101-110. Staunæs, D. (2004) Køn, etnicitet og skoleliv. København: Samfundslitteratur. Staunæs, D. (2005) 'From culturally avantgarde to sexually promiscuous - troubling subjectivities and intersectionality', NORA - Nordic Journal of Women's Studies, (11)2: 101-110. Steen, A.-B. (1993). Varieties of the Tamil Refugee Experience in Denmark and England. PhD Dissertation, University of Copenhagen. Stolle, D. (2001) 'Clubs and Congregations: The Benefits of Joining an Asso- ciation' in K.S. Cook (Ed.) Trust in Society. New York: Russel Sage Foun- dation, 185-201. Suleiman, Z. (1992) 'Somalis in Italy', Journal of the Anglo-Somali Society, 12- 13.

303 Taylor, C. (1994) 'The politics of recognition' in A. Gutman (ed.): Multicultur- alism. Examining the politics of recognition. Princeton, NJ: Princeton Uni- versity Press, 25-74. Taylor, C. & Gutman, A. (1994) Multiculturalism. Examining the politics of recognition. (2nd ed.) Princeton, NJ: Princeton University Press. Togeby, L. (1999) 'Et demokrati, der omfatter alle, der bor i Danmark?' in J.G. Andersen, P. M. Christiansen, T. B. Jørgensen, L. Togeby, & S. Vallgårda (Eds.) Den demokratiske udfordring. København: Hans Reitzels Forlag, 133-153. Torpe, L. & Kjeldgaard, T. K. (2003) Foreningssamfundets sociale kapital. Danske foreninger i et europæisk perspektiv. Aarhus: Aarhus Universitets- forlag. Touval, S. (1963) Somali Nationalism. International Politics and the Drive for Unity in the Horn of Africa. Cambridge: Harvard University Press. Turner, S. (2004) 'Introduktion', Den Ny Verden, (7) 7-13. Turner, S (2007). Burundians in Belgium: constructing, performing and contesting diaspora. Sørensen, N.N.(ed) Living Across Worlds. Diaspora, De- velopment and Transnational Engagement ,Geneva: International Organisa- tion of migration. Udlændingeservice (2006). Seneste tal på udlændingeområdet. København: Udlændingeservice. Udlændingestyrelsen (1999). Nøgletal på udlændingeområdet 1999. Køben- havn: Udlændingestyrelsen. Udlændingestyrelsen (2002). Nøgletal på udlændingeområdet 2002. Køben- havn: Udlændingestyrelsen. Udlændingestyrelsen (2004a). Fakta om asyl. www.udlst.dk. Udlændingestyrelsen (2004b). Hvad er Dublinkonventionen? www.udlst.dk. Udlændingestyrelsen (2004c). Udlændingestyrelsen - beretning 2003. Køben- havn: Udlændingestyrelsen. Udlændingestyrelsen (2005). Tal og fakta på udlændingeområdet 2004. Kø- benhavn: Udlændingestyrelsen. Udlændingestyrelsen (2006). Tal og fakta på udlændingeområdet 2005. Kø- benhavn: Udlændingestyrelsen. UNDP (2001) Human Development Report Somalia 2001. Nairobi, Kenya: United Nations Development Programme.

