Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Bleu blanc noir : assimilation trajectories, identity dynamics and boundary work of French Antilleans,… Ivemark, Biorn 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2017_november_ivemark_biorn.pdf [ 3.88MB ]
Metadata
JSON: 24-1.0355272.json
JSON-LD: 24-1.0355272-ld.json
RDF/XML (Pretty): 24-1.0355272-rdf.xml
RDF/JSON: 24-1.0355272-rdf.json
Turtle: 24-1.0355272-turtle.txt
N-Triples: 24-1.0355272-rdf-ntriples.txt
Original Record: 24-1.0355272-source.json
Full Text
24-1.0355272-fulltext.txt
Citation
24-1.0355272.ris

Full Text

 Assimilation Trajectories, Identity Dynamics, and Boundary Work of French Antilleans, West Africans, and their Children in Paris  by  Biorn Ivemark          A DISSERTATION SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  The Faculty of Graduate and Postdoctoral Studies  (SOCIOLOGY)          THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2017   © Biorn Ivemark, 2017 ii Abstract    This study compares the assimilation trajectories, identity dynamics and boundary work of French Antilleans, West Africans and their descendants in the Paris region. While previous studies have focused on the experiences of French Antilleans and sub-Saharan Africans separately or those of Blacks in France as a whole, this study engages in a more minute comparison of the experiences of West African immigrants and French Antilleans across two generations in mainland France. This comparison primarily aims to determine the role of the divergent civic, cultural and religious backgrounds of these groups alongside their largely shared racial characteristics in how they assimilate to French society across two generations. These variables are of particular interest given the salience of civic and cultural distinctions in France, while racial distinctions are notoriously downplayed. The main theoretical goal of the study is to assess the usefulness of segmented assimilation theory in accounting for the various assimilation outcomes of these groups. Drawing on 55 in-depth interviews complemented with wide-ranging statistical data, I explore the impact of cultural, religious and racial factors on the intergenerational educational and professional trajectories of both populations, analyze how these factors influence their identification patterns and assess how members of these groups seek to negotiate the various symbolic boundaries that they come up against, both in their relations to each other and to the majority population. The results suggest that French Antilleans have more favourable educational and professional outcomes than West Africans. Despite the importance of racial barriers for both groups, the findings also underscore the salience of cultural and religious forces as well as the identification dynamics and boundary work that both groups engage in. While some segmented assimilation mechanisms remain valid in the French case, the study also demonstrates the importance of empirically identifying societally specific assimilation barriers and cultural segments for the theory to retain its usefulness in other national contexts. iii Lay Summary    While previous studies have focused on French Antilleans, sub-Saharan Africans or Blacks in France in general, this study seeks to compare more specifically the challenges faced by West Africans, French Antilleans and their children born in Paris in becoming part of French society. A key goal is to assess the role that the distinct civic, cultural and religious backgrounds of these populations play compared to their often similar racial characteristics. This is particularly interesting to examine in France, given that civic and cultural distinctions are often put forward in daily life while racial distinctions are routinely downplayed. Drawing on 55 in-depth interviews and wide-ranging statistical data, I compare the educational and professional attainment of both migrant groups and their children. I then analyze how their differences and similarities affect their identities and the way they interact with each other and the broader society.       iv Preface    This dissertation is the original, unpublished, independent work by the author. The qualitative fieldwork of the study was approved by the UBC Ethics Board Certificate Number H11-01322.   v Table of Contents  Abstract ...................................................................................................................................................................... ii Lay Summary ............................................................................................................................................................. iii Preface ...................................................................................................................................................................... iv Table of Contents ....................................................................................................................................................... v List of Tables ............................................................................................................................................................ viii List of Figures ............................................................................................................................................................. ix Acknowledgments ..................................................................................................................................................... x 1. INTRODUCTION ..................................................................................................................................................... 1 I. Background and Rationale ................................................................................................................................. 3 Republican Universalism, Colour-Blindness, and Secularism ............................................................................................... 3 West Africans and French Antilleans ................................................................................................................................... 5 Intergenerational Assimilation Patterns .............................................................................................................................. 7 Black Identities and Boundaries ........................................................................................................................................... 8 II. Research Questions and Outline ...................................................................................................................... 10 Research Questions ........................................................................................................................................................... 10 Outline of the Study ........................................................................................................................................................... 11 PART ONE FRAMING THE STUDY....................................................................................................................... 13 2. HISTORICAL BACKGROUND ................................................................................................................................. 14 I. Historical Perspective on Immigration and Assimilation in France................................................................... 14 Immigration: From Europeans to Non-Europeans ............................................................................................................. 14 Assimilation: From 'Our Ancestors, the Gauls' to Identity Politics ..................................................................................... 16 Scholarship: From the Social to the Racial Problem? ......................................................................................................... 18 II. Colonial Expansion and the Creation of an African Diaspora in the Americas ................................................ 19 From Plantation Colonies to Overseas Departments ......................................................................................................... 20 West Africa before European Contact ............................................................................................................................... 23 French Penetration of West Africa ..................................................................................................................................... 23 After Independence: Labour Migration and Family Reunification ..................................................................................... 26 3. THEORETICAL FRAMEWORK ................................................................................................................................ 28 I. Immigrant Incorporation .................................................................................................................................. 28 From Classical to Segmented Assimilation Theory ............................................................................................................. 28 New Immigrants, New Circumstances ............................................................................................................................... 30 Segmenting Forces ............................................................................................................................................................. 31 Criticisms and Limitations .................................................................................................................................................. 32 Segmented Assimilation in France? ................................................................................................................................... 34 An Integrated Theoretical Perspective ............................................................................................................................... 35 II. Racial and Ethnic Identities and Boundaries.................................................................................................... 37 Conceptualizing Race and Ethnicity ................................................................................................................................... 37 Ethnic and Racial Identities ................................................................................................................................................ 39 Psychological Perspectives ................................................................................................................................................. 40 Racial and Ethnic Boundaries ............................................................................................................................................. 42 vi 4. METHODS AND DATA .......................................................................................................................................... 45 I. Presentation of the Data .................................................................................................................................. 45 Recruitment and Sampling Procedure ............................................................................................................................... 45 Data Collection and Analysis .............................................................................................................................................. 47 Ethical Considerations ........................................................................................................................................................ 50 Complementary Data Sources ........................................................................................................................................... 51 Demographic Characteristics of the Sample ...................................................................................................................... 53 II. Challenges and Limitations .............................................................................................................................. 55 Positionality and Fieldwork Challenges .............................................................................................................................. 55 Limitations of the Study ..................................................................................................................................................... 60 PART TWO ASSIMILATION TRAJECTORIES ...................................................................................................... 63 5. THE FRENCH CONTEXT OF RECEPTION ................................................................................................................ 64 I. Economic and Social Transformations .............................................................................................................. 64 Economic Restructuring and Social Mobility ...................................................................................................................... 64 Urban Decay and Social Marginality .................................................................................................................................. 65 Territorial Divides and Segregation .................................................................................................................................... 68 II. Discrimination and Public Attitudes ................................................................................................................ 72 Discrimination and Stigma ................................................................................................................................................. 72 Public Opinion .................................................................................................................................................................... 73 6. MIGRATION AND SETTLEMENT ........................................................................................................................... 76 I. Backgrounds and Motivations .......................................................................................................................... 76 African Backgrounds and Motivations ............................................................................................................................... 76 Antillean Backgrounds and Motivations ............................................................................................................................ 79 II. Settlement and Ethnic Communities................................................................................................................ 81 Settlement Experiences ..................................................................................................................................................... 81 Ethnic Communities and Solidarity .................................................................................................................................... 83 7. RAISING THE SECOND GENERATION ................................................................................................................... 88 I. Family Mechanisms .......................................................................................................................................... 89 Hypertrophied Father Figures ............................................................................................................................................ 89 Destructuring of Sahelian Family Systems ......................................................................................................................... 90 The Case of Sahelian Single Mothers ................................................................................................................................. 95 A Matrifocal Caribbean Model ........................................................................................................................................... 97 II. Neighbourhood Effects and Acculturation ..................................................................................................... 101 Neighbourhoods with Concentrated Social Pathologies .................................................................................................. 102 Sahelian Responses to Deviance ...................................................................................................................................... 106 Dissonant Sahelian Acculturation Challenges .................................................................................................................. 108 Consonant Antillean Acculturation Challenges ................................................................................................................ 113 III. Concluding Remarks ..................................................................................................................................... 116 8. SCHOOL, WORK, AND MARRIAGE ..................................................................................................................... 118 I. The Second Generation in School and in the Labour Market .......................................................................... 118 Divergent School Trajectories .......................................................................................................................................... 119 Schooling Experiences ...................................................................................................................................................... 120 Labour Market Inequalities .............................................................................................................................................. 125 vii Experiences in the Labour Market ................................................................................................................................... 127 II. Partners, Marital Options and Aspirations .................................................................................................... 134 African Partner Choice: From Kin to Common Culture .................................................................................................... 134 Antillean Partner Choice: Widened Exogamy .................................................................................................................. 140 III. Concluding Remarks ..................................................................................................................................... 142 PART THREE IDENTITY DYNAMICS AND GROUP BOUNDARIES ................................................................... 144 9. IDENTITY DYNAMICS ......................................................................................................................................... 145 I. Intergenerational Identification Patterns ....................................................................................................... 148 African Ethno-Cultural Identities...................................................................................................................................... 148 Antillean 'Double Culture' ................................................................................................................................................ 152 Becoming French Africans................................................................................................................................................ 154 Becoming 'Negropolitans' ................................................................................................................................................ 156 II. Second-Generation Identities ........................................................................................................................ 158 Eurocentric Identity ......................................................................................................................................................... 159 Afrocentric Identity .......................................................................................................................................................... 163 Creolized Identity ............................................................................................................................................................. 169 Conflicted Identity ........................................................................................................................................................... 174 III. Concluding remarks ...................................................................................................................................... 178 10. GROUP BOUNDARIES ...................................................................................................................................... 182 I. Mainstream Boundaries ................................................................................................................................. 182 Framing Antilleans and Africans ...................................................................................................................................... 183 Mainstream Boundary Work ............................................................................................................................................ 185 II. Antillean Boundaries...................................................................................................................................... 190 Mainstream Boundary Maintenance ............................................................................................................................... 190 Intergenerational Differences .......................................................................................................................................... 195 Brokering Mainstream Boundaries .................................................................................................................................. 197 III. African Brokering Patterns ........................................................................................................................... 200 Brokering Mainstream Boundaries .................................................................................................................................. 200 Brokering Antillean Boundaries ....................................................................................................................................... 202 IV. Concluding Remarks ..................................................................................................................................... 205 11. CONCLUSION ................................................................................................................................................... 208 I. Assimilation Patterns ...................................................................................................................................... 208 Summary of Assimilation Patterns ................................................................................................................................... 208 Summary of Underlying Causes ....................................................................................................................................... 210 II. Theoretical Considerations ............................................................................................................................ 214 Segmented Assimilation Theory in France ....................................................................................................................... 214 Racial and Religious Segments and Future Research ....................................................................................................... 217 Bibliography ........................................................................................................................................................... 221 Appendix 1. Interview Guide ................................................................................................................................. 232 Appendix 2. Consent Form .................................................................................................................................... 240    viii List of Tables    Table 1. Distribution of respondents by region of origin and generation ........................................................ 53 Table 2. Demographic characteristics of first generation Antilleans and Africans ......................................... 54 Table 3. Demographic characteristics of second generation Antilleans and Africans ................................... 55 Table 4. Religiosity of Sahelian and DOM migrants across two generations ................................................ 147 Table 5. Average identity chart ratings of first and second generation Antilleans ....................................... 148      ix List of Figures    Figure 1. Potential intergenerational assimilation and identification outcomes .............................................. 11 Figure 2. French possessions from the first and second colonial phases ......................................................... 19 Figure 3. Martinique and Guadeloupe in the contemporary Caribbean........................................................... 20 Figure 4. Contemporary West Africa and the Sahel region ............................................................................... 25 Figure 5. Geographical distribution and concentration of French-born and immigrant managerial/professional and worker households in the Paris Region in 2006 ............................... 69 Figure 6. Proportion of children who live in an immigrant household in the Paris Region ......................... 70 Figure 7. Identity spectrum of second generation Africans and Antilleans ................................................... 159 Figure 8. Identification patterns of second generation Antilleans and Africans ........................................... 179 Figure 9. Percentage of migrants and their children who have had an experience of discrimination in the past 5 years .................................................................... 187 Figure 10. Summary of African and Antillean assimilation patterns in the second generation ................. 213        x Acknowledgments    First and foremost, I wish to extend my gratitude to Wendy Roth for her skillful supervision of this research project and for guiding me along the sinuous path to its completion—her guidance, insight and support were invaluable. My deep appreciation also goes to my dissertation committee members, Gillian Creese and Daniyal Zuberi for accompanying me reliably throughout this process.  I also wish to thank Patrick Simon at the Institut National d’Etudes Démagraphiques for generously welcoming me as a visiting student at the institute in Paris.  I am deeply grateful for the generous financial sponsorship of the Council of European Studies, the Horowitz Foundation, and the Helge Ax:son Johnsons Stiftelse which greatly facilitated the undertaking of this study. The various teaching and research assistantships I had the opportunity to undertake for Neil Gross, Wendy Roth and Daniyal Zuberi were also very helpful in this regard as well as in the valuable experience they imparted.  I am much obliged to Nancy Foner for her discerning feedback on the manuscript, as well as to Dan Hiebert and Renisa Mawani for their helpful comments and observations.  Last but not least, I want to show my appreciation to the numerous family members and friends—too numerous to list here—who encouraged and supported me in various ways throughout these years. I particularly want to mention Anne-Dominique however—thank you for your encouragement, your patience, and for believing in me in my moments of discouragement and doubt.          1 1. INTRODUCTION   On May 10, 2008, the French Republic officially observed the 160th anniversary of the abolition of slavery.1 In Paris, the Conseil représentatif des associations noires (CRAN)2 organized a “march of freedoms” set to depart from Place de la République and head to Place de la Bastille, not only to commemorate the abolition of slavery, but also to denounce persistent problems with prejudice and discrimination. Another march organized by the Collectif des fils et filles d’Africains déportés (Coffad)3 was planned to depart from the same location at the same time, but was heading instead towards Place de la Nation (Nouvel Obs 2008).  Lucien,4 a retired French Antillean woodworker who had simply come to participate in the commemoration, remembers his reaction when he arrived at the square and found out that these marches would not only be heading in different directions, but apparently also involved different populations: If I had known I wouldn’t have gone. I saw one demonstration for Africans, and one for Antilleans. One went to Bastille and the other went to Nation… When I saw this I said to myself “we’re really buffoons”. If there’s a demonstration it should be for everybody. You shouldn’t have a separate one for Antilleans and a separate one for Africans. This vignette neatly encapsulates many of the key themes explored in this dissertation. First, these separate demonstrations highlight the downplaying of racial allegiances in the French conception of nationhood, despite the fact that experiences of discrimination and racial prejudice seem widespread in the country. Despite Africans’ and Antilleans’ shared experience of colonial rule, this incident is also a reminder of the very different legacies that the French imperial enterprise has left in its former colonies. Moreover, it hints                                                           1 Introduced by president Jacques Chirac in 2006 to observe the “memories of the slave trade, slavery and their abolition” – the specific date was chosen in reference to a law adopted on May 10th, 2001 identifying the transatlantic slave trade and slavery as crimes against humanity. 2 The Representative Council of Black Associations, a prominent black advocacy group created in 2005 which primarily struggles against discrimination, but has also made reparations for slavery and colonialism a part of its agenda. The organization officially represents the Rainbow Push Coalition, founded by Jesse Jackson.  3 The Committee of the Children of Deported Africans primarily works for the rehabilitation of the memory of the Africans deported during the transatlantic slave trade, as well as for moral and material compensations for deportation and slavery.  4 The respondents’ names have been modified to protect their anonymity. 2 at the distinct social positioning of two populations with different degrees of embeddedness in the national fabric, and is suggestive of the complex relationship that a shared racial background inevitably institutes between them both, as well as between them and the racially distinct mainstream population. Furthermore, through the prism of Lucien—put in a position where he fundamentally has to pick a side—it reflects not only the salient social boundaries between these two populations but also a strong desire to overcome them. Finally, these divergent marches also raise the question whether these two populations will themselves be heading in different directions over time, as their respective legacies would suggest, or if on the contrary they will converge around common experiences and aspirations as a result of their shared racial status. We will explore these themes by studying the assimilation trajectories, identity dynamics and boundary drawing strategies displayed by French Antilleans, West Africans and their descendants in the Parisian metropolitan area. Unlike previous studies—which have either focused on sub-Saharan immigrants, French Antilleans, or on the experiences of Blacks in France in general through a variety of analytical lenses—this dissertation intentionally compares the assimilation patterns of Black Antilleans and West Africans across two generations in the Paris region. While Antilleans from Martinique and Guadeloupe are French citizens and are linguistically, culturally and religiously similar to the majority population, West African immigrants from Mali and Senegal diverge more sharply from the French mainstream in these respects and are frequently stigmatized for their religious and cultural practices. However, despite greater racial mixing in the Antilles, both groups have a substantial overlap in their racial features. Since the French approach to immigrant incorporation tends to emphasize civic and cultural distinctions but refuses to formally acknowledge any racial differences between French citizens, comparing the assimilation patterns of these populations and generations will help to empirically determine if the sharp cultural and religious differences between them actually overshadow their racial commonalities, or if race actually remains the 'master status' that shapes their social experiences and identities. 3 Drawing on 55 in-depth interviews and descriptive statistical data, I first explore how these civic, cultural, religious and racial characteristics affect the intergenerational educational and professional experiences and trajectories of these populations. I then analyze how they influence their identification patterns, and assess how they seek to negotiate the various symbolic boundaries they come up against, both in their relations to each other and in French society as a whole.   I. Background and Rationale Republican Universalism, Colour-Blindness, and Secularism The French Republic is notorious for the abstract universalism that it inherited from the French Revolution, which founded its political model upon the equality of individual citizens before the law regardless of their religious, ethnic, racial or other group characteristics. This legacy has largely contributed to the elimination of religious matters from the sphere of public life, the refusal to recognize any ethnic, racial or religious distinctions among French citizens, and the forging of a national culture which its members are largely encouraged to perceive as colour-blind and universalistic. A central component of the French revolutionary ideals was what has become known as the principle of laïcité, i.e. the stringent form of secularism that has gradually eroded the influence of the Catholic Church on public life since the Revolution, reflected for instance in the elimination of religious instruction in the late 19th century and the separation of Church and State in 1905.5 As a large Muslim population has developed through recent immigration waves, the principle of laïcité has again come to the forefront of public debate, and is routinely invoked to regulate various expressions of Islamic religious life                                                           5 Rather than springing from a pragmatic desire to federate an existing religious pluralism, which for instance animated the founders of the United States, French secularism was instead shaped through a more fervent power struggle between revolutionaries inspired by enlightenment ideas, and the established Catholic Church. In the revolutionary era and its aftermath, Church property was seized, recalcitrant parts of the clergy were executed, the non-religious functions of the Church were taken over by the state, and several substitutes for the Christian religion, which would put more emphasis on philosophy and reason, were avidly experimented with. The fervent roots of this secularism explain why the state not only proscribes the funding of any religious activities, but that it also plays a more active role in emphasizing the need for expunging any presence of religion in public life. 4 in public space, the most notorious example being the controversy surrounding the headscarf (Bowen 2007; Laurence and Vaisse 2006). This antagonistic stance towards ethnic, racial and religious divides within the French polity has also favoured the promotion of a staunch form of assimilation that has successfully incorporated several large waves of primarily European immigrants by encouraging naturalization, acculturation and loss of ethnic distinctiveness (Hargreaves 2007).6  Although this ‘Republican model’ of immigrant assimilation has undergone substantial changes over past decades—as greater latitude has been given to the preservation of cultural specificities and racial discrimination has gained greater recognition—a substantial majority of the French population still remains attached to this model. Its defenders see in it the only means of safeguarding a cohesive society of equal individuals relatively free from the tribalism of ethnic politics, which many associate with an undesirable ‘Anglo-Saxon’ societal model. Its critics on the other hand see in this “French phraseology of universalism” the perpetuation of ethno-racial inequalities and discrimination as this universalism “serves before anything else as a shelter for an ethno-racial nationalism which refuses to say its name.” (Mbembe 2005: 152).7 My goal is not as much to take sides in these debates as to examine empirically the role that this republican model of integration may play in forging a sense of shared civic identity among the descendants of these populations or in contesting the legitimacy of social identities which they may not be able to disregard given the social barriers they give rise to in their daily lives. Given the deep historical roots that this republican approach to assimilation rests upon, it has created substantial practical, cultural and intellectual obstacles for scholarship dealing with race and ethnicity. Debates over discrimination in the past decade and a half have nevertheless slightly eroded the                                                           6 In fact, the French model of secularism and of assimilation are intimately linked and historically intertwined, as reflected for instance in the policies adopted towards the Jewish minority after the Revolution. While the revolutionary ideals had granted Jews equal rights as individuals it also required them to eschew their group affiliation to a large extent by relegating their religious activity and identity to the private sphere. Laïcité thus on the one hand opened a “neutral space” where Jews and Protestants were put on an equal footing with the Catholic majority, but on the precondition that their religious identity would not come to interfere with their adherence to the universalistic precepts that underpinned their French citizenship, which in practice required substantial cultural assimilation (see e.g. Hyman 1998). 7 All quotes from French language sources are my own translations.  5 hegemony of this paradigm (Simon and Sala Pala 2009) and a few works focusing on France's Black population have appeared in recent years, but have typically taken the form of essays (e.g. Durpaire 2006; Kouvibidila 2007; Zika 2008) rather than rigorous empirical analyses. Although important efforts have recently been made to encourage and develop research on this population (see Blanchard et al. 2011; Ndiaye 2008), no empirical studies to date have to my knowledge analyzed the assimilation patterns and identity dynamics that unite and divide the Black population in France and attempt to understand what these dynamics hinge upon, and what influence, if any, the ‘republican model’ may have on them. Nor have any studies to my knowledge engaged in any serious comparative analyses with the French Antillean population to determine the specificity of their experiences as quasi-immigrants and as racial minorities in mainland France.  West Africans and French Antilleans Although sub-Saharan Africans have been present in small numbers in France since the colonial era, their numbers only began substantially growing after the mid-1970s, primarily as a result of changes in migration policies. About 1.3 million first and second-generation sub-Saharans were estimated to reside in France in 2011, which represents around 11% of the country's immigrant population across two generations (Tribalat 2015).8 They primarily concentrate in the Paris region, where 61% of sub Saharan immigrants and 65% of their descendants reside (Beauchemin, Lhommeau and Simon 2016), and primarily hail from former French colonies in sub-Saharan Africa. The two largest sub-Saharan African populations in France have their roots in Senegal and Mali in West Africa.9 These early African migrants started coming to France in the 1960s to fill low-skilled industrial jobs in the Paris region. They are primarily of rural origin and have limited French                                                           8 This is by far the fastest growing population in France, and is likely to represent an increasingly large proportion of the immigrant population over time. According to the calculations of Michèle Tribalat, the sub-Saharan population in France increased by 90% between 1999 and 2011 across two generations (Tribalat 2015), and by an additional 43% between 2011 and 2015, which would actually bring its size up to 1.85 million in 2015 (Tribalat 2017) 9 Other major sending countries include Congo, the Ivory Coast, Cameroon and the Democratic Republic of Congo (Lessault and Beauchemin 2009).  6 language skills and low educational qualifications, and are also predominantly from Muslim backgrounds (Moguérou, Brinbaum and Primon 2016; Timera and Garnier 2010).10  By contrast, the Antillean islands of Martinique and Guadeloupe are part of a handful of French overseas departments and territories (DOM-TOM) that are an integral part of the French Republic,11 and have been part of France for longer than some parts of the country's mainland. Antilleans have been French citizens since the definitive abolition of slavery in 1848, and have since had full access to metropolitan France,12 where they were estimated to constitute a population of 372 900 with their descendants in 2011 (Marie 2014). Antillean migration to France was initially sparse and primarily hailed from the upper-middle class, but economic difficulties in the islands led to the creation of a specific migration policy in the 1960s that sought to attract low-skilled Antillean workers to fill labour shortages (Anselin 1990).  