304 UNDP (2004) QUESTS. Qualified Expatriate Somali Technical Support. www.undp.so.org./TOKTEN/quests.pdf. UNDP Somalia & World Bank (2003). Socio-Economic Survey 2002. Somalia. Nairobi: UNDP Somalia & The World Bank. UNHCR (1951) Convention relating to the Status of Refugees of 28 July 1951. Geneva: UNHCR. UNHCR (1995) Populations of Concern to UNHCR: A Statistical Overview. Geneva: UNHCR. UNHCR (1998) Refugees and Others of Concern to UNHCR. 1997 Statistical Overview. Geneva: Statistical Unit, UNHCR. UNHCR (2002) Refugees and Others of Concern to UNHCR. 2000 Statistical Overview. Geneva: UNHCR. UNHCR (2003a) 2002 Annual Statistical Report: Somalia. Geneva: UNHCR. UNHCR (2003b) Resettlement Statistics. Geneva: UNHCR. UNHCR (2003c) Refugees by Numbers 2003. Geneva: UNHCR. UNHCR (2005) Statistical Yearbook 2003. Trends in displacement, Protection and Solutions. Geneva: UNHCR. UNHCR (2006) Asylum Levels and Trends in Industrialized Countries, Third Quarter, 2005. Geneva: UNHCR. UNOCHA (2004) SOMALIA: IDPs, returnees desperate for assistance in Somaliland - UN. www.irinnews.org: UN Office for the Coordination of Humanitarian Affairs. UNOCHA (2005). Somalia: Remitttances - a lifeline to survival. Nairobi: UN Office for the Coordination of Humanitarian Affairs. UNOCHA (2006). Somalia: Year in Brief 2005: A Chronology of Events. Nai- robi: UN Office for the Coordination of Humanitarian Affairs. Utteh, H. A. (1997) 'The Plight of the Somali Refugees in Europe. With Par- ticular Reference to Germany.' in H.M. Adam & R. Ford (Eds.) Mending Rips in the Sky. Options for Somali Communities in the 21st Century. Law- renceville, NJ & Asmara, Eritrea: The Red Sea Press, Inc., 449-460. van Hear, N. (1998) New diasporas. The mass exodus, dispersal and regroup- ing of migrant communities. Seattle: University of Washington Press. van Hear, N. (2002a) 'From 'Durable Solutions' to 'Transnational Relations': Home and Exile among Refugee Diasporas', CDR Working Paper, (02.9) 1- 18.

305 van Hear, N. (2002b) 'Sustaining societies under strain: remittances as a form of transnational exchange in Sri Lanka and Ghana' in N. Al-Ali & K. Koser (Eds.) New Approaches to Migration? Transnational communities and the transformation of home. London & New York: Routledge, 202-223. Veney, C. (1998) 'The Formation of Ethnic Identity in Somalia: Implications for Refugees in Kenya', Scandinavian Journal of Development Alternatives and Area Studies, (17): 2-3. Vermeulen, F. (2005) 'Organisational Patterns: Surinamese and Turkish Asso- ciations in Amsterdam, 1960-1990', Journal of Ethnic and Migration Stud- ies, (31)5: 951-973. Vignéras, S. (1900) 'Chapitre Premier. Historique.' in Notice sur la Côte Fran- caise des Somalis. Paris: Imprimerie Paul Dupont, 5-16. Wahlbeck, Ö. (2002) 'The concept of diaspora as an analytical tool in the study of refugee communities', Journal of Ethnic and Migration Studies, (28)2: 221-238. Warsame, A. M. (2002) Queens without Crowns. Somaliland women's changing roles and peacebuilding. Uppsala: Life and Peace Institute. Werbner, P. (2002) 'The place which is diaspora: Citizenship, religion and gender in the making of chaordic transnationalism', Journal of Ethnic and Mi- gration Studies, (28)1: 119-134. West, C. & Fenstermaker, S. (1995) 'Doing Difference', Gender and Society, (9)1: 8-37. West, C. & Zimmerman, D. H. (1987) 'Doing Gender', Gender and Society, (1)2: 125-151. Willig, R. & Petersen, A. (2004) 'Work and Recognition: Reviewing New Forms of Pathological Developments', Acta Sociologica, (47)11: 338-350. Wilson, E. (1990a) 'Somalis in South Wales - the Early Days', Journal of the Anglo-Somali Society, 26-28. Wilson, E. (1990b) 'Somalis in the Royal Navy', Journal of the Anglo-Somali Society, 9-11. Wimmer, A. & Schiller, G. N. (2002) 'Methodological nationalism and beyond: nation-state building, migration and the social sciences', global networks, (2)4: 301-334. www.dr.dk (2005) Kritik: Kulturministeren misbruger kanon. Danmarks Radio. 27.09.05.

306 www.imiscoe.org. (2006) International Migration, Integration and Social Cohe- sion. www.somalnet.dk (2003). Somaliske Foreninger i Danmark. www.somalnet.dk/dansk/forening.htm. www.statistikbanken.dk (2006) Befolkning og Valg. København: Danmarks Statistik. www.transcomm.ox.ac.uk. (2006) Transnational Communities Programme. Østergaard, B. (1983) Indvandrernes danmarkshistorie. København: C.E.G Gad. Østergaard-Nielsen, E. (2001) 'Transnational political practices and the receiving state: Turks and Kurds in Germany and Netherlands', global networks, (1)3: 261-281. Østergaard-Nielsen, E. (2002) Politik over grænser. Tyrkere og kurderes engagement i det politiske liv i hjemlandet. Aarhus: Magtudredningen.

307 308 English Summary

This dissertation analyzes the relationship between processes of inclusion and transnational involvement. It explores Somali-Danish ethno-national associational engagement, oriented towards everyday life in Denmark and reconstruction in Somalia and Somaliland. The main research question is: How do Soma- lis in Denmark negotiate positions and struggle for recognition in Danish society and transnationally through associational engagement? This question is supplemented by three others: Firstly, why and how do Somali-Danes involve themselves in associations? Secondly, to what extent do the contexts of exit and reception shape their associational engagements? Thirdly, how do transnational engagement and diasporic mobilization relate to processes of inclusion in Dan- ish society? The dissertation thereby brings to the fore questions of inclusion, recognition, and positioning in different social spaces. The study is based on ethnographic fieldwork among Somali-Danish associational key persons and organizers in Copenhagen, thereby focusing on a specific group, not on the general Somali population in Denmark, as the Somali-Danish informants are among the most well-established Somalis in Denmark in terms of employment and education. This fieldwork has been supplemented with shorter fieldwork in Somaliland and London; altogether 50 informants were included. The dissertation is structured as a research journey, divided into three parts with departures, border-crossings and horizons. In the first part, Departures, I position and situate the dissertation theoretically and methodologically. In chapter 1, the study, the research question and the background of the study are presented and the overall theoretical and political discussions are briefly introduced. The theoretical discussions are elaborated in chapter 2, where I position the study meta-theoretically and present three theoretical mappings which together constitute the theoretical optic of the dissertation. The first mapping – simultaneous engagement – brings together theories of transnational practices, modes of migrant incorporation and migrant associations. This mapping also presents the analytical concepts of social fields, as proposed by Bourdieu and Levitt & Schiller, and Massey’s notion of social space. The second mapping revolves around the concept of diaspora. Inspired by Brian Axel and Avtar Brah, I suggest a focus on how the category of dias-

309 pora is turning to a category or identification as well as being in the centre of political claims. Finally, in the third mapping, I present contemporary theories of recognition and positioning. On the one hand, I focus on recognition as fuel for social struggles, taking departure in Axel Honneth and Nancy Fraser’s dis- pute on the nature of recognition. On the other, I supplement these theories with perspectives of positioning, performativity and cultural intelligibility inspired by Judith Butler, Sara Ahmed, and other feminist writers. In chapter 3, these theoretical reflections are ‘translated’ into methodological directions, underpin- ning the production of fieldwork, paying special attention to the notions of location-work and situated knowledges. I furthermore reconstruct the process of fieldwork and present the informants. In part II, Across Borders, I turn to the analysis based on the empirical material. In chapter 4, I present the context of exit, outlining the history of the So- mali-speaking region as seen through an optic of migration. I show that Somali migration is part of a long tradition of mobility, but that the global dispersion of Somalis has been especially spurred by the civil war. Apart from this dispersion, the Somali civil war has had two more consequences central to the dissertation: disintegration and fragmentation of the Somali Republic and Somali politics, and a transnational remittance economy where migrants in the western and gulf countries support the livelihoods of their relatives in the Horn of Af- rica. In chapter 5, I analyze the context of reception in Denmark. The Danish political and public debates concerning immigration, refugees and integration started to change in the middle of the 1990s, coinciding with the arrival in the country of a larger number of Somali asylum seekers. Somalis in particular have been portrayed as a group said to be ‘difficult to integrate’. I argue that this atmosphere created a context of identification and (non)belonging to Den- mark – a context of misrecognition – for the individual experiences within Dan- ish society. The informants articulated such encounters in different ways, ranging from categorizing stereotyping and discrimination as an exception, to perceiving it as a part of everyday life. Life in Denmark might thus add to the manifold losses caused by the civil war. Furthermore, changes in gender and family relations in Denmark and more generally in the West are debated among Somalis in Denmark, who often claim that life in the West is more difficult for Somali men than women. I suggest that this phenomenon can be seen as part of a gender (dis)order in which the discourses of ‘failed men’ and ‘too liberated