Antillean migrants have a higher labour force participation rate compared to other migrants, largely because of their ability to access public service sector jobs from which foreign nationals are barred. But just like other immigrants, Antilleans who arrived during this period have lower qualifications on average compared to the majority population, and primarily live in low-income housing projects with high concentrations of immigrants in the greater Paris region (Hargreaves 2007).13 As deindustrialization and mass unemployment have particularly affected these areas over the past few decades, this raises the question of the assimilation trajectories that their children are likely to follow. The job security that their parents enjoyed in the public sector is not available to them to the same degree, and they report experiencing comparable levels of racial discrimination as their sub-Saharan African peers (Beauchemin et                                                           10 This sets them apart from the more recent waves of immigrants from the African Guinean Gulf, who are predominantly Christian and came to France more recently with higher educational credentials and more favourable socio-economic backgrounds. Despite focusing solely on Malian and Senegalese immigrants and their children in this study, we will for the sake of convenience sometimes simply refer to them as “Africans” despite the fact that very different social and cultural dynamics might be at work among other parts of the sub-Saharan African population.   11 In addition to Martinique and Guadeloupe, the départements d'outre-mer (DOM) also include French Guyana, la Réunion and much more recently, Mayotte (since March, 31 2011). As all other departments in France, they are represented in the French National Assembly and the French Senate, and are largely governed by the same laws and regulations as the mainland. They are also part of the European Union, and use the Euro as their currency. The territoires d’outre-mer (TOM) comprise a number of islands that are less politically and institutionally integrated to the mainland.  12 Metropolitan France is the part of the country that is geographically located in Europe, i.e. the mainland and Corsica. Antilleans typically refer to it as “the metropole”, and particularly its white population as “metropolitans”. 13 They also concentrate highly in the Paris Region where three-quarters of French Antilleans are estimated to live (Giraud 2004) 7 al. 2010b), alongside whom they often grow up. Michel Giraud has warned that the Antillean population in France “has been slipping towards joining the most underprivileged sections of French society” (2004: 627), and similar concerns can certainly be raised regarding many descendants of sub-Saharan African immigrants in France, where school drop-out rates, unemployment rates, and crime rates among young men are higher than other children of immigrants (Brinbaum, Meurs and Primon 2016; Brinbaum, Moguérou and Primon 2016; Lagrange 2010).   Intergenerational Assimilation Patterns In recent years, sociologists of immigration have developed valuable theoretical accounts of the various paths along which immigrants are incorporated into their host societies. The segmented assimilation model makes the paths of upward or downward socioeconomic mobility contingent upon immigrants' acculturation patterns and on their position in a segmented labour market and a racialized social structure. Those trapped in the lower and informal sectors of the labour market, especially when racially distinct from the majority population, are at risk of experiencing a downward assimilation into an 'underclass' without prospects for upward mobility. Segmented assimilation theorists suggest that it is through limited acculturation and retention of ties to an ethnic community that racial minorities can avert this process of downward assimilation and increase their chances for upward mobility (Portes & Rumbaut 2001). This theory has been very influential on immigration research in the United States, but has still been limited in its application elsewhere despite the many empirical and theoretical benefits that could derive from such efforts. Studying divergent assimilation patterns in France has proved problematic as no racial or ethnic statistics are collected in the country. The few existing quantitative studies on segmented assimilation in France therefore remain tentative at best (Safi 2006; Silberman, Alba and Fournier 2007), but all invariably conclude that North and sub-Saharan African-origin youth are at risk of downward assimilation. These studies have raised several unanswered questions regarding the specific social mechanisms that may cause 8 this downward assimilation in France, which are thought to derive less from race and more from the experience of colonialism and/or more salient religious and cultural boundaries (Silberman and Fournier 2008). Efforts to identify the mechanisms that give rise to these outcomes however calls for more process-oriented qualitative work, which can subsequently inform more robust quantitative studies on the topic.14 This study will not solely focus on downward trajectories, as respondents have been recruited from a variety of social backgrounds. However, since it aims to identify the key mechanisms behind the assimilation trajectories of culturally diverse racial minorities in France, it provides an opportunity to empirically examine the salience of racial mechanisms alongside the supposedly more potent cultural and religious ones in determining the trajectories they follow.  Black Identities and Boundaries According to Michèle Lamont, the peculiar secular and civic conception of nationhood in France has created a sharp boundary between citizens who adhere to France's universalistic culture and foreigners who do not, leading to a stronger distinction between French citizens and immigrants. On the other hand, given the nation’s refusal to make racial distinctions, boundaries that would highlight Blacks as a group have simultaneously been delegitimized, especially as many Blacks initially hailed from the French Caribbean. However, with increased sub-Saharan African immigration, Lamont argues that this might be changing: This pattern of weaker boundaries toward blacks and stronger boundaries against immigrants (as compared with the United States) may be in the process of changing, as sub-Saharan African immigration grows… We may also be witnessing an accelerated process of the “blackening” of immigration, as more non-Caucasian immigrants come to France… Moreover, black Africans are joining North Africans at the bottom of the social ladder. Consequently, in the future, French definitions of social membership may come to be associated more explicitly with skin colour. It remains to be seen whether the republican myth and the presence of black French citizens from the Dom-Toms will remain powerful enough to trump the association between blackness and outsider status. (Lamont 2004: 151) Antilleans have been French for many generations and share a strong civic, cultural and religious common ground with other French people, whereas West African immigrants are commonly perceived as culturally                                                           14 Alex Stepick and Carol Dutton Stepick indeed argue that "it is the dialectic between such qualitative studies and the more representative, larger survey samples that will further refine the concept of segmented assimilation and contribute to its value as a typological framework" (2010: 1162) 9 'other' and stigmatized for cultural practices such as polygamy and female genital mutilation (Fall 2006; Poiret 1996). Antilleans are therefore often claimed to be perceived more favourably than sub-Saharan African immigrants (Gastaut 2000), but their phenotypical proximity to each other, and racial divergence from the majority population, raises the question whether this actually corresponds to their social experience. Comparing two generations of Antilleans and West Africans will allow us to empirically examine the extent to which cultural and religious differences actually eclipse race, or if race actually remains one of the prime factors that shape the social experiences and identities of these populations. Previous studies have shown how these two groups have crafted distinct African and Antillean identities in France (Beriss 2004; Poiret 1996). But as Michel Giraud has noted, this is less and less the case for many later-generation Antilleans: While a large proportion of the Antillese immigrant population in France continues to assert its identity by remaining loyal to its own culture, another large and growing fraction (among the young especially) is recasting its system of representation and values at the cost of a crisis in its symbolic framework. Individual groups which no longer fully identify politically or culturally with the migration source countries are beginning to appear. Yet, as minorities, they are still left on the edges of the host society. They increasingly regard this society as their own, but would like to join it on their own terms... France’s young Antillese are thus becoming ‘blacks’ in their own and in other people’s eyes (2004: 632). In a largely assimilationist country where racial discrimination seems widespread and where ethnic identities have little legitimacy beyond the first generation, the children of these two groups may in fact converge into thinking of themselves as a stigmatized racial minority despite the substantially different degree of embeddedness both populations had in the national community in the first generation.  However, given that Antilleans tend to occupy a distinct group position in France, and given the more substantial cultural and religious gap that the sub-Saharan African descendants may have to bridge—especially in light of the strict secularist ideal in French society which may be perceived as especially ill-disposed towards their religion—it is also possible that these expectations of racial convergence may be exaggerated and that distinctions between the Antillean and West African population are maintained to a large extent in the second generation.  10 II. Research Questions and Outline Research Questions In light of the discussion above, the following research questions will guide the study: (1) What social, cultural and racial factors have contributed to shaping the educational and occupational trajectories of second-generation West Africans and Antilleans, and what particular obstacles have they faced in this regard?  (2) In what terms do West African and Antillean immigrants and their children conceive of their racial, ethnic, national and religious identities, and what factors primarily underpin these identifications? (3) What boundaries are these two populations confronted with in their relations to each other and to the broader society, and to what extent do they reproduce or challenge them? Figure 1 provides a schematic representation of the most likely intergenerational assimilation outcomes for both groups, whether expressed in their socio-economic mobility, their social identities or their group relations. One possible pattern is that they both substantially converge towards the French mainstream leading to a greater blurring of group boundaries (Assimilation). A second possible pattern is that Antilleans gravitate away from the mainstream and towards Africans in the second generation as a result of shared experiences of exclusion from the mainstream (Racial Segmentation). A third possible pattern is that Antilleans maintain a cultural and civic distinction from Africans in the second generation, while they move increasingly closer to the mainstream (Cultural Segmentation). All three of these intergenerational patterns are likely to appear in various degrees in the data, but the question is to what degree and which social factors are more likely to be conducive to which outcome.    11   Outline of the study The study is organized in three distinct parts. The first frames the study, the second explores the respondents’ assimilation trajectories, and the third analyzes their identity dynamics and the boundaries they draw.   The first part of the dissertation comprises chapters two through four. Chapter two presents an overview of the history of immigration and assimilation in France, as well as an historical background of the region of West Africa and the Antilles where the populations under study have their roots. Chapter three then lays out the theoretical framework that informs the study. First, it discusses the relevant immigrant assimilation theories that will inform the analysis conducted in part two and explains how the study will seek to contribute to these theories with data from the French context. Second, it presents the theories of social identities and boundaries that underpin the analysis in part three. Chapter four presents Figure 1. Potential intergenerational assimilation and identification outcomes for second-generation French Antilleans and West Africans 12 the data used for the study, and how it was collected and analyzed, and discusses some challenges and limitations of the study.  The second part on assimilation trajectories also comprises four chapters. Chapter five details the particularities of the French context of reception and the constraints it puts on assimilation. Chapter six then explores the migration and settlement patterns of the first generation respondents. Chapter seven explores the family dynamics and neighbourhood effects that are influential upon second-generation respondents. Chapter eight then considers the second-generation respondents’ experiences and trajectories in school and the labour market, focusing particularly on their experiences of discrimination and differential treatment, but also their marital and partner choices.   Part three concentrates on identity and boundary dynamics, and comprises two chapters. In Chapter nine, I analyze the intergenerational identification patterns, and develop identity typologies for the second-generation African and Antillean respondents. Chapter ten then identifies the social boundaries that African and Antillean immigrants and their children confront in the host society. It explores how both groups seek to turn these boundaries to their advantage, and also how they themselves draw boundaries, not least among each other, and what underpins these symbolic dividing lines.  The final chapter concludes the dissertation and provides a broader discussion of its main findings.       13        PART ONE FRAMING THE STUDY      14 2. HISTORICAL BACKGROUND   This chapter aims to place Antillean and West African migration to France in its broader historical context. The first section provides a brief outline of France’s history of immigration and assimilation, as well as how academic research on assimilation, race, and ethnicity have changed over time. The second section then provides an overview of the history of Martinique and Guadeloupe in the Antilles, and of Mali and Senegal in West Africa, as well as the history of migration from these regions to mainland France. The goal is to highlight the historical ties between these geographical spaces and the experiences they have in common, but also the significant differences between and within them, as well as lay the groundwork for a proper understanding of the migratory processes that have shaped these communities in mainland France.    I. Historical Perspective on Immigration and Assimilation in France Immigration: From Europeans to Non-Europeans Compared to most other European countries, France has a long history of immigration. Large-scale immigration began in the mid-19th century as a result of the inability of internal migration flows from rural to urban areas to fill labour shortages in the expanding industrial sector. From 1851 to 1886, the country's immigrant population rose from 1 to 3 percent of the population, mainly from Belgian and Italian immigration. While many other European countries saw parts of their populations emigrate to America, France continued to receive immigrants from its neighbouring countries. When the United States began restricting immigration in the 1920s, the immigration flows to France began increasing rapidly with new economic migrants from Poland and Czechoslovakia in addition to workers from neighbouring countries. 15 In 1930, France was the country with the highest rate of foreign population growth in the world, and its share of foreigners was higher than the United States (Noiriel 1996, Weil 2005, Hargreaves 2007).  In the economic crisis of the 1930s, quotas were imposed for immigrant workers in some sectors, but immigration continued to feed the labour needs in other sectors. In addition to refugees from the Armenian genocide, Italians and Germans who fled Mussolini's and Hitler's ascent to power, large numbers of Spaniards migrated to France as soon as the civil war started and continued to do so after the war in increasing numbers. Under pressure from public opinion, the government proceeded to forcibly repatriate unemployed immigrant workers during this period, but also avoided interfering too much with immigration to accommodate the interests of some employers.  It wasn't until 1945 that France introduced an immigration policy which required immigrants to have employment in the country but also facilitated more permanent settlement. The policy didn't formally call for distinguishing immigrants by their origin, but in practice European immigration was favoured. The immigration flows from Spain kept increasing in the post-war period, in addition to a large Portuguese and North African population. Algerian immigration took off after the Second World War when the Algerian Muslims were granted French citizenship and were allowed to circulate freely on French territory. After the referendum granting Algeria self-determination in 1962, more than a million pieds-noirs (European colonists in Algeria) immigrated to France. Despite the fact that gradual changes have been made to the immigration policy for Algerians since, they still retain a privileged immigration framework compared to other nationalities. In addition to an influx of Moroccans and Tunisians, the North African population leaped from 2 percent of the foreign population in 1946 to 39 percent in 1982 (Hargreaves 2007; Weil 2005).  In the 1970s, the share of non-Europeans in the immigrant flows gradually increased, mainly coming from post-colonial migration from South-East Asia and Western Africa, but also from Turkey. The economic crisis of the mid-1970s led the French government to formally suspend labour immigration, and even attempt to organize the repatriation of a large share of North African immigrants, who were beginning to be seen as unassimilable. Official restrictions on labour migration notwithstanding, immigration has 16 continued apace—mainly through family reunification policies, despite initial efforts to discourage family reunification for Asian and African immigrants. About 120,000 to 150,000 immigrants from outside the EU now enter France every year, putting immigration levels at similar levels to what France experienced during the trente glorieuses—the three decade post-war economic boom—and a substantially higher level than that period for non-European immigrants (Tribalat 2013; Weil 2005). It has long been difficult to estimate the number of second and later generation immigrants in France, as the country grants citizenship to second generation immigrants and doesn't allow any collection of ethnic data on its population. A recent survey representative of the French population has sought to circumvent this, showing that 10% of the French population between the ages of 18 and 60 consists of immigrants and 11% of their descendants (Beauchemin, Lhommeau and Simon 2016).15 As much as a third of the French population under sixty is estimated to be descended from immigrants in the second or third generation (Tribalat 2015). A closer look at the origins and age structure of the children of immigrants show that 45% are from older European migrant waves (mainly from Spain, Portugal and Italy) and 31% have parents from North Africa, 11% from sub-Saharan Africa, and 9% from Asia. However, 88% of children of immigrants above the age of fifty are of European origin, whereas among children of immigrants under the age of twenty-five, 42% are from North Africa and 19% from sub-Saharan Africa. The youngest second generation population is of sub-Saharan origin, where 80% are under the age of 25 (Brutel 2017).   Assimilation: From ‘Our Ancestors, the Gauls’ to Identity Politics Despite their similarities in citizenship policies and immigration history, France and the United States differ fundamentally in their conceptions of state and society. While being American or Canadian can cohabitate with ethnic attachments of various kinds, ethnic retention is looked upon with far more suspicion in France, where it frequently qualified as “communitarianism” and seen as threatening to the social and political cohesion of the country (Horowitz 1992). The Jacobinist ideals inherited from the French Revolution                                                           15 In the Paris region these figures are substantially higher, as 43% of the population have a link to migration across two generations (Bidoux 2012) 17 emphasize the direct link between the citizen and the state, and are thus fundamentally inimical towards any communities or entities that might seek to mediate or influence it. As such, the country has historically emphasized cultural homogenization, a feat it has already largely accomplished on its own territory by eliminating regional cultures and languages (Weber 1977). This ideological framework, which strives towards individual equality through the imposition of common cultural and ideological assumptions, has fostered a stringent assimilationist policy towards immigrants and entrenched a perception of the national culture as intrinsically universalistic (Noiriel 1996).  However, the traditional French approach to immigrant incorporation has undergone significant shifts over recent decades. Following a backlash against assimilationism in the 1980s, when pragmatic policies and political clientelism for instance began accommodating specific immigrant community demands (Barou 1994; Kastoryano 2002), assimilationism subsequently reasserted itself again in the 1990s (Brubaker 2001; Favell 1998; Lorcerie 1994). In more recent years, anti-discrimination legislation has toughened, and some experimentation with affirmative action policies has even taken place (Simon and Sala Pala 2009). Other changes have also led to a decline of assimilationism in France. Numerous ethnic and religious advocacy groups have emerged, ranging from the Conseil représentatif des associations noires (CRAN) to the Comité contre l’islamophobie en France (CCIF).16 The position of Islam in French political life has also changed markedly with the creation of the Conseil Français du Culte Musulman (CFCM),17 as well as controversies centred on the Islamic headscarf, street prayers, and halal food. Additionally, the European Union’s more multicultural immigrant incorporation directives, which are automatically integrated into French legislation, have also contributed to erode French assimilationism. Changes in the political sphere have further contributed to this, most notably through the left’s turn away from its former working-class constituency which has historically functioned as the normative “melting pot” for newcomers, and its                                                           16 The Committee against Islamophobia in France. 17 The French Council of the Muslim Faith, created by Nicolas Sarkozy following the same principle as the representative organs for Jews and Protestants in France (Laurence and Vaisse 2006).  18 adoption instead of electoral strategies courting middle-class urbanites and various minority groups (Algalarrondo 2011; Tribalat 2013). As these developments indicate, France has become less assimilationist and stringently colour-blind in its emphasis over the years. However, these policy shifts have not substantially modified the French population’s perception of how immigrants should be incorporated.18 As such, most of the themes of the assimilationist repertoire dear to the electorate remain largely intact in political discourse, and a certain number of symbolic policies have been enacted along the same lines (see Favell 1998)  Scholarship: From the Social to the Racial Problem? Despite its long immigration history, France has thus not considered itself as a country of immigration until very recently. This self-perception has affected immigration scholarship as well. French sociologists who were interested in migration at the end of the 1960's were confronted by an almost empty field. European sociologists were more interested in relations between social classes than in their internal differences during this period, and the strong Durkheimian influence on the field in France left little room for analysis in ethnic terms (Réa and Tripier 2003). The same void could be found in historical research, which Gérard Noiriel (1996) claims is indicative of a "collective amnesia" that erased the country's understanding about the contribution of immigration to the renewal of the French population.   Although immigration research began developing in the 1970s and expanded in the 1980s, the field has remained affected by the French political and intellectual heritage, and most strikingly in its rejection of ethnic and racial variables of analysis (Chapman and Frader 2004; De Rudder 1999; Juteau 2006; Kastoryano 2004; Lagrange 2010). This has gradually changed in recent years, as debates about discrimination began cracking the egalitarian veneer that gave legitimacy to this “refusal of ethnicity” (Schnapper 1998). As such, over the past few years, new theoretical perspectives, concepts, and                                                           18 The 2015 survey of the Commission Nationale Consultative des Droits de l'Homme (National Consultative Commission on Human Rights, CNCDH) for instance reported that 87% of respondents considered it essential that foreigners who come live in France adopt French life habits – admittedly slightly less than 2012, when 94% expressed the same sentiment (CHCDH 2015). 19 terminology have been introduced to the field (e.g. Fassin and Fassin 2006; Poutignat and Streiff-Fenart 1995; Simon and Sala Pala 2009; Simon 2006). Despite these changes, issues pertaining to race and ethnicity are still frequently discussed euphemistically, and the collection of racial and ethnic statistics remain prohibited in the country (for an overview of the debates among social scientists on this issue, see e.g. Badinter 2009; Simon 2008).   II. Colonial Expansion and the Creation of an African Diaspora in the Americas French colonial expansion in Africa and the Americas followed similar patterns to that of other European powers, and occurred in two distinct phases: the first was primarily centered on North America and took place between the mid-16th and the early 19th centuries, when France lost all its major colonial possessions. Starting in 1830, a second phase of colonial expansion began, primarily centered on Africa and Indochina, which came to an end in the 1960s (see Figure 2).  Figure 2. French possessions from the first (light blue) and second (dark blue) colonial phases 20 From Plantation Colonies to Overseas Departments  Guadeloupe and Martinique were Spanish possessions since 1493, but the islands had not been subjected to any significant settlement until the French seized them in 1635. They were primarily populated at the time by the Caribs (also known as the Kalinagos), a South American-origin population who had displaced the longer-established Arawak population from the lesser Antilles (Hulme 1986). French coexistence with the Caribs rapidly deteriorated as colonization progressed, and after decades of skirmishes, wars and massacres, an agreement was reached in 1660, whereby the French would refrain from settling Dominica and St. Vincent, where the Caribs could retreat.19 The remaining Caribs in Martinique and Guadeloupe were gradually absorbed in the local population (Butel 2007; Meyer et al. 2016).                                                            19 The French nevertheless gradually settled on Dominica starting in the late 1600s, where they also imported a large slave population. The Island was later ceded to the British, who established a Carib Territory in 1903, which now is home to approximately 3.400 Caribs (Saunders 2005). Figure 3. Martinique and Guadeloupe in the contemporary Caribbean  21  During the first decades of French settlement, a variety of exotic agricultural products such as tobacco and cocoa were produced for the domestic market, for which engagés20 from the mainland were primarily recruited. The fall of tobacco prices in the early 17th century led many farmers to turn instead towards sugar cane production, stimulated by the technology and capital of newly settled Sephardic Jews fleeing Brazil as well as the example of neighbouring islands like Barbados, where large and lucrative plantations had been developed using cheaper African slave labour (Alleyne 2002; Butel 2007). Although slavery had been outlawed on French soil since the early 14th century,21 it was officially authorized in the colonies by 1642,22 where it was regulated by the Code noir from the late 17th century onwards (Meyer et al. 2016).  The French Revolution had a very disparate impact on the French Antilles. In contrast to the largest plantation colony of Saint-Domingue where both the whites and free persons of colour were eventually massacred by a far larger slave population, resulting in its declaration of independence under the name of Haiti in 1804, the Revolution had a more limited impact in the Lesser Antilles. Despite some limited slave uprisings, the white plantation-owning classes (known as Blanc-Pays in Guadeloupe and Béké in Martinique) continued running the affairs of the islands after the fall of the Ancien Régime. The plantation economy thus persisted until slavery was abolished in the mid-19th century,23 and only fully disappeared through crop diversification and the transition towards the industrialization of sugar production (Butel 2007).                                                           20 Engagisme was a form of indentured servitude in the Antilles. Plantation owners payed for the trip of the engagés, in exchange for which the labourer assented to work under servile conditions for a period of 36 months. After completing this three year period, the engagé was awarded a plot of land on which he could himself start a plantation. There were also forced engagés, usually convicts and Protestants who were sent to the islands to work off their penalties. The number of engagés gradually declined with the development of the sugar industry and the importation of African slave-labour. This quasi-feudal arrangement was abolished after the French Revolution, but similar types of contractual arrangements were reinstated after the abolition of slavery, primarily drawing labourers from South and East Asia to work in the Antillean sugar industry (Alleyne 2002; Butel 2007). 21 This law for instance was invoked in 1571 in Bordeaux, when a group of captured Africans were liberated by a court that stressed that France—“the mother of liberties”—did not tolerate such practices (Blanchard et al. 2011). This legislation was however curtailed in 1707 and abolished in 1717 (Meyer et al. 2016). 22 The French began to acquire slaves at St. Christophe and other islands as early as 1635. By 1654, there were about 12,000 slaves distributed across the Islands (Martineau and May 1935, quoted in Palmer 1996). 23 A law abolishing slavery was passed in 1794 by the National Convention, but only Guadeloupe was concerned as Martinique had temporarily come under British control (Blanchard et al. 2011). Napoleon reinstated slavery in the islands in 1802. 22 An intermediate caste of free persons of colour and mulattos had grown gradually throughout the years, who were prospering in commerce and the professions. They took on an increasingly prominent role in public life after the introduction of universal suffrage to the islands after 1870, and began replacing the settler class in political and civic institutions. A black political elite subsequently emerged in the 1890s—most notably in Guadeloupe, and largely out of frustration over the mulattos’ failure to adequately represent their interests—who eventually found common political ground with the white Béké caste.24  These upwardly mobile persons of colour, enthused by the egalitarian prospects of the French Revolution, also played a central role in brokering a full political and civic recognition of the islands by the mainland, which consistently took on the form of assimilation. At the onset of the First World War, this for instance expressed itself through requests for conscription and military service, which the islands’ status as colonies did not permit, but also through the longstanding ambition to be granted the status of French departments. This was ultimately fulfilled in 1946, which incorporated the islands into the political apparatus of the mainland as any other administrative unit of the Republic (Butel 2007; Dumont 2010). Following departmentalization, disillusion rapidly set in, as social, political and economic realities failed to meet expectations. Economic transformations in the islands resulting from the definitive move away from the plantation economy and the demographic boom of the post-war years had resulted in severe youth unemployment, which in turn spurred growing social unrest. In a radical attempt to address these problems, a massive migration policy coordinated by the Bureau pour le développement des migrations dans les départements d'outre-mer (Bumidom)25 was put in place in 1963, which also aimed to fill labour shortages in low-qualified positions in the public service sector of the mainland during the post-war economic boom. Around 160.000 people are estimated to have migrated from the overseas departments through the Bumidom policy, which was terminated in 1982 and replaced by labour market integration policies instead (Dumont 2010; Marie 2014).                                                           24 This racial stratification with an intermediary mixed-race class is common in the Caribbean (Alleyne 2002; Hoetink 1985). In the French Antilles, artisanal and commercial activities were monopolized by the mixed-race caste, which benefitted from a more favourable social position than darker blacks, either as slaves, or as ‘free persons of colour’ (see e.g. Ndiaye 2006: 43-44). 25 Bureau for the development of migration in the overseas departments. 23 West Africa before European Contact Before the arrival of the first European colonists, the African Sahel region26 was ruled by a succession of kingdoms and vast empires that prospered through trans-Saharan trade. This trade started developing in the 8th century and supplied North Africa with gold, salt and slaves after the Arab conquests. The Arab-Berber influence in the major trade hubs led to the constitution of a literate Muslim class, favouring the emergence of administrative bodies which were necessary for the functioning of such empires (Cuoq 1984). The first notable political entity of this sort in the region was the Ghana Empire initially established by the Soninke, which expanded as a commercial intermediary between the Berber caravans coming from the north and Malinke traders from the south. The Empire fell following the Almorovid invasion of the 11th century. The vast and prosperous Mali Empire emerged as a result of rapid and substantial Malinke conquests during the 13th century, but declined as early as the 15th century, when three of its main trade routes were cut off by the Touaregs to the North, the Songhai to the East, and the Fula to the West. On the Atlantic Coast, Portuguese ships were also beginning to provide coastal populations with goods previously acquired from the north, leading to the emergence of several coastal kingdoms that cut off the Empire’s only access to the sea. It was succeeded by the Songhai Empire, centered further to the east around the city of Gao, which began a substantial expansion to the North, South and West before being toppled by a Moroccan invasion in the 16th century that aimed to rekindle the decline in trans-Saharan trade resulting from the development of European commercial activity on the Atlantic coast (Lugan 2009).   French Penetration of West Africa Following in the footsteps of the Portuguese and other European nations, the French established their first trading outposts on the Senegambian coast in the early 17th century; the largest and most durable ones in Saint-Louis in 1659, and on the Island of Gorée in 1677. Although Saint-Louis and Gorée were partially used for the slave trade, they never took on the importance of the trade posts further south on the                                                           26 The Sahel is a semi-arid region bordering the Sahara desert to the south, stretching 3,300 miles across Africa 24 Guinean Gulf. The French purchased around one million slaves over the course of their participation in the slave trade, who were primarily deported to work in the sugar plantations in the Caribbean.27 This trade came to an end after the Revolution of 1848, when the short lived Second Republic definitively abolished slavery and extended political rights to its possessions. The same year, Senegal elected its first member of parliament to the French constitutional assembly (Meyer et al. 2016).  The interior territories of West Africa were only gradually explored and conquered in the second half of the 19th century. What is now Mali became a colony under the name of French Sudan in 1891. To avoid inter-state rivalry, French West Africa (Afrique Occidentale Française, AOF) was created in 1895, regrouping all French possessions between Senegal in the West and Niger in the East.  Traditional local authorities—largely religiously defined—retained significant local autonomy in these new territories in exchange for the abandonment of political ambitions and engagement in anticolonial struggle. There were however four communes that were granted the status of French municipalities in the 1870s and 1880s—Gorée, Saint-Louis, Dakar and Rufisque—whose residents became de jure French citizens with the same civil rights as their metropolitan peers, electing a municipal councillor and sending a deputy to the French Parliament (Meyer et al. 2016). The policy of assimilation introduced in these early Senegalese possessions conceptualized colonial subjects as blank slates which could ‘evolve’ under proper guidance to become full-fledged—albeit dark-skinned—French people. Although examples like Blaise Diagne—who became the first African deputy to France in 191428—certainly testifies to the reality of this policy,29 it remained limited to a small elite, and its proposed extension to the interior territories was met by local French and Creole30 resistance, which ultimately resulted in the                                                           27 The number of slaves transported by the French themselves is substantially lower however—estimated at 155 000—as they mainly bought slaves transported by other European powers, most notably the Dutch (Meyer et al. 2016) 28 The first black deputy in the French parliament was Jean-Baptiste Belley, who became deputy of Saint Domingue in 1793.  29 As in the case of the emergence of the black elite in the Antilles in the 1890s, Blaise Diagne was elected in 1914 by capitalizing on the resentment of the black indigenous population toward a putatively arrogant Creole (mixed-race) oligarchy. However, another important factor was the resentment towards the perceived French reluctance to assimilate the African elite, which was typically marginalized when retaining their so-called “personal status”, i.e. African or Islamic law (Boone 2006). 30 The Creoles were the descendants of French males in commercial or military positions and their temporary Senegalese wives. Their French ancestry awarded them French citizenship, and they typically distinguished themselves from the black indigenous population by emphasizing their loyalty to French society and culture (Brooks 2003).  25 aforementioned policy of “association” with local rulers, whose populations retained the “personal status” of traditional or Islamic law, disqualifying them de facto from acquiring French citizenship (Crowder 1962; Lambert 1993; Lewis 1962).31   After the two World Wars—in which Senegalese soldiers participated in significant numbers—the Union française was created, which granted citizenship to all colonial subjects. However, in contrast to their Antillean peers who were overwhelmingly pushing for assimilation and departmentalization, African elites                                                           31 This also shows the limits of the policy of assimilation and of French universalism in general in the colonial context. While French colonialism is often contrasted with British colonialism by invoking the former’s policy of making Frenchmen out of colonial subjects, this policy may have been written about extensively in theory, but was in fact very limited in practice—the most prominent example of it in fact being the four Senegalese communes which were already under French control during the French Revolution. Even in Algeria, which had been de facto territorially annexed to France and was thus governed by the same laws, the Muslims population was primarily governed by the code de l’indigénat (code of the indigenate) which gave them a separate legal status as French subjects under Muslim rule. This allowed them to retain their religion as aspects of it would be illegal under French law (such as polygamy) but they were unable to vote. If they expressed an interest, they could theoretically become French citizens and be subjected to French law, but in reality only small numbers acquired citizenship (Lacroix 2015).  