310 women’ form the backdrop for the enactment of gender ideals. Finally, I suggest that this situation could be analyzed with inspiration from sociological work on ‘strangers’, as a way of stabilizing ambiguous identities. In chapter 6, I turn to associational engagement. I outline the history of So- mali-Danish associations and analyze the social principles of organizations which unite and divide the associations: gender, clan affiliation, political orientation in relation to Somalia, and religious devotion. I argue that Somali-Danish associations constitute a field in Bourdieu’s sense, united by games of positioning and conflicts. I continue with a case study of the Women’s Association, focusing on issues of everyday challenges faced by women following voluntary ; Wetherell & Edley, 1999) . Furthermore, Khadija ind. Likewise I analyze the long-distance political orientations of the organizers and some of the members. I further explore how Somali-Danish activists articulated associational engagement as a learning process in relation to a new political culture of voluntary work. I argue, on the one hand, that associations offer opportunities for funding and for establishing platforms for actions and speaking positions for associational key persons: in this way, associations can serve as a means of attaining a voice and participating in Danish civil society. On the other hand, however, most associational key persons experienced jealousy, misunderstandings and conflictsIn chapter related 7, twoto their case engagement. studies of transnational involvement are analyzed: the constitution conference of the Development Organization of East Africa (DOEA) and the transnational umbrella organization, Somscan & UK. Taking a point of departure in these two cases, I analyze the articulated visions of development and change, as well as the mobilization of and identification with ‘the diaspora’. Both DOEA and Somscan & UK had reconstruction and democrati- zation as explicit aims, and both actors endeavored to challenge and overcome lineage affiliation as a principle of identification and loyalty. I show how there are pronounced expectations towards the so-called ‘diaspora’ that is supposed to exploit the educational, economic and political opportunities in the West. Likewise, in case of return, ‘the diaspora’ is supposed to contribute to local development, rather than return to the local way of life. In part III, Horizons, I collect the theoretical and analytical threads in the dissertation and conclude it. In chapter 8, I argue that associational engagements form spaces of recognition in which social struggles for recognition can be car-

311 ried out and culturally intelligible gender ideals can be enacted. In this way, associational engagement is used to negotiate and re-establish social positions and obtain recognition for social esteem and achievement at the group-specific level, guided by a practical sense of the game. I claim that the Somali-Danish associational field is part of an unequal overall transnational social field with different positions, and that there is thus no necessary contradiction between transnational engagement and inclusion in the country of residence. Finally, I analyze claims and mobilization made in the name of ‘the diaspora’. I show how they refer to a moral, transnational, and nation-like community of resources and agency, which is claimed, mobilized and recognized at different levels and by different actors. Diasporic identification and mobilization is thus also a way of struggling for recognition and for negotiating positions. In chapter 9, I conclude the entire dissertation. Firstly, I conclude that associational engagement offers a way of fulfilling senses of obligation and exploiting opportunities as well as enacting gendered cultural intelligibility, thereby functioning as a vehicle for negotiating and re-establishing social positions and social esteem at the group-specific level. Secondly, I conclude that associational engagement constitutes both a strategy of participation and a means of self- realization through the use of competences and qualifications, which otherwise can be difficult to exploit in Danish society. Associational engagement can thereby be a way of obtaining recognition for achievement and social esteem. Thirdly, I conclude that there is no contradiction between processes of inclusion and transnational involvement in relation to associational engagement. On the contrary, the most transnationally involved associational key persons were also amongst the most established in Danish society. Finally, I outline some of the implications of the dissertation and call for more collaborative, multi-sited and comparative research.