Figure 4. Contemporary West Africa (in green) and the Sahel region (dotted area) 26 gradually turned towards nationalism and independence instead. Despite the continued efforts to design federal structures that would keep the empire together, France was gradually pressured to grant self-determination to its African colonies. In contrast to Indochina and Algeria, where long wars of independence broke out, Senegal and Mali obtained their independence peacefully in 1960 (see Figure XX), along with all other French colonies in sub-Saharan Africa, despite the frequent maintenance of numerous commercial, military and cultural ties with France (Ageron et al. 2016).  After Independence: Labour Migration and Family Reunification  Sub-Saharan African immigration to France dates back to the colonial era, and first took place through the navy. As rivers were the primary means of entry into the inner territories in Africa, capable individuals from coastal populations were often incited to participate in these expeditions. Many of these were Soninkes and Manjaks who later came to work on transatlantic ships, gradually settled in French port cities like Marseille and Le Havre, where they typically worked as dock workers or handlers in the harbour. They lived in small male communities from the same village who settled in old buildings near the harbours, which also served as resource hubs that assisted newcomers in finding work and new migratory opportunities (Barou 2011).  Although some sub-Saharan African troops had taken up residence in France after the first World War, this African presence remained limited in the first half of the 20th century. The citizenship awarded to all residents of the Union française in 1946 in principle allowed for free circulation, but sub-Saharan Africans were overshadowed by the larger number of Muslims from French Algeria, for whom access to the mainland was easier and more affordable. After 1960, when migratory agreements were signed with the new independent states, Mali and Senegal retained existing policies of free circulation until 1963 and 1964 respectively, when entry into the country was made contingent upon an employment contract and a medical certificate. Although free circulation agreements were retained for much longer periods of time with other sub-Saharan countries who tended to send students and more qualified immigrants, the Sahelian 27 countries of Senegal, Mali and Mauritania primarily sent an illiterate rural population that was raising the concern of authorities (Barou and Rozenkier 2011). The limited work opportunities in the harbour cities led many Soninkes from Mali and Senegal to move to the Paris region in the early 1960s, where they began assisting young people from their villages of origin to join them by pooling family resources. Given the favourable economic climate during this period, these newcomers could easily find work, while the need for labour power led authorities to overlook their frequent illegal status, choosing instead to grant them amnesty on a regular basis. These migrants gravitated towards low-qualified employment in the automobile industry, construction, and waste collection—an area where they represented 90% of employees in Paris the mid 1970s. Given their improvised settlement in squalid abandoned buildings, the authorities encouraged specialized associations to convert these buildings into worker hostels. Despite these efforts, the traditional structure of community life was largely reproduced in these hostels, where a parallel economy and the first village associations also emerged (Barou 2011; Poiret 1996; Timera and Garnier 2010). The restrictions placed on labour migration in 1974 represented a major turning point for these migrants, as it heavily restricted the long periods they usually spent in their country of origin to stay with their families. France was also constrained by international treaties to adopt a family reunification policy shortly thereafter, and since many Africans from the Sahel could no longer balance their economic activity in France with their family lives in their home villages, they decided to bring their wives and children to France. The combination of these two policies in the 1970s marks the starting point of a more substantial migration of poor families from the Sahel region, which has proved far more challenging to manage and regulate (Barou and Rozenkier 2011; Timera and Garnier 2010).     28 3. THEORETICAL FRAMEWORK   I. Immigrant Incorporation As Alejandro Portes has noted: "In the absence of theory, what we have today is mostly an amorphous mass of data on immigration to different countries and a series of concepts whose scope seldom exceeds those of a particular nation-state" (Portes 1997: 819). Studies geared towards theory-building and theory-testing across national boundaries can make significant contributions to developing more comprehensive theoretical frameworks that account for the empirical variations in assimilation outcomes across different countries, which can in turn also encourage more cross-national comparative work within the sociology of immigration. This section will provide a brief overview of the most prominent immigrant assimilation theories, present their most salient claims and limitations, and discuss how they will be made use of in the context of this investigation.  From Classical to Segmented Assimilation Theory Sociological theorizing on assimilation goes back to the Chicago School’s studies on European immigrants during the first half of the 20th century (e.g. Park 1950; Warner and Srole 1945), but the first substantial contribution to assimilation theory per se came from Milton Gordon’s Assimilation in American Life (1964). Gordon identified seven distinct dimensions of assimilation: (1) cultural or behavioural assimilation, i.e. the adoption of the typical cultural and behavioural patterns of the mainstream of the host society; (2) structural assimilation, i.e. the entrance into mainstream group relationships in clubs, associations or institutions; (3) identificational assimilation, i.e. adoption of the mainstream's collective identity and sense of peoplehood; (4) marital assimilation, i.e. large scale intermarriage; (5) attitude receptional assimilation, i.e. absence of prejudice; (6) behaviour receptional assimilation, i.e. absence of discrimination, and (7) civic 29 assimilation, i.e. absence of value and power conflict across groups. Gordon thought that cultural and behavioural assimilation occurred first, but that it didn't guarantee that other forms of assimilation would follow. When structural assimilation happened on the other hand, he believed that all other dimensions of assimilation followed.  Despite the influence of Gordon's work, it was criticized for assuming a unidirectional incorporation into a monolithic American culture, for not recognizing the distinction between individual and group levels of ethnic change, and for failing to account for larger structural processes. Moreover, it did not address occupational mobility and economic assimilation, despite their key role in the assimilation process. Taking stock of these criticisms, Richard Alba and Victor Nee (2003) set out to rehabilitate assimilation theory, by redefining it simply as a gradual "decline of an ethnic distinction" over time, whereby "the occurrences for which it is relevant diminish in number and contract to fewer and fewer domains of social life" (Alba and Nee 2003:11). This definition covers the complex ways in which boundaries blur between majority populations and immigrants, framing it less as a one-way accommodation process and more as a bilateral negotiation. The concept of assimilation as employed throughout this dissertation will be defined as the process of a gradually reduced distance with the mainstream population, both in terms of the salience of cultural and ethnic distinctions and in the gap in economic achievement.32   As the second generation of post-1965 immigrants were coming of age, some sociologists began questioning the assumption of a gradual cultural assimilation and upward socio-economic mobility throughout the generations. Herbert Gans (1992) hypothesized a “second generation decline”, where the upward mobility aspirations of children of immigrants would be thwarted by limited skills, connections and employment opportunities. He argued that they could acculturate instead into other fractions of American                                                           32 I use the term “assimilation” in this study as I primarily build upon the American theoretical literature, where the concept is commonly used to refer to immigrant incorporation. It is therefore not to be confused with the French concept of assimilation which typically also implies a loss of ethnic attachment and identity in addition to a convergence with the mainstream. The concept of intégration, which has come to replace assimilation in contemporary French public policy and debate—and which has no particular requirement regarding the eschewing of ethnic attachment and identity—would thus come far closer to the way “assimilation” is used here.  30 society that harboured negative attitudes towards mainstream American culture, risking to hamper their upward mobility and trap them in perpetual poverty.  These ideas were extended and modelled by Alejandro Portes and Min Zhou (1993) in their conceptualization of segmented assimilation. In addition to the traditional assimilation path into the mainstream, this theory predicts two other potential assimilation paths which both assume that cultural assimilation can in fact be detrimental for upward mobility (Rumbaut 1994). One path consists in a downward intergenerational socioeconomic mobility deriving from the cultural assimilation into the adversarial subculture of a domestic underclass. An oft-cited example of this are the children of Black immigrants from the West Indies whose modest socioeconomic backgrounds led them to settle in dilapidated inner-cities where they are more likely to adopt the counterculture of their phenotypically related African-American peers, thus ruining both their aspiration and their disposition for upward mobility (Waters 1999). The other path presumes that the avoidance of intergenerational acculturation can encourage second-generation socio-economic mobility through the social capital and economic opportunities provided by the ethnic group, primarily characteristic of Chinese or Cuban communities in the United States (Zhou and Bankston 1998).   New Immigrants, New Circumstances Through deindustrialization and the growth of the service sector, segmented assimilation theorists argue that the U.S. economy offers fewer opportunities for upward mobility for the children of post-1965 immigrants, who are confronted today by a three-pronged segmented labour market, with a rewarding primary labour market demanding high educational credentials, a secondary labour market with unqualified and precarious dead-end jobs, and an ethnic economy offering mobility prospects through an ethnic community (Portes and Rumbaut 2001). In other words, children of low-skilled immigrants who lack the economic opportunities of a strong ethnic community face the challenge of bridging an educational gap 31 that took their European predecessors several generations to overcome (Portes, Fernández-Kelly and Haller 2005). These structural changes in the economy have heavily impacted American inner-cities, whose social dislocation has been described in great detail in the urban poverty literature (e.g. Anderson 1990; Massey and Denton 1993; Wilson 1987). Substandard schools and employment opportunities, exposure to criminality, violence, drugs, and a subculture that rejects the institutions and norms of 'mainstream' society are thought to put the children of immigrants who settle in impoverished inner-city areas at risk of being more socio-economically disadvantaged than their parents, and find themselves trapped in a stigmatized urban 'underclass' (Zhou 1997). The racial characteristics of recent immigrants and the structural barriers they confront are also claimed to set new immigrants apart from previous European waves. The racial distinctiveness of these new immigrants exposes them to more discrimination in the housing market, which channels them into problem-ripe inner-city areas and heightens the risk that their children will adopt the reactive subculture of stigmatized and impoverished native minorities, which they risk becoming undistinguishable from when they acculturate (Waters 1999).  Segmenting Forces Proponents of the segmented perspective on assimilation further argue that immigrants end up following different assimilation paths depending on their human capital, the context of reception, family structure and acculturation dynamics. Education, job experience and language knowledge provide a first set of sorting mechanisms for immigrants. Although those with higher levels of human capital are more likely to succeed socio-economically when they arrive in the United States, the context of reception can greatly affect these outcomes. Portes and Rumbaut (2001) distinguish between three different contexts of reception: government policy (where responses to particular groups can be receptive, indifferent, or hostile), public 32 opinion (which can be prejudiced, or non-prejudiced towards specific groups) and the ethnic community (which can be strong or weak). These various configurations facing different immigrant groups will affect the extent to which they can make effective use of their human capital in the host society. The composition of the family and whether it includes both biological parents can also be significant for second generation outcomes, as well as the different forms of intergenerational acculturation that can occur: consonant acculturation, where both the parents and the children acculturate to the host society at similar degrees, dissonant acculturation, where the children acculturate far more to the host society than their parents, and finally selective acculturation, where insertion in an ethnic community allows key aspects of the parents' culture to be preserved despite some adaptation to the host society. Portes & Rumbaut (2001) argue that dissonant acculturation puts children at risk of downward assimilation, as it undercuts parental authority and favours a 'role reversal' where the children adopt parental roles vis-à-vis their parents, who lack the linguistic and cultural skills to get by. Consonant and selective acculturation, the authors claim, typically allow for a preservation of parental guidance and authority, which may be decisive when the second generation confronts racial discrimination, challenges in school and in the labour market, but also when they are exposed to harmful influences of surrounding youth subcultures and their adversarial stance towards education and the surrounding society.   Criticisms and Limitations Since this theory was first formulated, it has been exposed to important criticisms. Reminiscent of the debates surrounding the 'culture of poverty' (for review, see Wilson 1987), the importance of the 'oppositional culture' of a native minority in explaining second generation downward assimilation patterns has been questioned, suggesting  class variables in general and the immigrant working-class experience in particular might have more explanatory power (Perlmann and Waldinger 1997). Furthermore, some have argued that oppositional cultures can also be found in the native white population, but that more severe 33 sanctions for deviant behaviours occur for the minority underclass youth because of racial biases inherent to U.S. society (Stepick and Stepick 2010). Attempts to empirically evaluate the downward assimilation hypothesis suggest that the case might be overstated, and that reproduction of parents' working-class status, or gradual advancement within the working class or towards the lower-middle-class tends to be the most common outcome for children of immigrants raised in inner-city areas (Kasinitz et al. 2008; Waldinger and Feliciano 2004). A study of second generation immigrants from a variety of backgrounds in New York City has shown that dissonant acculturation is the norm rather than the exception, but also that neither acculturation patterns nor levels of ethnic embeddedness explain the variation in mobility patterns, whether within one generation or from one generation to the next (Waters et al. 2010). Recent work assessing the influence of government policies, societal attitudes and pre-existing immigrant communities on disparate assimilation outcomes shows that a more adequate measuring of these factors eliminate their ability to account for disparate outcomes between groups (Waldinger and Catron 2016).  Regarding the purported benefits of ethnic retention, several studies suggest that it isn't the role of ethnic retention in shielding children from the harmful effects of the surrounding native culture that helps upward mobility, but rather the respect that parents command and the educational expectations they have for their children (Stepick and Stepick 2010). Furthermore the availability of the ethnic-adhesive path for many first-generation low-wage workers has been questioned, as in the absence of significant economic and other resources the community cannot provide opportunities for upward mobility, thus risking to further enhance marginalization from the host society instead (Alba and Foner 2015). These criticisms do cast doubt upon some aspects of the segmented assimilation model, most importantly regarding how widespread some of its predictions may be or the mechanisms which it hypothesizes. Nevertheless, the general framework of identifying enduring assimilation barriers and cultural segments in the host society that might negatively impact specific immigrant populations more than others, and how this may potentially intensify social segmentation remains a valuable theoretical pursuit. While using this framework in France will consist in evaluating the pertinence of some key mechanisms it 34 postulates in a different social context, it will also consist in examining more inductively which specific assimilation barriers and cultural segments may be impacting significantly the trajectories of the Antillean and West African populations under study.   Segmented Assimilation in France? A handful of studies have also looked at the relevance of the segmented assimilation model outside the United States (e.g. Boyd 2002; Safi 2006; Silberman, Alba and Fournier 2007; Silberman and Fournier 2008; Takenoshita et al. 2014). Those conducted in France have proved problematic as no racial, ethnic or religious statistics are collected in the country (Blum 2002; Simon 2008). The issue is further complicated by the fact that about 130,000 immigrants are naturalized every year (Weil 2005) and second-generation immigrants are automatically granted French citizenship when they reach the age of majority, making them statistically undistinguishable from the rest of the French population. These studies on assimilation trajectories in France nevertheless seem to agree that parts of the North African and sub-Saharan African immigrants are at the very least at risk of downward assimilation. Identifying statistical disparities between groups says very little about their causes however, as Maurice Crul and Jens Schneider point out: "Even if in other countries we find similar segmented outcomes, the mechanisms and institutional settings behind them will most probably be very different from those described by segmented assimilation theory" (2010: 1264). As presently formulated, segmented assimilation theory is indeed intimately linked to structural features unique to U.S. society, such as the presence of a native stigmatized racial minority and a historical experience that has made race a central criterion in social stratification. Although existing quantitative studies on segmented assimilation in France remain largely agnostic regarding the specific mechanisms that produce these outcomes, it has been suggested that they hinge less on 'race' and more on the experience of colonialism or more salient religious and cultural boundaries (Silberman and Fournier 2008).   35 An Integrated Theoretical Perspective Despite the broad set of assimilation outcomes posited by the segmented assimilation model, the question is whether its theoretical usefulness extends beyond a short-term description of divergent assimilation paths—which new assimilation theory suggests will eventually end up converging over time—or if it actually describes more permanent and enduring trends. Although assimilation may turn out to be the master trend over the long run—“bumps” notwithstanding—there is of course also the possibility that more enduring social divisions will remain and substantially affect future immigrant incorporation moving forward. The long-term accuracy of these theories can of course only be a matter of speculation, and what might be of greater importance are the causal factors they put forth for explaining these outcomes vis-à-vis the mainstream. As both theories were developed with different explanatory ambitions in mind, they emphasize very different processes: while segmented assimilation, as we have seen, is more attuned to identifying the contextual, structural, and cultural factors that produce incorporation outcomes that deviate from the mainstream path, assimilation theory focuses on the mechanisms at the individual, network and institutional levels that lead individuals to gradually meld with the mainstream. Building upon the behavioural assumptions of the new institutionalism in sociology, Alba and Nee see the behaviour of immigrants as deriving from a context-bound rationality that shapes their perception of costs and benefits within the surrounding institutional environment. Their rationality is constrained insofar as their perceptions of self-interest are culturally shaped (by customs, social norms, law, ideology, and religion), and their behaviour derives from rule-of-thumb heuristics based on incomplete information. Immigrants harness their individual human capital and the social capital of their group aiming to take advantage of opportunities in their institutional environments to improve their life chances. Assimilation is simply a by-product of this purposive action, "something that frequently happens to people while they are making other plans" (2003: 282).  36 Alba and Nee further specify a repertoire of proximate and distal mechanisms that can lead immigrants to blend to varying degrees into the mainstream. The proximate mechanisms, which operate at the individual and social network levels, derive from the forms of capital immigrants can mobilize and the rewards and punishments within close-knit groups that enable them to engage in joint action to achieve collective goals. The distal causes are embedded in the institutional environment which can elicit assimilatory or segregatory behaviours from immigrants depending on its inherent incentive structures.  This perspective has the advantage of providing a genuine conceptualization of human agency and of the structural environment in which it operates, specifying the degrees to which boundaries between immigrants and the mainstream are likely to blur as a result. As far as its modelling of human behaviour is concerned, new assimilation theory seems to provide a far more comprehensive—and perhaps most importantly, exportable—theoretical framework for understanding assimilation processes. Segmented assimilation theory on the other hand, while it hypothesizes very specific processes that are far more static and culturally contingent, suggests that there may be significant structural factors outside the purview of the standard assimilation mechanisms which may represent substantial hindrances to assimilation into the mainstream. The nature of these factors may need to be revised and adapted to new national contexts, but the goal of identifying potential structural and cultural factors that may be conducive to assimilation outcomes that deviate from the mainstream path clearly has a valuable theoretical purpose. Furthermore, the segmented perspective also has the advantage of recognizing more explicitly the cultural segments outside of the mainstream, and the influence these can have upon immigrant assimilation.  Both perspectives will therefore inform the theoretical framework for this study to varying degrees. While the mechanisms that underpin the behaviour of immigrants and their children will be seen as largely consistent with the more “agentic” perspective of the Alba-Nee model, a more explicit theoretical emphasis will be put on the salient structural barriers and cultural segments which might impede or even derail individual assimilation outcomes. Although the relevant factors and mechanisms that may lead individuals to gravitate towards or away from the mainstream will largely be determined in an inductive fashion, the analysis will also seek to assess the pertinence of some of the most salient hypotheses put 37 forth by segmented assimilation theory, most notably those pertaining to the role of intergenerational acculturation dynamics, of a putative ‘oppositional culture’, and the specific role of racial mechanisms in the assimilation process.    II. Racial and Ethnic Identities and Boundaries Racial and ethnic identities and distinctions are pervasive in virtually all human societies. They are constituted through complex social dynamics, involving micro-level processes of identification and ascription, as well as the mobilization of group-specific resources and attributes in the face of a large variety of external social circumstances. But race and ethnicity and their respective dynamics are also informed by hierarchical valuations derived from perceived social, economic and cultural differences, as well as the socially inherited understandings of these differences, the historical legacy of group relations, and the power asymmetries that have driven them.  Conceptualizing Race and Ethnicity  The ambiguity of the concepts of ‘race’ and ‘ethnicity’ has often been emphasized. In this study, race will be defined as a set of physiological characteristics that link a person to a broad geographical ancestry, while ethnicity will be understood as a collective identity built around a perceived common descent and history, represented and sustained through a set of symbolic elements and practices (Cornell and Hartmann 2007). As these definitions suggest, race and ethnicity overlap substantially. Racial characteristics can be a central criteria for ethnic group membership just as ethnic groups can have very diverse racial make-up, and there is always a plurality of ethnic affiliations within broad racial categories.33                                                           33 While religious groups can transcend and aggregate several ethnic and racial groups, religion can also be understood as a subcomponent of ethnicity. Although it may in some cases be less “assigned” than an ethnic belonging, given that individuals may be able to distance themselves more easily from a religious inheritance than an ethnic one, this religious heritage typically plays a 38 However, ethnicity is always associated to a shared sense of ‘groupness’, whereas race is a far broader and more obdurate category to which differing meaning can be attributed, fundamentally affecting its salience in social life. In light of this, I resolutely distance myself from perspectives that conceptualize racial categories as a priori social groups with tangible interests and identities, seeing in it the same "pseudo-scientific operation" that Weber identified in the direct and immediate link that Marx drew between class and class consciousness (Giddens 1980: 43).34 While recognizing for analytical purposes that racial and ethnic criteria are fundamentally cognitive constructs that divide up the social world (Brubaker, Loveman and Stamatov 2004), and that the social significance attributed to these phenomena should be a matter of empirical investigation (Brubaker 2009), I also distance myself from the radical constructivist assumptions that often go hand in hand with these approaches, which put an excessively strong accent on the fluidity and malleability of racial and ethnic phenomena.35 By recognizing both their fundamentally obdurate nature and the fluidity that can characterize them, I take an intermediary stance between constructivist and primordialist positions, acknowledging the visceral and enduring importance social actors consistently attribute to ethnic and racial attachments. Beyond these definitions and considerations, my general approach to the complex and multifaceted racial and ethnic phenomena in this study will be resolutely empirical. As such, I seek to situate them within their highly specific socio-historical context, and approach the way they are concretely experienced and expressed by the social actors themselves. While this micro-oriented focus on the concrete relational processes underlying ethnic and racial phenomena subjects a large number of 'givens' to empirical analysis, I simultaneously seek to avoid attributing an excessive autonomy to agents and disregarding broader                                                           fundamental role in shaping a large part of the cultural practices, customs and behaviours that members of an ethnic group take for granted. As with race or ethnicity, religion can also be assigned from without, be a motive for social stigma, rejection or prejudice, but operates more on cultural, philosophical or metaphysical assumptions than the more obdurate racial or ethnic ones. 34 I incidentally agree with Bourdieu's assessment that Marx "provided false theoretical solutions––such as the affirmation of the real existence of classes––for a true practical problem: the need for every political action to demand the capability, real or supposed, in any case credible, to express the interests of a group" (Bourdieu 1998: 31). 35 Following for instance Andreas Wimmer, who also attempts to “move away from the constructivist credo that is currently shared by most authors writing on ethnicity” and their “routine assertion that ethnicity is constructed, contextually variable, contested, and contingently eventful”, emphasizing that “not all ethnic boundaries are fluid and in motion, not all are cognitively and emotionally unstable, contextually shifting, and continuously contested.” (Wimmer 2013: 204) 39 historical, social structural and cultural forces that these dynamics are inevitably caught up in and deeply shaped by.  Ethnic and Racial Identities In an attempt to capture the fluidity of ethnic identities, Stephen Cornell and Douglas Hartmann distinguish between “thick” and “thin” ethnic and racial ties, depending on the extent to which they organize social life and action. Racial and ethnic bonds can involve more or less shared interests, institutions and culture, and the more they do, the “thicker” they are. The “thickness” of ethnic and racial identities also varies depending on the nature of the broad social arenas or “construction sites” that these groups confront (ranging from politics, labour markets, residential space, social institutions, to culture and daily experience) and the resources these groups bring to bear upon these contexts (the group's pre-existing identity, its size, internal differentiation, social and human capital, and its symbolic repertoires). Cornell and Hartmann’s framework will be useful for assessing the 'structural' differences and similarities that both groups and generations confront that might affect the salience of their ethnic and racial identities. Richard Jenkins (2008a; 2008b) attempts to integrate ethnicity into a broader theory of social identity and gives a more micro-sociological and social psychological account of the complex dialectical process linking internal processes of identification and external processes of ascription. At the individual level, drawing on Cooley and Mead's distinction of the "I" and the "Me", he argues that racial and ethnic identities first emerge in childhood through the delicate interaction between the child's individual needs and the socialized demands largely inculcated by significant others and the cultural framework they operate within. The self-image that results from this process is mobilized in social interactions, and is in turn affected by the public image resulting from others' interpretations of and reactions to the individual's behaviour. Power differentials also allow some groups to categorize others more potently and to impose a public image that can influence other groups’ self-image and the social, educational and professional aspirations open to it—an element that is often missing in many social anthropological theorizations of ethnicity. 40 Jenkins's conceptualization of social identity formation as "a dialectical synthesis of internal and external definitions" (1994: 19) occurring within the framework of individual cognition, social interaction and institutions, provides a solid theoretical framework for investigating the racialization, ethnicization and nationalization processes at work among the groups under study (Brubaker 2009). This framework accounts not only for the subtle and multi-level interplay of identification and categorization in racial and ethnic identity formation but also for the role of power differentials on these dynamics, and thus provides a potent balance between "an objectivist physics of material structures" and "a constructivist phenomenology of cognitive forms" (Bourdieu and Wacquant 1992: 5).  Psychological Perspectives Psychological theories of ethnic and racial identity development in early life provide additional insights into the process through which ethnic identities typically develop (Erikson 1968; Marcia 1966; Phinney 1990). The ethnic identity development model suggests that ethnic and racial identities unfold through a variety of distinct stages. According to this perspective, after an initial stage where ethnic identity is unexamined (either by lack of concern or by passive acceptance of others' perspectives through socialization), a period of crisis pushes individuals into a phase of exploration where they seek to make sense of the meaning that their ethnicity has for them. This can be followed by a stage of resolution, where they come to a sense of clarity about its meaning, which in turn can result in commitment, where a sense of belonging and positive feelings about the ethnic group are achieved (Umaña-Taylor et al. 2014). This general model for racial and ethnic identity development, with its focus on the need to come to terms with the meaning that a racial and ethnic identity carries for the individual will provide a useful general frame for conceptualizing the identity dynamics of the Antillean and African respondents of this study. Although this model does not necessarily preclude that individuals may come to re-examine their identities later in life and revisit earlier stages of the process, other models have sought to address identity changes that are less tied to specific phases of the life course, and are more focused on specific triggering 41 events. A notable one in the context of this study is Nigrescence theory (Cross 1991) and its related formulations, such as Racial identity theory (Helms 1990), which outline a series of stages (or statuses) identified in the identity development of African Americans. In the pre-encounter stage, the individual adopts a worldview that idealizes whiteness and devalues blackness, before a triggering event such as discrimination pushes the individual into the encounter stage where pro-white views are abandoned. In the immersion-emersion stage, blackness is idealized and whites denigrated, before the individual settles on the internalization stage, where a feeling of inner security is attained and where Blackness is integrated to the sense of self (Schwartz et al. 2014). This identity theory, which is more tailored for Black Americans specifically, puts a greater emphasis on social triggers that make race salient, and is useful in addressing the issue of internalizing the perspective of other racial groups on an individual’s self-perception. This will provide a more pointed theorization of identity dynamics that might also be widespread among the Black population in France.  Another research tradition which has been influential on the ethnic identity model described above also merits some closer attention as it sheds light on some of the underlying motivations behind identification and the shifts it can undergo. Social identity theory (Tajfel 1981; Tajfel and Turner 1986) posits that individuals strive to achieve positive and distinct social identities. According to this theory, individuals have multiple social identities that are more or less psychologically salient and that derive from their knowledge of being members of given social groups or categories, as well as the emotional significance they attach to these memberships. After individuals have categorized themselves and others in groups, they assess the intrinsic attractiveness of their ingroup by comparing it to relevant outgroups on a series of valued dimensions. If group comparisons happen to result in unsatisfying social identities, they will either favour alternative social identities (individual mobility), seek more favourable value dimensions or intergroup comparisons (creativity), or openly challenge the status hierarchy (competition). The strategy they adopt will depend on the degree to which they identify with the identity, the perceived permeability of group boundaries, and the perceived stability and legitimacy of the status hierarchy (Ellemers 1993). Individual mobility and creativity are only viable for individuals with low levels of identification with their 42 ingroup (Mummendey et al. 1999). When individuals identify strongly with a social identity, its negative social evaluation becomes less tied to their self-evaluation, and they are more prone to assert its value or distinctiveness through collective strategies of social competition (Ellemers, Spears and Doosje 2002). This theory intricately nests group identity dynamics within the specific nature of group relations, and will thus be particularly useful for assessing the identity dynamics between African and Antillean respondents, whose group relations are one of the central foci of our analysis.  These strands of research provide a more comprehensive theoretical framework for evaluating the experiences that respondents have of their racial and ethnic identities growing up, and the particular challenges they may face in elaborating a coherent and positive sense of self out of their diverse cultural ties and their racial characteristics.  Racial and Ethnic Boundaries The boundary metaphor can be traced back to the work of the anthropologist Fredrik Barth (1969), who argued that ethnic distinctions do not necessarily derive from substantive cultural differences between groups, but rather from the social boundaries that they erect and maintain between each other.  This boundary approach has been influential on the ethnicity literature over the last two decades (e.g. Bauböck 1998; Lamont and Molnár 2002; Sanders 2002; Wimmer 2008; Wimmer 2009; Zolberg and Long 1999). It has also informed novel theorizations of immigrant incorporation (Alba and Nee 2003) by seeking to explain how "bright" boundaries that starkly distinguish social groups become more "blurred", i.e. more ambiguous and less fraught with social consequences (Alba 2005). More specifically, as Andreas Wimmer points out, the boundary approach has helped to imagine social landscapes in which ethnic divides are culturally meaningful, consequential for the allocation of resources and the distribution of life chances, and historically continuous, all the way allowing an observer to describe and imagine how they might move across a landscape, become porous and inconsequential, be crisscrossed by other, more meaningful boundaries, or perhaps even dissolve altogether (Wimmer 2013: 4) A first important point to stress is that the social boundaries that regulate the processes of exclusion and inclusion in any society do not appear ex nihilo, and are largely shaped in a path-dependent manner from 43 historically inherited cultural, legal, and institutional materials (Favell 1998). An example would be the strict secular injunctions and the denunciation of “communitarianism” in France, which have for instance emerged—political opportunism notwithstanding—following a political, cultural, civic and intellectual lineage which largely determines the potential spectrum on which existing religious and ethnic boundaries can shift and what grammar they can usefully employ to that end. Boundaries are thus also subjected to important constraints when seeking to modify them or impose new ones. Wimmer (2013) furthermore provides a general account for how boundaries are made and unmade through individual action. He argues that the boundary-making strategies that individuals pursue for a variety of reasons may become influential depending on existing institutional incentives and social networks, as well as on the individuals’ position in economic, political and symbolic hierarchies. He claims that when various individual strategies converge on specific categories and boundaries, a consensus is likely to emerge and certain social boundaries might thereby gain in salience.  Massey and Sánchez (2010) concentrate more specifically on how immigrants attempt to negotiate the boundaries that they encounter in the host society. Drawing on Fredrik Barth’s insights, they argue that it is through the complex process of self-definition and ascription by others that boundaries between natives and immigrants are “brokered”. This brokering is undertaken in two distinct ways: first through framing, where positive or negative meanings are attributed to one’s own or to another ethnic group, and second through boundary work, where an ethnic group engages in practices that either promotes contact and access to resources or inhibits contact and skews access. Through this two-way process, immigrants “broker the boundaries to help define the content of their ethnicity in the host society, embracing some elements ascribed to them and rejecting others, while simultaneously experiencing the constraints and opportunities associated with their social status” (Massey and Sánchez 2010: 15).  A common emphasis in this literature has been placed upon how boundaries serve to maintain processes of ethnic closure that aim to deny outsider groups access to valuable material and symbolic resources. The approach adopted here will more explicitly assume that the object of immigrant boundary brokering also includes social psychological motivations put forth in the theory of ethnic and racial identity 44 development and social identity theory, i.e. achieving a positive and distinct social identity, where the value of the group is enhanced and where negative attributions and social identity threats are deflected. Much research on social boundaries has also distanced itself from approaches that give explanatory power to the “cultural stuff” within these boundaries. Despite my interest in how distinctions and categories are drawn, and how boundaries are negotiated and brokered, I will also take a more explicit interest in the “cultural stuff” that these boundaries enclose, as this also plays a determining role in shaping expectations, behaviour patterns, as well as loyalties and demands placed upon individuals. I believe this oscillation between a boundary approach and one more centered on the cultural assumptions of ethnic groups provide a more holistic perspective on the social processes under study.   45 4. METHODS AND DATA   The data used in this study primarily derives from in-depth interviews complemented by quantitative and observational data. This chapter describes the recruitment and sampling procedure, explains how the data was collected and analyzed, and presents the detailed demographic characteristics of the sample. I also discuss some issues pertaining to positionality and reflexivity, the challenges and limitations of the project, as well as the ethical considerations of the research and the precautions these dictate.    I. Presentation of the Data Recruitment and Sampling Procedure The main source of data consists of semi-structured in-depth interviews with 55 Martinican, Guadeloupean, Senegalese and Malian migrants to the Paris region and the metropolitan-born children of these populations. The selection criteria for first-generation respondents was to be born and raised in these geographical areas, and to not have migrated to mainland France before adulthood. Second-generation respondents were recruited when they had at least one parent who was born and raised in these countries or islands, when they were themselves born and raised in the Paris region, and were above the age of 20. The interviews were conducted between August 2011 and January 2015.36 The duration ranged between one and five hours, but most lasted between one and a half and two and a half hours.   Respondents were recruited using a variety of means. The first participants were found using personal networks and advertisements disseminated in residential areas, specialized stores, and hairdressing salons, at cultural events and on internet forums. I noticed that individuals who freely                                                           36 A complementary interview was also conducted in March 2016. 46 responded to ads were more likely to have a particular motivation to participate, an “agenda” of sorts where they selected me more than I selected them. One respondent recruited through this channel was for instance a Malian undocumented immigrant who wanted to use the opportunity of the interview to hopefully get a wider hearing for his particular plight. Most respondents who eventually participated in the study were recruited through referrals from other study participants, but I made an explicit point of systematically seeking out a variety of social networks to diversify the respondents as much as possible.  Although qualitative studies are by their very nature non-representative of the broader population, I wanted the samples to be reasonably in keeping with the broad traits of the population, and thus avoided for instance just interviewing Christians among the Senegalese, or just Jehovah’s Witnesses among the Antilleans. I sought to constitute samples that were stratified by gender and class, and that had an even balance between Martinicans and Guadeloupeans as well as Senegalese and Malians. An important selection criteria was also diversity of experience (Trost 1986). If respondents mentioned a family member or an acquaintance that could provide a perspective that seemed novel or deviated from the material I had collected thus far, I prioritized those interview opportunities. In rare cases, this sometimes meant interviewing respondents that only partially fit the selection criteria but could provide unique perspectives on the research questions. Some examples of this are participants who had arrived during childhood instead of being born in France, had grown up moving back and forth between their area of origin and mainland France, some who were of mixed race, or even one who had been adopted. These individuals, who were in a sense both insiders and outsiders to the group were often particularly useful insofar as they not only had acquired a more reflexive relationship to their group identity, but also had a clearer understanding of how group boundaries were drawn and how salient they were (Johnson 1990). Given the centrality of family dynamics to the study, I also tried as often as possible to interview parents of the second generation respondents, or children of the first-generation respondents. I also interviewed the sibling of a second-generation Antillean respondent and siblings of two second-generation African respondents, which provided precious insights into the inner workings of the family, the relationship between siblings, as well as an alternative perspective on the same family environment. 47 Data Collection and Analysis The interviews were often conducted following the same pattern. I always started with some basic demographic questions (place of birth, marital status, number of siblings, etc.). This information was later added to a spreadsheet which was useful for comparing the participants’ characteristics and backgrounds. Starting with these questions also gave opportunities to build rapport with the respondents and transition more naturally into speaking about their family and upbringing, which could be very personal—and in some cases emotionally laden—for some respondents. The first substantial part of the interviews dealt with the respondents' backgrounds and educational and professional trajectory. I first probed their family background and upbringing, before asking about their neighbourhood and their key peer groups during their development. Depending on the rapport I had been able to develop with respondents, I would explore these issues in different orders and depths. I then gradually transitioned towards their school trajectory, with a specific focus on the educational tracks the respondents chose and what prompted these choices. I then explored their trajectories in the labour market and their experiences with discrimination and differential treatment in general. This part of the interview aimed to map out the socio-economic circumstances and challenges the respondents' faced, and the dynamics underlying their mobility patterns. I introduced the second part of the interview with a chart to set the basis for a discussion about their social identities. The chart provided a list of racial, ethnic, national and religious identifications (and space for additional identities they might have found important) and asked respondents to rate the importance these had for them on a scale of 0 to 5.37 The chart allowed me to map out their self-perception and allowed me to ask why they chose to rate some identities as stronger or as weaker than others. In most cases it provided a useful starting point for discussing their attachment to various parts of their social identities, and what may have contributed to giving them that salience. Any life experiences they might                                                           37 Antillean respondents were asked to rate the extent to which they identified with being European, French, Antillean, Martinican/Guadeloupean, African, Black, of a particular religious identity (specify) and any other identity that might have been important to them. African respondents were asked to rate their identification with being European, French, African, Malian/Senegalese, with an African ethnic group (specify), with being Black, and with a religious identity (specify) and any other identity they might have wished to add.  48 have mentioned in the first part of the interview which could have contributed to shaping these identities was also brought up here. At this stage, I usually shifted the discussion towards group relations. I asked Antilleans about their experiences with Africans, and Africans about their relations with Antilleans. I also asked both groups about their relations with the majority population, and focused upon how they felt perceived and categorized, both by the majority and the other minority group. This line of questioning helped me gather information on the 'cultural repertoires' they draw on to justify the boundaries they draw (Lamont 2000), and I paid particular attention to the salience of racial, ethnic, cultural and religious considerations in these boundaries. I concluded most interviews by asking about the respondents' views on some general social issues relevant to the themes of the interview, and also asked some additional demographic questions which were more tactful to bring up at the end of the interview, such as their parents’ profession.  When the respondents suggested it, the interviews were conducted in their homes. I generally preferred this as it gave an opportunity to collect rich observational data, provided material for additional questions, and gave a better sense of how they lived their lives. Furthermore, the home could often provide us with more privacy for the interview. In some cases family members were present, which could introduce a different dynamic and make some topics more difficult to discuss, but this was often compensated by obtaining multiple points of view on the same topic. In many cases however, the interviewing took place in public spaces: coffee shops, restaurants, or even parks, or semi-public places like African worker hostels, which also provided rich ethnographic material. I used an interview guide (see Appendix 1) with pre-formulated questions that covered all the sections and themes I have just outlined, which I fine-tuned over the course of the first interviews, to add new questions and remove or reformulate those that were less effective. Although I aimed to cover all the themes in the interview guide, the interviews were conducted quite loosely to make the interaction more natural. They were thus given much leeway to explore additional topics that might have been particularly relevant in their own experience. 49 I took detailed field notes after each interview, which gave a summary of the most salient themes that it explored and allowed me to connect them to the theoretical and substantive issues of the study (Denzin and Lincoln 2000; Weiss 1994). The field notes also allowed me to record some of the important information that might otherwise be lost, and also gave me an opportunity to reflect on how the interview was conducted, how I might have affected the interview in a negative way, or simply which parts of it may have worked and which didn’t.  I compensated the respondents with €20 for taking the time to participate in the study. This sometimes facilitated the recruitment of respondents, and allowed me to speak to people who may otherwise not have chosen to participate. I thereby avoided only interviewing people who saw their participation as an end in itself, who were perhaps also more comfortable in voluntarily taking on the role as a ‘spokesperson’ for their community. The interviews where financial compensation played a large part in the respondent’s decision to participate were generally of a poorer quality, but it was nevertheless important to use this compensation as it also added more respondents from working-class backgrounds with limited schooling whose perspectives were in many cases valuable for the study.  All the interviews were conducted in French, and were fully transcribed in the original language – only the interview quotes used in the dissertation were translated. The interview transcripts were coded and analyzed following the procedures detailed by Strauss and Corbin (1998: 101-142), using the Atlas.ti software package for qualitative data analysis. A first phase of open coding aimed to descriptively categorize the interview material before a second phase of coding which used more analytical codes derived both from the theoretical constructs that informed the study and the underlying social mechanisms emerging from the interview material itself. Throughout the first phase of coding, I created several memos where I jotted down notes for various patterns that were not adequately covered by existing codes and that did not yet have codes developed to cover them. These memos were subsequently useful for developing theoretical codes. I also created primary document families for each group and generation, which were useful for filtering and comparing the material. A final step of coding consisted in developing code families that thematically 50 categorized all the codes and memos. This helped structure the data and design the content for the various chapters of the dissertation.  Ethical Considerations I provided all respondents with a consent form (see Appendix 2) that specified the general purpose and themes of the interview. The form guaranteed the respondents full confidentiality and detailed the procedures that ensured this. The form mentioned the approximate duration of the interview and pointed out that the respondent was at liberty to end the interview at any time. In addition to my own contact information, the document also provided the contact information of the UBC Office of Research Services in case participants had any questions or concerns about the study or their rights as a participant. Before every interview, the respondents were asked to carefully read and sign the form before the interview, and were given a copy for their own records. I always walked them through the details of the consent form and asked if they had any questions about the interview or the study before they signed the form. All the audio files and transcripts were tagged with ID numbers, and pseudonyms are used to refer to the respondents throughout the dissertation. All materials related to the study were kept on a password-protected hard drive, and were not made accessible to anyone else. I have also taken precautions in editing out any potentially identifying information in the dissertation, such as very specific and unusual biographical information. I also took ethical precautions when recruiting respondents, and was careful to inform people that I could not contact anybody without their prior consent. This sometimes made me highly dependent upon the diligence of prior interviewees in getting in touch with other respondents, but it made sure that their privacy was respected. I took some additional precautions to minimize any potential harm that could result from participation in this study. Since upbringing, migration experiences, experiences of racism and discrimination can be emotionally laden topics for some respondents, I remained sensitive to the 51 participants' emotional reactions throughout the interview. In the rare case when a respondent teared up over the course of our conversation, I asked if he or she wanted to change the topic, take a break or even stop the interview. I avoided probing insistently on topics when I sensed that they were making respondents too emotional.  In the few cases where the French language skills of the respondent were insufficient to ensure that full informed consent had been obtained, I relied on other family members of respondents to explain every point mentioned in the consent form. I was careful to avoid violating any cultural norms with the African respondents, but circumstances where I had such concerns were very rare. I also suggested making the results of the study available in the form of a synopsis in French to those who were interested.  Complementary Data Sources In addition to this qualitative data, I use quantitative data drawn from the Trajectoires et Origines (TeO) survey (Beauchemin, Hamel and Simon 2016) to identify the broad patterns of intergenerational social mobility, discrimination experiences and social identity dynamics within the Antillean and West African populations in mainland France. The TeO survey was conducted jointly by the Institut National de la Statistique et des Etudes Economiques (INSEE) and the Institut National d'Etudes Démographiques (INED) between September 2008 and February 2009 on a random sample of 21.000 respondents, and is representative of the French mainland population between the ages of 18 and 50. In addition to respondents from the majority population, the sample thus includes respondents born abroad and in French overseas departments, as well as their descendants born in mainland France. The survey broadly aims to study the access that immigrants and their children have to a variety of social and economic resources in France, and therefore inquires about their educational attainment and their labour market position, their experiences of discrimination in school and in the labour market, as well as other areas of social life. Furthermore, the 52 survey includes questions about respondents' social identities, their religious observance, the origin of their partner, as well as other questions that are useful for assessing the respondents' socio-economic and cultural integration. The qualitative approach, well-suited for identifying complex social processes and mechanisms, is often criticized for not being generalizable. Although I consider this methodology to obey to a different scientific logic—capable of generalizing on processes and mechanisms, but not on their distribution or the frequency of their occurrence within a population—this quantitative data will provide a more solid grounding for the study by allowing me to contextualize some of my findings within the general patterns of the population as a whole, and also contrast these with the majority population and other immigrant groups. This data also has the advantage of giving greater validity to some of my findings. Insufficient sample sizes precluded against isolating specific African nationalities in the survey, meaning that Senegalese and Malian respondents are grouped with the rest of the Sahel region, including Mauritania, Gambia, Guinea-Bissau, Guinea, Burkina Faso, Niger and Chad. However, given the random sampling of the survey and the relative size of the Senegalese and Malian communities in France, these two nationalities nearly constitute three quarters of the respondents in the Sahelian sample (Beauchemin, Lhommeau and Simon 2016).38 Furthermore, given that these migrants share significant historical, structural, cultural and religious characteristics in common and were part of the earliest wave of sub-Saharan migrants—overwhelmingly low-skilled rural populations hailing from the valley of the Senegal River—this amalgamation is appropriate for our African respondents.39 The same issue pertains to the data on Antilleans, who for statistical reasons are aggregated into the broader category of overseas departments (DOM)—including Martinique, Guadeloupe, Guyana and La Réunion, where Antilleans represent about two thirds of the sample.                                                            38 See online appendix: http://www.ined.fr/Xtradocs/teo/annexes/Annexes_Chapitre_1.xls 39 Migrants from the Guinean Gulf and Central Africa came to France more recently and are generally more highly skilled than their Sahelian counterparts. 53 The survey data will only be used as a secondary data source for the study, primarily for descriptive and comparative purposes, and for contextualizing and strengthening the findings from my interviews. It will primarily be used to describe the family structure of the populations under study, their intergenerational social mobility, their experiences with discrimination, their endogamy rates, as well as their social identities.  Demographic Characteristics of the Sample The sample includes 55 respondents that are evenly divided by region of origin and by generation. The Antillean sample is slightly tilted toward Martinican respondents while Malians predominate slightly in the second generation sub-sample. Table 1 below summarizes this distribution of respondents.  The demographic characteristics of the fist generation respondents are summarized in Table 2. In the African first generation, a slight majority have their roots in Senegal (57%) and the gender distribution in the sample is balanced (53% male). Most respondents come from a Muslim background (77%) while the rest are Christians. The ethnic backgrounds are varied, but from the most to lest frequent, they identified as Bambara, Peul, Soninke, Jola, Manjack, Wolof, Serer, and Malinke. The sample has a sizeable number of both young and older respondents: 46% are under 35, and 38% are over 50. This age distribution also affects the respondents’ civil status: 31% are single and have never been married, 38% are married and have never divorced, and the remaining 31% are divorced, whether they have remarried or remained single since. The number of divorcees in the sample is also slightly above the population average, and these are  African Mali Senegal Antillean Martinique Guadeloupe First Generation 13 6 7 14 9 5 Second Generation 14 9 5 14 8 6  Table 1. Distribution of respondents by region of origin and generation 54 overwhelmingly women. The average age of arrival in France is 27 years. As I sought to sample for socio-economic diversity, the respondents also vary in their educational attainment: 46% have a secondary school degree or less, while 38% have a university degree.  Most Antillean first generation respondents have their roots in Martinique (64%) and is tilted towards women as three quarters are female respondents. The sample is balanced in age, having both fairly recent arrivals and more established migrants: 42% are under 35, and 42% are over 50. This partly explains why most are unmarried and childless (50%), but a sizeable contingent are nevertheless divorced or separated single parents (29%) whereas only a minority are married or in a stable long-term relationship with children (21%). Even though the age distribution skews the civil status patterns compared to the population averages, the differences between the African and the Antillean sub-sample in this regard remain consistent with the broad matrimonial population patterns. As respondents as a rule came to mainland France at a young age, and two came in late childhood, the average age of arrival in France is only 17.7 years. The educational attainment is varied, albeit slightly skewed towards the well-educated: 36% have a secondary degree or less, while 57% have a university degree. Around a third of the respondents were parents of other respondents.  Among the second generation Africans (see Table 3), most respondents have their roots in Mali (64%), and the sample is disproportionately male (85%). The most common ethnic backgrounds were Soninke and Peul, followed by Bambara, Toucouleur, and Malinke. All respondents were either born in France or came during their first years of life, with the exception of one respondent who came at the age of 12. The  Gender Age Structure Civil Status Education  Female Age at arrival Age Under 35 Over 50 Never married Never divorced Divorced Secondary or less University African 47% 28.8 42.3 46% 38% 31% 38% 31% 46% 38% Antillean 78% 17.7 44.7 42% 42% 50% 21% 29% 36% 57% Table 2. Demographic characteristics of first generation Antilleans and Africans  55 ages range from 21 to 42 years, with an average age of 27.8 years across the sub-sample. Only about a fifth have a university degree, the remaining 80% have a secondary degree or less. About a third of respondents had parents who were also interviewed.      In the second-generation Antillean subsample, 43% of respondents had at least a parent from Guadeloupe, but more than half of the Martinican-origin respondents had mixed backgrounds (mostly with a French parent, but one respondent had an African parent). The respondents are almost evenly divided by gender (57% female), and the ages range from 21 to 42, with an average age of 31.5 years. About 29% have a secondary degree or less, while 43% have obtained a university degree.   II. Challenges and Limitations Positionality and Fieldwork Challenges I anticipated that the study would most likely pose several challenges that were important to reflect upon. The most obvious one perhaps being how my positionality––i.e. my own racial and ethnic characteristics, social class, gender and so on––would impact my ability to build rapport with some of the respondents, gain access to some information or social settings, and to simply be able to collect rich data.  Qualitative researchers that study groups from which they are separated by a considerable social distance have understandably been led to question the quality of the data they collect (e.g. Duneier 2004),  Gender Age Structure Education  Female Age Under 25 Over 35 Secondary or less University African 15% 27.8 43% 14% 79% 21% Antillean 57% 31.5 29% 29% 29% 43%  Table 3. Demographic characteristics of second generation Antilleans and Africans 56 and some have even argued for more or less strict forms of insider epistemology as a result, where 'racial insiders' would have a clear advantage over 'outsiders' in the field (Collins 2000). However, fieldwork experiences of 'racial insiders' do not always lend much support to this view (e.g. Twine 2000), suggesting that these researchers, despite their advantages on some levels, also face other sets of challenges, and that they can even be at a disadvantage on several criteria compared to 'outsiders' (Young 2004). But regardless of whether we are insiders or outsiders to the groups we study, we are never spared a reflexive evaluation of how we are perceived by our respondents, and we need to evaluate to the best of our ability how this may affect the data we collect and how we interpret it. The issue of how I was perceived was always something I took into consideration, and it was clearly in many cases an obstacle. In some cases this seemed directly attributed to race. In one case an Antillean respondent wanted to refer me to a Malian family who lived next door who she thought would be an interesting case for me, but then added that they wouldn’t be willing to talk to me because they “hate whites”. Obstacles like these were of course difficult to overcome. Sometimes, my race was simply a matter of surprise given that the theme of the study led respondents to expect a black person. Mostly this led to polite inquiries about why I chose to study this particular topic, but in one instance an Antillean woman who I came to visit for an interview gasped when she opened her apartment door, exclaiming: “but you’re white!”. Others were more explicitly concerned about my intentions or biases, which would probably not have been an issue if I were a racial insider. When I asked Paul, a 29 year-old Martinican computer engineer if he thought that not being part of the group might constitute an obstacle for studying it, he replied: Yeah, I think so. Yeah absolutely, because when I asked my mate… to contact you, he said “no, we don’t know what he wants to write, maybe his agenda- his political point of view is the complete opposite of ours, and maybe he will use arguments to show that Antilleans isolate themselves, and that they want to remain in their condition or something like that”. So yeah—because it’s true, there is prejudice against the other…  This particular respondent stood out from the others through his more intense “colour consciousness” and high sensitivity towards group generalizations and stereotyping, which might have made him an unusual case. But his friend’s reaction nevertheless suggests that some members of the community believed that 57 my being white meant that I would be likely to inadequately portray them and therefore dissuaded them from participating in the study. Large class divides could sometimes complicate the interviews and my ability to read the interviewees. Although most interviews with young people from modest backgrounds with low levels of education went well, in one particular case it seemed clear that I failed to engage the respondent adequately and dispel the tension that I sensed from the beginning. However, it was hard to tell if the standoffish posture, the brief answers to my questions, and the avoidance of eye contact were due to some sort of hostility, to my inability to create rapport, to the unnaturalness of being in an interview situation speaking about his personal life with a stranger, to the fact that he felt compelled to participate because of the person who referred me to him, or if the social distance between us just distorted my perceptions. The language and culture barrier could also interfere with the interview situation. I conducted one interview with two Malian women who had both been living in France for ten years, Khadija spoke French fluently, whereas her friend Awa barely spoke the language at all. Despite the efforts of Khadija to translate the questions and answers for Awa, she answered very briefly and gave very little material for probing, the whole exchange with her remaining painfully superficial. As this interview occurred fairly early in my fieldwork, it was more difficult for me to ask pointed questions that might perhaps have been more fitting to her specific experience—which in this case clearly illustrates the disadvantage of being an outsider to the group. Another interview was memorable for the problems created by the language barrier. This respondent, Harouna, brought me to a nearby Malian worker hostel, despite not living there himself, into a cramped two-person bedroom filled with 5 or 6 other young men. This interview was particularly strained because the respondent’s French language proficiency was too low to adequately grasp and respond to many of my questions, regardless of how simply and straightforwardly I tried to formulate them. His friends, who were sitting around us throughout the interview (mostly watching a portable television), were either not much more proficient in French or were simply not willing to be of assistance. This experience also suggests another barrier that derived from how I was perceived. After trying to 58 encourage his friends to participate at the end of the interview—mentioning the 20 euro compensation—he failed to persuade any of them. As Harouna explained to me when we left the room to visit the hostel, they were worried about somebody collecting information about them, as many had entered the country illegally or overstayed their visas, and were concerned about the authorities finding out.  Being male may have made it more difficult to speak to more (and perhaps different) women in the African sample, as I had more difficulty recruiting women from this demographic, while I was turning down interviews with men. This might be due to a variety of factors, but should be kept in mind as one of the limitations of the sample.  Although reactions to my race, gender, and social background could obviously not be helped, I think this was partially counterbalanced by the fact that my social identities could come across as “fudged” to many respondents (Lamont 2004). As a Swede who settled elsewhere in France with his parents at a young age, my accent could not be associated to some familiar local demographic that my respondents might have had feelings about, and my non-French name and affiliation to a foreign academic institution made me less conspicuous and more of an outside observer. Judging by how forthcoming respondents were about their own lives and experiences, and how potentially tense situations were in general easily diffused, I believe most interviews were handled quite successfully. The social issue I probably struggled with most was in my attempt to find the right balance between formality and casualness when addressing my younger respondents, at least insofar as needing to choose pronouns: either the colloquial “tu” (used among friends, social equals, and most young people) or the more formal “vous” (appropriate for addressing adult strangers and social superiors). When substantial class barriers were present, I was hesitant to use the “tu” form, even if this seemed more natural to me. When I did do so from the start with some younger respondents, this could sometimes be interpreted as impolite. One respondent who was slightly younger than me—whom I had an excellent interview with and who later convinced a close friend of his to talk to me as well—was discussing over the course of the interview how respect was lacking in contemporary France, and illustrated his point by saying that he wouldn’t just say “tu” to people he did not know out on the street, which I took as a disapproval of my 59 mode of address. In other early interviews, despite accepting my suggestion that we could say “tu” to each other, many continued saying “vous”. I suspect that this unequal form of address was something that many respondents from these backgrounds were not unfamiliar with in circumstances such as these, but it consistently made me uncomfortable and often led me to clumsily veer back to saying “vous”. The tensions surrounding the use of these pronouns only occurred in the context of large class disparities, where I foolishly thought I could bridge such a substantial social divide by attempting to implement an informal mode of interaction (Mauger 1991).  Speaking about racial issues across a racial divide in a society where racial distinctions are illegitimate also led to some peculiar experiences. The middle-class respondents were usually those who seemed most uncomfortable discussing these themes, while those from more ethnically diverse working-class areas in the suburbs were far more used to it and relaxed about it. However, the spontaneous use of explicit racial categories was quite rare. The most common was to use more socially acceptable euphemisms, like the more neutral English word “black” or the slang word “renoi” instead of Noir, and the Antillean category of métropolitain or français de souche for “white”. When somebody used the term blanc it was not uncommon that they prefaced it with some apologetic signal, marking its illegitimacy. One Antillean woman used the term very liberally, but did so deliberately to defy what she saw as an unfair double standard in how the terms ‘black’ and ‘white’ were used: “When I say ‘white’, the person I’m facing will feel offended because I say ‘you’re white’—but you call me black, why am I not allowed to call you white?”.  There were probably some advantages as well in being an outsider. First of all it was not assumed that I knew, putting me into the role of the "socially acceptable incompetent" (Lofland et al. 2006) which allowed me to ask a broader set of questions and access taken-for-granted information less problematically. Furthermore, since the relationship between Africans and Antilleans was a central aspect of the study, not being part of either group was a better position to be in when asking about their group relations. I was often concerned that I might be seen as a meddling, divisive figure that would seek to ‘divide the black 60 community’ with my line of questioning, but this was not my experience at all. In most cases people seemed fine expounding on this, unless they had no experience with it or had not given it any particular thought. Social distance between researchers and their respondents is a reality in all empirical research, and even though it clearly varies in degree, its complexity cannot be circumvented by simplistic reifications of the social world. Researcher positionalities provide both opportunities and limitations, and to some extent it is possible to maximize the former and minimize the latter through appropriate reflexivity, self-presentation and relational skills. Some individuals might be at a greater advantage than others for undertaking a given study, but it cannot simply be predetermined by simple reference to abstractions such as race, gender or class, especially when researching in different national and cultural contexts. Indeed, race, class and gender are culturally laden and relational categories that vary in salience and configuration depending on the context, and the fact that a foreign researcher happens to share a number of these 'objective' attributes doesn't necessarily mean they will immediately be accepted by the population they study or have insight into their experiences. In this regard, my familiarity with French society and its cultural codes was perhaps a greater ‘advantage’ in my research—at least on the second generation—than the ‘disadvantage’ that may have come from not sharing the same 'racial' attributes as my respondents.  Limitations of the Study There are limitations to the study that are important to consider. The most important criticism that can be levelled towards the study is that there is a disequilibrium at the heart of the research design, which largely derived from the nature of my research questions. Given my ambition to study the general identification patterns across both populations and generations, where a variety of class backgrounds were sought, I was led to recruit several childless first-generation participants who were more qualified and had arrived more recently than the parents of most second generation respondents I interviewed. This, in turn, was detrimental to the study of assimilation mechanisms, as it diminished the number of first-generation respondents that could shed light on family patterns that were salient for second generation respondents. 61 Furthermore, identifying the salient assimilation mechanisms within these communities sometimes necessitated an exploration of some cultural factors that seemed central in the interview data. I ventured into this area with the utmost care, and supported any tentative conclusions in previous research, but being an outsider here might clearly have been a disadvantage as well. Being able to study more respondents within the same family, and/or more first generation respondents that were perhaps more representative of the parents of my second generation respondents would have been more helpful for solidifying these findings.  The broad ambitions of the study may also have come at the price of losing depth in the various subsamples. Studying two diverse geographical areas, broken down into two generations, and also stratified by class and gender provided useful data regarding the general patterns within the population, but obviously made it difficult to go into greater depth into the specific case of say middle-class second-generation Antillean women as the samples become so small. This could of course have been remedied with a larger sample, or with a modified study design. As mentioned earlier, the second-generation African sample in particular suffers from a greater gender imbalance than the others, due to a greater difficulty in recruiting young French-born African women. Whereas the preponderance of males in this sub-sample might have been useful for understanding mechanisms of downward mobility; it had the disadvantage of putting a lesser emphasis on young women’s experiences. Fortunately, the few interviews I did conduct with this demographic were of good quality, and I also often asked male respondents about the trajectories of their sisters.   Another limitation is the absent voice of the mainstream population from these group dynamics, despite it being present in all the experiences of the interviewees. Interviewing white French people could have better helped assess how they conceptualized ‘otherness’ and ‘Frenchness’ and what their social inclusion and exclusion of these two population groups hinge upon, which would have been useful for making a more nuanced assessment of how boundaries were drawn. It could also have provided an account of whether the mainstream population feels framed in a particular way in their social interactions with these two populations.  62 Finally, given that segmented assimilation is notorious for its hypothesized ‘downward assimilation’ path, it is worth pointing out that the study is not designed to empirically ‘test’ the specific modalities that putatively drive ‘downward assimilation’ among these communities. This would essentially have required a male-only sample exclusively drawn from the most socially disadvantaged parts of the African and Antillean populations in France, with a heavy ethnographic focus on the social dynamics of one, or a few specific areas. Several studies with this general type of focus already exist (e.g. Lapeyronnie 2008; Mohammed 2011; Wacquant 2008), although they lack these explicit theoretical ambitions and rarely focus on the specificities of different communities such as those that I am interested in here. By contrast, this study aims to conduct a broader investigation into the assimilation mechanisms, social identities and group relations—across the class spectrum, in two generations, and for both genders—in order to identify the role played by my respondents’ racial, cultural and religious backgrounds in these social processes. Given this interest in social characteristics that contribute to erecting structural barriers, the study will investigate the relevance of some of the key claims of segmented assimilation theory for interpreting the findings, but does not have the ambition to rigorously ‘test’ the mechanisms it hypothesizes for its ‘downward path’.   