312 Dansk resume

I denne afhandling analyseres somalisk foreningsengagement. Afhandlingen be- lyser det overordnede spørgsmål, hvordan somaliere i Danmark forhandler positioner og kæmper for anerkendelse i det danske samfund og transnationalt, set igennem foreningsarbejde. Baggrunden for denne problematik er på den ene side den blandede modtagelse, somaliere har fået i Danmark som ’svære at integrere’, det høje antal somaliske foreninger i Danmark, samt forestillingen om foreninger som et særligt dansk og ’integrerende’ fænomen. På den anden side indgår mange somali-danskere i transnationale relationer og ser sig selv som en del af en større somalisk ’diaspora’. Dette spændingsfelt sætter fokus på spørgsmål om inklusion, anerkendelse og positionering i flere forskellige sociale rum, og disse spørgsmål udforskes i afhandlingen. Det overordnede forskningsspørgsmål er suppleret med tre supplerende spørgsmål, der angår, hvorfor og hvordan somaliere i Danmark engagerer sig i foreninger, hvilken betydning de somaliske og danske kontekster har for sådan- ne engagementer, og endelig hvordan transnationalt engagement og diasporisk mobilisering relaterer sig til inklusionsprocesser i det danske samfund. Afhand- lingen er baseret på etnografisk feltarbejde blandt centrale somaliske forenings- personer i København, aktive i foreninger med særligt fokus på integration i Danmark og genopbygning i Somaliland og Somalia. Dette feltarbejde er yder- ligere suppleret med kortere ture til Somaliland og London, og i alt 50 informanter danner grundlag for studiet. Afhandlingen er inddelt i tre overordnede dele, struktureret som en teoretisk og analyse rejse med opbrud, grænsekrydsning og horisonter. I den første del positionerer og situerer jeg afhandlingen teoretisk og metodisk. I kapitel 1 in- troduceres studiet og forskningsspørgsmålene. I kapitel 2 positionerer jeg afhandlingen meta-teoretisk og præsenterer tre teoretiske kortlægninger. I den første kortlægning samles teorier om transnationale praksisser, inkorporation og migrantorganisationer, såvel som begreber om felt og socialt rum. I den anden præsenteres diasporateori, særlig inspireret af nyere forskning om diaspora som claim og identitetskategori. Endelig, i den tredje kortlægning, kombineres aner- kendelsesteorier med feministiske teorier om positionering og kulturel genkendelighed. I kapitel 3 ’oversætter’ jeg disse teoretiske tilgange til metodologiske

313 retningslinjer, der guider dataproduktionen og den analytiske proces, såvel som jeg præsenterer feltarbejdet og informanterne. Afhandlingens anden del består af analyser, baseret på det empiriske materia- le. I kapitel 4 præsenterer jeg den somaliske kontekst, med særlig fokus på mi- grationshistorie og borgerkrigen. Udover at somaliere nu er spredt over hele verden, har borgerkrigen betydet disintegration af den somaliske republik og konsolideringen af en transnational økonomi, hvor somaliere i Vesten og Gol- fen sender penge til deres familier på Afrikas Horn. I kapitel 5 analyserer jeg modtagelsen i det danske samfund. Siden midten af 1990’erne, hvor det største antal somaliske asylansøgere kom til Danmark, er somalierne blevet portrætte- ret som en gruppe, der er særlig vanskelig at integrere. Denne overordnede modtagelse danner baggrund for de individuelle møder med det danske samfund. Hvor nogle af de somalisk-danske informanter i undersøgelsen giver ud- tryk for, at de ofte mødes med stereotype forestillinger om somaliske flygtninge og er udsat for diskrimination, fremhæver andre, at diskrimination på det per- sonlige plan er en undtagelse. Ligeledes er ændringer i køns- og familierelatio- ner meget debatterede blandt herboende somaliere, og somaliske mænds situation er italesat som særlig vanskelig. Jeg foreslår, at somaliernes situation kan analyses med inspiration fra sociologiske teorier om ’fremmedhed’, som det omgivende samfunds måde at stabilisere flertydige identiteter og tilhørsforhold. Kapitel 6 handler om somalisk foreningsengagement som et felt - i Bour- dieu’s forståelse -, kendetegnet af positioneringer og konflikter. Jeg opridser de somaliske foreningers historie i Danmark og analyserer de underliggende orga- nisationsprincipper: Køn, klan, politisk orientering i forhold til Somalia og So- maliland, og Islam. Derefter fortsætter jeg med et casestudie af en kvindefor- ening, hvor jeg fokuserer på hverdagsudfordringer blandt almindelige medlemmer, der følger frivillige danskkurser. Social kapital og lang-distance natio- nalisme blandt medlemmer og foreningsledere analyseres også. Endelig bredes blikket ud til somaliske foreninger generelt med fokus på læreprocesser og til- egnelsen af ny politisk kultur. Jeg argumenterer for, at foreninger giver muligheder for økonomisk støtte og for at etablere tale- og handlepositioner for foreningsledere og talspersoner. Foreningsengagement kan således være en måde at deltage i det danske samfund på, selvom foreningsledere ofte er udsat for misforståelser og konflikter i relation til deres engagement. Analysen af foreningsarbejde fortsættes i kapitel 7. Her analyserer jeg to casestudier af transna-