63        PART TWO ASSIMILATION TRAJECTORIES      64 5. THE FRENCH CONTEXT OF RECEPTION   As a necessary preamble to analyzing the assimilation patterns of my respondents, this section provides an overview of the social, economic and institutional context that immigrants are faced with when they settle and raise their children in the Paris region, with a special focus on the contextual variables emphasized by segmented assimilation theory.   I. Economic and Social Transformations Economic Restructuring and Social Mobility The restructuring of the economy and the limited economic growth in France since the 1970s has deeply affected low-skilled workers in France. Between 1980 and 2007, 36% percent of the workforce in manufacturing was eliminated, which amounts to 1,9 million jobs (Demmou 2010). In the Paris region alone, 650 000 manufacturing jobs disappeared between 1975 and 1988 (Mohammed 2011). The democratization of schooling in the 1980s, and the corresponding inflation of degrees has also left young working-class males with limited education much less well-adjusted to the labour market, which has furthermore become dominated by a low-skilled service sector and highly credentialed professionals and managers.  Simultaneously, attempts to reduce unemployment have primarily consisted in multiplying temporary contracts while maintaining the stringent protection that long-term contracts benefit from, producing a two-tiered labour market that overwhelmingly penalizes the young (Le Barbanchon and Malherbet 2013), who are now two and a half times more likely to rely on temporary work than they were 25 years ago (Peugny 2013). Moreover, these shifts in the economy and the labour market have not been accompanied by reforms to the French corporatist welfare state which has increased inequalities by 65 privileging established occupational and statutory groups over the poor, and has primarily penalized the young as well as immigrants and their descendants by making their access to the labour market more difficult (Smith 2004). Although social mobility in France was high during the four decades following World War II, most of it occurred between the 1950s and the 1970s, and it has largely stabilized since 1977 (Peugny 2013). Whereas the impact of socio-economic backgrounds on mobility prospects decreased for those born between 1930 and 1960, it has rather increased for following generations (Lefranc 2011). Furthermore, both downward mobility in the intermediary and higher occupational categories and stagnation in the working class has become more widespread compared to the baby-boom generation (Peugny 2007). The structure of the French educational system has also been shown to contribute to social segmentation by reinforcing the advantages of families with greater economic and cultural capital. In addition to a below OECD-average public spending on primary education, French families also spend the most on private tutoring in Europe, the access to which is of course economically skewed. Additionally, the two-tier higher education system—with an elitist track constituted by the Grandes Ecoles and a non-elitist mainstream university system—increase the education gap between a small elite and a middle class seeking higher education, and further hampers social mobility (Ben-Halima, Chusseau and Hellier 2014).  Urban Decay and Social Marginality The deleterious impact of these social and economic policies are most visible in working-class neighbourhoods in the periphery of large urban areas––often referred to as cités––whose residents were typically employed in large manufacturing industries. Starting in the mid-1980s youth unemployment in particular increased drastically in these areas, and immigrant households were typically hit first and hardest in this process. In addition to the economic and social impact of deindustrialization, a series of housing policies in the late 1970s also encouraged established residents to move out from these areas. These residents often 66 played a central role in socializing immigrants into French working-class life, through their activity in trade unions and the communist party, whose sharp declines in membership also hit two important pillars of immigrant integration (Hargreaves 2007). As many moved out, vacant apartments were typically sought out by low-skilled immigrant newcomers, further increasing concentration of ethnic minorities (Masclet 2003). Through a complex and drawn out process on several levels of social life—economically through joblessness and welfare dependency; culturally through ethnic concentration and the loss of social and cultural embeddedness in the French population; socially through the deterioration of schools and increased crime rates but also through the harmful effects of high turnover rates on social capital—these neighbourhoods have gradually developed various forms of social marginality and stigma. The social problems generated by these developments have largely perpetuated themselves over time despite state intervention and have even worsened after the economic downturns of past decades. Today, the 1296 neighbourhoods in the country that have been singled out for urban renewal policies (which concentrate 4.8 million people, i.e. 7.5% of the country's population) recorded an average unemployment rate of 26% among the population aged between 15 and 64, while it is only 10% in other neighbourhoods. Youth are clearly the most affected group, as 36% of the population of these areas under the age of 30 are NEETs, compared to 17.4% in the rest of the country. Although the least qualified have the highest levels of unemployment, and that their unemployment rates do decrease with education, residents of these areas with over two years of university education nevertheless have unemployment rates that are three times higher than their counterparts in the rest of the population. Furthermore, over two thirds of the inhabitants in these urban areas receive some form of welfare benefits, compared to 45% of the population of metropolitan France, and 64% of welfare-recipient households in these areas are characterized by low incomes, compared to 33% in metropolitan France (ONPV 2016). Despite the state's continued investment in these urban areas, the fragile mechanisms that previously ensured the incorporation of immigrants into the mainstream working-class have been substantially damaged. In many of these areas, where school drop-out levels and youth unemployment rates are two to three times higher than the national average, a handful of disenfranchised youth without 67 qualifications and prospects on the labour market turn more or less intensively and durably to criminal activities and drug trafficking, rejecting dominant norms and institutions. Nowhere is this revolt towards the surrounding society more obvious than in their tense relations with the local police, often considered as a rival gang (Mohammed 2011; Sauvadet 2006). In some areas, this often proportionately small group of youth have successfully taken control of public space and largely imposed their norms of collective life on the neighbourhoods they inhabit, and have thereby in effect inverted the social and generational order (Bronner 2010; Lapeyronnie 2008). It might thus not be wholly inappropriate to speak of the existence of an 'oppositional culture' in many of these areas, which could potentially affect the assimilation trajectories of the children of immigrants who are already faced with the many social, economic and institutional disadvantages in these poor urban areas.  Sociological comparisons between the impoverished U.S. inner-cities and the French cités have stressed the many deep differences between these two urban formations, despite the many surface similarities they may share. Loïc Wacquant (2008) points out that both have been ravaged by deindustrialization and mass unemployment, have high concentrations of ethnically marked populations, similar age structures and household compositions, and that they both share a similar bleak and oppressive atmosphere and have salient stigmas associated to them. Despite these similarities, Wacquant maintains that the French cités can not be compared in sheer size, degrees of poverty, criminality, or ethnic homogeneity to dilapidated U.S. inner-cities. Furthermore, he also points out that in contrast to the United States, the state hasn't completely deserted these working-class areas in France, which are rather the target of ambitious urban renewal programs. Although Wacquant is correct in pointing out the flaws in these frequent comparisons, the experience of poverty, crime and general insecurity is a relative phenomenon, and it is precisely this lived experience and the meaning attributed to it that shapes the prevailing attitudes towards these areas and their inhabitants, not objective differences in degree or in scale compared to other societies.    68 Territorial Divides and Segregation More broadly, at the level of the Paris region as a whole, the formation of these urban entities has been accompanied by other patterns of disaggregation which have generated and increasingly polarized social landscape. A recent report from the Institut d'Aménagement et d'Urbanisme d’Ile-de-France40 has shown that over the past thirty years, the increased number of immigrant households has led to an extension of their geographical spread and a diversification of their social backgrounds. In fact, the proportion of immigrant worker households diminished by 10% since 1982, their share among all immigrant households dropping from 44% to 25% in 2006. The geographical distance has decreased between immigrant households from all occupational groups with the exception of immigrant managerial and professional households which have remained about the same. However, by contrast, the geographical divide between occupational categories has increased among French households since 1982, as managers, professionals and business owners are now far more removed from blue-collar and service workers than they were three decades ago. Furthermore, the residential distance between French and immigrant households has increased across all categories over the same period with the exception of immigrant managerial/professional households, who have actually come closer residentially to all French households. The French working class have primarily distanced themselves from immigrant worker households, but this distanciation is less stark than it is for other occupational categories in the French population41 (IAU-IDF 2013). As Figure 5 illustrates, the pattern we see appearing in the broader Paris region is a clustering of managerial and professional households with little immigrant/French-born differentiation in central Paris and in the inner and outer western suburbs, a clustering of working-class immigrant households in the northern inner suburbs, while French workers largely aggregate in the outer periphery of the metropolis,                                                          40 This Institute of Planning and Urbanism of the Paris Region 41 These patterns have been assessed by calculating ‘indexes of dissimilarity’ that measure “the proportion of households of a given group (immigrants, occupational category…) that would have to change communes in order to obtain the same geographical distribution as the group taken as a reference (all households in the Paris region, French-born households, etc.)”. The highest indexes of dissimilarity in the region are found between native managers/professionals and immigrant workers (0.464), between native managers/professionals and native workers (0.445) as well as between immigrant managers/professionals and native workers (0.444). The index of dissimilarity between native households as a whole and immigrant worker households increased from 0.220 in 1982 to 0.328 in 2006, suggesting an avoidance of stigmatized suburban or social housing areas with large concentrations of immigrants. 69  Figure 5. Geographical distribution and concentration of French-born and immigrant managerial/professional and worker households in the Paris Region in 2006  70 primarily in the east. Although this certainly highlights the spatial separation between the highly-skilled French population and the working-class immigrant population, it also shows the limited influence of immigrant status on residential divides within the higher occupational strata and the spatial convergence of immigrant households with high levels of qualification and the rest of the French population. It also points to the increased class polarization within the French population and the greater separation between the French and immigrant working-class compared to higher occupational categories.42 Although one should be cautious of making an ethnic reading of these figures—as the native category also includes several second-generation immigrants born with French citizenship—the Institute’s separate mapping of the geographical distribution of children of immigrants (see Figure 6) shows that they are primarily concentrated in the inner suburbs, especially in the north and south.                                                            42 These patterns also lend credence to analyses that emphasize the increasing economic, cultural and political divide emerging between large metropolitan areas that concentrate a high-credentialed native population and a low-skilled immigrant-origin population that reap the economic rewards of globalization, and a more peripheral and economically disenfranchised class of low-skilled natives that have been pushed out to more economically stagnant and peripheral small-towns and rural areas (Guilluy 2013).   Figure 6. Proportion of children who live in an immigrant household in the Paris Region 71 A closer look at the ethnic disparities among immigrants further nuances the picture. In the Paris region, Turks and sub-Saharan Africans remain the most spatially segregated groups. Although segregation rates of sub-Saharan Africans and Maghrebis decreased slightly between 1975 and 1999 (Safi 2009) they increased between 1999 and 2006 (IAU-IDF 2013).43 Many studies on the topic unfortunately suffer from not adequately controlling for class backgrounds, but as Préteceille (2009) has for instance pointed out regarding DOM-TOM migrants, their segregation index as a whole in the Paris region is higher than that of unskilled industrial workers (the most segregated occupational category), despite that fact that only 16.6% of the DOM-TOM population in the metropolis were unskilled industrial workers.44 Disfavoured neighbourhoods are typically avoided by all groups—suggesting that there is no uniquely ‘white flight’ from these areas—but when controlling for social and demographic variables, it nevertheless remains much harder for sub-Saharan Africans to leave poor urban areas than other groups, and they are four times more likely to settle in these neighbourhoods than the French mainstream population (Pan Ké Shon 2009). The sub-Saharan African immigrant population in the Paris Region has increasingly spread out geographically over time, but mainly within the most socially disfavoured municipalities, and the growth of the sub-Saharan African population in these municipalities has coincided with a decrease of the French-born population over the same period (IAU-IDF 2013). Regarding immigrant concentration levels as a whole, the proportion of immigrant-origin populations in some neighbourhoods has grown exponentially, as in Clichy-sous-Bois where it has risen from 22 to 76% between 1968 and 2005.45 Although only 25% of immigrant-origin youth under the age of 18 in France were of non-European descent in 1968, they represented 75% of this youth in 2005, primarily from North and sub-Saharan African backgrounds (Aubry and Tribalat 2009).                                                            43 Lagrange suggests that very different patterns may characterize Sahelian and non-Sahelian Africans, the latter typically being more skilled, and thus probably more geographically mobile than the former. 44 The explanatory power of class in accounting for these territorial concentrations seem to have decreased over time, as the socio-economic characteristics of immigrant households explained 38% of their ethnic concentration in the Paris region, but only 25% in 2006 (IAU-IDF 2013). 45 These figures only include the children born of one or two immigrant parents, and therefore don't include third generation immigrants or children born of immigrants who have obtained French citizenship. 72 II. Discrimination and Public Attitudes Discrimination and Stigma  In addition to these processes of segregation and concentration of ethnic minorities in poor neighbourhoods, many African-origin French citizens are affected by greater socio-economic exclusion than other groups. The odds of unemployment among North African and sub-Saharan African second-generation immigrants is 2 to 4 times higher than the majority population, and this discrepancy remains after accounting for education levels, social background, age and family composition. Second-generation immigrants also experience higher job insecurity and rely greatly on subsidized employment schemes compared to the mainstream population (Meurs, Pailhé and Simon 2008). A "testing" study of discrimination towards children of North African and sub-Saharan African immigrants in low or medium-skilled occupations in France has shown that they are, at similar levels of qualification, about three to four times less likely than their peers of French descent to receive the same number of positive replies from their attempted contacts with employers (Cediey and Foroni 2007).  Even though highly-skilled second generation immigrants typically face lower levels of unemployment than their peers with lower qualifications, they also abandon their studies before obtaining a diploma at 2 times higher rates than their peers from the majority population, and have more precarious forms of employment with lower salaries. They also often occupy positions they are overqualified for, which is especially true for women (AFIJ 2007; Beaud 2003; Brinbaum and Guégnard 2011; Silberman, Alba and Fournier 2007). Self-reported discrimination rates are also high for non-European origin populations. Typically between 20-25% of North and sub-Saharan African immigrants and their children report having been unfairly refused a job in the last 5 years, as opposed to only 5% of the majority population; these levels reach over 30% for individuals of African origin when they have been unemployed in the past 5 years, which is 20% higher than the equivalent figure for the majority population (Lhommeau, Meurs and Primon 2016). For these African-origin groups, the cumulative stigma of class and race is also reflected in the fact 73 that they are two to three times more likely than other non-European-origin groups to report their place of residence as one of the reasons they are targeted by discrimination (Silberman and Fournier 2008). Furthermore, many African-origin youth––especially when male and socialized in poor urban areas46—are commonly subjected to 'identity checks' by the police based on their physical appearance and their demeanour, and are often exposed to stigmatizing encounters with the surrounding society (see e.g. Lapeyronnie 2008). This daily-life experience of being the carrier of a set of potent stigmas feeds the sense of rejection from, as well as the concomitant rejection of the surrounding society (see Anderson 2008). Even though most immigrants in France typically hail from rural and low-skilled backgrounds, and that this has a strong and direct impact on the differences in socio-economic achievement of second generation immigrants in France, racial characteristics and ethnic origin are thus far from irrelevant in their social experiences and their mobility patterns, which is in line with what segmented assimilation theory would predict.  Public Opinion As members of the majority population are not part of this study, and their experiences and attitudes play a central role in the assimilation process, the broad trends in public opinion can be useful for assessing the barriers toward boundary blurring between the majority and the groups under study.  The Commission Consultative des Droits de l’Homme (National Consultative Commission on Human Rights) produces annual reports for the French government on racism, antisemitism and xenophobia in the country. The surveys draw from representative samples of the mainland population, which may therefore not be able to directly address the views of the population of French descent, but has the advantage of presenting the general climate of ethno-racial relations in the country and how it evolves over time.                                                           46 Although both genders are affected by unemployment, lower wages and being overqualified for their jobs, young women are much more likely to succeed in their studies, be employed and be more socially and geographically mobile than their male counterparts (Lhommeau, Meurs and Primon 2016). 74 In 2013, three quarters of respondents considered that there were too many immigrants in the country (56% in 2015; henceforth, unless otherwise mentioned, all figures in brackets are from 2015),47 58% saw immigration as the prime cause of insecurity (42%), and 77% believed that immigrants only came to France to access social benefits (60%). Furthermore, only 22% agreed with the statement “it is primarily French society that doesn’t give people of foreign origin the means to integrate” (24%), whereas 68% agreed that “it is primarily people of foreign origin who don’t give themselves the means to integrate” (49%). With regard to children of immigrants born in France, 35% of respondents considered them as not really being French (26%).  The general sense of group relations and social cohesion is quite strikingly negative, as 83% of respondents considered in 2013 that there were tensions between the populations of different origins that compose French society (55% considering that these populations lived in parallel to each other, and 28% that they lived separately), and 60% agreed with the statement that “one doesn’t feel at home in France as one did before” (50%). With regard to racism, 84% consider it to be widespread in France (85%) and although 38% consider that nothing can justify racist reactions (40%), 61% maintain that certain behaviours can sometimes justify racist reactions (59%). To the open question of who the primary victims of racism are in the country in 2015, more than half mention either Muslims, Arabs or North Africans, around 19% mention Blacks or Africans, around 23% broadly mention foreigners or immigrants, and 22% mention French people or whites. With regard to self-reported racist sentiments, 39% of respondents declared that they were “not at all racist” in 2013 (53%). This view was most commonly held in the Paris Region, in neighbourhoods with more than 9% foreigners, among managers and professionals as well as the most educated segments of the population. Despite all this, 71% held that one could succeed in France regardless of one’s skin colour (74%).                                                           47 The figures from the 2013 report are among the most dismal for the state of ethno-racial relations in the country, whereas the opposite was true for the 2015 report, despite the two terrorist attacks that took place in the capital that year. Even though these trends were foreshadowed in 2014, there are reasons for concern over the reliability of the most recent figures, as the attacks affected the data collection for the report, which was moreover conducted by two different institutes over two different periods. I nevertheless report figures from both years to give a better sense of the spectrum and fluctuation of opinion on these issues. 75 Regarding the perception of the extent to which immigrant groups cut themselves off from the mainstream, 23% of respondents considered Blacks to be a separate group in society (14%), 31% that it was a group that was open to others (39%), while 46% considered that they didn’t particularly form a group (45%). In 2015, the same question was also asked about Antilleans, showing that 9% saw them as a separate group, 41% saw them as a group open to others, and 43% saw them as not particularly forming a group, suggesting that Antilleans are seen as being less distant from the mainstream than Blacks in general.  The attitude towards Muslims seems double-edged, as a large majority sees Muslims as ordinary citizens while simultaneously feeling ambivalent about Islam. Although 76% of respondents agreed that French Muslims were just like any other French citizens, 49% nevertheless saw Islam as a threat to French identity in 2015 (CNCDH 2014; CNCDH 2016). Other poll data on the public perception of Muslims confirm this ambivalence. An IFOP poll from 2012 for instance found that 67% of respondents thought that Muslims and people of Muslim origin were not well-integrated, primarily putting the blame on cultural differences and a refusal on the part of Muslims to adapt (IFOP 2012). In another Le Monde/IPSOS survey from 2013, 74% of respondents described Islam as an intolerant religion which is incompatible with the values of French society, and eight out of ten respondents believed that the religion sought to “impose its mode of functioning on others” (Le Bars 2013). As this overview suggests, there are deep structural and sociological fault lines in the Parisian metropolitan area which are in line with many of the structural assumptions of segmented assimilation theory regarding barriers to labour market integration, dilapidated residential areas with adversarial cultures, and negative public opinion towards immigration and Islam. These factors are likely to play an important role in the assimilation trajectories of immigrants and their offspring.    76 6. MIGRATION AND SETTLEMENT   Although both Antilleans and West Africans started migrating to mainland France in large numbers in the 1960s and were primarily driven by economic motivations, there are some significant differences in the migration motivations of both groups. These primarily derive from the differences in living standards and local economic opportunities, as well as the differential access they have to mainland France. Antillean respondents were also more typically driven by individual economic aspirations and were more oriented towards the host society, whereas African respondents had more collective aspirations and a greater orientation towards their societies of origin. Many African respondents also settled in France in order to send remittances back home and support their communities of origin. Among respondents who migrated more recently there were several examples of more high-skilled migrants driven by a desire to advance their careers, especially among Antilleans. Despite the barriers that both groups may face in their settlement experiences, there are nevertheless clear differences in their magnitude. Although both groups lack a formal ethnic economy that they could potentially benefit from in France, they differ significantly in their involvement in ethnic associations and their orientation towards the region of origin, with Antilleans seeming to register lower degrees of social capital.   I. Backgrounds and Motivations African Backgrounds and Motivations In contrast to the Senegalese in New York for instance who mostly come from urban backgrounds seeking adventure and opportunity (Kane 2010), most early-wave Sahelian migrants to France were rural Soninkes 77 and Peuls from the valley of the Senegal river whose migration was a collective enterprise. They often came to France at a young age, and were only able to pay for their trip through the financial support of their kin. In order to reimburse their debts to their family and to those who looked after them in the worker’s hostels where they typically settled, as well as assist their fragile rural communities of origin, they were expected to begin sending funds back to their family as soon as they found employment. Over time, they were expected to regularly return to get married, if possible several times, and father multiple children, with the aim of moving back home in old age, having achieved the desirable status of a wealthy polygamous patriarch who had provided for the well being of their whole community (Barou 2011; Timera 2001). This traditional migration pattern has become more diversified over time however, as the respondents in our sample illustrate. They are led to relocate their lives to France for a variety of reasons, which primarily depend on whether the migration is driven rather by necessity or by opportunity and if it is initiated by the individual or the household. Moussa provides a typical example of a migration pattern that is household and necessity oriented. He comes from a poor rural area in Mali, has two young children, but was also expected to provide for his five sisters and his mother. He learned various trades in Mali, but these were insufficient to provide for the needs of the family. Several family members—including his own parents—had worked in France and been able to send remittances back home. This led his family to decide that he would try to work in France to help support them. Moussa arrived in France on a temporary tourist visa and had been residing in the country illegally for seven years at the time of the interview. He was hired by a construction company for 8 euros per hour, and tried to send at least half of his earnings back to his village every month: “With five sisters plus my mother, it’s impossible to live well in our country, somebody has to go abroad to show solidarity with the family… even if it is illegal”.  The household and necessity can inform the migration decision at the other end of the socio-economic spectrum as well, as illustrated by the case of Gérald, a 51 year-old who worked as a philosophy teacher in Senegal before moving to France a decade ago. Despite his general satisfaction with his life circumstances in Senegal, his daughter suffered from a rare genetic condition that was impossible to treat 78 in the country, which led his wife to seek treatment for her in Paris. As her condition requires specialized interventions on a regular basis, Gérald quit his job in Senegal to join them in Paris, where he is now trains employees in a company: I didn’t intend to migrate, but the reality of the medical system—considering the health situation of my child and my family—was such that I was a little compelled to come, because over there the health system isn’t developed to the point that it can deal with or at least treat such rare pathologies. So that’s what determined that we came here to France, for the health treatment. And then the situation of my wife here, to meet her needs I was compelled to join her here so the family could live in better conditions. There is also a subset of respondents who have come to France through marriage or through family reunification, such as Fatima, a 60 year old lab assistant from Mali who ended up joining her husband after he had settled in France while she was finishing her degree in biology in the Soviet Union, or Awa, who came to France from Mali as a 17 year-old through an arranged marriage. This migration pattern was most common among female respondents. There are cases in the sample where migration is far more individual and opportunity-driven, essentially consisting in looking for a better and more appealing life. Ousmane came to France from Senegal as a 22 year old: “I said to myself, like everybody else, like every young person who gets ahead, I said ‘why not me’?” Despite staying in France illegally for several years, he has now obtained French citizenship and finished a training program as a plumber. Marietou was working in the health service of the Malian army, and simply saw an opportunity of bettering her life: “I came to France for a month-long leave, and as all Africans, believing France was the Eldorado, I stayed… thinking I would have more comfort. I didn’t think about… the difficulties I could run into”. She also emphasises how this decision was her own: “as opposed to many people from my country, my decision was personal. My decision to come to France was my own, I wanted it, and the decision to stay was my own too.” Patrick, a 32 year-old basketball player from Senegal, was driven to migrate by a desire for professional advancement as well as to travel and see the world:  Our dream, in Africa, is to travel, you know, and go where what you do pays well you know, For instance, for sports in Senegal—football, basketball—as far as training goes it’s all good, but financially, the teams don’t have any money you know, so our dream is to give all we’ve got so we can have a contract with either the Arabs—over there they pay well—or in Europe; and so that’s our dream… I was competing in Senegal, then I had a contract with a Spanish team, I came, they housed me, fed me, all that, and then they couldn’t any more you know. So I couldn’t stay there, because I didn’t have family there, so I came here.  79 Some respondents in the sample were also sponsored by their families to study in France. Mamadi, a 50 year-old security manager in a museum, came to Paris from Senegal as a student in his early twenties: “my father could afford to help me to pursue my studies… he saw that I studied well—and he had already travelled, he had been to Europe with his friends—so they talked [about it] and they said ‘you can send him’, so I came.” Another case is Ibrahim, a 33-year old from Mali who had studied economics for four years in Algeria before applying for enrolment in several Master programs in France. He was admitted, and met his wife during his studies. He now works as a neighbourhood planner in Paris. As these vignettes suggest, the motivation for migration among first generation migrants are quite varied and depend heavily upon the socio-economic circumstances before migration and the perceived opportunities in France and the social networks they can make use of there.   Antillean Backgrounds and Motivations Since the departmentalization of Martinique and Guadeloupe in 1946, there is no particular obstacle for Antilleans settling in mainland France. Older respondents in the sample came through economic necessity in the 1960s and 1970s while the younger ones are more driven by a desire to pursue their studies or advance their career. Marcel is the oldest respondent in the sample. He came to mainland France from Martinique in 1952, where he had struggled to make ends meet as a dock worker before seizing the opportunity to seek employment on the continent: “I had the opportunity to come here, and given the difficulties I was having there, I came. Since I came during the trente glorieuses,48 I found work right away. And I’ve worked all the time, until I retired.” Martine, a 60 year old administrative clerk from Guadeloupe, came to France with her husband to seek employment in the late 1960s. She sets herself apart from those that came during the same period through the Bumidom policy:49 …the people that came before benefitted from what they called the Bumidom... to work in certain areas: hospitals, the postal service, the subway, the national railway a little bit, and also in construction, so they knew                                                           48 The three-decade post-war economic boom. 49 The administration that organized the migration of DOM workers to France from 1963 to 1982 (see chapter 2). 80 what they were coming for, there were no surprises. I wasn't coming in that framework, I chose to come, and chose where I wanted to work, it's completely different. We spoke about it with some people who came during that period, yes, they knew they had come to work in the hospital or in the postal service or in the subway... But it's true that most were also people who hadn't really had very much schooling, so they were looking at working as hospital agent who do the cleaning, or perhaps as nursing aids. Other respondents came through yet other channels, such as Lucien, a 60 year-old retired woodworker from Guadeloupe who came to mainland France through his military service: “The army was mandatory at that time. So even if you didn't want to, you had to do it... I was lucky to come to the mainland, but there were some that went to Guyana or Martinique... And after the service, you had to fill out an application to stay and work.”  Although the gap in living standards has decreased between the islands and the mainland over time, the greater educational and professional opportunities still motivated many of the younger Antillean respondents to move to mainland France. Eugénie came from Martinique as a 20 year-old after discussing her educational options with the administration of her high school: “I was looking for something and they told me ‘but they have that program in Créteil.’ I said ‘but that's 8000 kilometers from here,’ and they said ‘yes, but with [policy of] territorial continuity, whether it’s at Créteil, Valenciennes, or Marseille, it doesn't make any difference.’” Josette had finished a Master degree in biology in Guadeloupe and had done an internship in Canada before she considered moving to mainland France. She was primarily dissatisfied by her labour market opportunities back home, and thought moving to mainland France would be beneficial for her career. She has now found employment in Paris in a completely different field—web traffic management—and is considering staying in her job for at least a couple of years. When I came here, it derived from a need - the need to change air, the need to find something I'd like to do. I was teaching, I wasn't necessarily happy, well, I didn't see myself doing that for the rest of my life. And I was in a kind of precarious situation, I was mainly working undeclared, so it wasn't really—at some point you want some stability... When you have a certain level of education you expect things to unfold a certain way. As the experiences of these younger respondents illustrate, their migration decisions are undertaken with the goal of increasing their human capital or furthering their careers. The geographical distance is largely compensated for by the institutional integration of the islands in the national fabric as well as specific perks 81 like congés bonifiés, i.e. fully reimbursed plane tickets every three years for public sector employees and their families, which substantially modify the motivations and attitudes towards migration.     II. Settlement and Ethnic Communities Settlement Experiences Both Antillean and African first-generation respondents faced many similar challenges and barriers during their settlement process in France, but Antilleans were generally more favoured by their high rates of public service employment, as employers were responsible for providing them with housing.  Both African and Antillean migrants benefit from earlier migration waves insofar as they usually already have friends and family members in the capital which will typically house them for short or long periods while they seek employment and alternative housing opportunities. Given the frequent use of existing networks, they often initially settle in areas where they already have family members before looking for other housing opportunities. Few respondents have experienced overt forms of discrimination when looking for housing, and in the public sector many employees are offered housing as a matter of policy, but many sense that they are treated differently in the open market. Gérald, from Senegal, explains: …in the private sector we had enormous difficulties finding housing. As soon as we applied, when the person met us they told us ‘no, it won’t work out’ but we would never really know why… I thought that the reason we might have had so much trouble finding housing—because we were rejected several times—could be because of our origin, or because of representations they might have had that we wouldn’t be good tenants.  It was much easier for Gérald and his wife to find government-subsidized housing, and his latest experience looking for an apartment after separating from his wife had been better, but he relied on a work colleague who knew the property owner and had recommended Gérald to him. Ibrahim also had trouble finding housing when he was about to start his Master program. When asked if he had the feeling he had been discriminated against when looking for housing, he responded: “Well that’s for sure, it’s a secret to nobody here.” But just as Gérald it was far easier for Ibrahim to find 82 housing through government-sponsored student housing rather than in the free market. He thinks Africans are particularly disadvantaged because of cultural differences: It’s much more complicated in the housing market—and especially the rental market—for an African… I found that for many of my friends who were here, who wanted to find an apartment, it wasn’t easy… Maybe culturally they consider that we Africans aren’t good with upkeep, or aren’t very- and there was that other factor that they would blame Africans for generally: that we were too loud, because when one had an apartment everybody would meet up at his place, and we would talk late into the night, we would laugh, we would do all sorts of things, and be very susceptible when somebody would say “be quiet”… or “make less noise”. I don’t know, maybe it’s cultural, but it’s also due to the fact that few of us managed to find an apartment where we could get together from time to time… Antilleans respondents who worked in the public sector were provided housing by their employers. Christophe explains how this process worked for him in the national telecommunications company: “the housing when I started working was all France Télécom… you have a service for that… you fill out a form, [you say] where you want to be in Paris… and then you wait until something is free, and generally it frees up pretty fast”. By contrast, Martine, who mainly sought to work in the private sector had more difficulties finding suitable housing when she first arrived from Guadeloupe. She thinks it is mostly due to the strong demand for housing rather than pervasive discrimination however, at least in government-subsidized housing: “I think that even with the best of intentions, there isn't enough social housing to absorb the demand for it, so of course... there are a lot of people who cohabit, who are roommates... In the Paris region, it's difficult, very difficult.” Pétronille, a 29 year-old Martinican woman who studied biology with Josette in Guadeloupe, also moved to France after being unable to find work in the Antilles. She remembers that finding housing in Paris was particularly difficult, and that the fact she was Antillean could be an obstacle: “I got a job in an international company… I began looking for housing, and I found out that I was Antillean and that it was a problem—people didn’t want to rent to Antilleans because we didn’t have any guarantors here.” In the end, Pétronille was helped by the Association des Jeunes Travailleurs (Association of Young Workers) who eventually helped her find government-subsidized housing in the 15th district: “It’s really ‘relations of stretched out hands’ that allowed us to make ourselves a place here, because otherwise it was checkmate, we’d have gone home right away.” Pétronille had a similar experience when buying furniture for her 83 apartment, which she wanted to pay in several installments. After providing the furniture company with her bank identification, she was told that the fact that her bank had its address in the Antilles was going to be a problem. Her bank in Martinique helped her create an account in Paris, and also told her that there were companies that used software to filter bank addresses and would discriminate against Antilleans because they were known as bad payers.  Ethnic Communities and Solidarity Ethnic economies are thought to play an important role in providing employment opportunities for immigrants and their children when such opportunities are lacking in the mainstream, which can at best provide them with opportunities for socio-economic mobility, and at the very least, give them an adequate means of sustenance that gives structure to their lives and those of their families. Neither Antillean nor African respondents suggested that they had any such resources to lean on in Paris. Over the past three decades, the French government has sought to create incentives for self-employment through small business creation in the socio-economically deprived neighbourhoods with large immigrant populations. The Agence pour la création d’entreprise (Agency for Business Creation – APCE) has recorded a regular increase of business creation among immigrants since the late 20th century, but few of these were initiated by Sahelian Africans. Figures from the TeO survey have shown that immigrants from the Sahel and their children have the lowest rates of self-employment (3%) of all groups from immigrant backgrounds, for instance lagging substantially behind those of Portuguese (9%) or Turkish (16%) origin (Lagrange 2013). As such, Sahelian Africans lack an ethnic economy that is sufficiently dense to create significant opportunities for their members and serve as a buffer in situations of economic adversity. In addition to direct economic opportunities, a dense ethnic community provides social capital which can also be beneficial for the assimilation of newcomers and their children. The restrictions that forbade foreigners from creating associations were lifted in 1981, which rapidly led to the development of large numbers of immigrant associations. Malian and Senegalese immigrants are no exception in this regard. 84 Fatima points out that “there are many associations of people who come from the same village, there are many Malian associations”.  Francis, a Senegalese IT worker in his 40s, has recently gotten involved in the association of his rural community of origin by occasionally attending meetings and providing financial aid to them. He explains that the Senegalese have a dense network of associations, which are mainly focused on the interests of very specific ethnic or local groups, and are used to develop infrastructure or community development projects:  …[the Senegalese] are pretty tight-knit, but it’s also complex, it’s by ethnic groups and regions. There must be a Senegalese association somewhere, but I don’t think it’s very well attended. But on the other hand there is the association of this region, of that city, and then there are the associations of the ethnic group in that city. For instance you can have the association of the city, and within that association you’ll have many other associations. I know that exists as well.  These associations, known as dahira, in practice function as prolongations of the family that has remained in the country of origin. They solicit funds for a variety of infrastructural and other community development projects. Among all immigrants in France, Africans from the Sahel region are among the most oriented towards their societies of origin. They are the most likely to have a residence in the country of origin (22%), to provide regular financial aid to a household outside France (39%) and to contribute financially to a project in the country of origin (21%) (Lagrange 2013). With regard to the general sense of solidarity, Francis thinks it is quite strong among the Senegalese overall, but also points out that Senegalese immigrants who have lived in France longer tend to display less personal investment towards their fellow nationals compared to more recent immigrants, suggesting at the very least that patterns of ethnic solidarity among fellow nationals in France gradually erode over time:  When you see an old Senegalese, who’s been in France for 30 years, 40 years—or even 20 years like me—you go to him and you say “Mister, I come from Senegal, but I do not know where to sleep” he will take out his wallet... he’ll give you the addresses of social workers, he’ll give you addresses for hotels, and then he may give you twenty, fifty euros, and that’s that. But on the other hand if you see a young Senegalese and you tell him “Mister, I’m Senegalese, I don’t know where to sleep” he’ll tell you “come with me, you’ll sleep in my home, you’ll eat there for a little while”… So the longer people stay here, the more different they become. Other respondents are more critical about the attitudes and intentions of their fellow nationals in France. Patrick, a 32 year-old Senegalese basketball player who only arrived in France a year ago believes there is 85 too much competition and jealousy between Senegalese immigrants and considers that they are primarily driven by their own interests:  They pretend to be tight-knit, but they’re not tight-knit. That’s why I don’t want to get involved in the associations. Because the African is- every time he’s just looking out for his interests, you know. They say ‘we’re going to set up an association for this-and-that’, but afterwards there are always problems there… The Senegalese who are here are different from the Senegalese at home… because here people tell themselves ‘this is Europe, everybody fends for themselves’. Even the Senegalese that hangs out with you only hangs out with you because he has an interest to do so… Because actually we all have the same objectives, we all came here to look for [opportunities] so jealousy reigns… You can have a good friend back in Senegal, but if he comes here you’ll see him change… That’s why I sometimes… refrain a little bit from socializing with them. Given that no migration barriers exist for Antilleans, that their lives in France do not revolve around supporting their communities of origin, and that they don’t face the same linguistic, cultural and institutional obstacles as their African counterparts, it is unsurprising that their associative activity doesn’t display the same patterns of cooperation and competition, but instead mainly revolves around traditional cultural activities and festivities. Although some respondents reported attending Antillean gatherings with variable frequency, many described the ethnic community as highly fragmented, without any structures that can compare with those of other communities, and were often quite critical towards existing structures. Paul, a 29 year-old computer engineer from Martinique for instance explains:  There isn’t really any cohesion, I don’t think there is. Two Antilleans will meet, they will get along “we’re Antilleans, we understand each other and stuff, it’s cool”… if there’s a problem, if you’re in trouble I’ll help you because you’re Antillean and stuff, but there’s nothing organized, there aren’t any structures… that will amplify the phenomenon. Marcel, who has been retired for many years but who hosts a weekly radio show on current events and who acquired a strong social and political consciousness throughout his working life, is very disillusioned by the tenor of Antillean associative life, which he finds far too focused on food, music and dance, and would like to see it get more involved in cultivating Antilleans and advancing their interests: Most who got involved in those associations just did so to fill their pockets, saying "we'll do this, we'll do that" but in the end they do nothing. And when the winter holidays begin, then they do balls and stuff like that. That's it… There was one that I supported... that tried to raise the level of the Antillean community, and it lasted for a while, as then it was the same old thing: lack of funds, no venue, and it costed too much... as soon the Antillean community wants to do something good, everything is put into place to stop it from evolving. 86 In addition to the Antilleans being closer to French individualist norms than their African counterparts, the acculturation processes they are exposed to in mainland France change their patterns of ethnic solidarity even further, especially when compared to other groups. Josiane, a 47 year-old nurse’s aid from Guadeloupe thinks Arabs and Africans display much more in-group solidary than Antilleans, whom she thinks have modelled themselves excessively after the individualistic norms of the mainstream: The Arabs are much more united. The Africans [pause] they were united at first, I find they're beginning to drift apart. You see, it's each for his own now... The Antilleans have never been united... It's always “mind your business and don't come bother me”. I remember, when I was a kid,50 next to where we lived... there was a whole community [of Antilleans]... the kids would all hang out together, but our parents wouldn't. We would go to each others houses, but the parents would stay at home... In Guadeloupe, we're very neighbour-neighbour, you know, "hey neighbour, give me a hot pepper!"... Here, our parents come here, they see that "oh, so this is what France is like!", so for them, people don't care about other people...  Josiane has also noticed that many Antilleans are now entering retirement homes, whereas traditionally their children would care for them during their old age back in the Antilles: “Africans aren’t like that, the Arabs neither. I tell you, I’m a domestic nurse’s aid—the Arabs aren’t like that, for their fathers or mothers it isn’t like that. They take turns: today it’s you, tomorrow it’s me, the day after it’s you”.   Vincent—a French-born 25 year-old MBA student whose father comes from Martinique—thinks that the superficial sociability that Antilleans generally display actuality hides a jealousy and resentment of others’ success which fundamentally hinders any true cooperation. Interestingly, he thinks the second generation is less affected by this:  …if we meet in a social gathering, like at a party, and you’re Antillean and I'm Antillean, maybe yes, you'll try to find out "so where are you from?" ... but when it comes to helping each other... no. Because there is this Antillean mentality, which I hate, it's jealousy. They see the other move forward—it’s a problem for them. Instead of helping each other, we'll criticize each other more than anything else. If I move ahead, it's all good, but if you move ahead more than me, it's no good... I think that the Antilleans, well, like me, who are born in France, have more- solidarity... there's no jealousy between us. Jenny—a 30 year-old Martinican who works in a pharmaceutical company—thinks Antilleans can be too insular in their social lives, and only socialize with other Antilleans “I think there are many Antilleans who only hang out with Antilleans here. I think it’s a pity”. Despite the fact the two thirds of her social circle is                                                           50 Josiane came to France at the age of 11 and is speaking here of a neighbourhood in France. 87 made up of Antilleans, she claims that “I’ve already been called ‘acculturated’ because I didn’t socialize that much with the Antillean community... I don’t reject either my culture or my origin, but do I therefore really have to go to every Antillean party and only socialize with Antilleans?” As this overview suggests, Antilleans display more individualistic tendencies than their African counterparts, both in their migration patterns, settlement experiences, and in their links to the region of origin. Patterns of in-group solidarity among African migrants are primarily transnational and geared towards their kin group in the first generation, whereas Antilleans’ involvement in associations in more geared towards cultural events and festivities. Neither Antilleans nor Africans have any notable ethnic economy which they can make us of to improve their social mobility as a group in mainland France. As we noted earlier, Antilleans nevertheless have access to many advantages through public sector employment, which such as subsidized housing and travel. Taking stock of these background characteristics, we now turn towards the influence of familial and neighborhood factors on the assimilation patterns of the second generation of both groups.   88 7. RAISING THE SECOND GENERATION   In a meta-analysis of dozens of studies on the assimilation patterns of a variety of second generation immigrants in the United States, Eva Morawska (2009) identified the shared micro and macro-level structural and agentic circumstances that explained enduring differences in their socioeconomic achievement. She first mentions family characteristics and the home environment: the human capital of parents, the rules of behaviour and expectations children are socialized into, and the tensions around issues of obedience, sexual relations, and parental expectations regarding life achievements. Morawska also points to the characteristics of the neighbourhood—primarily its class and ethnic composition, level of social cohesion, and the behaviours it encourages and discourages—as well as the peer group, which also represents an important reference point for role-models and normatively expected and sanctioned behaviour and aspirations. She emphasizes the key role of the school environment as well, its quality, role models and the expectations it places on students. Race and gender—with their enabling or hindering social implications—and the structure and dynamism of the local labour market as well as the obstacles it erects for advancement also play a determining role in assimilation prospects. Finally, she highlights the agency of second generation individuals: their life orientation, human capital and personal aspirations. We examine all of these aspects of the assimilation process across in the two chapters that follow, with a particular emphasis on the specific difficulties and obstacles that second generation respondents have faced. The first section of this chapter will deal with their family characteristics and home environment and the second with their neighbourhood and peer groups. The next chapter will focus on their schooling and labour market experiences as well as their marital choices and perspectives. The impact of class background, race and gender will be discussed throughout both chapters.    89 I. Family Mechanisms The family is one of the most significant social arenas for shaping the assimilation prospects of second generation immigrants. The family’s human capital will influence their income as well as the residential opportunities and thus the quality of the parenting environments they can provide for their children. The intricacies of the family structure—whether children of immigrants grow up with both parents, but also the number of siblings they have and the size and involvement of their extended families—are also likely to greatly influence social and economic outcomes of the second generation. Furthermore, families transmit some of the most foundational values, beliefs, rules of behaviour, and general expectations to their children, which largely derive from immigrants’ cultures of origin. The role that cultural heritage may have in helping children adapt to the host society needs to be determined empirically. Since the Sahelian and the Antillean samples diverge sharply in their family patterns—both in terms of structure and of culture—we begin by discussing the impact these have had for both groups.  Hypertrophied Father Figures One of the most salient aspects of family life that stood out in nearly all interviews with second generation Africans was the overarching authority of the family father. Modibo, a 21 year-old cook of Malian descent explains that he couldn’t fathom disobeying his father: “Even if I know he’s wrong, I lower my head and say ‘yes.’ Then I go outside and avoid him, or I go to my room”. This behaviour pattern may even lead him to accept demands that he otherwise wouldn’t: “he might for instance need something and I don’t dare to say ‘no’ to him, [even] knowing that by doing that thing for him I’ll have problems of my own.” This demand placed upon children that they unilaterally subject themselves to paternal injunctions even led some respondents to make very significant life sacrifices. Mamadou, a 28 year-old second-generation Malian who works as an airport security agent, for instance recounts how his father systematically opposed the many opportunities he was given to become a professional football player—a longstanding dream of his—despite the fact that several coaches and agents had visited his father to ask 90 him to sign professional contracts for Mamadou. As many other respondents, he described his deference to his father’s position as a simple matter of respect: “in the end, given that… there’s this respect before anything else, I listened to my father”. Mamadou elaborated on this, explaining what going against his father’s authority would have meant, and why this was a price he was not willing to pay: …I wanted to be a professional sportsman, but oh well, that’s how it is. [pause] And yeah, this respect towards the parents—it’s very important [he laughs]. If I were disrespectful I would already have left a long time ago. I would have had some regrets because to lose contact with your parents frankly, it’s like losing everything—honestly, yeah, you can’t move forward in life anymore. [Y]ou have the advice of your brothers and sisters, but you always need your parents’ advice. That’s for sure. Mamadou illustrates the rigidity of the expectations placed upon him and how following his own path would have meant breaking contact with his parents. As we will see throughout the chapter, this absolute paternal authority that is typical of many Sahelian families has significant consequences at several levels of the assimilation experiences of the second generation.   Destructuring of Sahelian Family Systems In contrast to modern European family systems, the traditional family structure in Sahelian Africa is patrilineal and patriarchal with strict hierarchies of age and sex. Gender relations are highly codified: polygamy is legal and widespread in Mali and Senegal,51 and women are typically assigned to a subservient role with very limited autonomy.52 The extended family, particularly male kin on the father’s side, play a significant role in the education of the children. Furthermore, children are assigned to age groups—called                                                           51 An analysis of the data from the 2009 Malian census has shown that 30% of men and 42% of women live in polygamous marriages in the country. The practice is less common in urban areas (26% of men and 34% of women) and in the northern regions. The proportion of men in polygamous marriages in Mali have decreased gradually in all age groups above 35 years of age since the 1970s, but have increased sharply among younger age groups since the 1990s (Coulibaly Diamoutene 2015). The 2002 census in Senegal indicated that 23% of men and 50% of women lived in polygamous marriages, also with a higher rate in rural than in urban areas (ANSD 2006). Regarding the perception of the moral acceptability of polygamy in these countries, a recent Pew poll has shown that in Senegal, 8% see the practice as morally wrong, 86% as morally acceptable, while in Mali, 11% think it is morally wrong, and 74% see it as morally acceptable (Pew Research Center 2013). 52 There are however important differences between the status of women in different ethnic groups. In contrast to Fula or Soninke families where women are usually tightly controlled, solidarity is vertical and loyalties are patrilineal, this is less pronounced in Serer, Wolof, or Manjack families, where the mother maintains important prerogatives in education and the management of family assets (Lagrange 2010). 91 fedde—where the oldest sons vow to protect the young, while the young owe them respect and obedience in return (Kayongo-Male and Onyango 1984; Kula-Kim 2010; Nicollet 1992; Wane 1969). In sharp contrast to the individualistic parenting practices in France, Sahelian parents thus traditionally rely to a much larger degree on the extended family and the broader community for childrearing. Francis, a Senegalese IT worker in his mid-40s fondly describes how children can roam freely outside of parental oversight: “you go out with your kid, but everybody comes to pick him up—he belongs to the society. So you go out, your kid is three years old, or five years old, and you don’t worry about him. It’s not like here.” As much as he loves this aspect of childrearing in his society of origin, he recognizes that it cannot be maintained in the absence of the social structure and culture that underpin it. But he thinks many Sahelian immigrants from poor rural backgrounds who maintain their traditional educational practices without the environmental support-system of their society of origin run into many problems with their children in the far less cohesive and social-capital saturated environments which they inhabit in France: Back home it’s the society that makes it like that. As I said, when you have a child, it’s the child of everybody, and... you educate him... like everybody else does, and push him to do only good things... But here, what we found here is that it’s a catastrophe for immigrants... but I think it’s because the immigrants who came here didn’t know how to read, they didn’t speak the language, they worked and then they brought their wives here, who could neither read or write, they had children, and then they did their best to give them the education from where they came from. And it perturbs the children. In the end they don’t succeed, they abandon the education. And then it becomes very difficult.  Thomas and Znaniecki (1958) described a similar pattern of destructuring and anomie within the Polish immigrant community in the United States, which resulted from uprooting families and individuals from the communities that ordered their lives and transplanting them in an alien cultural and social environment which they are hard-pressed to adapt to. It is precisely this loss of prior forms of social organization which seems to particularly affect many Sahelian families on the lower end of the socio-economic spectrum, fundamentally altering and straining their internal dynamics. Hugues Lagrange, who has closely studied the challenges faced by African families in several marginalized urban areas in the periphery of Paris and Nantes (Lagrange 2010; Lagrange 2013) has argued that the higher delinquency and school drop-out rates that have been identified among Sahelian young men partly result from the shattering of the coherent and 92 cohesive extended family which played such an essential role in socializing the young in the country of origin:  Within Western societies, families of African origin are akin to a moth-eaten cloth. The bonds that structured the relation between the sexes and the generations have been partially destroyed… In Africa, a successful education is guaranteed by the coherence and redundancy of advice and expectations from the male circle of the mature generation towards members of the young generation. But that circle was broken through emigration, and the attempts to restore it using the oldest sons have failed. Accordingly, the crisis owes as much to the alteration of the workings of normative systems as it does to the gap with Western norms. This dismembering primarily weighs on the migrant groups for whom kin plays an important role—those who were, in Africa as well, in the most traditional configurations: rural populations with little or no schooling. (Lagrange 2010: 202-3) Given the large gap between the family models and norms of the sending and receiving countries, it is particularly difficult for Sahelian families to compensate for this loss of embeddedness in the extended family. Since strengthening the nuclear family unit around the marital bond would require overhauling too many deep seated cultural assumptions, the fundamental family dynamics often remain largely unaltered despite the absence of the social structures that are supposed to underpin them. With often absent working fathers, Lagrange argues that women are often left to their own devices in parenting without being granted the same authority over their children as the father and his male kin. This, he argues, is prone to lead to dysfunction when raising large families with limited resources in adverse social settings. Furthermore, the large sibling groups characteristic of Sahelian families have been shown repeatedly to be more involved in gang activity and delinquency (Mohammed 2011). Lagrange suggests that the strong internal solidarity and cohesiveness of siblings from Sahelian backgrounds can explain their greater propensity to develop deviant behaviours. Furthermore, the status and authority of the father is threatened at a fundamental level in the host society, given his typically low occupational status, the fact that his children quickly become more familiar than he is with the workings of the surrounding society, and that the power he is used to wielding over his wife is far less secure than in the society of origin. Lagrange argues that these threats to patriarchal authority are often compensated for by increased authoritarianism, which creates problems of its own.   93 Given the importance of parental authority, many first generation Africans express a strong sense of dispossession of the means of controlling their children (Segalen, Aouici and Gallou 2011). Many respondents complain of not being able to discipline their children as they do in Africa. Patrick, a Senegalese 32 year-old, explains that many parents are wary of pursuing these practices given the very different attitudes towards corporal punishment in France. And given that those from less educated backgrounds are often unable to control their children by other means, they are more likely to slip out of their control: …their parents were people who didn’t go to school, you see, and when they come here they are told “here in France we don’t hit children” you see, so therefore they are afraid to educate their children like we educated them in Africa. Because in Africa we hit them… you go to school they hit you so you learn; you go outside, you do something wrong, they hit you even if they’re not your parents… That’s how we were educated. And our parents when they come here… they let their children do what they want. When the child is young, you can command him, but when he grows up you can’t do any more for him. I’m not saying it’s everybody, I’ve seen others who are- who have a good education, because their parents educated them well you see, but others—half almost—their children are something else.  Echoing other respondents, Patrick thinks that many parents are afraid because they know that “if they hit their child, then he will denounce you, and then you’ll have problems”. Despite the fact that respondents so frequently mentioned the restrictions on corporal punishment, what is probably most significant is the absence of legitimacy granted by some children to the normative framework of their parents, which a combination of limited communication and corporal punishment rarely manages to change. This just leads the father’s authority to be seen as tyrannical and unjust, and instead increases a sense of injustice and resentment despite superficial compliance (Mohammed 2011). It is this subtlety that for instance explains the somewhat contradictory positions of Aminata—a 63 year-old Senegalese mother of 10—who despite frequently relying on corporal punishment for raising her children, comments that “people who tell you ‘I couldn’t educate my child because I didn’t hit him’, it’s not true, you don’t educate a child by hitting him, you actually make him a failure by hitting him”. She complains however, like many other respondents, that there is not enough dialogue and communication between adults and children in their families: “We Africans… our problem is that there isn’t enough dialogue between us and our children”. This can pose particular problems when children acculturate to other norms than those that parents seek to enforce as we will see later when we look more closely at acculturation patterns in the second generation.  94 Although there were several examples of involvement in delinquency among my respondents and their siblings, it was difficult to assess the role that sibling groups and their internal dynamics may have had in this. However, the large sibling groups in Sahelian families often did seemed to increase parental acculturation, not only because large numbers of acculturated children rooted the parents far more into the country and the culture, but also because each child gradually pushed parental boundaries further towards the norms of the host society, thereby increasing parental acceptance of some aspects of the surrounding society to which they were otherwise previously opposed. Abdoulaye, a 22 year old computer technician with parents from Mali who was raised in a family of ten children, has noticed that the younger ones have had much more freedom than he did at their age. He doesn’t think that this is specific to his family: “I even see that… with many other families actually. I’ve noticed that the more you move ahead in the generation and the more freedom they have”. Other respondents corroborate this, and often see it as a very positive development, sometimes even betraying a slight jealousy over some freedoms their younger siblings can enjoy which they never did. Sané explains that her parents  are much closer to us [now], my brothers take the liberty to say much more things, to free themselves, when we’re together we talk… there’s a bit more open-mindedness, keeping in mind that we’re Muslims after all… even if [my father] knows that my brothers have girlfriends, he lets them be for the most part I’d say… even if the religion forbids it… But there’s no guarantee that he’ll still speak to us if we convert to Christianity, on that level there is still reluctance.  Abdoulaye believes these changes in parenting derive from the parents growing softer as they age, but also thinks that some cultural factors may also be at play: …often in African families what happens is that they take good care of the first child so that he can take care of the others. You see? So that's what was supposed to happen. But the problem is that my brother’s not supposed to do my parents' job. With us it worked because we were very close—around the same age—but when they began going to school, he already had his first child. So he didn't have time, he couldn't take care of two families. The space that this freedom opens up can be double edged however. In fact, Abdoulaye claims that the more hands-off attitude of his parents’ towards his younger siblings directly explains why some of them have gotten into trouble over recent years. He takes the example of his younger brother who dropped out of school at an early age and has been condemned for repeated offenses: 95 …as a result of the acquaintances they made, it went a little bit off the tracks, but it’s strange because they didn’t discipline them in the same way as us, in the sense that we got straightened out right away. But [with them] they were a lot more easy-going, they—well they talked, which is a good thing, but they talked too much! [he laughs] In the sense that I don’t know, I talk once, twice, three times, if I see that you don’t get it I discipline you right away—with them it’s 5, 6, 10 times! And up to this day they haven’t done anything. But I was like: “now we really have to do something!” you know? [he laughs] and in the end after doing nothing it all derailed. That’s how it is, it’s that simple.  Regarding Lagrange’s claim regarding a putative lack of maternal authority, mothers certainly seemed to exert less direct authority over their children, and were often presented as being much easier to talk to—and often serving as an essential intermediary between the children and the father in this regard—but several mothers who had largely raised their children by themselves provided striking counterexamples to this pattern.  The Case of Sahelian Single Mothers Most respondents who struggled with the exacerbated authority of their fathers often spoke of their mothers as more flexible and accommodating, but the respondents who were raised by single mothers clearly did not seem to suffer from an authority deficit. Aminata—a headstrong Senegalese woman who decided to move to France with her youngest son at the age of 50 when her husband’s parents had pressured him to take a second wife—is emblematic of this. She raised her son all by herself in very precarious circumstances, but has always been very involved in his life and has always been a strong authority figure for him, almost to the point of depriving him of agency. When a bank clerk recently asked if her 22 year-old son was over the age of legal majority, she replied “as long as I’m alive, he will not be over the age of majority”. Soraya—a 40 year-old real-estate agent born in France of Malian parents—remembers how careful her mother was to make sure nothing happened to her children: “As a woman who lives alone, she was very careful with her children… so it was especially important that we daughters didn’t get pregnant… and the boys should above all avoid stealing, doing drugs or stuff like that”. Soraya’s mother for instance still walks her daughter to the bus every morning despite her protestations (she is in her late teens) and doesn’t leave her until she is in the company of somebody she trusts. 96 This degree of parental control could potentially be seen as hampering children’s adjustment to the surrounding society. Lagrange for instance argues that it is in fact an excess rather than a lack of authority that causes many of the negative outcomes among Sahelian-origin youth. However, it is often a necessary measure for many parents to keep their children out of harm’s way, especially in the case of single mothers who raise their children in difficult neighbourhoods. In the communities that he studied, Lagrange found that single-parenthood was only of limited or no significance for school achievement, but that it was on the other hand associated with higher delinquency rates in adolescence. However, this link between single parenthood and delinquency only characterized European, North African and Antillean youth, but hardly had any incidence on Sahelian African youth.53  Single-mother households are thus by no means all comparable, and single-parenthood shouldn’t be seen as an explanatory variable in itself. As Aminata’s story illustrates, Sahelian single-mother households are shaped through mechanisms of their own, resulting in very different configurations. Given the stigma that generally accompanies female celibacy in the Sahel, which several cultural practices—such as polygamy, levirat,54 or the use of “pretend” husbands for social occasions—are meant to avoid, single-mother households are often the product of particularly resilient women who have challenged and escaped patriarchal control. Furthermore, while French single-parenthood rates are typically higher in ZUS areas55 than in the rest of the country, the reverse seems to be true for Sahelian Africans, as Lagrange found in his study that only 15.7% of Sahelian households in ZUS areas were single parent headed, while 22.5% of them were in the rest of France. He also points out that the operations of Aide Sociale à l’Enfance,56 a social program that helps destructured families and children in distress, indicate that Sahelian families are far less often the subjects of interventions compared to French-origin families, who are more prone to experiencing social problems like alcoholism, domestic violence, or psychiatric disorders with the                                                           53 On the other hand, children of Sahelian immigrants from traditional households were more than three times as likely as children from mainstream French and European families—and almost twice as likely as children from North African families—to engage in delinquent behaviour, suggesting that other factors other than single motherhood are at play (Lagrange 2010) 54 An Islamic tradition whereby the younger brother of a deceased male is supposed to marry his widowed wife. 55 Zones Urbaines Sensibles—literally “sensitive urban zones”—are neighbourhoods earmarked by the government for specific urban renewal policy interventions. 56 Social Assistance to Children 97 deleterious impact these have on children’s safety and well-being. Sahelian female-headed families do thus not seem to be as widespread, nor as closely associated to poverty or social pathologies as French and Antillean families are.  A Matrifocal Caribbean Model In contrast to the traditional structure of Sahelian African families, a far more common pattern in Antillean families are households centered on the mother and her children, where often absent men play a more marginal role (Smith 1996). As Pétronille—a 29 year-old Martinican—points out: “back home women are at the center of everything” whereas she thinks the exact opposite is true in African families. Despite the intense debates over the origins of these family patterns57 and the multiple preconceptions regarding them (Condon and Byron 2008) their widespread nature and consistency over time is well-established (Mulot 2013). According to the TeO survey, around 40% of DOM mothers in metropolitan France raise their children as single mothers, compared to only 18% of the majority population. Furthermore, second-generation Antilleans in mainland France have almost the same levels of single motherhood as their parents (Lagrange 2013).58 Among my own respondents, 91% of second-generation Africans grew up with both parents, but this was only the case for about half of the second generation Antilleans, including those who were raised with stepfathers. As Marie-Ange—a 41 year-old cashier from Guadeloupe—bluntly phrases it: “the fathers are never there you know, Antillean fathers are never there”. Josiane—a Guadeloupean nurse’s aid who raised her son largely by herself in a cité in the northern outskirts of Paris—ties the low school performance of many young Antilleans that live in her area to the                                                           57 The links between these family patterns and the heritage of slavery have often been stressed in the literature, emphasizing the fact that marriage between slaves was discouraged or directly opposed because of the undesired regulations that this imposed on plantation owners. However, many other causes have been shown to be influential, such as the decline of religiosity in the islands, the refusal among slaves themselves to get married, as well as the preservation of matrimonial patterns common in the West African areas of origin (Gautier 2000; see also Giraud 1999). 58 These rates do not seem to have been affected in a major way by the migratory experience, nor have they changed markedly over time, as 30% of Guadeloupean heads of household were women in 1900, and about one third of households with children in Guadeloupe were reported as growing up with an absent father between 1954 and 1982 (Gautier 2000). Furthermore, these families are typically overrepresented in the socio-economically challenged segments of the Antillean population. In Martinique in 2010, 83% of children whose parents were unemployed lived in a single-parent households (Marie and Breton 2015). While single-parenthood rates are around 35% in Guadeloupe as a whole, they represent half of households in urban areas targeted for urban renewal, twice the rate of similar areas in metropolitan France (ONPV 2016). 