314 tionalt engagement: En stiftende konference for en somalisk udviklingsorgani- sation (DOEA) og en transnational paraplyforening (Somscan & UK). Både DOEA og Somscan & UK har udvikling og genopbygning som mål, og begge aktører ønsker at udfordre klanlogikken i forhold til identiteter og loyaliteter. Ligeledes beskriver de begge deres engagement som en del af den somaliske ’diaspora’. Der er udtalte forhåbninger om bidrag til udvikling af Somaliland og Somalia fra ’diasporaen’, som forventes at udnytte mulighederne i Vesten i form af uddannelse, økonomiske ressourcer og læreprocesser fra demokratiske styreformer. I afhandlingens tredje del samler jeg de teoretiske og analytiske tråde i en diskussion og konklusion. I kapitel 8 argumenterer jeg for, at foreningsengagement skaber sociale rum, hvor anerkendelse og kulturel genkendelse kan opnås, hvad jeg kalder for spaces of recognition. På denne måde kan foreningsengagement bruges som en måde at forhandle og (gen)etablere sociale positioner og anerkendelse, styret af en praktisk ’sans for spillet’ snarere end overlagte strate- gier for at opnå prestige. Anerkendelse finder dog primært sted i den somaliske gruppe – både i Danmark og transnationalt – hvor det somalisk-danske for- eningsfelt er del af et større og transnationalt socialt felt. Endelig analyserer jeg, hvordan ’diasporaen’ italesættes og mobiliseres. ’Diasporaen’ henviser til et moralsk og transnationalt ressource- og handlingsfællesskab, der italesættes og mobiliseres på forskellige niveauer og af forskellige aktører. Diasporisk identi- fikation og mobilisering er dermed også en måde at kæmpe for anerkendelse og forhandle positioner på. I kapitel 9 konkluderes afhandlingen. For det første konkluderer jeg, at foreningsengagement er en måde at imødegå forpligtelser og udnytte muligheder på, såvel som at opnå kønnet kulturel genkendelighed. Foreningsengagement kan derved være et middel for at forhandle og (re)etablere sociale positioner og agtelse på det gruppespecifikke niveau. For det andet slutter jeg, at foreningsengagement er en deltagelses- og selvrealiseringsstrategi i form af muligheden for at bruge kvalifikationer og kompetencer, hvilket ellers kan være svært i det danske samfund. Foreningsengagement kan derved være en måde at opnå anerkendelse på. For det tredje konkluderer jeg, at der ikke er noget modsætnings- forhold mellem inklusionsprocesser i Danmark og transnationale praksisser i forhold foreninger på det institutionelle niveau. Tværtimod er de mest transna-

315 tionalt engagerede foreningsledere også blandt de mest veletablerede i det danske samfund.

316