98 fact that they were raised by single mothers who can’t attend to their children as much as they would want to—a phenomenon she thinks is very widespread within the Antillean community: I failed at everything in school. And when I got my son... I told him: “I’m not asking for much: a high school diploma”. A high school diploma – he hasn’t even been able to bring me a trade school certificate! He messed up- he wasn’t fit for school. But many are like that, you know. If you look closely, many are – many young Antilleans. I don’t know many who have a high school diploma... They drop out of school when they’re sixteen. As soon as school isn’t mandatory any more, they stop going. What are you supposed to do? You can’t force them to go, you can’t be behind them all the time – I work at night, I work during the day... I was alone with [my son]. It’s not easy, I think that’s why I only had one child. Because it’s not easy raising a boy by yourself. And many Antillean women are alone with their children... Antilleans don’t get married. In my mother’s generation, there aren’t many who are married... My two grand mothers weren’t married, nor was my mother. In contrast to her mother and grandmothers, Josiane decided to get married, but her experience dramatically illustrates the downward trajectory that can open up for children from broken households. She met her first husband—a white Frenchman—when she was already pregnant with her son from a previous relationship with an Antillean man who “didn’t want to take his responsibilities”. She stayed married with him for ten years before they divorced. She then remarried with an Antillean man who has been incarcerated for the past 15 years. She tried to educate her son less harshly than how she had been raised herself, but this did not work: We Antilleans have a pretty strict education you see… I didn’t want that education for my son, so I trusted him. I trusted him, but I always had an eye on him because I know how young people are. And I put him in boarding school, so he wouldn’t [hang out] in the cité, pff… It didn’t help at all. He began not going to school—I work, I can’t watch over him! And people say “it’s the parents’ fault, it’s the parents’ fault”, it’s not the fault of parents, I know it’s my son’s fault, I warned him. So he began smoking drugs, then he began dealing drugs, then he began assaulting people, and there—all that, little by little. He was 19 when he went to jail. He had his 20th birthday in prison. And now he’s turning 21 and he’s still in prison. Although Josiane’s experience is certainly unique in the Antillean sample, her story depicts quite starkly how socio-economic and residential conditions can negatively intersect with family formation patterns and educational challenges in the disruption caused by migration. Lagrange claims that relevant parallels can be drawn between Antillean households in situations of poverty in metropolitan France and the harmful dynamics identified in the “culture of poverty” thesis (cf. Small, Harding and Lamont 2010): Such an interpretation can… account for essential traits of how the most marginalized fraction of Antillean families operate in metropolitan France. In these families, who are a minority component of the ZUS population, single-parenthood is inscribed in the continuity of the matrifocal household, which translates the prevalence of the mother-child filiation over partner bonds. This family situation, when combined with low 99 resources, creates educational problems no less important than those than derive from Sahelian authoritarianism, but more similar to those experienced by certain European families. (Lagrange 2010: 230-1) Some respondents mention similar themes in their upbringing as their West African counterparts, especially regarding parental authority, lack of communication, as well as the traditional reliance on the broader community for childrearing, which is disrupted by migration and requires parents to adapt to the far more individualistic lifestyles in metropolitan France. Echoing this theme, Ingrid—a Guadeloupean-origin 37 year-old secretary for the ministry of foreign affairs—points out that her parents weren’t able to reproduce the education they themselves received back in the Antilles because of cultural differences between the islands and the mainland, which led them to have to devise new methods of childrearing that would be better suited to the new environment, which led them to be more restrictive and controlling than back home:  ...in the Antilles, there is this notion that everybody knows everybody, so in a sense, even if children play outside, the neighbours keep an eye out, watch over them, and so on, but in the metropole, it’s a lot more ‘every man for himself’. So they have to be more careful you know, and be harsher... I had the impression that we had the strictest parents of the neighbourhood, it was crazy, we couldn’t do anything, but it’s actually true that when they were kids they could do more things because everybody knew where they were... if there was the slightest problem, they could go see the neighbour, who could take them home, you know, and so it wasn’t easy for them to have been the first to arrive in the Paris region, to put in place a way to raise the children... but they never talked to us about it. Another common trait many Antilleans share with their African counterparts is the relationship that the young have to parental authority and to adults in general. As Josiane explains, echoing many of the practices identified among African respondents: “I was educated by my grandmother, where you’re not supposed to talk back to adults. When an adult speaks to you, you had to bow your head down and do as he said... even if they’re not family – as long as the person knows you, if an adult speaks to you, you’re not allowed to complain, you’re not supposed to speak”. As Pétronille explains, this is also reflected in more cohesive families, which put more emphasis on solidarity than on autonomy:  We have a family environment that is much stronger. So we have the bad side where we have everybody on our backs, and at the same time the good side where you're also supported paradoxically, by this same family community. It's super important... Then there's this sort of- in fact, we're very harsh in the way we're supposed to behave in our families. We find that people here are very free with regard to their families actually, that there sometimes is a certain lack of respect of the family hierarchy, whereas over there, each and every one has his place. You're not supposed to move. 100 Alice, a 30 year old math teacher with a French mother and a Martinican father, who was born and raised in Paris thinks the difference between her parents is instructive in this regard, where she feels her father has much more in common with friends from immigrant backgrounds: When we talk with friends, whether they are Arab or black, they say that it’s really a cultural difference, it’s not the same relation to parents… I can allow myself to talk back to my mother, we talk, there’s an egalitarian relation when we argue you know, but my father—well, now I begin to do it because I’m a little older, but when I was a child I would never have talked back to my father, it’s impossible, while I’ve already argued with my mother on equal standing if you will, and I think that’s cultural. My aunts also on my father’s side—impossible that I talk back to them, I can’t even imagine it, and I call them “auntie”, while on my mother’s side I call them by their first names. Just as among the Sahelian respondents, it was not uncommon for first generation Antilleans to complain about the restrictions placed on disciplining children in France, arguing that it gave their offspring too much power, and that it led them to lose respect for parental authority. Josiane thinks this is changing in later generations however: “the young Antillean mothers now—I can see it, I have several of them around me—it’s not like before, like in the generation of my mother”. Several second-generation respondents pointed out that they didn’t perpetuate the practice—neither Ingrid nor her siblings have for instance perpetuated this practice despite its prevalence growing up: “It’s something that wouldn’t even have crossed our minds”.  As this section has shown, several cultural differences between Antillean and Africans play a determining role in the assimilation process of their children. These cultural backgrounds are associated with different family structures and dynamics which are influential in many ways upon second generation outcomes. The turmoil that migration can cause in the internal dynamics of these family systems—particularly for West Africans given their stark gender and age hierarchies as well as their reliance upon the extended family for childrearing—can also moderate or exacerbate typical risk factors for second generation adjustment. As such, while segmented assimilation theorists would for instance underline the comparative disadvantages that children from single-parent households may face (e.g. Zhou 1997), the specific cultural characteristics of West Africans—ranging from their large families to their practice of polygamy and their tight control of women—are not only correlated with higher rates of delinquency in 101 ‘intact’ families, but also with single-mother households that produce far less deleterious outcomes than Antillean families for instance. However, a more nuanced appraisal of these family dynamics requires an analysis of how they interact with the local environment, which is why we now turn to the impact of neighborhoods, peer groups, and the acculturation patterns of the second generation.    II. Neighbourhood Effects and Acculturation Neighbourhoods that concentrate poverty and social problems have been shown to impact individual outcomes negatively (Wilson 1987; for a review see Edin and Kissane 2010). These neighbourhood effects can exert considerable influence on the assimilation patterns of many Antillean and African second generation respondents, but the extent to which they do also depend on the family mechanisms discussed earlier. As we will see these can offer some unusual coping mechanisms for dealing with children who become insubordinate or turn to delinquency. The prevalence of these patterns would seem to depend on acculturation to a large extent. Acculturation almost inevitably generates a symbolic rupture between the generations, and the intergenerational conflict it leads to has been a central theme in the immigration and assimilation literature as far back as Warner and Srole (1945) and Thomas and Znaniecki (1958). Segmented assimilation theory suggests that such conflict can be conducive to downward assimilation. The negative environment that some neighbourhoods provide, and the oppositional culture of influential peer groups within it, can certainly affect assimilation trajectories. These acculturation patterns differ between Antilleans and Africans, and lead to different types of vulnerabilities and protective effects for both groups.     102 Neighbourhoods with Concentrated Social Pathologies Many respondents raised their children, or were raised themselves, in relatively difficult areas. We have previously described some of the most salient socio-economic processes that have given rise to the housing estates where endemic problems of precarious employment, chronic joblessness and welfare dependency coincide with high crime rates and incivilities. Although some second generation respondents have fond childhood memories from the cités where they grew up, and wish to emphasize the positive aspects of their neighbourhoods—especially in light of the overwhelmingly negative stigma attributed to them—many parents complain about the feeling of insecurity they experience there. Marie-Ange, who was born and raised in Guadeloupe, has noticed changes in street crime that she’s witnessed over the past few years:  It’s changed a lot in 6 years, 7 years. It’s just… little youths that we see every day, and we don’t know why the municipality lets this happen… this kind of gratuitous violence – no, I experience it, they’re right in front of us, they do it, it’s unbearable… [N]ow it’s calmed down a little, but you know, they just go somewhere else anyway. But no, even if it’s happening everywhere, still, in Saint Denis it’s really something special… We just let it all go on. I mean, when the mayor says “but you know, those poor youngsters“—that kind of talk really has to stop now, it’s enough now, it’s enough.  Marie-Ange acknowledges the complexity of the problem, and thinks it derives from many factors: the social and economic circumstances some parents face, the shortcomings of parents themselves, a greater sense of entitlement among the young, as well as the deviant incentive-structures in the neighbourhood and broader society:   ...those young people want things made easy for them, they want everything... without making any efforts, without working. There really are those who deal drugs, or they’ll be lookouts all day... I don’t condemn all parents, because there are parents who work in the evenings, who don’t have a choice, who raise their children by themselves – where it sometimes also works out just fine – and then there are others who can’t handle it anymore, who’ve thrown in the towel, who’ve completely given up... I don’t know their lives. I know cleaning ladies who have three jobs, and to get by they have to leave their children at home alone, and when they’re not there, the children get bored – I’m talking about children who are 10 years old, 9-10 years old – they get bored, so they go out with friends, and in the mean time the mother is working and doesn’t know what her kid is doing. But that’s how it is.  Other respondents are more uncompromising regarding the responsibility of parents for the development of these social problems, emphasizing their lack of investment in or authority over their children, while also pointing to the negative incentive-structures of the host society, that encourage children to be impolite, materialistic and ungrateful. Aminata raised her son in a neighbourhood that has seen repeated 103 instances of rioting, and is earmarked for a substantial urban renewal program, but was proud that she had managed to protect him from the negative influence of this neighbourhood: We had one bed, we slept in it, when he woke up in the morning he went to school, I went to work, in the evening we saw each other again, and he was there in front of his TV or his computer. No, it’s not the [fault of the] system... Do you think I’ll leave my children, or my child of 10 or 11 in the street until 11pm? And that I’ll sleep peacefully? No. No. Idriss [her son] is 22 years old and I don’t sleep before he has come home... When you hear that a 22 year old woman was killed, another youngster of 21 was found dead, you can’t. You can’t sleep knowing that your children are in the street.  Ethnographic studies in these neighbourhoods have shown the role that local cliques of young men play in providing their members—who have often failed in school and only have dismal employment prospects—with an alternative means of self-valuation through a status system that rewards aggressive, transgressive and fearless behaviour, allowing them to exert greater power on their surroundings and gain access to symbolic rewards (Mohammed 2011; Sauvadet 2006). These groups are frequently highly anchored in their neighbourhoods, and often occupy a prominent and influential position in public space, which they in many cases come to dominate and regulate (Lapeyronnie 2008).  Aminata’s son Idriss confronted precisely such a normative system when arriving in France at the age of 12, where individuals were pressured to partake in all sorts of more or less serious transgressions to acquire approval and esteem from local peer groups. At first he remembers being afraid: “when you saw guys throw rocks at the police and everything... you were scared” because in his home country “you can’t even think of doing that” given the lower tolerance within the population for such behaviour and the harsher policing practices. He explains how he got used to it over time, however: “it was always like that there, you threw rocks at the police, you went with your buddies to burn cars and everything. But all that is just to be accepted in the group. You have to go through that, otherwise they’ll think you’re a pussy and all that... It’s afterwards that you understand that it’s actually useless. Those aren’t real friends”. Aminata nevertheless recognizes the importance of the fact that her son spent his childhood in Africa: “I was lucky to bring my son here with a basic education. And when he came... I didn’t let Idriss lose himself. I didn’t let him lose his basic education either. And I also taught him not to be ungrateful.” It is likely that the education her son received in Africa helped his mother to raise him in France the way she 104 did. Had Idriss grown up in the same neighbourhood where they first settled when moving to France, the local environment and his peer groups would most likely have been able to have more influence upon him. Patrick, a recent Senegalese migrant, also thinks being socialized abroad largely protects people from getting involved in this particular subculture: If you take me to a dangerous area, and I don’t know, all children sell drugs or whatever... I’m an adult, I know what is good and what is not good you know, and I know where I’m from, and I know what I want you know, I won’t follow those youths, because we don’t have the same mentalities... but if you have a child that is 8 or 7 years old or whatever, you put him in those neighbourhoods, you’ll see, one day he’ll hang out with them and you never know. This is precisely what happened to Djibril—a 23 year old professional rugby player with roots in Mali—in his final year in high school. He had always been a good student and had never been into any particular trouble but started spending time with some other young men in his neighbourhood who had dropped out of school and had followed different trajectories. This gradually led him to begin selling drugs, until he failed his Baccalaureate and decided to put a stop to it:  And do you think [your parents] were worried about you at some point? Were they concerned about bad influences? At one point, yeah, the year that I failed my Baccalaureate... that year they were worried. I skipped a lot of classes, got warnings from the school, got bad marks... that year they were pretty worried... Then I took charge of myself, and told myself that if I continue like this it will end up nowhere, so- And was this just related to your studies or was it your acquaintances?  My acquaintances, that’s for sure. My acquaintances, yeah. At one point, well, we decided that we needed money… and gradually, that’s how it happened. [T]hey had quit school for the most part when they were 15-16 years old, I was 18 and I was in high school. I told myself “the Baccalaureate is no big deal, it’ll be fine, you’ll get it”, so you see them hang out, you hang out with them, they’re well dressed, they have brands and you don’t, so little by little it’s—you don’t even notice it—you’re told “if you know someone... you can give him this and sell it to him” so he gives it to me, I sell it and at the same time I add a little bit to the price... and little by little, there you have it. And then you think, you think about the day they’ll kick your door in at 6am, and knowing my parents I thought “wow, no, now it’s getting—now it’s too much” so you sort yourself out, you tell yourself “this isn’t for me”. David, a 32 year-old administrative assistant with a Martinican father and a French mother, had similar experiences that lasted longer and only ceased after a religious conversion. He describes in great detail the cultural environment of the difficult cité he was raised in outside of Paris. Given the widespread contempt for the police in the area where he was raised, having a police officer as a father was probably the most challenging aspect for being accepted when growing up:  105 ...I was despised, that’s for sure. In that type of situation, when you don’t have enough distance, you want to show people that you’re not like your father, since in the end, in the common way of thinking at my age, a good police officer just didn’t exist. The police is the enemy, so you obviously want to set yourself apart and say in a sense: “my father is a police officer, but I’m something else”. So I flirted with delinquency, but without really being a full-blown delinquent. But I’ve been around a few.  David provides a compelling account of the pressure he sensed from his surroundings growing up, that led him to be influenced by the oppositional culture of his neighbourhood without ever fully surrendering to it. As he explains, it simply generated too much cognitive dissonance to go any further:  I grew up in a universe where I quickly understood, as early as middle school, or maybe even earlier, that there were two sides: the side of victims, and the side of oppressors. And it was hard to stay between the two. Difficult not to be an oppressor, while still being respected by the true oppressors. So it was difficult. I have been many times a victim of delinquency. So at one point I began shifting – but given the education I had, I didn’t feel at ease. My conscience was still strong enough to bring me back to my senses. David’s account seems to frame the education he received from his parents as a sort of mediating force between himself and a hostile environment which he would otherwise be tempted to give in to simply to spare himself from being on the receiving end of its harshness. Many other respondents echo this view, like Vincent, whose father is from Martinique and whose mother is from a mixed Arab and African background.59 His father decided to move the family out of the neighbourhood where Vincent had lived until the age of nine, in the northern suburbs of Paris: “since it was a little heated there… my father wanted something more calm, and wanted us to grow up in a nice surrounding” Vincent thus grew up in a small town in western France, but returned as an adult to visit the cité where he grew up: The first year when I came to Paris I went back to where I had lived before, to see how it had changed, and we ran into people we had been to school with, but I didn’t remember them. And my buddy told me “well that guy’s a junkie, that guy is this, that guy is that”. They do nothing with their lives. So the people around me with whom I’ve stayed in touch have more or less succeeded, but the people with whom I was in school, they haven’t really succeeded, they’ve become junkies, drug dealers, stuff like that. So honestly I don’t know... but I think I wouldn’t have turned out so bad if I had stayed in Paris because there was that framework... there was that education, there was that framework that said: “that’s not what life is about, my son” [he laughs]… there’s that respect and that education that I think ensures that I wouldn’t have deviated. I don’t think I would have deviated.                                                            59 Despite being mixed, Vincent primarily thinks of himself as Antillean. He only grew up around his Antillean family as his mother has broken off all contact with her African family members.  106 As many other respondents, Vincent also points to the importance of the education he received, but also his parents authority in enforcing it, and says he is immensely grateful for how his parents educated him and his siblings. His experience also points to the important role of the peer groups that people associate with in the neighbourhood, given that those with whom Vincent socialized growing up—whom he stayed in touch with during his years outside of the Paris region—have all done well for themselves. The reasons why individuals get involved in specific groups can’t solely be explained with sociological factors, given for instance the large differences one finds within the same sibling groups. Individual personality most certainly matters, as well as ability to succeed in school or other arenas that favour social mobility, but it also derives from the quality of family relationships. Regardless of which factors are most influential, the prominence of such adversarial peer groups creates a very challenging parenting environment.  Sahelian Responses to Deviance Instances of family destructuring in France can sometimes lead Sahelian family fathers to change their migratory strategies. As we have seen, the labour migration restrictions and family reunification policies of the 1970s led many Sahelian immigrants to choose to relocate their families to France. But the development of insubordinate behaviours among their children as well as the greater tendency of women to gain autonomy and file for divorce have led some Sahelian family heads to grow increasingly dissatisfied with their migratory arrangements. According to Jacques Barou (2001), new strategies have been developed that allow men to sustain their villages of origin while making sure their children will later continue providing financial support for the village. Barou explains that they overcome the restrictions on immigration by marrying their first wife in Mali, bringing her to France, fathering a child with her and sending her back to Mali shortly thereafter. In subsequent marriages, they follow the same procedure. As French-born infants, these children will be given French citizenship at the age of 18 despite having lived their whole lives in Africa. They will thus be given a traditional education that protects them from youth insubordination 107 and female emancipation, while at the same time allowing them to settle in France at a later age and thus continue providing for their villages back home. A common occurrence for dealing with the insubordination of children is to simply send them to live with family members in Africa for an unspecified period of time. Patrick has noted that there are numerous Sahelian families in his area who have sent children back after they developed behavioural problems which the parents simply couldn’t manage or stand anymore: There are lots of them here. I’ve seen lots and lots of them here. They’re sent back to Africa, for instance to Senegal and Mali, to show them how things work over there. As soon as parents bring them there, they give them nothing, they let them get by all by themselves, you know. They leave them there for 5 years, or six years, or seven years, and if they come back here they’ll understand you know. Automatically they’ll change. To show them that what they went through here isn’t life, you know. It isn’t life. So we need to talk to them every time, but you can’t talk to them because they think they’re more intelligent than we are, because we live in Africa. But they’re wrong, because we see how they lead their lives, what they’re doing isn’t good, it’s not something that will build their future. Then they’ll end up in jail, and then they can’t do anything you know. Several respondents had children or siblings who were sent back to Africa during their childhood, or had even been sent back themselves. A good example of how such a decision can be taken is the case of Fatou and her oldest son Hamady. Fatou remembers how worried she was about some local men who were taking an active interest in her son, on whom she thought they were having a very negative influence in the neighbourhood in a south eastern banlieue where they lived at the time:   I simply wanted to preserve my children from everything that was going on. I wasn't blind after all; I saw the drug problem and other things, the crime that was increasing in the neighbourhood… I didn't hesitate to put myself on medical leave to look after my children. I saw there were too many new friends around... they didn't seem interesting to me, the way they spoke... I didn't find it was valuable for my children... I always tried to keep my children away from those kinds of people... I didn't hesitate to go confront one of their ringleaders outside... when he was taking my eldest son around with him everywhere––he must have been 32 or 35 that guy, and my son was only sixteen.  Seeing her son develop these suspicious relationships led her to eventually move further outside of Paris to a small town in a rural area. Unfortunately, these individuals stayed in contact with her son, and regularly sought him out to do favours for them despite the geographical distance. Fatou explains how this led her to opt for more radical measures at that point:  It didn’t stop there, so I sent him to one of his half-brothers that I had raised, and so he was very well informed about the situation and he took care of him, and he put him to work with him, since school was really “I go today, tomorrow I don’t”… and finally he sent him back to me because it was just impossible. Well I didn’t 108 overly complicate things: I put him in the first plane and I sent him to Mali to stay with my big sister. So there was a part of the family… that took care of his well-being: food and stuff, and I sent money for his pocket money and to pay for his school. It was a private school in partnership with France. Well, he came back with a fiancée and then: married, family father, computer programmer—not bad! [she laughs] Well, no I mean I really fought. You can’t have children and just leave them like that, and then go to parties here, parties there… that’s not how I was raised. The experiences of these respondents who were sent back to Africa seem mixed. Some have experiences in their surroundings that have gone remarkably well, truly offering children a “second chance” (Blaedsoe and Sow 2011), while others seem much less successful. Malik, a 33 year-old second-generation Malian who works in an auto-parts company says “I know parents who have done it. Generally, there are some who have come back and who are worse—some became well-educated, it depends. It depends on the person”. He thinks this practice was something that was common for his generation, but that it has largely been curtailed since: “now you have laws that protect [children]—you know, you can’t send away a child like that”.   Dissonant Sahelian Acculturation Challenges The substantial cultural differences between Sahelian African parents and their French-born children make patterns of dissonant acculturation the norm. Whereas there is evidence for this leading to negative outcomes in some instances in the second generation, parental expectations and authority could also successfully minimize these outcomes.  Many parents complain about their children becoming rude and disrespectful compared to how they turn out when educated in Africa. Mamadi—a 50 year-old Senegalese man who has raised his children in France—explains that many parents feel a profound sense of loss when their children turn out so different from what they expected: “It happens that parents say to themselves that it’s a little bit like they’ve lost their children... I know quite a lot of families who can have disputes with their children: the way they talk back to them, the way they speak, the vulgarity... there are children who are very rude... It isn’t easy. The parents say to themselves ‘but we haven’t educated them like that’”.  109 Fatima—a 60 year-old Malian woman who has raised several children in France—underlines the many difficulties that parents’ low social status generates for acquiring the respect of the second generation when the latter are highly acculturated to French norms: We aren’t able to educate our children here as we do in Africa. The child obeys his parents in Africa. And the teacher—the child respected the teachers, even more than his own parents. It’s what I told my own children, you have to respect your teachers, they give you knowledge... but some of them don’t understand, and go scream at the teachers in school. [O]ur children don’t have that respect towards us or towards their teachers... the children even see that they speak better French than [their parents], in the end they don’t respect them... [H]ere as soon as [the child] sees that the parents go cleaning outdoors, do manual labour while he goes to school, he won’t have any respect for his parents, that’s also why they aren’t able to channel their children. Sometimes those Malians who are here, they speak to them, but they don’t listen... they aren’t able to impose themselves on their children. And the children as well, they see that they haven’t gone far in school, they don’t respect them at all.  The social norms of the host society, which second-generation immigrants largely take for granted growing up, will inevitably constitute the prism through which they evaluate their parents’ status and thus the legitimacy of their authority. Many of these parents often occupy the lowest rungs on the occupational ladder and often lack the cultural and social capital that would allow them to impose their cultural framework on their children efficiently, and that which their children are taught to value in school further distances them from the cultural framework of their parents and can lead them to lose esteem for their authority and see it as illegitimate instead of axiomatic. As a first-generation migrant, Patrick illustrates this loss of esteem through the many negative experiences he has had with French-born second-generation Africans in his neighbourhood: “even we have problems with them because they call us ‘blédards’”—a disparaging characterization of rural African immigrants who haven’t yet acculturated to urban life in the West—“I've been with Africans, whites, and all that, I've never seen anybody who said 'blédard' to me before”. Patrick thinks that if they knew more about their parents lives and experiences, they would be more humble and less disrespectful: “[their parents] were back in villages where there was no electricity, no water, all that… if you knew the history of your parents, I think you’d behave properly towards all those people”. In addition to the possible lack of awareness of parental experiences, these behaviours clearly seem to derive from deeply internalizing key values of the host society, arguably through the remnants of a colonial hierarchy, but perhaps more 110 importantly through the valuation of conformity to urban lifestyles, behaviours and attitudes, which end up in a cultural devaluation of parents.  This suggests that there are patterns of dislocation that result from a dissonant acculturation in Sahelian families, in accordance with the predictions of the segmented assimilation literature. However, as a result of parents’ ties to their societies of origin, the expectations they placed on their children and the authority with which they enforced these expectations, many second generation respondents retained significant ties to their culture of origin despite also acculturating to the host society. This could lead to friction and tough compromises, but not necessarily to the dislocation described above.  Many respondents nevertheless insisted on conflicting values, desires and ambitions between them and their parents. Sané for instance remembers how frustrated she was during her teenage years because of the many restrictions she was subjected to compared to her peers: …my friends… were allowed to go out a little bit more than my mother would [allow] me… they don’t let us girls in Africa do that much, you can’t just allow yourself to say “after school I’ll go out for a stroll”, there’s always an ulterior motive… I didn’t understand why my mother told me “no, you aren’t allowed to go out”, so I just skipped class, so during a certain period instead of going to school I went and had fun and then I went home as if nothing had happened… I signed all my papers myself when I had detention. So at times my father didn’t see anything. It was only when [the school] phoned him, but otherwise he wasn’t aware of everything that was going on… Sané’s brothers were too young to intervene at the time, which they otherwise might have been likely to do. So as Sané’s experience suggests, dissonant acculturation can in fact contribute to subverting parental authority in some specific contexts. Here for instance, where there are significant cultural differences between the family and the mainstream, some key expectations that parents place on their children may simply not be accepted by the children as legitimate when their acculturation has led them to evaluate those expectations negatively. Dissonant acculturation may lead children who don’t have any significant authority figures among their siblings to avoid or circumvent these expectations, given that parents lack the cultural competence both to make culturally appropriate interpretations of their children’s aspirations and behaviour, but also to perceive and appropriately remedy their attempts to subvert these expectations. 111 However, the hypertrophied authority in Sahelian families, and the strength of some of the cultural norms and expectations that derive from it, nevertheless make such sidesteps fairly unusual among our respondents, as a frontal challenge of parental norms often seems unthinkable given the serious consequences it is fraught with. Interestingly, there are patterns in Sané’s experience that tell a very different side of the story, where she instead sought significant compromises that satisfied both her desires and those of her parents. In the following passage, she remembers her frustration over her mother’s constant attempts to control her in her early teens, which gradually set off a process that led her to get married through her own volition at the age of fifteen: When I began in high school, me and my mother hated each other… she thought I has was in a downward slide, but I wasn’t… I was looking for boundaries… and they weren’t able to set any… When I went out—even dressed properly—for her there was an ulterior motive: boys… but no, really, I went out to have fun with my female friends… And when I met my boyfriend—my husband actually… we were together and it was going really well… and since we have religious beliefs and that I’m a Muslim, we knew we couldn’t be with each other outside of marriage. Of course, we asked the right questions, and I spoke to my father about it… and of course he wasn’t happy that we were together… [but] my father asked the right questions, and I was afraid to get married because I wasn’t ready in any case to take responsibility for a household, and I was lucky in a way because my husband was undocumented, so he didn’t have an apartment, he didn’t have a job, and I knew I wouldn’t live with him… so from then on, I got married, and there, we were married, I could do what I wanted! [she laughs]… I didn’t tell my friends I was married, it was just my boyfriend... I knew we wouldn’t live together, and that it would be like my “boyfriend”—and I won’t hide that I thought “if it doesn’t work out I’ll divorce”… But today it’s a choice I don’t regret… I got married at the right period in my life, it allowed me to get closer to my parents, to my family.  Sané’s experience illustrates how children of West African immigrants can find compromises with the cultural values and expectations of their parents, where an excessive acculturation would risk profoundly harming or even rupturing ties between parents and children. These compromises can require children to make concessions that can seem quite significant, but they seem attractive because they allow them to resolve some of the tensions that build up from acculturation and a distancing from parental norms. Sané remembers how her parents reacted when she discussed getting married with them: My mother looked at me—she was doing her prayer, I still remember—she gave me one of those looks… because for her I was getting married just to have fun… I was getting married to have sex—I’ll say the word—for her I was getting married for that. I wasn’t getting married for that. I was getting married because I knew [my fiancé] was giving me something, and that I wasn’t allowed to be with him if I wasn’t married religiously, and that I needed that so I could feel freer. Because as soon as I got married I felt free and my father told me “you do what you want”.  112 She realizes deep down that this choice was a compromise however. When I later asked her to fill out an identity chart to rate the salience of her social identities, she spontaneously added “free” to the identity list, suggesting she felt a sense of emancipation from rigid family norms. She rated this emancipation a 3 out of 5. When asked why, she replied: “You can’t do anything you like, you know. Five would be a lie”. Many respondents nevertheless complained about what they saw as unreasonable demands that their parents placed upon them. Given the centrality of the home village, parents often expect that children will share – at least partly – this fundamental orientation towards the society of origin. Modibo, a young French-born cook with Malian roots, has travelled a few times to Mali with his parents, but he feels like his father “hasn’t understood that we and him aren’t the same” and complains that “he expects too much” from him and his siblings: “I think he just wants us to build a castle for him so he can take it easy.” His father sends remittances on a regular basis to his family back home, and Modibo suspects that he wants him and his siblings to begin doing so as well: “He’s never said so to me, but I think he [expects me to]”. Most of the tension between Djibril and his father tends to revolve around the expectation that he should perpetuate this practice: He considers that… he has, during his whole life… cut his arm off every month to send [money] over there, to look after… his brothers who are back home sleeping, who have three wives, pray all the time and are just taking it easy. Since his father said “the family shouldn’t get poorer after I’m gone, or end up in the street, you have to take care of them” his whole life has been a sacrifice, and he thinks I’ll do the same thing… he thinks I should cut my arm off as well, and I consider that it isn’t my duty, but he thinks it is. And he doesn’t even thank me, even when I make an effort. For him it’s normal, it’s really- there’s nothing unusual in me giving away half of my pay check, it’s just normal. It’s my role basically.  Despite the difficulty that some respondents may have of making these financial sacrifices, some do give part of their pay on a regular basis and find it relatively natural, as Mamadou for instance does: “frankly I don’t forget where I come from, because I already send—‘already’, it’s almost all the time—we send money to our family, we’ve built houses, it means that it’s not for nothing”. As these examples show, the negative effects of dissonant acculturation patterns can in many cases also be attenuated when parents succeed in inculcating the legitimacy of their expectations and authority in their children, especially when both generations are willing to make some compromises.  113 Consonant Antillean Acculturation Challenges Because of the more limited cultural differences between the French Antilles and the mainland, differential intergenerational acculturation patterns do not play as dramatic a role in the families of Antillean respondents as they do in African families. Antillean parents also have less expectations of cultural retention vis-à-vis their children. This can often facilitate their assimilation into the mainstream, but can also make some second-generation Antilleans more receptive towards other cultural norms that can for instance be widespread in their neighbourhood.  In a sense, Antillean parents and children are closer to a form of consonant acculturation, where they both largely internalize mainstream French norms. As Ingrid explains, her parents were always very keen on making sure their children didn’t stand out, and were well-adjusted to the surrounding society: …they are very much into appearances. You have to appear—normal… So that’s really super important in the family actually, and I’m worried that I may have transmitted that to my daughter… because it’s been drilled in since childhood… there shouldn’t be anything that can be reproached to you when seeing you or when listening to you—or reading you… It prevents us from being natural. Because the natural scares us… So what’s wrong with us when we’re being natural? What happens? What’s the problem? Are we going to do or say something that will show everybody that we’re Antillean and not French? Even though she concedes that this effort to fit in can stem from a concern over one’s reputation in the tight-knit Caribbean communities of origin, she also thinks it is indicative of an insecurity over not being seen as fully French. Although this strong accentuation of mainstream French norms can have a favourable effect on the assimilation of the second generation into the host society, it can also promote a sense of anomie, and when raised in suburban housing estates in the Paris region, can make them more prone to identify with cultural elements from their local environment. Josette for instance explains that many second-generation Antilleans from modest backgrounds who are born and raised in disadvantaged Parisian suburbs acculturate more into the cultural ferment of their neighbourhoods than they preserve an attachment to their culture of origin: “They don’t necessarily have all these little things… from the culture… so they’re a little bit… they often call themselves 114 banlieusards,60 it’s the banlieusard mentality you know… rather than ‘we’re Antillean, children of Antilleans’… they don’t manage to define themselves in that way”. This probably also explains that many second-generation Antilleans who grew up in disadvantaged neighbourhoods seemed prone to convert to Islam. Three of Lucien’s sons have for instance converted to the religion, and he thinks this is frequent among young Antilleans in the area: “I don't know why they went into Islam. Suddenly they all went into it... it's someone in the neighbourhood that filled their head with that stuff”. David—who also grew up in a neighbourhood with a large Muslim population—makes the same assessment: “I have Muslim friends who have grown up in Catholic families, who are Antillean and who have become Muslim—superficially sometimes”. He thinks this is mainly due to the social environment in which they grow up, as a result of “their acquaintances, or maybe sometimes also the fact of going out with a Muslim girl”. Josiane’s son has for instance been dating a Muslim girl from his neighbourhood and has ceased eating pork, but as far as she knows he hasn’t yet converted to Islam. She thinks it has become a trend: “now it’s become fashionable among young black Antilleans—they’re all Muslims”. She believes part of this comes from their ambivalence towards their own culture: “in the end, I have the impression that young Antilleans in Paris, in France, they want to deny their own culture to put themselves in another culture”. Eugénie argues that “there are perhaps people who at some point need more landmarks, and a religion with more structure”. Josette echoes this by pointing to the strong religiosity in the Antilles that is not adequately fulfilled by the Catholic Church in mainland France which leads second generation Antilleans to be more drawn more easily to Islam, but also to evangelistic churches as they provide a stronger sense of community: “there are also many who convert to evangelism, so there are these two—these movements are very strong, where… you have this sense of fraternity”. Jean-Louis, who was born in Seine-Saint-Denis of two Guadeloupean parents, provides an example of an Antillean convert. He grew up in a cité predominantly populated by Arabs with a substantial African minority, but hardly any Antilleans. Although he visited Guadeloupe when he was eight years old, he has                                                           60 I.e. “from the banlieue”, where banlieue (suburb) is used as a euphemism for stigmatized suburban housing estates with high concentrations of immigrants.  115 no memory of it at all. He converted to Islam at the age of 17. Despite seeing himself as French and Antillean, he primarily thinks of himself as a Muslim, and his religious affiliation serves as the prime structuring force for his social and cultural life. He follows all the dietary restrictions, has ceased drinking alcohol, fasts every year and regularly attends the mosque, and also aspires to marry a Muslim woman and transmit his faith to his children. He has no particular involvement in his culture of origin, which many of his religious practices would in any case impede, but his mother has largely come to terms with it despite being very upset at first: When [she saw] in the long term that my behaviour didn’t worsen, and that on the contrary it got better, and that it was making me into a good person, you know, and then she sees the evolution, I got my diploma and so on. I minded my own business, nobody influenced me, I haven’t changed my behaviour, nor my acquaintances—I mean I didn’t bring home any bearded men or veiled women, you know what I mean, which means that currently it’s working out quite well. Even though Lucien has accepted his sons’ conversion and that he doesn’t feel personally distanced from them as a result, it has made some religious discussions more tense, and he also feels it has cut them off from their culture of origin to a large extent, not least because of dietary restrictions: “I have to make separate meals for them… if my children don’t eat pork I respect it… [but] if there’s not a little piece of salted pork in a good Antillean meal, it’s not a meal. This is our culture”.  David thinks many Antilleans who display these peculiar acculturation patterns convert to Islam because they grow up alongside sub-Saharan Africans with whom they identify with as young blacks: “most Africans who are in France and who have French citizenship… are Muslim… So the Antilleans who socialize with Africans here—after all, I know a few—well, birds of a feather flock together, and at some point you want to intensify the communion you have with your friends, and in the end, you get on board”. Conversion to Islam seems to be a fairly common occurrence among second generation Antilleans who grow up in neighbourhoods with substantial Muslim populations. As a result of sustained migration—particularly from North Africa—in some areas and neighbourhoods around Paris Islamic culture can take on the role of a majority culture into which other populations acculturate. This does by no means only affect Antilleans. Lucien’s sons were for instance all married to white converts, suggesting that it affects 116 other residents of these urban areas.61 Although these patterns only seem to affect the second-generation Antillean respondents who grew up in poor suburbs with high concentrations of immigrants, it is an unusual instance of dissonant acculturation which is likely to impact their social mobility. While one might suspect a greater exposure to discrimination or stigma as a result, it also often seems to provide structure and purpose to their personal and public lives, which suggests that it may also potentially have positive effects on their life chances.     III. Concluding Remarks This chapter has highlighted some of the most salient differences between the structure of Antillean and African families and their internal dynamics. Despite some evidence of ethnic retention having protective effects on second-generation Africans—even in contexts of single-parenthood—there are also signs of it creating adaptation difficulties that can have deleterious effects on second generation assimilation. Although Antilleans may be favoured in some ways through their greater cultural proximity with the mainstream population, there is also evidence that this cultural proximity, combined with higher levels of single parenthood can put them in a greater position of fragility and lead to unfavourable social outcomes for their children. Respondents who were raised in difficult neighbourhoods were often faced with high rates of delinquency and peer groups with an oppositional culture in public space. This could serve as an attractive outlet for respondents with limited educational achievement and poor prospects on the labour market. However, it also often put significant pressure on parents to avoid that their children would develop insubordinate and non-adaptive behaviors or be caught up in illegal activities through these peer groups.                                                           61 Given the limited sample size it was unfortunately difficult to ascertain whether there were any gendered patterns here—if conversion for instance was more appealing for young men in these areas than it was for young women, and if it was correlated in any way with their school or labour market trajectories. 117 Several respondents described experiences of second generation Africans who had been sent back to Africa by their parents as a result, and despite the debatable regulating effects that these methods may have had, sometimes the mere threat of them could serve as a buffer for these negative outcomes. The acculturation patterns between parents and children were different in both groups. African respondents experienced greater dissonant acculturation given the larger cultural differences between their country of origin and the host society, whereas Antillean respondents generally experienced a more consonant acculturation. The dissonant acculturation patterns between the generations among African respondents often seemed to be attenuated to a large extent by culturally salient expectations and strict patriarchal authority. Antilleans could in some cases also experience dissonant acculturation when they adopted local cultural segments that set them apart culturally both from the mainstream and from their islands of origin. The interview data does not clearly suggest that dissonant acculturation patterns per se have a negative impact on second-generation outcomes. If anything, it is rather the loss of parental control in a context where children risk getting involved in delinquency and crime. Dissonant acculturation did not necessarily mean a loss of respect for parental wishes and concerns. Similarly, there was little evidence from the interview data that consonant acculturation in itself had a clear protective effect on second generation Antilleans. Taking stock of these family mechanisms and neighbourhood effects, we now turn to the trajectories and experiences of our respondents in school and in the labour market.         118 8. SCHOOL, WORK, AND MARRIAGE    The first section of this chapter compares the trajectories of Antilleans and Africans in school and in the labour market. It draws from our qualitative data to highlight the experiences of our respondents in the educational system and on the job market, with a particular focus upon the obstacles they have faced. The second section explores another important aspect of their assimilation patterns, namely their partner choices and marital aspirations.    I. The Second Generation in School and in the Labour Market A 2010 report from the Court of Auditors points out that France is the OECD country with the highest level of educational underachievement at the age of 15, where the disparities in results between students have increased the most, and where the impact of social backgrounds on school results is the most elevated (Kepel 2012a). Since 2003, the OECD Program for International Student Assessment (PISA) has registered a stable number of highly performing students, but a sharp increase in the number of underperforming children in France. Furthermore, the 10% of students who performed least well saw their results decline by 23 points between 2003 and 2012. Children from immigrant backgrounds in France are 2.3 times more likely to be among the least performing compared to those without migrant backgrounds (1.7 on average in the OECD), and the disparity between the performance of children from immigrant backgrounds and their peers from the majority population in the sciences is 87 points (53 on average in the OECD)—while this disparity decreases to 50 points among second generation immigrants, it is substantially higher than the 31 point difference of the OECD average. (OECD 2012; OECD 2015)  119 Divergent School Trajectories Brinbaum, Moguérou and Primon (2016) have shown that more than 40% of second-generation Sahelian children repeat at least one class in elementary school, while this only affects a third of DOM children, and a quarter of the majority population. In the case of the majority population and DOM children, the gender disparity is only slightly more favourable for girls, but in the case of Sahelian children, the gender gap spans more than ten percentage points, as only 30% of Sahelian girls repeat a class in elementary school. However, when social background, parental education level, family structure and language proficiency are controlled for, these differences are considerably reduced for all these groups.  After successfully passing an exam at the end of middle-school, children are faced by decisive orientation choices in the French school system.62 For children from the majority population, and those from Sahelian or DOM backgrounds, males tend to be quite evenly divided between those who choose academic tracks and those who choose vocational tracks (although there is a marginally stronger bias towards vocational tracks among Sahelian and DOM children), while girls tend to favour academic tracks, especially in the case of girls from DOM backgrounds, where 63% choose this orientation. However, among those who fail the middle-school exam or who fail to obtain any additional diploma, there are larger gaps between second-generation school children and their peers from the majority population. In fact, among the 18-35 year population, while 17% of males and 13% of females from the majority population acquire no further diploma (and 9% and 8% respectively without any diploma), this was the case for 18% of males and 10% of females from DOM backgrounds (13% and 2% without any diploma), and for 29% for males and 17% for females from Sahelian backgrounds (19% and 6% respectively without any diploma). These rates reveal how school drop-out rates are higher for males and for Sahelian-origin students relative to DOM-origin students. When controlling for social background and family structure, the disparities between DOM children and the majority population disappear, but they remain for young males from most other                                                           62 Around 20% of an age group always leaves the educational system without a diploma, which represents around 150 000 people (Kepel 2012a).  120 immigrant backgrounds. After controlling for the educational trajectory, the concentration of children of immigrants in the school, private tutoring efforts, and orientation choices the disparity for children from sub-Saharan backgrounds remain,63 which suggests other causes for explaining these patterns. These could range from characteristics pertaining to the youths themselves, to unwanted orientations, or even discriminatory treatment in school.64  Schooling Experiences All respondents were asked about their school experiences and trajectories, which can help shed some light on these disparities and other variables that might not transpire in quantitative studies. As most others in his class, Malik didn’t know what he wanted to do after middle-school. He thought he would try doing something similar to his father, so he settled on a diploma in automatized system maintenance. Looking back, he regrets this: I remember that we generally didn't know what to do... And then, well, we chose things, things that didn't suit us. Over time we realized that we could have done better, something different... [W]hen you want to choose the academic track, it's as if they want to stop you... "No, you're not suited for the academic track, you don't have the qualities". You're not smart enough basically. They say something like that, and well, everybody believes it, so we all chose the vocational track. After obtaining his diploma, Malik wanted to pursue a program where he could study while doing an apprenticeship, but no high school would accept him, which he thinks results from the bad reputation of his former school. So he went to several temporary work agencies and began working instead.  Ismaël—an unemployed 23 year-old with parents from Mali—says that a school career counsellor explicitly pushed him towards a vocational class in the final year of middle school, despite the fact that he had sufficiently good grades to continue on the academic track: “I don’t know why they put me there—they said it’s either that or you repeat a class. I didn’t want to repeat a class since I had already done that                                                           63 They also remain for children of North African and Turkish immigrants. 64 Some evidence suggests that the occupational position of the father in the country of origin can explain part of this variation in school outcomes between groups (Ichou 2013). Neighbourhoods have also been shown to affect school performance. Goux and Maurin (2007) showed that repeating a grade at the end of junior high-school increased markedly when the other adolescents living in the same neighbourhood had also been held back a grade, and that adolescents’ educational advancement is negatively impacted by the proportion of families with limited educational credentials living in the area. 121 in elementary school”. After they put him in the vocational class it was impossible for him to return to the academic track, despite having good grades. Souleymane—a 31 year-old audio-visual technician with roots in Mali—spent a large part of his youth in a rural area outside of Paris after his mother Fatou had relocated the family there to provide her children with a safer environment. Being the only black child in almost every class in middle-school was a negative experience for him, especially since he had attended a much more multi-ethnic elementary school in his former neighbourhood and never gave much thought to his race. His recollection shows how perceived discriminatory treatments can impact school trajectories: "in middle-school it was a difficult period for me because there was of course prejudice on the part of teachers and school personnel… In class as soon as there was a problem I was the one who was blamed, even though I had nothing to do with it". As he tired of being disproportionately singled out, and felt unjustly treated, he claims that he purposefully became more disruptive in class and only did the strict minimum to not be held back. It was at the end of middle-school, when it was time to choose his orientation that Souleymane believes he was specifically targeted by a member of the school staff. One day he was supposed to defend his choice of orientation in front of the school board, and an administrator was supposed to pick him up in class and escort him to the meeting room when the board was ready to receive him, but the administrator came to escort all the other children in the school except Souleymane. He knew that this administrator personally disliked him and “really was a racist”. Given that he didn’t show up, the school board arbitrarily put him on a professional accounting track. Souleymane was furious, but decided “I’ll go into accounting, and I’ll show them that I can do better than end up in a professional accounting track”. He successfully finished his two years, obtained good grades on his professional high school diploma, allowing him to attend a preparatory class to integrate an academic track. He succeeded, and ended up obtaining the baccalaureate he originally intended to do. He feels a profound sense of injustice as a result of this experience, and wanted to send a copy of his diploma to the school to mark his sense of revenge, but eventually decided against it. He later pursued university studies in cinema for several years. 122 These experiences are not limited to Africans. Some Antilleans also mentioned feeling that they were sometimes pushed into sub-standard tracks as a result of a certain prejudice. Esmeralda—a 39 year-old communications officer with parents from Martinique—spent her first years of schooling in a predominantly white area in central Paris before her parents moved to Saint-Denis, an inner suburb to the north of Paris with a large immigrant population, where she went from first in her class to below average, despite her sense that the level was lower in Saint-Denis: “I didn't adapt” she claims, but also explains that “people began stigmatizing there too--that's the great paradox: it's here [in Saint-Denis] even though there's a very varied population that people began stigmatizing”. At the end of middle school, since she was slightly below average, the school advised her mother to put her on a track to become a nurse’s aid instead of an academic track, as they were convinced that she would fail her middle school exam. Her mother decided to oppose the official stance of the school, and insisted that they wait for her result on the exam: “and there they tell her ‘no, you can dream, but she won’t get [her middle school exam], that’s for sure’ It was completely crazy, when my mother told me all this it just fills me with rage”. She succeeded at the test, followed an academic track, and went on to obtain a four-year university degree in fine arts. Esmeralda thinks she was treated as she was because of her race, but also suggests that it was specific to her schooling experience in Saint-Denis. Some respondents, like Calixte—a 42 year-old kitchen chef with parents from Guadeloupe—suggested that there was frequent discrimination in grading as well: “in school, when there are exams, there is a difference in grading you know—but of course, you can’t prove that most of the time”.  Some educational obstacles also derived from culturally specific family dynamics which were often observed among second-generation African respondents. The education of Baïdy—a 26 year-old traffic warden with roots in Mali—was for instance affected by a strong sense of duty towards his family. Despite having successfully completed two years in a five-year sound engineer program at a technological university in eastern France, Baïdy began questioning his emphasis on his studies in light of his family situation: “I said to myself—especially thinking about my father, who was getting old: ‘well, should I continue, or should I quit?’… since I’m the oldest son, I asked myself ‘what am I supposed to do’ you know”. Baïdy eventually 123 quit his studies and returned to Paris to look for work. This suggests that family duties and expectations can lead some respondents to choose short vocational tracks or otherwise end their studies prematurely.   The more specific expectations that parents would have towards some second-generation African respondents could particularly affect girls given the very stark gender roles in Sahelian families. Soraya for instance explains how her schooling was affected by the household chores she was expected to do: …I lived with a single mother and was the oldest of the girls [and] in African families, you lean very much on the children—especially my mother. So often I would help her with the cleaning, the ironing, all that, so sometimes the classes took the back seat… so at times, yes, I repeated a class, but not because I felt that I wasn’t capable, it’s just because I didn’t have time to study or concentrate on my work like a girl of my age.   Another example of a culturally specific influence on second generational educational chances comes from children who were sent back to Africa during their childhood. As the oldest siblings, Djibril and his sister were sent back to Mali during their childhood at the initiative of their grandfather, who wanted them to retain their culture and get to know their family—the idea being that they would thereby follow in the footsteps of their father by continuing to financially support the village back home, and hopefully also serve as examples to their younger siblings. Djibril was sent back to France after three years when he contracted an eye allergy that could not be adequately treated in his village, but his older sister stayed in Mali for a few additional years. Djibril feels he has turned out alright despite those years in Mali as he both came at a younger age and left the country earlier, but his sister was also affected by the substantially different circumstances she faced as a result of her gender: “She’s a girl, I’m a boy… we didn’t have the same daily life. She had to go get water at the well, do the cleaning, and then there’s especially the gender mutilation that she experienced there”.65 In light of these experiences and “those long years outside of school… you don’t have the same future perspectives afterwards”. Although she managed to obtain a professional degree when she came back to France, Djibril believes that her experience in Mali “has closed doors for her”.  Some Antilleans and Africans also commented on the content of the school curriculum. Older Antillean respondents who had grown up in the Antilles or done part of their schooling there pointed out                                                           65 Djibril claims that the genital mutilation that his sister experienced in Mali was done at the initiative of his grandparents there, and that his parents in France were not informed.  124 that they hadn’t been adequately taught the history of their islands in school. Sabrina—a 40 year-old market analyst who grew up in Martinique—remembers that in her elementary school history book, there was hardly anything about colonization and slavery… Before, you were taught the history of France, not of the Antilles”. However, the issue about the place given to colonial history and the history of slavery in school programs was a far more recurrent theme for second generation African respondents. Ismaël has a strong feeling of historical injustice that he feels isn’t being addressed, resulting in a psychological distancing or even disaffiliation with school: “in elementary school they spoke about everything… but they never spoke about what France did to Africa, well, we already know, they did 400 years of slavery”. Another important aspect of the schooling experience of several respondents from both groups is their avoidance of negative schooling environments for their children. The mother of Iba—a 33 year old data analyst with roots in Senegal—decided to pull her children out of the local public school after they moved from a small apartment in a comfortable suburb in the Western outskirts of Paris to a larger house in a more multicultural area in the northern suburbs: My mother didn’t appreciate the middle school where we went, I know my brother went three months to the school and… after that it was just private schools… Because she found that… the teachers didn’t pull the best students upward… the teachers concentrated more on the students with problems, rather than those who didn’t have any, and my mother didn’t want us to fall behind compared to others, so she put us in a school where we’d be in a more structured setting.  Choices such as these are of course not accessible to everyone, but some parents are prepared to make financial sacrifices to provide better educational settings for their children. Marie-Ange—a Guadeloupean woman who raised her children in mainland France—for instance recounts how her son was concerned about the quality of the middle school he began attending, which led her to send him to a remote private school despite their limited financial resources: …when he came home from his first day at school he told me “listen mom, if you want me to make it, come see what’s in my classroom, I don’t want to stay there”. Ok, so I went, I saw, and I said ‘ok, we’ll put you in another school’. So we put him in another school, and it’s working out just fine… Financially… it’s a little difficult, but we don’t regret it at all.  Many of these experiences suggest that a substantial part of the difficulties faced by respondents of both groups derived from their class background and the area in which they were raised. However, there also 125 seemed to be a widespread experience of differential treatment based on their ethnic origin and/or racial background, which could affect how their school work or their future career prospects were assessed. This seemed to be generally more severe among second generation African respondents than their Antillean peers. Also, second generation Africans could experience educational obstacles as a result of the cultural specificities of their families, which was not the case for Antilleans.   Labour Market Inequalities Previous studies have consistently shown the higher prevalence of unemployment among immigrants and their descendants compared to the majority population. Men from the majority population have an average unemployment rate of 8%, but the rates are much higher for sub-Saharan Africans,66 who have unemployment rates of 15% in the first generation, and 27% in the second generation. Women from the majority population have a 9% unemployment rate, while for sub-Saharan African women rates are the opposite of those found among men, as 21% are unemployed in the first generation, but only 12% in the second.67 By contrast, the rates of employment among Antilleans are even higher than in the general population, at 75% for men and 71% for women, compared to 68% of men and 60% of women in the general metropolitan population (Brinbaum, Meurs and Primon 2016). Part of this derives from the presence of a young high-skilled Antillean population that diverges markedly from the socio-economic background of those who came to France during the BUMIDOM. Among 20-34 year-old Antillean-born Guadeloupeans with a university degree, 53% thus resided in mainland France in 2010, and 49% of that                                                           66 As well as for Algerians and Turks. 67 These unemployment figures can be explained by traditional gender roles in both generations. In the first generation women are more likely to be married and care for children and the household. But as is often the case in migrant families with traditional gender norms, these norms also affect the school and labour market outcomes in the opposite way in the second generation. As Nancy Foner has for instance pointed out, in these families the "gender inequalities that tie girls to the home and reward female obedience and passivity, end up helping them to succeed academically" given that they spend more time at home on school work, whereas boys are encouraged to be more independent, and are more likely to end up spending time on the streets and neglecting their studies. These gender disparities are also reflected in the labour market, as "second-generation women... not only have better educational qualifications than their brothers, but they are often preferred for... white-collar service jobs" (Foner 2000: 237, quoted in Morawska 2009) 126 same age group who were employed that year were employed in metropolitan France. As such, the Antillean-born population in that age group who resides in mainland France have roughly the same levels of qualification as the rest of the population, whereas those who reside in the Antilles are twice as likely not to be university educated and over twice as likely to have no diploma beyond middle-school (Marie 2013). The first generation overall remains overrepresented in public sector employment where they are more prone to occupy lower levels of responsibility than the majority population (Recoules 2012). When controlling for socio-demographic characteristics, children of DOM migrants have similar chances of being employed as the majority population, both in the case of men and women. Female children of DOM migrants however experience difficulties finding stable employment after their studies, a phenomenon that controlling for social and familial characteristics doesn’t fully account for (Brinbaum, Meurs and Primon 2016). Self-reported experiences of discrimination in employment are high among many first and second generation immigrants, and Antilleans are no exception: in the last 5 years, 30% of DOM migrants and 40% of their children report having experienced some form of discrimination. This is also the case for nearly half of sub-Saharan Africans and their children (Beauchemin et al. 2010a). Intergenerational social mobility patterns show that occupationally, children of immigrants find themselves in an intermediary position between that of their parents and that of the majority population. Whereas 75% of sub-Saharan African immigrants were industrial workers, this is the case of less than 8% for the second generation. While 15% of sub-Saharans were service workers in the first generation, they are 40% in the second. Also, the proportion of those who occupy higher and intermediary positions in the occupational structure increase from less than 10% in the first generation to more than 30% in the second. In the case of DOM migrants their distribution in the occupational structure remains roughly similar across both generations, with the exception of the shift from industrial activity to the service sector. Whereas there are minimal disparities in salaries between the second generation and the majority population, the greatest disparity is in gaining access to employment or a promotion (Meurs, Lhommeau and Okba 2016).  127 By examining the experiences of our respondents in the labour market, we will be able to get a clearer picture of how they have perceived their difficulties accessing employment, getting promoted, and their experienced with various forms of discrimination.  Experiences in the Labour Market Although the children of immigrants may on average occupy more favoured positions on the labour market than their parents, the disparity between the generations is stark when it comes to simply being able to find work. The migrants who came to France during the trente glorieuses generally found it very easy to find work. Martine remembers how easy it was to find employment when she came to metropolitan France from Guadeloupe in the early 70s: “Until 75, I mean you found work very easily. I could have left my job today and found work the next day or the day after. And when I say that, I mean in all sectors”. For Martine, experiences of discrimination primarily occurred in specific areas:  When I was almost sure, anonymously, that it would work out fine, as soon as I had the face-to-face interview, I knew right away that I wouldn’t be retained. Because they called my name—my name didn’t have any foreign ring to it… but as soon as I was there—oh, suddenly things were different. So I’ve felt that… But that was in quite particular areas… [f]or instance when I wanted to work in a bank… I’d say in all the “shop window” positions, in reception, where you’re in contact with the public. Very much so.  Among the African first generation respondents, the labour market experiences of the respondents who do not reside in France legally are naturally in a very particular position on the labour market, where strict necessity primes over other considerations. Moussa, who has been doing demolition work for a few years now, explains that “as I don’t have any other better work possibilities, I accepted to work in these conditions to pay my rent, to eat”.  Mamadi came from Senegal as a student in the early 80s, finished a degree in mathematics, but didn’t manage to successfully pass his teaching exam, which left him in a precarious position as a substitute teacher, which was difficult to sustain after he got married and had children. He eventually found work in the national museum consortium, where he gradually specialized in security work. He does feel that his ethnic background has harmed his career, especially with regard to promotions, but thinks the scarcity of 128 work and the intense competition this generates goes a long way in explaining the difficulties he has had, pointing out that even other Africans discriminate in the workplace. During his graduate studies in a smaller city outside of Paris, Ibrahim—who came from Mali to France as a student—had difficulty finding the mandatory internship to finish his degree. Even though he was never directly brushed off because of his background, it was nevertheless striking to him that he was treated differently from his peers: “I did job interviews alongside other students from the same university, and I had better grades… I had more motivation… but I wasn’t chosen, so that was very, very strange”. He eventually had to find an internship in Paris, which allowed him to finish his degree. Although his experiences with discrimination when he studied in Algeria were far more significant, they were also more overt, and he was surprised that it was so prevalent in France:  Before being confronted by reality… I wouldn’t have believed it, I wouldn’t have said it was possible, but after having met… some Africans or students of African origin that I found here, the common stance really was: “you’re African, you can obtain your diploma without doing an internship because everybody knows you won’t get an internship.” They’ll say “we’ll find a sort of internship where there are companies that we know are aware of the difficulties that African students can face, so they accept to take them as interns, but not as real interns, you basically just come to consult some files and you’ll put them in your [thesis]. Gérald who was a former teacher in philosophy in Senegal, couldn’t rely on his African diplomas to find work, so he finished a Master degree in adult education and now designs training programs for adults. However, although people are very kind and considerate with him, he doesn’t feel recognized at the level of his qualifications and experience: “I’ve seen people where I work who have more responsibilities, but I have more diplomas… I tell myself that maybe it’s because I’m African… I think that with my experience and also my diplomas I could have had far more responsibilities”. Fatima—a Malian woman who obtained her biology diploma in the former USSR—has also confronted many difficulties finding work that fit her qualifications, but she thinks this also derives from there being many people with similar training from Eastern European EU-countries whose diplomas are recognized in France, which puts her at a significant disadvantage.   The difficulty of finding suitable employment mainly seems to affect the first generation respondents with high qualifications or those who aspire to upward mobility. Many African first generation 129 respondents with less qualifications had a much easier experience on the labour market. Khadija is a 32 year-old who worked in the healthcare system of the Malian army before she came to France at the age of 22. She hasn’t had any difficulty finding work: “I’ve had to struggle for all sorts of things, but I haven’t had any trouble at all finding work”. Marietou, a 26 year-old from Senegal who works as a nanny says the same thing: “I haven’t really struggled since I came here… I stayed about 2 months without work, generally I’ve found easily”.   The second generation respondents have a distinct experience from the generation of their parents. The greater scarcity of employment, their higher levels of education, aspiration to equal treatment and upward mobility, their acculturation which may at times not be conducive to finding employment, but also the fact that they cannot attribute differential treatment to anything else than overt discrimination or racism probably all contribute to this difference.  Experiences of discrimination were mentioned by many respondents, and these could take many forms. Echoing Ibrahim’s experience at university, Malik for instance remembers what he was told when he looked for an internship in high school: “the teacher said ‘be careful, we see discrimination every year, if you say your name is Mr. Cissé or Mr. Traoré—oh no, we have no more openings’. In high school they already warn you”. Some respondents even mentioned overt and explicit experiences of discrimination. Hamady recalls: “one employer even told me clearly… at least it has the merit of being honest: ‘no, I’ll never take any blacks in my company. Maybe you’re hard-working, maybe you aren’t, I don’t care, I won’t take any blacks’”. Usually, experiences of discrimination in hiring or in the workplace are more subtle, and more difficult to prove unambiguously. In any case, Malik often feels that he is treated differently in the workplace:  In the workplace you notice it. When you’re of foreign origin people are suspicious. When you do a task, they are always checking on you. There are those who don’t even say hello. I saw it at [a large auto manufacturing plant] when the higher-ups come they say hello to people, but when they see you they don’t. Well, I just say “I come to do my work whether he comes and says hello or not”… I think it’s gotten worse. That’s what I feel. Especially since the riots of 2005, it’s gotten worse. 130 Many respondents reported having difficulties gaining access to the labour market. Souleymane has always had difficulties finding employment, after finding small and undeclared jobs he began noticing he was treated differently: “when I went to look for work with a friend of mine who was white of course, they took him right away and they refused me, for the same job—even though there was no difference, equal competence and all that”. Souleymane worked for temporary work agencies for a while before seeking al